Open Thread, September, 2010-- part 2 — LessWrong

5

Open Thread, September, 2010-- part 2

by NancyLebovitz

17th Sep 2010

1 min read

5

5

Open Thread, September, 2010-- part 2

1Vladimir_Golovin

1Vladimir_Golovin

5Craig_Heldreth

12Scott Alexander

2Jonathan_Graehl

3datadataeverywhere

9Scott Alexander

0Richard_Kennaway

4Scott Alexander

0Scott Alexander

20datadataeverywhere

7datadataeverywhere

2datadataeverywhere

1datadataeverywhere

0Douglas_Knight

1Vladimir_Golovin

10datadataeverywhere

8Vladimir_Golovin

8Vladimir_Golovin

5Vladimir_Golovin

0Scott Alexander

6Scott Alexander

11jacob_cannell

8Richard_Kennaway

-2jacob_cannell

1Douglas_Knight

0Douglas_Knight

0Douglas_Knight

3datadataeverywhere

2datadataeverywhere

2datadataeverywhere

0datadataeverywhere

0datadataeverywhere

0datadataeverywhere

10jacob_cannell

0Jonathan_Graehl

0Jonathan_Graehl

0Jonathan_Graehl

0Jonathan_Graehl

10Mitchell_Porter

10Scott Alexander

0Vladimir_Nesov

0Scott Alexander

1randallsquared

7randallsquared

2Risto_Saarelma

1Vladimir_Nesov

0Jonathan_Graehl

2Richard_Kennaway

0Jonathan_Graehl

5datadataeverywhere

2datadataeverywhere

-1Jonathan_Graehl

0Jonathan_Graehl

3Mitchell_Porter

0Mitchell_Porter

1Mitchell_Porter

1Douglas_Knight

0Douglas_Knight

0Jonathan_Graehl

1Mitchell_Porter

2Douglas_Knight

0Douglas_Knight

0Oscar_Cunningham

New Comment

877 comments, sorted by

Click to highlight new comments since: Today at 8:59 AM

Some comments are truncated due to high volume. (⌘F to expand all)Change truncation settings

[-]Will_Newsome15y370

Condensed Less Wrong Wisdom: Yudkowsky Edition, Part I

Mysterious Answers to Mysterious Questions

Ask "What experiences do I anticipate?", not "What statements do I believe?"
Your strength as a rationalist is your ability to be more confused by fiction than by reality. If you are equally good at explaining any outcome, you have zero knowledge.
The strength of a model is not what it can explain, but what it can't, for only prohibitions constrain anticipation.
There's nothing wrong with focusing your mind, narrowing your categories, excluding possibilities, and sharpening your propositions.
For every expectation of evidence, there is an equal and opposite expectation of counterevidence.
You can only ever seek evidence to test a theory, not to confirm it.
Write down your predictions in advance.
Hindsight bias devalues science: we need to make a conscious effort to be shocked enough.
Be consciously aware of the difference between an explanation and a password.
Fake explanations don't feel fake. That's what makes them dangerous.
What distinguishes a semantic stopsign is failure to consider the obvious next question.
I

... (read more)

[-]wedrifid15y110

Please make this a post. It is is a valuable resource that I would like to have accessible.

1Will_Newsome15y

Should I add the other sequences first, you think? It's already too long, and that'd double the length or more.

6komponisto15y

Actually, I say put in on the wiki.

1Will_Newsome15y

That's a much better idea, I think.

5JGWeissman15y

Doubling the length would be fine for an article. Include links back to the detailed original articles.

6ata15y

Agreed. Slogans/quotes are fine as reminders/summaries of points that are explained and defended in more detail, but not as substitutes for them.

5lessdazed14y

From video dialogues: * How do you know the costs of your irrationality if you're irrational? * We're here to talk about rationality, which is the art generated when you want something more than your particular mode of thinking * Well, if you expect the future to be just like the past, calling that "realism" isn't going to save you from the fact that you’re guaranteed to be wrong. * ...there are specific propositions, right? You can't just bundle all the propositions together and slay them with one mighty blow that consists of one thing you can do wrong if you believe this bundle of propositions. * Curiosity requires ignorance and the ability to relinquish your ignorance, and I see you attaching a lot of importance to your ignorance here. * This sounds to me more like a mistake you are making in your model of the world than something you could actually do to the world itself. * If you want a precise practical AI, you don't get there by starting with an imprecise practical AI and going to a precise practical AI, you start with a precise impractical AI and then go to a precise and practical AI. * You can make mistakes even if you think you have a precise theory, but if you don't even think you have a precise theory you're completely doomed.

4arundelo15y

One thing you need is a paragraph break (a blank line) before and after the list. The source code should look like this: *Mysterious Answers to Mysterious Questions* * Ask "What experiences do I anticipate?", not "What statements do I believe?" * Your strength as a rationalist is your ability to be more confused by fiction than by reality. If you are equally good at explaining any outcome, you have zero knowledge. [Edited to add italics to the subheading.]

0Relsqui15y

How do you put in the editing characters literally? Backslashes?

2mattnewport15y

Markdown.

2Relsqui15y

Saved; thank you.

0Will_Newsome15y

Thanks much.

1[anonymous]15y

Also see divia's post for a spaced repetition database that could be helpful for internalizing these important thinking patterns.

1Vladimir_Golovin15y

This is an excellent effort. Please keep it up.

1Will_Newsome15y

I will do so, though I'm worried about what happens after I finish the "Yudkowsky Edition"; there's lots of other Less Wrong wisdom but it's further spread out and I feel like if miss some if it then people will be sad. But at any rate I plan on going back and getting the links to the posts for all the above and then doing the other sequences, then posting the whole giant thing at the top level. Over the next few days I have a different and probably higher utility post to help Louie Helm write, though.

[-]Relsqui15y210

Would anyone else like to see a new demographic survey done? I'm interested in how LW's userbase has changed since the last one (other than, well, me).

6[anonymous]15y

Yes.

5DanielVarga15y

Yes. And please include the questions "Are you a male with very long hair?" and "Are you by any chance named Vladimir?", because we need more data to investigate some strange issues.

1[anonymous]15y

Most of my social group (comp sci, into metal/alternative/gaming, slightly misanthropic) went through a period of having long hair. So it is unsurprising to me.

1Vladimir_Golovin15y

Just to complicate things a little: in addition to being a Vladimir, I was a male with very long hair.

2[anonymous]15y

Yes.

0[anonymous]15y

Definitely.

[-]Will_Newsome15y200

OkCupid thread, anyone?

I was thinking that those of us that aren't shy could share our OkCupid profiles for critique from people who know better. (Not that we have to accept the critiques as valid, but this is an area where it'd be good to have others' opinions anyway.)

If anyone wants to get the ball rolling, post a link to your profile and hopefully someone will offer a suggestion (or a compliment).

Also, I bet cross-sexual-preference critique would be best: which for most of us means gals critiquing guys and guys critiquing gals. But I realize the LW gender skew limits that.

[-]mattnewport15y170

I was interested to see what discussion this post would generate but I'm a little disappointed with the results. It looks like further evidence that instrumental rationality is hard and that the average lesswronger is not significantly better at it than the average person without a particular interest in rationality.

I'm going to throw out a bunch of suggestions for things that I think a rationalist should at least consider trying when approaching this specific problem as an exercise in instrumental rationality. I anticipate that people will immediately think of reasons why these ideas wouldn't work or why they wouldn't want to do them even if they did. Many of these will be legitimate criticisms but if you choose to comment along these lines please honestly ask yourself if these are ideas that you had already considered and rejected or whether your objections are in part confabulation.

One obvious reason for not trying any of these things is that the issue is just not that important to you and so doesn't justify the effort but if you feel that way ask yourself how you would approach the problem if it was that important to you. I haven't tried all these things myself. I rejected som... (read more)

[-]Alicorn15y100

Presumably people who are 'not photogenic' are not made of some different type of material that reacts differently to light than photogenic people. The problem must either be a lack of good quality photographs or an issue with uncomfortable body language when being photographed.

The camera also adds (visual cues that make it look like it adds) weight, and messes with color. My best friend just got married and had lots of photos taken of her and her husband. He looks fine because he starts out skinny as a rail and his coloration works in the photos. But in the very same photos, she develops a blotchy complexion and her hair color looks unnatural and gross. And while she's not fat, the extra ten pounds on the glossy photo nudge her a little that way. Her body language looks fine in photos (and if she were tensing up, wouldn't she also look tense on video? Video of her looks much better), and the quality of the photographer or camera can't be the issue because in the very same photograph her husband looks exactly like himself in real life and she looks weird.

8mattnewport15y

I don't know exactly what the problem might be with your friend's wedding photos but in general the problem of how to make people look as good as possible in photos is quite well understood. There's an entire industry devoted to doing it. I can list several technical errors that can appear to add weight or mess with color but these kinds of things are not unsolvable. Part of the skill of a good photographer is avoiding these problems. Photoshop can also be used to fix specific problems with colour reproduction. I would bet that an experienced portrait photographer could identify what went wrong with your friend's pictures to produce a less than satisfactory result by examining them for a few minutes. I suspect there may be genuine cases where certain people seem relatively less attractive in still photos than in person but this may be due to aspects of their personality or behaviour which the camera cannot capture. I doubt there is anyone however whose perceived attractiveness is not increased by a good photo relative to a bad one and in photography much of what constitutes 'good' has been figured out over the years.

2mindspillage15y

"Knowing what to do" and being able to do it well are different. I have had good photographers take pictures of me. They have used the appropriate lighting and angle and helpfully tried to coach me in what to do--how to pose, how to smile, what to wear. And indeed, their pictures turn out better than most snapshots. That doesn't mean that I am able to use their advice effectively--to hold an unforced smile and keep my eyes open at the same time, avoid tilting my head funny, or not look frustrated and impatient after the 50th shot. It's a difficult skill for me, and while I expect I could be better with practice, it's not high on my list of desired skills to improve. So I'm not photogenic. Which means merely that I don't have these skills now and don't pick up on them quickly, but when you look at my awful pictures it's no different than if photogenic-ness were some immutable inherent quality. (But I'm off the market anyway. On the upside, my partner was pleasantly surprised when he first met me in person that I was better-looking than my photos suggested.)

5Craig_Heldreth15y

xxxxx

5mattnewport15y

One of the reasons that instrumental rationality is hard is that acquiring good data is hard. Imperfect data is generally better than no data however and there are other sources where you can find research into some of the same questions that OkCupid covers. Most of the advice in their 'Don't Be Ugly By Accident' post is just standard stuff for portrait photography for example which any book on photography would cover in great detail.

3Relsqui15y

I agree about OKT, as I noticed elsewhere. I also agree with Alicorn about the glasses, if that's practical and if at-a-glance attractiveness is sufficiently high priority for you. Thanks for that. ; ) I don't feel I can remark on the way most women use their OKC profiles, because I don't read many of them and I try to stick to the extraordinary ones. But I can say that there are tons of men out there who are clearly parroting what they've been told will attract women, trying to come off as the perfect knight in shining armor while successfully avoiding showing any hint of personality. The effect is to make it seem like they're trying to attract the similarly generic woman so they can get married and have generic children. I recommend looking up how to write the accents; some of these words change meaning without them. A common example is that "año" means "year" and "ano" means "anus." Not that any sane reader wouldn't know what you meant, but it's worth knowing anyway. Some verbs change with accents in ways which are much more subtle: "estudio" is first person present and "estudió" is third person preterite. A few specific errors, if you're interested: I'd use "estudiaba" rather than "estudié" because it refers to an ongoing process, rather than a single event in time. (By contrast, one might say "empezé estudiar español en el grado segundo," because one began to study at one point in time.) I think you made a typo writing "en la escuela"; I would probably have written "a la escuela" (at school, rather than in school), but I'm not sure you're actually wrong. It might just be a style choice. Similarly, I'm guessing "facilidan" is meant to be "facilidad." Your "vecindad" is singular, so it "tiene" many Spanish-speakers, not "tienen." And while "muchas personas" is technically correct, it's the equivalent of saying "many persons" in English--more common would be "mucha gente" (many people). I'm not fluent either, so I can't promise that's exhaustive, but I've s

3Alicorn15y

I would advise you to wear smaller glasses if that is possible given your eyesight.

1[anonymous]15y

There are fun quizzes and they tell you stuff about your personality. That's why I registered; my half-assed profile explicitly says I'm with someone and to message me only if you're interested in platonic friendship. I have made friends with one really nice couple, though.

3[anonymous]15y

I know why I'm not photogenic: 1. Bad posture (which I can fix when I'm standing at a mirror, but which shows up a lot on candid pictures.) 2. Trouble with facial expressions (I'm not sure how to put this ... I'm not good at knowing how my face looks, and I have a dumb expression in most pictures. The general effect is "chipmunk.") 3. Small total volume of pictures (neither I nor my friends are in the habit of taking lots of pictures of each other.) One of my defects is -- I'm not sure if there's a shorter way to put this -- knowing what my body position would look like to an observer. It's why I can't do something like, say, golf: you'll tell me to change my form and I won't understand what I'm doing wrong because I can't "see" myself. I think that photogenic people and performers, apart from being physically attractive, are really good at "seeing" themselves.

[-]Relsqui15y100

I think that photogenic people and performers, apart from being physically attractive, are really good at "seeing" themselves.

I'm not sure I agree with this--or rather, I'm not sure this is the best model of what's going on. My impression has always been (and this fits with my photo-taking advice elsewhere in this thread) that you don't learn to see how you look when you're doing something right--you learn how it feels to be in the correct position to do it. That is, someone who's watching you might say "your back is curved, straighten it," and you can straighten it, but you still don't see what they see. You just find out what it feels like to have a straight back, and can try for that again later. I've never played golf, but I'd be surprised if good golfers are thinking about what they look like when they're putting. I'd expect them instead to recognize the feeling of being in the correct posture from having done it before.

4mattnewport15y

This kind of self awareness would be a good starting point to fix the problem if you decided it was important enough to you. There are various things you can do which plausibly claim to improve body awareness (I've heard the Alexander Technique mentioned around here though I don't know anything about it myself) and good body language can be learned to some extent. Even if you don't think it's worth the effort to work on these things however, if you go to a good professional portrait photographer they should be able to help you address these kinds of problems and get some good pictures. Portrait photography isn't my main area of interest but I've read some books that cover the basics and they generally talk about techniques for getting the client relaxed and comfortable in order to minimize the effects of awkward body language and about things you can tell a person to do that will help them position themselves in a way that will produce good photos.

3[anonymous]15y

I like your ideas. Although some become harder to enact the less frequent your desired partner type is, which seems to be a problem for some people. I'll note that if you are only willing to spend limited time on it and have the choice between improving general attractiveness and A/B testing profiles, I would pick the former. I'm currently aiming for the increased salary and improved fitness. I don't hold out much hope for OKcupid, I think I'll do better just getting out more to the sorts of events that the people I am interested in might go to.

[-]Vladimir_M15y160

I've never used any dating websites, but people who care about that sort of thing should note that the advice they'll get this way may have a very low, or even negative correlation with what actually works. I don't mean to say that people will consciously write misleading things -- just that, for various reasons, they may not work with a realistic idea of the thought process of those who are supposed to be attracted by the profile in question. To get useful advice, the best way to go is to ask someone of the same sex (and preferences) who has successfully used some such site to share their insight.

[-]wedrifid15y120

the best way to go is to ask someone of the same sex (and preferences)

Definitely. It is often recommended to guys to focus almost exclusively on advice from other males rather than women until you have reached a level of understanding such that you can reliably distinguish between people talking about what the world 'should' be like rather than what the world is. Having a certain level of social presence also makes it more likely that people will refrain from trying to foist the rules for boys on you and actually give honest assessment's of preferenes.

[-]Scott Alexander15y120

I don't know who erisiantaoist is, but I cannot believe he actually started his profile with the "I am slightly more committed to this group’s welfare..." quote. If I were at all gay I would date him in a second just for that.

...although I am generally surprised at how anxious people in this group are to signal transhuman weirdness, especially transhuman weirdness that only one in a few thousand people would understand or find remotely sane. Do you really have access to such a high quality dating pool that you're looking for people who will be impressed instead of confused when you name-drop AIXI and state your intention to live forever?

[-]Will_Newsome15y200

In my case I'm not looking for a transhumanist match: there are lots of really smart, interesting girls out there that haven't heard of transhumanism, and I have a lot more things to talk about than ethics. (Seriously, who talks about bioconservatism on the first date? Once a girl said on a first date, "There are so many books, it's the only thing that makes me really sad that I'm going to die." And I was like, "Oh, don't worry about that. We're not that far off from solving death, you'll be fine. I personally work on the problem; trust me, you'll have trillions of years to read your books." I was still able to get a second date.) I just do it because I think it's hilarious that I can come across as insane and still get girls by being relatively laconic. Social normalcy screens off epistemic oddity, to an extent. I'm not socially normal but I do an okay job at emulating it most of the time.

So those that would take inspiration from my profile, note that the nerdy parts aren't optimized for success. At all. I just like typing AIXI. AIXI AIXI AIXI AIXI AIXI...

[-][anonymous]15y160

Maybe they're just sick of half-heartedly dating folks they don't click with.

I had a brief relationship with a nice boy who had never heard of most of the things I'm interested in -- he wasn't an intellectual type. Nothing against him, but it was surprisingly disappointing. After that, I thought, "Okay, in the long run I'm going to want a deeper connection than that."

[-]Relsqui15y200

Maybe they're just sick of half-heartedly dating folks they don't click with.

The first few versions of my profile were geared to show off how geeky and smart I was. This connected me to people who spent a lot of time playing tabletop roleplaying games, reading fantasy novels, and making pop culture references to approved geeky television shows, none of which are things which interest me particularly.

Eventually I realized that I am not actually just popped out of the stereotypical modern geek mold, and it was lazy, inaccurate, and ineffective to act like I was. Since then I've started doing the much harder thing of trying to pin down my specific traits and tastes, instead of taking the party line or applying a genre label that lets people assume the details. In that way, OKC has actually been a big force in driving me to understand who I am, what I want, and what really matters to me. A bit silly, but I'll take it.

Here's a question I've been pondering a lot: What are good questions to use to actually learn something about a person? (If you suggest "what kind of music do you listen to" ... you're fired.) If they're not the same for everyone, and I expect that they aren't, how do you find them?

[-]orthonormal15y110

What are good questions to use to actually learn something about a person?

"What's something you believe, that you'd be surprised if I believed too?"

(I've yet to try this in a romantic context, but when meeting new friends it usually leads to a good conversation– the more so for ruling out first-order contrarian beliefs that they'd expect me to share.)

3Relsqui15y

Oh, that one is excellent. I might try that on some of my current friends. Although ... I wouldn't recommend using it on someone who dislikes debating or defending their beliefs (or on someone about whom you do not know that). If they're right, you have an immediate source of conflict which if taken personally could nip the new acquaintance in the bud.

9arundelo15y

I want to upvote this more than once. For a while (a long time ago) I asked people "When did you first realize you were different?" Once a young woman I was on a date with said "But I'm not different"!

0Relsqui15y

Weird. How did that turn out?

1arundelo15y

Well, I knew it was a bad sign, but I didn't want to write her off based on just that one thing, so I did end up asking her out again. After two or so dates she realized she wasn't interested in me but apparently didn't have the guts to tell me that, so she stood me up and pretended she didn't see or hear me (!) when I went to her place of work. (We were both pretty young -- just out of high school.)

8Relsqui15y

Ouch. Classy. One thing I've learned from having mostly male friends who sometimes complain about their dating lives is no matter how much of an asshole I feel like when I turn someone down, it's much less than the asshole I'm actually being if I don't tell them. This is why I always respond to new messages on OKC that aren't outright rude, offensive, or all textspeak (and even then I sometimes do). Although, I will admit that, after an exchange has alerady gotten started, if I've been signalling "not interested" over a few messages (responding with minimum polite reply and not asking any new questions), and the other person persistently doesn't get it, I may just trail off.

[-]MBlume15y100

One thing I've learned from having mostly male friends who sometimes complain about their dating lives is no matter how much of an asshole I feel like when I turn someone down, it's much less than the asshole I'm actually being if I don't tell them.

Strongly agree in general...

This is why I always respond to new messages on OKC that aren't outright rude, offensive, or all textspeak (and even then I sometimes do).

I might be unusual here, but I actually consider online dating to be kind of a special case. My usual strategy is to message anyone I'd want to go on a date with, and then forget I sent the message. This means if my mailbox turns pink, it's a pleasant surprise, and there might be a date in the offing. Finding instead a polite rejection is a bit disappointing.

That is, IRL, if we're friends/acquaintances, and I'm politely/vaguely suggesting that I'd like us to date, and you're picking up on that, it is totally good for you to shoot me down, so that I can quit wasting mental/emotional energy. Online where things are more explicit, the only waste of mental/emotional energy is when I'm logging in to find a message, only to find a rejection.

1Relsqui15y

I'm actually the same way, but I think a lot of people aren't. Or even if they are, it's kind of draining to put so much energy over time into writing messages and get nothing out of it. Being politely turned down is at least an acknowledgement. In practice, I almost never get to the point of having to explicitly turn someone down, though. Either one-sided clicks don't happen that often, or when they do, the other person gets the hint I'm not interested.

7CronoDAS15y

I wish more people knew this.

6Alicorn15y

Should I really message everybody back and tell people to go away if I don't want to talk to them? If some guy in his thirties or forties sends me a message asking a question that I clearly answered in my profile and we have only a 40% match, what do I even say? "You are out of my age range, apparently did not read my profile, and have a low match percentage with me; please go away"? This sounds like an invitation for him to argue with me!

3Relsqui15y

This policy may not scale to the point where I get a large volume of messages from people I have zero interest in engaging with; I wouldn't know, because that's not the case. However, in response to your example message, I might say no more than "That's in my profile" or a one-sentence answer with no followup question or other remarks. If he doesn't take the hint, I can either argue back if I feel like being snarky, or ignore him afterwards.

4[anonymous]15y

Perplexed said something smart, but here are my drunk brainstorming ideas: Do you believe in God? Why or why not? (people are usually willing to answer this, but they'll get offended in a hurry if you argue) What experience in life have you learned the most from? What was your favorite subject in school? Do you still follow it? Do you think it's worthwhile to give to charity? Which ones? Do you give to them? (see God question caveat, and the last question is extra offensive if they have to say no) What do you think about him? (Mention someone you both know personally or indicate a person who can't easily overhear) What do you do when you don't have to do anything else?

3Relsqui15y

I like some of those, particularly the last one--I've seen something similar, which was "what could you talk about for hours?" I'm reluctant to ask "most," "least," or "favorite" questions, because almost nobody has good prepared answers, except to the trivial ones like "favorite color." Which is not an effective question for getting insight into someone's worldview.

3Alicorn15y

I like asking people what their favorite playing card is, and acting confused if they don't have one (but about half the time they either already had one or are willing to make one up to play along).

0Relsqui15y

That's funny. I don't particularly have an opinion about playing cards but I thought of one immediately when I read that.

1Alicorn15y

My favorite is the ten of diamonds; what's yours? :)

7CronoDAS15y

Mana Maze.

0Relsqui15y

Har har.

3Relsqui15y

Jack of hearts. Why yours?

1Alicorn15y

The ten of diamonds is the "good ten", and worth two points (more than any other single card) in cassino, my favorite card game. Why the jack of hearts?

1Relsqui15y

Highly ranked enough to be human, but humble among them. Hearts connotes sociality (on that test I lean towards diamonds, but if we're talking about real life rather than gaming I'm much more a heart).

1Perplexed15y

Actually, the most informative prepared question is probably this one:

6Relsqui15y

And then what do you do if they say "no"? :) Personally, I'm annoyed by people who say "you can ask me anything, go ahead" (especially in response to the "most private thing" prompt on OKC). It's a way of putting the burden of making conversation on the other person, instead of sharing it with them.

5[anonymous]15y

Meh... The more I think about it, the more I love prompting your interlocutor to gossip about a third person. You can potentially learn: -How interested she is in other people -How willing she is to talk about other people behind their backs -How much she knows about other people -What qualities she values in other people -Whether she tends to judge people generously or not plus whatever actual facts you may manage to glean about their relationship with said third party! It's great! It's kind of ethically shady, but I am drunk and willing to overlook that for the sake of argument. Edit: Hey guys I fail at lists sorry

0Cyan15y

Under the comment box to the right, you'll find a little help link.

-2[anonymous]15y

Edit: Ok grumpy people, let me clarify: drunk people don't format good and don't really spend a lot of time reading instructions, I know it's there, tyvm.

4Perplexed15y

I don't think that a prepared interrogation is the way to do it. Instead, I think you need to listen carefully to what is said in casual conversation ("What kind of music?" works fine!) and then ask followup questions to draw them out. Example: "You don't like country? Me neither. What is the thing about it that annoys you most."

0Relsqui15y

Oh, I think I was unclear--I meant the kinds of questions you'd want on a profile page or a matching survey, not for in-person dates. In the latter case I completely agree with you. :)

3[anonymous]15y

I'm not completely in the geeky mold either. But if you literally take a random sample of young men in my area, I will not get along with most of them. There's some sense in filtering. To learn about people, I usually get them started on their interests. It only really works for people who have interests (enthusiastic about a hobby or career) but do you really want to date someone who doesn't? I only feel I know someone when I know his personal philosophy, but that usually takes time to come out.

0Relsqui15y

Of course--the alternative to self-labeling isn't sacrificing your personal criteria. I might not have understood you correctly, because I'm not sure whether this is agreeing or disagreeing or a tangent. That actually makes me flinch a little, because I've spent a lot of time on OKCupid thinking "It seems like everybody else has defined things which they go out and practice and spend money on and share with their friends. I don't have anything like that. Am I just too boring? : \" I suppose that's more not having a discrete hobby than not having interests--but my interests are much more about the way I live than what I go out and do. I don't think I actually am boring, but I'm afraid I read that way, because I don't have a thing which I do. Indeed. I might have said "worldview" instead, but they're probably different angles on the same idea.

1CronoDAS15y

I usually go with "What kinds of things do you like to do in your free time?" although I don't know if I'm really the one to be making suggestions on this topic, considering how little I go out. One reason I ask that is because I fit into the category of and I'm hoping to find people with similar interests.

1[anonymous]15y

If you had a year where you didn't have to work*, what would you do? *You had lots of money and a job to go back to after

0Relsqui15y

That's a really good one. Oddly enough, I think the "if you won ten million dollars" question works well for the same reason--they give you a sense of someone's priorities besides the necessities. (Someone asked me that on Formspring a while back, and I had fun taking my time and thinking about what I really want.)

0[anonymous]15y

I think typical questions like "what kind of music do you like?" are counterproductive, annoying, and cliched. Conversation should be more like dancing and less like filling out a form. To understand a person, you must (simply) read between the lines of what they say and how they act. Study their body language and tone of voice. Notice, when the conversation reaches a crossroads, what path they choose to follow.

6erratio15y

I would upvote this multiple times if I could. Having mostly non-mainstream interests sucks for dating.

5orthonormal15y

Remember, these profiles are selected from the pool of "people who link their OKCupid profile on Less Wrong, and ask advice of the LW pool".

2ata15y

I admit that my profile does gratuitously signal transhuman weirdness... I'm okay with that not because I'm expecting a large number of people to be impressed rather than confused/repelled by that, but because I'm more interested in meeting the (smaller number of) people who are interested/impressed/okay with that.

2Relsqui15y

I think the point is to be aware that you may also be driving away people who don't mind it, or might be interested if they learned a little first. Or people who think it's weird, and will judge you based on it if it's the first thing they learn about you, but will shrug it off and agree to disagree if they've already learned to like you first.

2ata15y

Yeah, I've considered that. For me it's just a little bit hard to avoid talking about under "What I'm doing with my life", because that is what I want to do with my life. But I was aiming to signal something like "interesting and plausible enough to contact him and find out more" — maybe I'm overshooting that and landing in "he sounds crazy; ignore" territory for most people, even people who might otherwise be convinced of the correctness/desirability of the things I mention there. P.S. ...I just recognized your username! You messaged me a few months ago when I lived in Berkeley and I had a bit in my profile that said that I expected I'd usually be too shy to message people but not too shy to reply. Turned out I was too shy to reply. Sorry :( I wish I had replied. You sound cool.

1Relsqui15y

I should note that I was just responding to the previous comment here; I hadn't seen your profile yet but now I have to go see. Oh, yeah! I remember you. No worries. :) Your description of your sexuality sounds just about like the mirror image of mine. It's tricky to describe, isn't it? I took a crack at it in my "message me" section but I don't know how good a job it does of explaining. Anyway, about the strength of your statements re: singularity et al ... I think the only part that I would expect to make someone o_O is the throwaway mark about humanity probably not surviving this century. I don't think most humans believe that to be the case. Whether they've ever really thought about it or learned why anyone else thinks so is not the point; they don't, so that might come off as crazy. Up to you whether to care.

1Alicorn15y

The up-front declaration that you are Not Interested In Marriage Or Children Or Probably Even Exclusivity thing is new; what made you decide to add it?

1ata15y

Because I don't want to lead people on if they're looking for those things. Do you think it would be better to leave that out of the profile and just let my responses to the relevant matching questions filter for that sort of thing? (I actually need go back and change some of those to more than "a little important"...)

1Alicorn15y

Well, it does a good shutting-down job, but you should definitely also fix your questions - you're still at 97% with me even though I'm a fan of all three things.

0ata15y

But do you think it's worded in a way that repels people with whom I could otherwise at least be friends? I don't want to signal that I'm hostile to people who are looking for those things, only that I'm not personally looking for them in a relationship.

1Alicorn15y

I don't actively look for Just Friends on OKC, so I'm not sure how informative I can be about the search patterns of people who do. I didn't get an impression of hostility, just a clear "Dead End" sign.

0Relsqui15y

Last I checked, straight men aren't that into gay dudes. :P

0[anonymous]15y

I took it as moderate evidence that that guy is Clippy

[-]Alicorn15y120

May as well... Me.

Great profile!

witty, + 5

sci-fi and singularity geek, +5

draws a webcomic + 5

wants kids, +2

degree in worldbuilding, awesome, +10

cooking, +3

vegetarian, -2 :(

likes fancy pretentious cheese, +1

I'm too old, live on another continent, and my wife is next to me right now, -50

5CronoDAS15y

Thank you. Now I have a face to go with my pointless Internet crush! :)

1komponisto15y

Aargh! Surely you know that only low-status people use the preposition "for" in this context! High-status people say "in".

1Wei Dai15y

This is curious. Do you have an explanation for why "for" is associated with low-status and "in" is associated with high-status (in this context)? Are there other similar linguistic phenomena?

4komponisto15y

Hm, I suppose I could attempt one. I think my current best guess would be along the following lines: High-status people will tend to have a richer cache of stored expressions; in a given situation, they are more likely to be able to precisely reproduce a previously-heard expression appropriate to the context, rather than having to make up a new expression on the spot. This is especially so if the idea being expressed is one that high-status people think about more often than low-status people do. Consequently, a high-status person will be more likely to remember the phrase "graduate school in philosophy" in detail, including the specific information about "in" being the preposition used; whereas a low-status person, who (at least at first) may not have as much occasion to speak about graduate school in philosophy, may only have something like "graduate school [preposition] philosophy" stored in their mind. As a result, when they first need to use the expression, they will have to spontaneously choose a preposition, and the choice they end up with may not be the same as the one in the existing expression commonly used by high-status folk. But now, when the low-status person next uses the phrase, they will have a tendency to remember the preposition they themselves used the last time; so this new expression will then spread among their low-status associates, and will become the standard cached version of the expression for low-status people.

0Relsqui15y

I don't know whether this is one, but I expect that some such expressions have different "correct" forms on either side of the Atlantic.

0komponisto15y

Yes, of course status levels are not the only source of linguistic variation; there's also geography, and other things also. Note however that high-status language varies less by geography than low-status language. Also, British English (at least "Southern British Standard") sounds higher status to me than American English in general, so I would find it surprising if an expression that struck my (American-English) ear as low status turned out to be a high-status British form. I would expect the reverse -- that is, something that sounds low-status to a British speaker being a high-status American expression -- to be more common.

0komponisto15y

Sticking to prepositional shibboleths, another one that comes to mind concerns professional athletes and sports teams: Low status: "X is on the Texas Rangers." (Generalized from the use of "on" in expressions like "whose team are you on?") High status: "X is with the Texas Rangers." (Standard expression used in the specific context by sports journalists, etc.)

0erratio15y

One of my best friends, who is far more intelligent than I, sometimes says 'could of' instead of 'could have'. My point being, having language pet peeves is fine and normal, but whether people make those errors or not is a really poor indicator of status (even though hearing 'could of' and 'for all intensive purposes' drives me up the wall)

5komponisto15y

Some very intelligent people also wear T-shirts instead of suits. Nevertheless, it would be preposterous to deny that wearing a suit is a meaningful status signal, or to claim that clothing is a "really poor indicator of status". Signaling mechanisms aren't perfect, and yet they're still signaling mechanisms anyway.

7mattnewport15y

In engineering / software circles, wearing a T-shirt rather than a suit is a kind of countersignaling.

3komponisto15y

Indeed; and no doubt linguistic countersignaling also occurs in some communities. (Example: politicians.)

1erratio15y

Spoken language is much more difficult to change on purpose than clothing is, so in my view it has much less value than clothing choices as a status indicator.

4komponisto15y

In that case it should have more value as a status indicator -- harder to fake.

1erratio15y

I think my main disagreement with you here is in whether unconscious or conscious signalling has higher value as a method of determining actual status. I would argue that choosing to put on a suit is actually a better determinant of high status than an accent that indicates I grew up in Dumbistan, because status is something you obtain as opposed to something you either have or don't.

5Perplexed15y

I think that all talk of status flirts with the Mind Projection fallacy. Status is almost entirely in the eye of the beholder. A high status person is one who conforms to the ideals of the status-judge. So, you are both right as to what signals status, as long as you really mean "what signals status to me".

1kodos9615y

How can you even tell the difference? It seems like unless you're speaking in a very formal, deliberate manner, they're pronounced nearly identically.

1erratio15y

It must be a dialect thing, I can clearly hear the difference. I wish I couldn't.

0Alicorn15y

I don't think I've actually heard "in" used there, and it sounds awkward to me. I guess I'm just low-status. Oh well.

0komponisto15y

You have, many times. Trust me. "I'm a graduate student in philosophy". -> "I went to graduate school in philosophy" -> "Philosophy? I went to graduate school in it."

4gwern15y

The Googles would seem to agree: * http://www.google.com/search?q=%22graduate+school+in+philosophy%22 * http://www.google.com/search?q=%22graduate+school+for+philosophy%22

0Relsqui15y

Oh man, you read like an alternate universe version of me where I stayed in school and kept reading fiction. Or, if it's not quite that strong, you at least send off a lot of "in my tribe" vibes. I'm almost exactly the same way. I don't need only one at a time, but I like small groups or pairs. The closer I am to the people in question, the more of them I can handle at once. I'm building a list of axes (that's the plural of "axis" here, not "axe") along which humans are distributed for social/relationship purposes, and I definitely think "what size of group do you prefer to socialize in" is important enough to be on it. (It might actually have been the first one.) I really want to try your cauliflower soup now. I have a rant about a set of "most private thing" answers that includes yours. You're not obliged to care, but there it is.

3Alicorn15y

I just don't know how to organize things I talk about into "more private" and "less private". I just have "too private to mention" and "not private". So while I'd gladly pick something and put it there if I had such a spectrum, I don't know how to make facts about myself line up in a spectrum. Suggestions? Soup. I also add a parsnip or two to each batch lately, and I tried it with a turnip once but didn't think it was an improvement. I would have added these to the post but I have forgotten how to log into my blog, it's been so long since I've used it.

3Relsqui15y

Oh, that's a good reason. On a side note, I'm curious--have you made a deliberate choice to use the UK style of punctuation for quoted words at the end of a sentence (period outside, not inside)? If so, why? The only thing I can think of is to look for intersections of the set "other people might consider this too private to post" and "I do not consider it so." Anything in both of those is probably at least private enough to be interesting. Alternatively, I consider your other use of that box (adding something which didn't fit anywhere else) to be entirely valid. Thanks!

5Alicorn15y

I put punctuation inside quotes only when it is part of the quote. For example, I'll put an exclamation point inside quotes when I note that I sometimes greet people by saying "Hi!". (But then I put a period after that.) I am not conscious of this being a UK thing; it's just how it makes sense to me. Can you give me examples? The ones I've seen in the wild have not had any clear analogues to myself.

2RobinZ15y

I use the same notation, and have seen other people report the same for the same reason.

2NancyLebovitz15y

I punctuate the same way, and for the same reason. I suspect it's a geekishness thing.

6arundelo15y

Guy Steele & Eric Raymond (don't know which wrote this part): Here or here. (The first link is to the copy on ESR himself's site, but the quotes are messed up.)

2komponisto15y

Me too. The mathematician Paul Halmos was an outspoken defender of this.

1Relsqui15y

Yeah, this is certainly logical, and like RobinZ I've seen other people do it for the same reason. It happens to also be what's taught in UK/world English, as opposed to US English, or at least such is my understanding. It startles me when I see it, though, presumably because the US rules were so thoroughly ingrained in me as a child that it seems to mismatch an otherwise good impression of someone's grasp of grammar. In case it's not clear enough, I don't think less of you for using that syntax; I'm just probiing my reaction to it. Uh, hm. I'm not sure, actually; I don't know where your comfort zone lies! Sex is usually a good bet; I've also used that field for things I'm not embarrassed by but can be shy about mentioning because some people are dicks about it. Things I can remember having there at some point, or could have, include: * I don't know how to ride a bike (which was true until I was 23). * I don't drink coffee or alcohol--not because of any code of conduct, just taste. * I'm pretty difficult to bring to orgasm. No one has ever done it within a couple weeks of a first attempt. There are flaws in these; the second one makes me sound boring, and the latter two both result in dumbasses saying things like "you just haven't tried the right drink/dick yet." But they're examples of things in the personal-but-not-too-personal set for me.

0Alicorn15y

I know how to ride a bike, but don't have and have never had a driver's license - but that isn't private at all. I don't drink coffee or alcohol either, but one's clearly stated in the sidebar and the other seems irrelevant and also not private at all (and I answered a question about it). And my analogy to the last case would seem to invite messages that, as you say, are not welcome.

0[anonymous]15y

It is a well written and interesting profile. I'd go with it being a bit more concise, but that is my personal taste. You make it clear that you are unapologetic about your liking of Stephenie Meyer, I don't think you need the section about liking kids movies. It seem somewhat redundant (I'm not implying that Twilight is a kids movie, simply that both make the point that your taste in media is your own, not societally driven).

0[anonymous]15y

It's a great profile. My preference is for more concision, but not everyone's the same about that.

0Will_Newsome15y

IMO I think your third picture (the wedding one) should be your main one, it's a very good photo.

2Alicorn15y

But I'm not wearing a bandana, and therefore I don't look like myself!

7Will_Newsome15y

You totally do look like yourself! It's not like you'll show up for a date and the person will be like "WHENCE THE BANDANA, LIAR?! YOU ARE NOT THE DROID I WAS LOOKING FOR." And then you'd be like "WTF" and that'd be that. Where was I? Oh, right, sleep deprived.

0Alicorn15y

Can I get second opinions on the relative goodness of the bridesmaid picture sans bandana and the more characteristic bandana-containing pictures? Will's raving and I can't trust his judgment :)

5Perplexed15y

I definitely prefer the bandana. But then I'm an aging hippie, so my judgment is suspect. Nice profile, btw. God, you guys make me feel old.

1[anonymous]15y

They made me feel old (I'm 31), now I feel less old.

3Emile15y

I find the bandana one more attractive (the one with the dark sweater, not the one with the big wooden, I mean, metal thingamajig), you look more natural and comfortable than the wedding one. (Does this mean I'm an aging hippie too? I do have a beard.)

5Alicorn15y

It's not wood, it's an actual metal bat'leth.

3ata15y

That one is my favourite too.

1Alicorn15y

Well, sheesh, now I don't know what to do. My own preference is the closeup of my face, so I'm going to switch back.

3[anonymous]15y

Honestly, I think most people expect you to put up your best pictures in the fore. If you don't they might think your average pictures are your best pictures, and you look worse in person. The world of dating is not at all about honesty sigh

3Alicorn15y

These are my best pictures. I'm terribly unphotogenic; these three (and a redundant bat'leth picture) are just about the only ones I can stand to publicize.

3Relsqui15y

For what it's worth, I like them. You have a good smile, and the smile is what makes or breaks a picture.

1Alicorn15y

Yes, and in most photos, I look like a complete doofus - these are the good smiles, not just random smiles :P

3Relsqui15y

Well your good smiles are adorable. So there. :P

1mattnewport15y

If you have a friend with an interest in photography and a decent camera ask them to take a few pictures. A good camera and some basic photography skills make a huge difference.

0Alicorn15y

As far as I know, I do not have such a friend.

1mattnewport15y

Well if you acquire one bear it in mind. I have a Panasonic GF1 which happens to be the camera OkCupid found took the photos with the highest average attractiveness and a photo I have of myself taken with it (a random snapshot by my sister who's not majorly into photography) got the highest rating of all my photos on their My Best Face test even though I don't think it's the most flattering picture of me from the ones I tested. Some nice depth of field and bokeh in the background seem to swing it for a lot of people though.

0[anonymous]15y

I meant the comparison of your bridesmaid one to your non-bridesmaid ones. People, generally, will make an initial judgment based on the first picture. Of course if you are not interested in general people, it doesn't matter so much.

1RHollerith15y

I think both bandana pics (specially the grey bandana one) are better than the bridesmaid one.

1Relsqui15y

I like bat'leth most, other bandana second, wedding third. But if you're going with the non-bat'leth bandana photo because you can't get close enough in on that one to frame your face the way you'd like, as it seems like you are, I trust your judgment.

0[anonymous]15y

Bridesmaid over bandanna. I love the little braidy-hair thing. Doesn't matter that you don't look like that all the time -- I don't wear satin leopard-print dresses all the time, but it's a good photo.

0[anonymous]15y

Bridesmaid over bandanna. I love the little braidy-hair thing. Doesn't matter that you don't look like that all the time -- I don't wear satin leopard-print dresses all the time, but it's a good photo.

1Alicorn15y

All right, I swapped 'em. That photo also has me in makeup, which I also never wear and was obliged to permit for the wedding... The braid is actually really cool. The hairdresser taught me how to do it - it builds into its procedure the fact that humans have two hands, and doesn't require you to have three or four, which is nice :)

9Apprentice15y

Now, when I've read LessWrong in the past I've always thought to myself, "what a nice friendly community, populated by geeks like myself". But looking at all those profiles I'm amazed how outdoorsy adventurism seems to be a big part of your self-image. SarahC "hiking and trail-running the Wasatch Mountains this summer was a blast" Yvain "insane adventure ... mountain-climbing in the Himalayas" mattnewport "Mountains make me happy. I love snowboarding and hiking on them." Nisan "I love exploring" [pictures in exotic mountainous locations] JGWeissman "I am an active member of the UCI Sailing Association" Relsqui "You want to go do something with me. Ride bikes" I've got nothing against mountains, I've even been up a few and enjoyed it. I'm not fat or anything, I even exercise semi-regularly. But if I were to make a profile like this it would never occur to me to emphasize those aspects of myself. Is this known to be attractive? I personally find these profiles intimidating. I would be more awkard and less comfortable meeting any of you after seeing those profiles than just after seeing your LW contributions. To me, those profiles have a tendency towards coming across as perfect-and-a-bit-bland or goody-two-shoes, which is simultaneously intimidating (I'm not perfect) and not exciting to me. I like people that have some decadent, flawed or 'evil' aspects. Saying you have an evil sense of humor is something, but it's a bit non-specific. Disclaimer: I married my high-school girlfriend. I have zero experience with dating or dating websites. I probably have no idea what I'm talking about. I would possibly be qualified to give advice about maintaining a successful relationship but I know very little about starting one. My tastes in people are also probably not typical, I have a fondness for the romantic and the mysterious - exhaustively detailed profiles are inherently something of a turn-off to me. Edit: Rereading my comment, I think I hit the wrong tone. I meant

7[anonymous]15y

Well, a lot of people like physical activity in a partner -- it says something about physical attractiveness, and also a sort of energetic outlook. As for the predominance of outdoorsy activities -- honestly some of it is a class/culture thing, but so what. I've also noticed that there's overlap between math and mountaineering -- sort of the same kind of "coincidence" as math people liking Bach. Geeks tend to be drawn to physically demanding individual pursuits: running, climbing, cycling, and to a lesser extent weights. (Swimming ought to qualify but I've never met a lot of geeky swimmers.)

1Apprentice15y

I see, that's interesting. (I run and swim - but swimming is very popular in my area.)

1mattnewport15y

I live in a city surrounded by mountains that hosted the Winter Olympics this year so snow sports and hiking are pretty popular around here. There's probably some cultural context that interferes with the signalling going on in this case.

2Apprentice15y

Ah! So basically I fell for the fundamental attribution error. My explanation (dispositional): This person is a sports fanatic since he practices such an exotic high-commitment sport. True explanation (situational): This person lives in Vancouver, where this taste in sports is commonplace.. Since you'll be looking for a person in your own city, this misunderstanding of signals presumably won't be a problem.

4Relsqui15y

I find it really interesting that I'm included in that, but I think I understand why. I absolutely share your intimidation by really outdoorsy-oriented profiles. It makes me feel like the person would always want to be off doing things I didn't know how to do, wasn't fit enough for, or just wasn't interested in. I don't place my mentions of cyclling in that category, for two reasons: 1) Bicycling as a primary mode of transportation is common among people my age in my city. Cycling isn't a hobby for me; my bike is my car. I rarely ride just for pleasure. 2) Because of the first point, dating another cyclist--or at least a bike-friendly person--is actually a practical matter for me. If I date someone who has a car and no bike, and we travel together, I'm relying on that person for transportation wherever we go, and cannot leave by myself if I want to. This is inconvenient at best, and potentially scary at worst. We could travel separately, but drastically different speeds make that a bit of a nuisance ... plus, traveling with my date is just nicer. This is not to say that I wouldn't date someone who didn't use bikes for transportation. But it's easier, and if it's going to be an issue I'd like to establish that up front. Besides ... it's a filter for the sorts of people who might think "Ugh, why are these stupid bicyclists riding in the road?! Don't they know it's just for cars?!" ;)

4Apprentice15y

You describe it well - I get tired just reading this stuff :) I see - fundamental attribution error again. The true explanation is more situational than dispositional.

0Relsqui15y

Yeah, but I found it really interesting to look at it from that perspective, so thanks for bringing it up.

2Jonathan_Graehl15y

Outdoors-adventure stories/pictures: I also enjoy such activities in moderation (e.g. I play beach volleyball several times a week), but doing so is seen as an attractive quality (evidence of "spirituality", health, attractiveness, willpower, etc. in comparison to the stereotypical couch potato or computer nerd). So you should expect people to sell that part of themselves to the extent that it exists in any quantity.

9[anonymous]15y

If people haven't you should read the OKtrends blog, especially about what makes a good picture (good quality camera etc). Also a bit about what people are looking for dating wise would be useful. Generally that is too dull to put in the profile itself, but required for advice I think. My profile is currently moth-balled, and not really designed to be attractive (I did get accused of being narcissus, which put me off making it more attractive). I found it horrendously addictive interacting with people and looking at the breadth of humanity and their hopes and dreams. If not people are nosy, I will divulge my okc identity via private message. I am also quite happy to give out awards, if people want them. Or just look at your visitor list for the guy with the round blue sunglasses on.

9Will_Newsome15y

I and 100 others second the recommendation to read OKtrends: simple, frank statistical analysis. It's not the most useful information, but I was really surprised at their frankness when discussing touchy subjects: religion, race, attractiveness. They're very down-to-Earth in their analysis. Good point.

8[anonymous]15y

I like the OkTrends blog; the analysis is more casual than scientific, of course, but the content is stuff that I was really interested to learn. Also, as a more general point, I think that non-scientific but enormous back-of-the envelope calculations (like OkTrends) are a good supplement to tiny but scientifically designed studies. If the big&sloppy approach and the little&precise approach give completely different results, that's cause for further investigation. Pre-Internet, we didn't have the opportunity to do that much big&sloppy statistics, except very crude things like the census; but today, I think it might be useful to double-check the results of small experimental studies (e.g. in medicine) with self-reported stuff from an enormous population sample.

3datadataeverywhere15y

I agree with the big & sloppy point, but would also like to point out the parallel between the OkTrends blog and the Mythbusters.

4Relsqui15y

While OKTrends does contain a lot of interesting, well-explained, and often entertaining statistics, I would be very cautious about mentioning it in the same breath as anything to do with how to attract people. The folks at OKC can describe the kind of photos which get a user lots of messages--basically, pointing the camera down your cleavage if you're female, cropping your head out in favor of your abs if you're male--but a fat lot of good that does you if you have a small chest or a belly. (Or if you want to get messages other than "UR HOT WANNA GO OUT 2NITE.") Similarly, knowing that "vegetarian" is a first-message keyword which disproportionately leads to conversation is not all that helpful if you aren't one. Relatedly, their data is collected from a large and surprisingly mainstream userbase; if LWers are as atypical as we say we are, very few of those users are going to be similar to whomever you're trying to attract. Maybe you know a geeky, intelligent woman who wouldn't roll her eyes at the headless ab shot, but I don't. If you don't believe that the target audience of OKC is very different from LW's readerbase, read the comments on any OKTrends post. They're not all inane, but ... (I suppose I'm making the assumption there that LWers looking for dates are more interested in someone smart and with common interests than someone who relies on having tits to get attention. If I'm wrong, feel free to disregard.) By the way, just in case you don't feel insecure enough yet, OKC claims to be quietly segregating its users by hotness. I've seen it theorized that that was a publicity stunt or a sneaky way to pull back inactive users, which seems quite plausible to me, but doesn't make the stunt any less scummy. Don't get me wrong; I like and use OKC. Just remember that, no matter how clever and statistically sound their algorithms are, most of their data still comes from people who think that what checkout stand magazines say about people, dating, and sex is actu

9wedrifid15y

I have no problem with dates who are smart and have common interests and rely on tits to get attention. :)

5wedrifid15y

That seems to be a desirable outcome and one that I expect would be the natural outcome from applying statistical measures to interaction patterns. I expect and prefer OKC to provide matches that are most likely to lead to rewarding interactions. These do tend to be more likely between people of approximately equal hotness.

0Relsqui15y

That's conventional wisdom, certainly, but I'd be interested in seeing an actual study.

2mattnewport15y

Who You Find Attractive is Based on How Hot You Are. I couldn't find a published study with a couple of minutes but I recall seeing experimental evidence for this before.

1Relsqui15y

Thanks, that was interesting. One of his conclusions has a funny side effect: if you care a lot about how hot people are, you're probably pretty attractive. ;)

0wedrifid15y

Have you tried google? I've seen several but do not have the links on file.

3wedrifid15y

Does that help? I would have expected leaving the head there to go with the abs would work better. If I see pictures of just breasts then I wonder what is being hidden (and aside from that find the expressions on a girls face and the style of hair to be potentially attractive.)

4Relsqui15y

I thought they'd mentioned that in OKT once, but I just went back and looked and didn't see it. So maybe I made that up. Her personality. >_>

2[anonymous]15y

As I recall, "body" as opposed to face pictures actually hurt your chances, statistically.

2luminosity15y

Interestingly, since receiving the mail saying I'd now be seeing hotter people in matches etc I have noticed a distinct fall in attractiveness of profiles to me, whether looks or profile based.

2mattnewport15y

I'm pretty sure it's some kind of stunt since I got that email. I'm curious why you think the idea is 'scummy' though? Given how skewed first messages are towards physically attractive women (the same effect doesn't seem to exist for men) it seems like some kind of mechanism for balancing this effect would be useful. I've thought about a dating site where you have to 'pay' more (probably points of some kind rather than real money) to contact the most contacted people to try and reduce this effect but I'm not sure how you'd pitch it so as not to offend people. It would help counteract the trend for the hottest girls to be overwhelmed with messages and the average girls left receiving few messages though I think which could potentially be good for everyone.

2Relsqui15y

Not segregating people by hotness, emailing them to tell them so. It's the equivalent, by omission, of sending a bunch of their users a message saying "you're not pretty enough." It's a message which saturates our culture, and I'm not a fan, whether it's stated outright or not. This would solve the balance problem from a technical perspective but not a human one. If you set a site up this way, the value of an incoming message would be proportional to the cost of messaging you. If you're "cheap," an incoming message is just as likely to mean the messager couldn't afford anyone better as that someone's interested. If you're "expensive," every message means interest ... but you get fewer of them than you might elsewhere. Nobody wins. I'm definitely interested in better algorithms for matching people up, but I don't think that particular idea is viable.

2mattnewport15y

This is definitely a problem. I think you'd need to somewhat disguise what was going on so people didn't feel they were being 'priced'. I'm not sure you appreciate the dynamics of messaging on these sites. The hottest girls get vastly more messages than anyone else, more than they can possibly read and reply to. The problem they face is filtering out messages they might actually be interested in from the noise. For these users fewer messages is better, particularly if the messages are higher quality (which they will probably tend to be if they are expensive to send). Meanwhile less physically attractive girls and most guys get few or no unsolicited messages. The system should help increase the number of messages they receive. They may indeed be receiving messages from people who 'couldn't afford anyone better' but they are getting messages and chances are the messages they receive will be from more realistic matches. As wedrifid pointed out, people tend to end up in relationships with people of roughly equal attractiveness. All the average guys who send dozens of messages to the hottest 1% of girls who they have little hope of success with might consider messaging someone whose profile interests them but who is of more average physical attractiveness. I may not have made it clear that the idea would be that the recipients receive some or all of the cost of messaging them. This way the most in demand users would be able to 'afford' to message people of similar attractiveness but wouldn't be deluged with messages from people 'beneath' them who they likely have little interest in.

4Relsqui15y

Agreed. No, trust me, I know this part. :p But I see what you mean; my afterthought about how it would reduce their message quantity missed the point. I wasn't assuming that would be the case, although this would mean that hot people would mostly receive messages from other hot people. Which, according to your link, would be preferable for them. Interesting. Okay, I retract my immediate rejection, at least enough to admit that I'm curious about how this would pan out.

0[anonymous]15y

The post I linked to was purely about the camera and how a picture should be taken, not what a picture should be taken of. I doubt that we are sufficiently weird that we like grainy, poorly lit, low resolution images. But yes, I'm still not quite sure how we should give advice in general, not knowing what the other people are looking for.

2Relsqui15y

True, but they have another one about the other thing. Hmm. I've sometimes made a point of trading profile evaluations with people who know me well in person--the idea being "does this fit what you know of me?" This isn't directly useful for our purposes, but I wonder if that's actually a good criterion, and if so, what we can learn from it. I suppose what that method checks for is for impressions we get from the way people write things which give a significantly different idea than they intend. Even little things like the same word repeated in nearby sentences can convey thoughlessness or inarticulacy. Or the combination of different profile pieces, written at different times, might suggest a strong emphasis on something which isn't that important to the subject. What if we went through each others' profiles, line by line, and verbosely described the impressions and implications we read from it? In other words, explicitly state the details we're filling in between the lines. Might that be a productive form of profile critique between strangers?

2wedrifid15y

That's a really good idea.

1[anonymous]15y

Possibly. Extreme diplomacy needed though, you would have to be careful that your negative associations with specific traits didn't leak too much into the language used. I'll give it a go. 1) You give the impression you are mainly interested in game design (both in the what you are doing and message me if). Would you also like talking about the design of other non-game things? 2) Your profile doesn't indicate an interest in the real world (MMO + guitar + books). Are you interested in societal problems and their solutions (or lack thereof)?

0Relsqui15y

Thanks, that's exactly the kind of thing I was looking for--this should be more clearly on a back burner, since it's not a primary thing. I was thinking of rewriting the "message me if" soon anyway, so that'll probably go away. This is a good example of your point about negative associations--using "real world" to mean "the big picture/humanity as a whole" instead of just, well, occurring outside of a virtual reality. So I read that as "you have no life," which would offend me if I didn't think it was a misunderstanding. Some of them. Some I find interesting problems in their own right, but in general I don't spend cycles on things I don't feel I have any power over. Which charity is most efficient is not really an issue you worry about when 10% of your income is still approximately $0 (and I'm not swimming in free time to donate, either). With luck, I might in some years be settled enough to have the luxury of addressing the big picture. In the meantime, it doesn't seem productive to spin my wheels about it.

2[anonymous]15y

Will try harder in the future :) I'm probably odd in thinking that parties/clubbing are much forms of escapism as computer games. Neither are bad inherently either, a little escapism is a fine thing. Even if you don't have power over them, knowing what is going on big picture wise can be useful: what education to pick (growth/shrinking industries), where to live (do you expect oil prices to go up significantly in the near future, if so then minimize commute time) etc . If these things aren't interesting to you and you wouldn't want to have conversations about them, then it doesn't matter.

1Relsqui15y

Yes and no. I see your point, but I think being social and developing those skills is healthy and productive. Some video games can also teach you skills that are broadly applicable, but I don't know of many. I was so unwilling to acquire an education in something not personally important to me that I put off going to college for about six years, until I knew what it was I was interested enough in to choose as a specialty. So, while I can see why someone might choose an education based on industry, it's not really for me. I don't have either a car or a driver's license. This might conceivably change at some point, but I haven't really needed them yet, and I don't really expect to get into a position where I do any time soon. Don't get me wrong. I do read some current events, and when I get curious I'll go study a bunch of history to put it into context. (A while ago I spent three days on Wikipedia reading about every major conflict in the middle east since Israel's statehood, because I was sick of only having a general "there is conflict there" sense without understanding why.) But doing so is not a major part of my life.

3Wei Dai15y

The author of that blog seems to give advice based purely on statistical correlation without making any effort to distinguish between correlation and causation. See this post for example.

0[anonymous]15y

Well the author of the blog is trying to get people interested in the website, so it should be viewed as marketing more than advice. Advice just sells better than pure stats. It is still the only source of statistics on what goes on behind the scenes of okcupid, so it should provide some useful information, if you are interested in mainstream dating. It is also interesting purely from a sociological point of view, which races message the other races. It also provides the people on okcupid a common topic of conversation, if they are moderately interested in statistics.

3Wei Dai15y

No need to be so defensive. I'm not saying "don't read that blog". I'm just annoyed that the author doesn't do a better job analyzing the data (that presumably only he has access to).

3[anonymous]15y

Hmm, wasn't trying to be defensive. It was a valid criticism of the site. I just was worried that I had given the impression that I thought it was a place for advice! Relsqui kind of thought that, so I thought it needed a disclaimer. I wished to give the reasons that I had recommended it under, to dispell any lingering impression that I might be advising people to talk about zeus by proxy. I prefer to look at it from a half full perspective. Most other web businesses don't release any analyses at all.

1Wei Dai15y

Ok, got it. Thanks for the clarification.

8[anonymous]15y

Oh, why not. http://www.okcupid.com/profile/sadielou13

6Relsqui15y

I moused over your photo thumbnails to see what whpearson was talking about and thought, "Man, who does that smile remind me of? Oh, it's Kaylee! That's a Kaylee smile." And then I read the actual first line of your profile and cracked up.

3thomblake15y

I didn't realize there were Lwers in the New Haven area; I should've expected that to change once school started up again. Tom McCabe is nearby too, I reckon. There should be a meetup, but I'm not allowed to organize those anymore.

By the power vested in me by the cult of CEV, I hereby grant you permission to organize LW meetups.

2Will_Newsome15y

Relative LW newcomer (I think) Will_Sawin should be where Tom McCabe is, I think, if Tom is at Yale and Yale is near New Haven.

2Alicorn15y

Why aren't you allowed?

1thomblake15y

I totally botched the one at the H+ summit that I had taken responsibility for, so I declared that I am no longer allowed to organize those.

9Scott Alexander15y

Don't feel bad. You can't possibly be worse than me. I accidentally arranged a London meet-up with two days notice on Easter in a bookstore that was closed at the time (and Tomas, Richard, and Ciphergoth showed up anyway, to their great credit).

2wedrifid15y

Meanwhile everyone else was busy at church celebrating the rebirth of Christ? At least it isn't your religiously significant tradition you clashed with. :)

0Richard_Kennaway15y

The Starbucks along the road was open, so we celebrated the global capitalists and their rascally way of making money by selling people what they want to buy.

0jacob_cannell15y

Wait, so how was that bad?

2jacob_cannell15y

Unsolicited advice: Your third pic with the red dress is the best - but only 3? You could use a few more. Aspartame is evil! Ahh you and alicorn make me feel old. I would never have guessed so many people where were under 25.

2[anonymous]15y

I prefer the third picture to the first, although that may just be the clothing colours and fewer jpeg artifacts. You might want to crop the third to remove the plastic cups and extraneous people. Other than that, I think you have a good mix of factual info to quirkiness in your profile. You might think about re-ordering your categories of interests in terms of what is most important to you. If people are scanning then they might switch off after reading something that isn't so important (movies?) before reading something that is.

1Interpolate15y

I would make your third picture your display picture. I don't have many specific names/titles listed in the favourites section, but I found that I received more messages when I did.

1wedrifid15y

Things you couldn't do without: Aspartame. Nice. :) I seem to recall someone laying into superstition surrounding aspartame risks here a while back. Including rather adamant skepticism about a friend's belief that it was the cause of her difficulties. Was that you or do we have another fan of artificial sweetener? :)

4komponisto15y

You were thinking of LauraABJ.

0wedrifid15y

Ahh, yes, that's the one.

2[anonymous]15y

No, I'm a different person. I've never researched the health risks specifically, but I'm generally skeptical of the mindset that "we're not healthy because of all the chemicals we put in our bodies!" I'm also in favor of taking small risks that greatly increase one's quality of life. I hadn't thought about it much, but I guess aspartame says a lot about me.

0[anonymous]15y

No, I'm a different person. I've never researched the health risks specifically, but I'm generally skeptical of the mindset that "we're not healthy because of all the chemicals we put in our bodies!" I'm also in favor of taking small risks that greatly increase one's quality of life. I hadn't thought about it much, but I guess aspartame says a lot about me.

1Kevin15y

Reiterating advice I gave someone else -- you would look better with less blurry photos.

1komponisto15y

Out of curiosity, why do you consider Schubert lieder to be controversial?

0[anonymous]15y

People either love 'em or hate 'em. I have known folks who absolutely can't get through one song.

0komponisto15y

Interesting. Living as I do in a Second-Viennese-School-and-beyond sort of world, I guess I'm just not used to thinking of Schubert as a "controversial" composer.

0Vladimir_M15y

The third picture looks like you're standing at attention. I'd recommend a more relaxed-looking pose!

2[anonymous]15y

Oh god. I have heard several picture criticisms here and from people I know personally. The fact is that I really don't photograph well. It's hard enough to find pictures where I have a nice expression and don't look fat. Good posture and good camera quality -- not gonna happen.

4[anonymous]15y

I'd message you on the strength of those pictures/profile (assuming I was in the right location/age group). So I wouldn't get too down-hearted about people criticising them. This is a thread to suggest improvements, so we are looking for what could be improved.

0Vladimir_M15y

The picture is funny in a cute way. It looks like you had a severe case of that semi-conscious reflex to assume a "proper" position when someone points a camera at you, so you ended up striking a pose straight from the drill manual. :-) Getting rid of that misleading instinct is a good first step towards making oneself more photogenic. In any case, look at it as a cost-benefit question. How much do you care about this particular mode of public self-presentation, and how much effort for procuring good photos does this justify? (The easiest way to get some good photos is to hang around those irritating people who relentlessly snap their digital cameras next time you go to a party, and then just ask them to email you the pictures and make a selection out of the huge pile. The best way is to ask someone who has some basic photography skills to snap a few pictures of you.)

3Relsqui15y

Tacking on some more photo advice: When I was taking my senior formals for my high school yearbook, the photographer kept egging me on to smile more. It got really annoying. I WAS smiling! Finally, just to make him knock it off, I plastered on a giant stupid really fake grin. He said "perfect!" and shot it. They came out fantastic.

0Alicorn15y

Hey, you live near me! Neat!

0[anonymous]15y

I didn't know that! Let's meet up some time.

0Alicorn15y

I PMed you my contact info so we can set this up offsite; did you get it?

0[anonymous]15y

yep.

6Relsqui15y

Observation about online dating which didn't fit anywhere specific under this thread: it's a very strange sensation to be simultaneously aware of having specific preferences about unimportant traits (height, baldness) and that they're irrational. I've noticed that if I'm looking at the profile of a 32-year-old man who is balding, I feel like he's too old for me; another 32-year-old who is not will not give me that feeling. This annoys me a little. I don't care about baldness, but apparently some part of me does.

6Relsqui15y

Are you intending to make a top-level about this thread? I think there's some really interesting stuff in it: profile optimization techniques, whether and how you can glean advice from a statistical analysis of other peoples' results, and non-dating applications of learning to write a good profile (e.g. self-knowledge). I'd be interested in trying to distill the ideas into a useful post, but it's your thread, so I consider you to have right of refusal. Also, to you or anyone else: agree/disagree that this subject merits a top-level?

4Will_Newsome15y

I haven't really followed the thread at all, but I grant permission to anyone to do what they want with it; it's their karma. I have a few LW posts that I'm already busy with for the next 5 days or so. I believe you haven't written a post yet, Relsqui? I think you should do it. If you want, write a draft and set up a piratepad and link to it from here so that other people can contribute and edit.

4Relsqui15y

May well do; thanks. I have two more tests to study for this week (one down already), so, uh, it won't be today. But I'll see if I have time later in the week to assemble comments. I've seen a few collaborative text sites mentioned here--etherpad, and now piratepad. Anything particular I should know about choosing between them, or just try 'em and see what I like?

2Alicorn15y

Piratepad is just an instantiation of the Etherpad software.

1erratio15y

There's also the shiny new discussion area just under the site header /points to link

0Relsqui15y

Oh, yeah! Good point.

1Relsqui15y

Posted in discussion. If it gets a positive response, I'll shunt it to the main.

5WrongBot15y

My profile. I met my first girlfriend via OKC, but that was ~5 years ago. Haven't had much success there lately, but I've effectively become much more selective since then (due to polyamory's relative rarity), so that isn't too surprising. Very curious to hear what people think about how I've presented myself, as I'm dubious that my learned skills in that domain have transferred to internet profile writing.

0knb15y

Well, you've got preselection going for you, I guess.

4Scott Alexander15y

Ah, sure. People (especially women), give me what help you can: http://www.okcupid.com/profile/ScottAlexander

7Interpolate15y

I get the sense that your profile content doesn't do you justice - perhaps you could afford to be more arrogant? No one you want to meet would find you boring. I like most of your pictures, but I would include a few where you look more friendly and approachable, eg. pictures of you at work.

7Relsqui15y

That's a brilliant piece of advice about the attitude to take when profile-writing. Don't worry about looking good to people you aren't interested in! It doesn't matter what they think! Given the potential objections below about Yvain's work photos, a photo of him socializing might also fit the bill. Nothing says "I'm approachable" like "Look! These people clearly approached me, and I didn't bite them."

4Alicorn15y

Yvain is a med student. Pictures of him at work might have him wearing a mask and up to his elbows in blood. Which would be interesting, but not approachable ;)

1Interpolate15y

I realise he is a med student, which is why I suggested "at work". Maybe this is a personal quirk, but people in surgical scrubs exude compassion and approachibility to me. Conversely, pictures of people at work in an office setting usually seem impersonal and/or trite to me.

1Alicorn15y

My dad (a doctor, although he doesn't practice as a physician and does consulting instead) wears scrubs as pajamas, so I don't trust my intuitions about what they mean one bit.

7Relsqui15y

1) I don't want to alarm you, but there's a tiger next to you. 2) I second the comments about your pictures. Get someone with a real camera to take some for you, smile when they do it, and then crop/balance them. 3) You made me giggle a few times; points for that. 4) I do too message men! 5) You get to goad the reader into asking you out and still have it sound like a joke exactly once. The second time it's desperate. (You do it once in the "message me if" and once in a photo caption.) Where in CA are you from? :)

0Scott Alexander15y

2) I am mostly using those pictures to signal that I do interesting things; generally, the more interesting the thing I am doing, the less likely I have a good photo of it (eg when trekking in mountains I became unshaven and poorly groomed and hideous and can only take pictures by holding the camera in front of my face). I'm not really sure how to fix that. Should I get rid of all my interesting pictures? 5) Meh, good point. I'm from Orange County, but haven't lived there in a while.

6Relsqui15y

Nah, you could just supplement them with tidier ones.

4gwern15y

Perhaps I am alone in this, but the 2nd/middle picture looks terrible - as if you hadn't washed or shaved your head in a week.

0[anonymous]15y

I figured that signaling that I do interesting things was worth me almost always being poorly groomed when I do them, and that people would generally understand that grooming takes second place when you're trekking. Do you think not?

2Kevin15y

I think the blurriness of your pictures makes you look less attractive. There was an OK Trends blog about this: http://blog.okcupid.com/index.php/dont-be-ugly-by-accident/

1[anonymous]15y

I like it. I like the humor, I like the varied skills and interests.

0pwno15y

Really like your "you should message me if" section. I would word it a bit differently, but definitely intriguing.

0arundelo15y

So the picture with the tiger is really you!

3[anonymous]15y

3Relsqui15y

I'm willing to help cross this divide and check out some dudes or ladies-into-ladies on OKC. I think having seen a lot of profiles with the same errors over and over is a bigger asset than the ovaries, though. Also, unless we really have no criteria other than "is male" or "is female" (and we deserve to have more precise standards than that), it's going to be hard to distinguish between "your profile doesn't attract me" and "you don't attract me." I'm not sure how to account for, well, taste. To clarify with an example, I'm happy to give that ball a kick and put my profile out there to be critiqued. But there are some qualities that are important to me which I think are underrepresented among LW's users (like having good interpersonal/social skills), and some that I'd expect to be common here which I'd prefer to avoid in a partner (like dedicating an above-average proportion of time to academic or professional pursuits). So if I find out that my profile isn't appealing to people who don't fit those preferences, is that a good or a bad thing?

2CronoDAS15y

I don't think the "head back" picture is a good pose.

0Relsqui15y

I don't think it's great either, but it's the only photo I have of myself that isn't actively misleading about my appearance. And even that one's getting out of date. (A little bit of hair growth is a big difference when it's very short.) What I actually need to do is take some new pictures. :) Or shave my head again, which I'm considering.

3ata15y

Here's mine. I like how I match over 90% (and usually 96%-98%) with almost everyone else here.

3luminosity15y

97%. But I guess that's really not too surprising. One of the things I like about okcupid is how it's easy to make the match rating act as a warning. Make it ignore choices that differ form yours but you don't feel are important, and put high value on the ones that do. I can usually tell at a glance if someone is religious or highly socially conservative.

2ata15y

If anyone's interested in giving advice on the "actually looking attractive" aspect of OkCupid profile optimization: I am wondering what to do about facial hair. This is what it currently looks like, and I don't like it very much at all. Possible options include going back to something like this, or aborting Operation Facial Hair altogether. (My Best Face ranked the first aforelinked photo as my worst, and the other one as my best (which surprised me a bit). That is some evidence about what people prefer, but not very strong evidence, because there are several other differences between the photos that could account for some of the variation.)

1Alicorn15y

I vote for abandoning facial hair. And your photo with the hat is totally the best.

0Relsqui15y

I prefer the second one, but I tend to prefer beards in general.

0WrongBot15y

You have the same first name and initials (I'm assuming) as an ex-boyfriend of mine. You're also about the same age, live in the same general area, and look quite a bit like him. So that was slightly startling. Also, yeah, 96% match rating. There seems to be an LW value cluster that's packed pretty tightly, relative to general personspace.

0wedrifid15y

Yup 96% here. :)

3[anonymous]15y

It is also worth noting that making a good profile is not the only thing you can do. If you socialise on the site, you get more exposure in general and people might see you if they are browsing the other people's journals or looking at the recent activity section.

3Will_Newsome15y

Autothexis at OKC. Critiques are welcome. Profile's been pretty successful. I think the main limiting factor is my age, but even then I've managed to hook up with a few cute girls from Cal without putting in too much effort.

0Relsqui15y

40% match. Wow. Picture me backing away slowly.

3Will_Newsome15y

Haha I noticed that. It's rather odd. A girl I've been dating for the last few months and the one I like the most has about a 40% match with me and it's incredibly hard to see why. I guess I should directly compare our questions or something. It was at about 80% until they switched up the Matching mechanism: I have no idea what they did, but I think it was for the worse. Maybe not though.

2[anonymous]13y

You are probably using it wrong.

3wedrifid15y

I'm bayesian_prior. The profile I have there is successful but I haven't looked at the text I wrote there in over a year. Critiques are welcome.

3Nisan15y

You may want to expand the sections "The first things people usually notice about me" and "The six things I could never do without". The former is a chance to provide a random detail about yourself. As for the latter, "eudaimonia" is a great answer, but if I didn't know the transhumanist connotation and looked it up on Wikipedia, I'd conclude that you must be a fan of virtue ethics :) ETA: Oh, and a while back OKCupid got rid of the "I am _, __, and _" section, and incorporated the text into your self-summary. You may or may not like that.

0wedrifid15y

Thanks. Virtue ethics? Gah! Must change that.

2Matt_Simpson15y

I never thought I'd say that LW caused me to join OKCupid. Here's my profile: http://www.okcupid.com/profile/TheMattSimpson Comments? Suggestions? It's probably obvious that I'm mainly interested in hearing what the LW ladies have to say, but if you men know of such things I'm all ears.

2[anonymous]15y

Ok, want to ask advice about something ... touchy. Like most girls' profiles, mine says nothing whatsoever about sex, to avoid attracting armies of pervs. But sexual incompatibility actually can be a deal-breaker for me, and I'd imagine I'm not alone in that. I don't have anything complicated going on -- I just think that sex is nifty, and I want some way to avoid winding up with people more prudish than myself. Is there a delicate way to hint at that?

6jimrandomh15y

Just say "No prudes". Why make it more complicated?

5wedrifid15y

Nice double entendre Including the phrase 'sexually compatible' seems to send the right signal without being excessively crude. You can also use the question system to filter this kind of thing. There are questions that explicitly handle these sort of preferences and also the ability to select a set of questions as mandatory. Apart from that you can do a search through all the sex questions and answer every one of them. That is what I did when I lost the 'more desiring of sex' picture on my profile and wanted to reclaim it. This will (obviously) lower the match ratings of prudes and also allow you to see [comparisons on various relevant criteria[(http://www.okcupid.com/profile/sadielou13/compare/bayesian_prior). Even if you haven't explicitly mentioned anything about prudishness you can make a reasonable inference about relative prudishness of potential matches by looking at a get a good indication of that from looking at "kinkier", "more old fashioned", "more desiring of sex" and "more moral". (I just noticed that my rating puts me at +31% on 'pure'. WTF? I must be confused about what purity means since I could have sworn I was no such thing!)

3Will_Newsome15y

The girl I'm primarily dating put sex as one of the six things she couldn't do without. That didn't put me off. It also screens off crazy social conservatives.

2luminosity15y

I've seen plenty of women's profiles that mention sex. I imagine if armies of pervs were a terrible problem that probably wouldn't happen as much? If you don't want to risk that though, my recommendation would be to leave it off entirely, deal with it in messaging or in person. It seems to me that if you don't want to state it outright, you're likely to be misinterpreted.

2[anonymous]15y

I'd keep it impersonal and put mentions of it with words that aren't associated with sex normally or are negatively associated with sex. "Sex is nifty" is good in that it doesn't associate you with the sex, so doesn't illicit mental images. Nifty is also an odd word to associate with sex. "I don't have a religious view on sex" is also unlikely to get the pervs going. Both of these are a bit boring through. Failing that, to keep it more light hearted you might want to try adding in some sexual double entendres? Hard to balance the right level of subtlety and understandability though.

1Relsqui15y

I agree with wedrifed; answering questions is a good way to express this and match on it without putting it up front in your profile. By the same token, you can sort the public answers of a potential match to see if you're compatible in that way.

1Duke15y

Somehow I doubt that the part about needing a guy to have lots of sex with you is worth mentioning.

2Kevin15y

www.okcupid.com/profile/kfischer

1cousin_it15y

Late to the party! I'm cousin_it there too. Visible only to logged-in users.

1fburnaby15y

Looks like I'm very late to the party. Just in case, I'm rock n stroll .

1JoshuaZ15y

This is interesting to read as someone who is not on OkCupid. Simply reading this thread makes me more inclined to sign up under the basis that input from LW is much more likely to make such a profile useful and less likely to make my profile simply be further damage to everyone's signal to noise ratio. Must think about this.

3JoshuaZ15y

Ok. Replying to self to keep things organized. Have made a profile. Some of the match questions are really poorly phrased. For example, consider the question "Have you ever had a sexual encounter with someone of the same sex?" which has four possible answers: * 1 Yes, and I enjoyed myself. * 2 Yes, and I did not enjoy myself. * 3 No, and I would never. * 4 No, but I would like to. This neglects some fairly obvious other options such as "no, and indifferent." Many other questions had similar problems. The question about contraception was confusing since it wasn't clear whether abortion was considered contraception for this purpose. I assumed that it was not but I could see someone interpreting it as referring to any form of birth control.

4mattnewport15y

This is because they are largely user submitted and not actively filtered by OkCupid staff.

7AdeleneDawner15y

It doesn't help that they're limited to four answers of relatively short length, either.

3Alicorn15y

I'd change "religion" in the sidebar to agnosticism or atheism. If people are sorting for religion (which is most of what that affects - anyone who scrolls enough to look it on your actual profile page can also see the essays about your Jewish cultural background), this will screen off anyone who's looking for agnostics/atheists and let through anyone who's looking for religious Jews. Basically a solid profile; if you were in my age range and lived nearby, I might have pinged you.

0JoshuaZ15y

Thanks. Updating per your advice.

0JoshuaZ15y

Hmm, so any chance I could get feedback on this profile?

1MBlume15y

I'm Sgt. Pepper I've had precisely one positive outcome on OKC, and that was a friend who led me to cool activities, not a hookup. Still try from time to time though.

1[anonymous]15y

My profile. I initially joined OkCupid for the tests and stayed for the journals community, so it isn't really optimized for attracting dates/matches. I have made a few valuable acquaintances through it, however. As an aside: Why haven't more people mentioned LW in their profiles? ETA: Despite my not using OkC primarily for its intended purpose, critiques are very welcome.

2knb15y

Drop the Myers-Brigg type indicator (very last season), and replace it with the hip and stylish Five factor model. Or your profile will just end up attracting a bunch of Jung-fanboys who want to talk about their dream journals. :3

1Nisan15y

Okay, here's mine. I think I can offer advice to both men and women here; a big part of writing a good profile is simply writing well.

3[anonymous]15y

I really like this; good pictures, and, yes, good writing.

0Nisan15y

Thanks!

1Perplexed15y

I created a profile just for purposes of this thread. Don't know if I will finish and maintain it. I'm not really looking for someone right now. But feel free to critique as if I were.

2Relsqui15y

Your six things made me laugh. But your "private thing" answer annoys me--conveniently, I have already written out a rant about why.

0NancyLebovitz15y

The word you're looking for is probably xenophilia.

1mattnewport15y

I deleted mine. Too depressing.

0mattnewport15y

This discussion prompted me to give OkCupid another try. Profile's still a work in progress, haven't filled out the books/movies etc. section yet.

4[anonymous]15y

I'm going to assume you are interested in a wider range of women than the average (on lesswrong) and point out a profile that I think is successful, and might be useful to you. If you are specifically looking for gender atypical women, then probably ignore this. Take this guy. Things he does in his profile worth noting (but not necessarily copying) 1) Implies he gets lots messages 2) Implies that people might want to stalk him 3) Shows he can organise events IRL * If you look at his journal, lots of people are talking to him, lots of girls are flirting with him. Those sorts of things are a lot harder to engineer than an attractive looking profile, so are good demonstrations of value. You stand out from the crowd. I've met him in real life and he does seem successful with women. He is easy going and friendly, but intensely cocky and flirty. *Going to an okcupid meetup is a good way of increasing your exposure on the website as well. You will be featured in diaries + pictures and if you make a good impression you'll get compliments and people to follow and comment on your journals. It can be grey hat if that is the only reason you do it though...

0mattnewport15y

I think that's probably accurate. I can see why the profile you link is successful. It's funny and well written. It also signals high status without being too direct or explicit about it. I'm not confident in my ability to pull off the same degree of funny over a similar amount of text however. I tend towards more brevity because I find I get bored of most long profiles (this one mostly avoided that) and I assume most women have the same reaction but I'm not sure if that's a mind projection fallacy. Humour does seem effective in profiles but one of the reasons its a good signal is it is hard to fake and I'm always wary of trying and failing, more so in writing than in person.

3Relsqui15y

If you figure this out, tell me. Your food section is making me hungry ... but pickled onion monster what? Your photos are quite good, and the captions made me giggle. I can see why you picked the one you did for the top picture, but I really like the smile in the last one. Smiles are very attractive. For what it's worth, I think your profile could be a little longer, but the length of the one whpearson linked is excessive. Information content > brevity, but clarity > length. I agree with this. I find humor to be a mild attraction boost, but obviously deliberate attempts at humor which flop are a pretty big turnoff. (For the record, being funny constantly, even successfully, also is--I need to know someone is capable of taking anything seriously.) This is the problem with a lot of dating/profile advice which observes that people who are __ get lots of dates/messages. It's easy to say that, looking at the data, but much harder to squeeze any kind of practical advice out of it, since "be __" usually isn't. That said, if you do add more content to your profile and want an opinion on e.g. whether the humor works, I'd be happy to give it another eye.

0mattnewport15y

Monster Munch. It's kind of an in joke for anyone who happens to have lived in England - it's a well known brand there but pretty much unheard of anywhere else. For anyone who gets the reference the juxtaposition of a list of healthy unprocessed foods with a completely artificial processed snack might raise a smile. Thanks. The top one tested best on My Best Face on my old (now deleted) profile. I haven't compared it against the bottom one which is pretty recent. Smiles being good seems to be the conventional advice but OkTrends suggested otherwise in one post. I'll rerun the My Best Face test with the newer pictures included. I might take you up on that. I'll probably fill out the empty sections and refine what's already there over the next few days.

0Relsqui15y

As noted elsewhere, I don't put a lot of stock in their recommendations, but I admit I'm only the one data point.

0mindviews15y

I'll join in the fun - any suggestions appreciated. My profile is currently limited to OKC users, though. I wish there were more LW ladies in SoCal who were easier to find...

4Alicorn15y

I don't recommend leaving this as a surprise. The question it prompts is "Why? What's wrong with his taste in books? Is he embarrassed or something?" And while these questions might go into a message to you from the girl of your dreams, they might also send her scurrying away.

2mindviews15y

I tried something different and added a link to this section. Any comments on how that works?

2Alicorn15y

This is better.

0JGWeissman15y

Here is mine Like several others here, I am looking for someone who at least isn't turned away by my transhuman values. My profile has been moderately successful in generating dates, but not much long term.

6RHollerith15y

I would leave all that out. You already said you work as a software developer and that you have a technical degree. If a female programmer wants to know more, she can ask. The convention in these profiles is to be conversational. If you were introducing yourself to a stranger in person, and the fifth and sixth sentences out of your mouth were the above, most women would peg you as a hopeless nerd. The probable reason this section is titled "What I'm doing with my life," is that most profile writers need to be nudged into listing actual accomplishments, but you've already established yourself as an accomplished person in the previous section, so maybe use this section to describe your hopes for the future and what you've been doing to improve yourself. I understand that you want to discourage readers unaccepting of transhumanism, but maybe spell out how your transhumanism would affect your future partner: "Any woman that gets involved with me would have to understand that a part of my income will go to philanthropic ends -- and unusual philanthropic ends which are ridiculed by some at that! Also, I have arranged for my body to be cryonically preserved on my death, and I would prefer for my partner to assist me in that goal. At the very least, she would need to promise that she will not interfere with my arrangements." Oops. Did not mean to write so much. I hope this is not unwelcome advice?

[-][anonymous]15y110

"Also, I have arranged for my body to be cryonically preserved on my death, and I would prefer for my partner to assist me in that goal. At the very least, she would need to promise that she will not interfere with my arrangements.

I would actually recommend against saying that, even though it may be factually true. I would not attempt to dissuade my man from cryonics if he were interested, but if I saw this in an OkCupid profile, I would think, "This guy is ALL ABOUT transhumanism and probably boring." Seeing as I'm more used to the idea than most women, what turns me off will probably turn off lots more potentially compatible readers.

Edit: I agree with the above that you don't need to go into so much detail about the programming though.

1JGWeissman15y

Done. I agree with Molybdenumblue about the transhumanist stuff though.

0RHollerith15y

I do too on reflection.

0luminosity15y

Why not? Mine

0erratio15y

If you ever move to Sydney I would be highly interested in meeting you. So I would have to say, looks like your profile is working :)

0luminosity15y

I guess I must be doing something right then. And I'll bear that in mind should circumstances change and I end up in Sydney. :)

[-]datadataeverywhere15y200

I recently mentioned Dempster-Shafer evidentiary theory another thread. I admit, I was surprised to get no replies to that comment.

Is there any interest in a top-level post introducing DST? Bayesianism seems to be cited as a god of reasoning around here, but DST is strictly more powerful, since setting uncertainty to 0 in all DST formulas results in answers identical to what Bayesian probability would give. I would like to introduce Dempster-Shafer here, but only if the audience will find it worthwhile.

3Will_Newsome15y

I would very much like an introductory post. Hopefully LW's best statisticians would reply and a good discussion would get going. As far as human epistemic rationality is concerned, though, does it ever help to try to 'move in harmony with the DST' the way it seems to help when people try to 'move in harmony with the Bayes' by e.g. making sure to take into account prior probabilities or the like? Are there any qualitative applications of DST to one's beliefs that are more powerful or more elegant than naive Bayesian heuristics? Also, it is just a confusion to ask what the equivalent of Bayesian networks are for DST? I'm definitely not a statistician but the application of verbalized statistical heuristics has helped me a lot in the past. It'd be cool if you could cover these questions in your post, but if not I still think you should write it. Pleaaaaaase.

3gwern15y

I'd like to read about it, but less about the details than why we should prefer it to Bayesianism. For example, how would proponents deal with Cox's theorem?

2ata15y

I'd be interested in a top-level post — I had been meaning to ask about it, in fact. I've read about DST briefly and had been wondering (1) if in practice it is actually useful for anything you can't do as easily with pure Bayes, and (2) if they are theoretically isomorphic or if Bayes is strictly a special case of DST.

7datadataeverywhere15y

In my opinion, using DST usually adds unnecessary complexity to problems that can be sufficiently solved in a Bayesian framework. Then again, I think that the same thing can often be said of descending from a Bayesian to a Frequentist approach, which is to say that most problems are simple, and properly using any framework is enough to get a good answer. See neq1's post that inspired my original comment. That said, I work on problems that I have solved both from a Bayesian perspective and from the perspective of DST, and I have found the former lacking. There are at least a few problems that I feel like DST is much better at. If you search Google Scholar for Dempster-Shafer and look at results in the past few years, you'll notice a really clear trend for using it to extract information from noisy sensor data. That's what I use it for, and seems to be a strength of DST. As to your second question, I think it is in the realm of possibility that Bayes can be used to construct DST, but I don't know how and if it is possible, it is certainly more difficult than going the other direction. In some sense, DST is meta-Bayesian, because PDFs of PDFs of priors can be specified, but doing that with a strictly Bayesian framework misses the set-theoretic nature of Dempster's Rule of Combination, and results in a weaker theory, that among other things, still doesn't handle contradictions any better than Bayes does.

2Perplexed15y

I would like to see it. I tried to get some idea of what it is about by Googling but failed miserably. I should disclose up front that seeing Zadeh's name associated with it did not give me a warm and fuzzy feeling of anticipation.

2datadataeverywhere15y

What?! Seeing Zadeh's name should definitely give you a "fuzzy" sensation! ;-P

0Perplexed15y

Usually I am conscious of my subconscious puns. Dada Dada everywear.

0lukeprog15y

Yes please!

1datadataeverywhere15y

I've been working on this, but have been swamped with other things. Among other things, I've reversed my position on ata's question about the equivalence of Bayesian probability and DST, and am reasonably sure that I can demonstrate that DST can be derived from, for instance, Jaynes' formulation of probability. For related reasons, I've grown to distrust most underlying formulations of DST, and while I still trust the method (it works, after all), I think standard presentations have gotten it Wrong. Which of course means that I feel the need to get it Less Wrong before I post it here, making this a much more monumental challenge.

1lukeprog15y

I appreciate your commitment to quality!

0satt15y

I certainly wouldn't mind. I saw your other thread and hoped someone else would reply to it and get a conversation rolling. I've heard a tiny bit about DST, and it sounds interesting, but I know too little about it to comment on it myself.

0AdeleneDawner15y

I was surprised that that didn't get any responses, too. I don't usually contribute to discussions of probability, and probably wouldn't comment on such a post, but I'd certainly be interested in reading it.

[-][anonymous]15y140

I'm considering writing a post on anapana meditation (sometimes also known as mindfulness meditation or breath meditation) that gives instructions, tips for practicing, and details costs and benefits. I would probably be drawing primarily from personal experience (but I'd still appreciate any references to relevant studies). At the end of the post I would invite others to use my knowledge to give this form of meditation a trial. Is something people are interested in? For those who are interested, is there anything you would like to see addressed?

My meditation history: I learned anapana and vipassana meditation from a 10-day course at one of these centers in 2006. After a year of false starts I was able to keep up a daily practice of 1 hr from 2007-2008 and 2 hrs from 2008 until the present. During that period I also took an additional 4 (or 5?) 10-day courses.

3Bobertron15y

I'd be interested in mindfulness, concentration and relaxation. How they relate to each other and anapana meditation. Should one practice something in addition to anapana meditation if one is interested in concentration or relaxation?

0[anonymous]15y

Thanks for your feedback. I'm planning on focusing on mindfulness and concentration, as well as the relationship between the two. I think I'll be able to address how those two are related to relaxation benefits. Sorry but I don't really have a good answer to your question. I haven't explored many other practices with this aim. Sometimes I find that trying to put all one's attention into simple repetitive tasks where one isn't personally invested (like doing dishes) is relaxing, and can help maintain concentration during the day. This can be a form of meditation somewhat akin to walking meditation. The benefit I get from this might be dependent on my daily sitting meditation practice though.

2[anonymous]15y

For those who are curious, I've decided to write the post and am currently working on a draft. I'll ask for feedback in a separate comment when I'm done.

0Will_Newsome15y

Sweeeeet. You might want to consider putting it in an etherpad somewhere for a quick review (I think piratepad might be the most common, I dunno). If you link to it from here you should get 2 to 10 people to look at your draft.

So, hello. Has anyone here ever experienced spontaneous and sudden evaporation of akrasia alltogether? I'm asking mostly because this is exactly what happened to me, on 26th of August this year.

I didn't do anything special. I had tried taking cold showers every now and then for a week earlier, and started taking some nutrient pills around that time, that's pretty much it. Then, that morning, I suddenly started working on the projects I had planned and thought of.

That may not sound all that dramatic, but I haven't introduced myself yet. I have been my whole life a rock-solid underachiever. After elementary school doing homework was not enforced, so I gradually stopped doing that. University doesn't care if you participate in lectures, so I didn't. All my academic effort happens roughly one day before any given exam. It's not that I didn't like the subject I study, or that I didn't want to do it. I just couldn't. There was a mental block that totally prevented me from using my free time for any of my projects, things I wanted to work on.

So that day, I spontaneously figured that I gotta study one thing in order to be prepared for the next academic year. So I did. I figured my room was... (read more)

4Morendil15y

The phenomenon of plateaus in learning is fairly common, and in Go specifically if you look at a graph of how a population of players are distributed in terms of ranks, it has several noticeable discontinuities. So the Go jump isn't too surprising. The one with akrasia is, because I wouldn't think of that as something you were learning.

2realitygrill15y

My particular jumps in Go strength came after reading more new theory, trying to integrate it into my game for a while (and failing), and then frustratedly taking a break. I'd come back a few kyu stronger, and games would consolidate this into maybe another stone or two's increase in strength. This happened enough that I stopped feeling frustrated (rather just stuck) and proactively took breaks. Usually about a week would do it.

1byrnema15y

After reading the post about Go, I started playing for a while ... and I was so bad so consistently I quit playing (out of discomfort with opponent discomfort, rather than frustration). Has anyone had the experience of playing several games with no improvement? On the last online game I played, the person I played against paused to ask me if I understood the rules. The next question was, 'do you understand that you're supposed to surround my stones?' or something like that. I know that I have trouble making any plans spanning several moves, and when I do make 'plans' I can't recall them after activity on another part of the board..

5Jonii15y

There is a proverb in go: "Lose your 100 first games as fast as possible", mostly because there is little use in making plans before you have some sort of intuition of what's going on on the board. So if you just play and lose without thinking too much about it, you get to the fun part faster.

0byrnema15y

Nice quote. My interest in playing Go has been rekindled, thank you, and I agree with Relsqui that it applies to many contexts.

0Relsqui15y

Thank you, I've been trying to remember the actual wording of that proverb. A Go player told it to me years ago and it's nice to remember whenever I'm new to, and sucking at, anything. "This is a necessary frustrating hard part; keep going and you'll move through it."

4Morendil15y

Sure, every Go player hits a plateau in their playing career, in fact more than one. As a proto-Go player (and I think of myself as a dabbler still), I used to have a specific computer program as my sole opponent: IgoWin, a tiny 9x9 Go program which I still think makes a pretty good introduction to the game. When you start out as a total beginner IgoWin gives you 5 handicap stones, which on a 9x9 board makes for a real easy games. Then, as is traditional, every time you win a game it takes away one handicap stone, until the games start to become challenging again. My first ever plateau in learning Go was when I got IgoWin to play me even, taking Black. As White I was totally unable to beat it. No matter how much I played (for weeks on end) I couldn't figure out how to get past that. After a while I broke down and asked a friend of mine who I knew was a player to teach me the basics. Two or three games with him on a 13x13 board later, I was able to clobber IgoWin as White in even games. I started playing human opponents exclusively, and later learned that bots like IgoWin are considered a bad way to learn, because Go playing programs so lack an understanding of the game that playing them will give you bad habits. So the idea is that there are plateaus, but there are also plateau-busters. For me, playing with a "real" opponent who knew what he was doing was one. Another was reading in-depth reviews of games at the GTL. Yet another was to do lots of tsumego from graded problems books. Also, reading important books like In the Beginning or Attack and Defence, which help make sense of the whole game and give you a conceptual framework in which to plan your moves.

0byrnema15y

Thank you. Since I'm embarassed about wasting the time of real people (though I've read I shouldn't be), I think playing a computer is for me. Even if I learn bad habits, it'll be a way to 'get through' my first 100 games. From playing one online computer game of Go, I find that I enjoy playing and losing to a computer because I know the computer is infinitely patient and doesn't analyze my mistakes or wonder what I'm thinking about for so long. So my question is; did you find that the experience you gained from the computer outweighed any bad habits you learned, or would it have been preferable to meet with your friend right at the beginning?

2Morendil15y

I definitely got over my bad habits pretty fast. I think I did, anyway.

0[anonymous]15y

My mistake was definitely using the full board.

0Douglas_Knight15y

My mistake was definitely using the full board.

0realitygrill15y

I would try playing as much as possible on KGS, using 9x9s and 13x13s in the beginner room. Just try to get a feel for (and understand) common patterns. People are quite friendly and willing to do reviews. Personally, I think 9x9s and Igowin are just boring, and don't teach you much. So try to get past this. Lots of the fun in Go is trying to implement new ideas. 9x9s are generally too constrained for this.

1Jonii15y

I have dedicated much of my life for overcoming akrasia, especially after I found LW I've been working hard(well, thinking hard and planning hard mostly), does that count as learning?

0realitygrill15y

More details? I may try to replicate this.

0Morendil15y

It might. The difference between that and Go seems intuitively large, but I can't properly articulate why.

3jimrandomh15y

This exactly mirrors my experience with correcting an unrecognized thiamine deficiency with sulbutimine. You had a micronutrient defiiciency (possibly but not necessarily the same one I had), and those pills you mentioned taking fixed it. Problems with brain biochemistry disguise themselves as psychological problems. This is a big deal and it needs more attention, because I think a lot of people are struggling with problems like this and don't know that fixing it is even possible.

1ata15y

Indeed. What's worse, a lot of them will spend years (and countless dollars) in therapy trying to talk their way out of their biochemical problems...

2ata15y

Huh. I could use a miracle like that. What were the pills you had recently started taking?

0Jonii15y

Sorry, I don't know. Omega 3, vitamins, stuff like that, but I threw the package away. It was some sort of combination box of all those supposedly important nutritients you need to take daily, with quite a lot of different stuff in it.

2wnoise15y

Are you going to be able to buy the same mix once this runs out?

0Jonii15y

Sure. I might comment here after I buy more of that.

2Jonii15y

Dunno if this is all that interesting story, but some people found it interesting, or at least inspiring, so I thought I should share. shrug

1Vladimir_Golovin15y

I experienced this several times. Last time I had this feeling was just a week ago. There were at least two more occasions, both during this year. And I'd very, very much like to know what causes this state, and how to induce it at will.

0[anonymous]15y

Congratulations! Not that I want to generalize from one example or anything, but what were the nutrient pills?

0Nisan15y

Thanks for sharing. I've had akrasia evaporate for short periods of time before, but the cause was always identifiable: * Taking THC once cured me of an internet addiction, and that solved my akrasia problem at the time. * Another time, my akrasia was due to problems in my social network. A positive change in my social network removed that problem.

0Relsqui15y

Are you comfortable elaborating on the problems in your social network? I've had some very frustrating issues which could be described the same way, so any related data is of interest.

0gwern15y

Abrupt changes due to chemistry are not unheard of. Maybe you could stop taking any and all vitamins/supplements to see what happens - if you dare.

0Jonii15y

I haven't been taking those all that much lately, once every 4 days or so for the last two weeks.

0gwern15y

Deficiencies can be small - chemicals, after all. If you aren't taking them all that much, then it's an even cheaper experiment.

0Jonii15y

It would take longer, which is my main concern. Gotta see if this remains stabile for a reasonable amount of time until I can go testing around. For the last few weeks, it's been far from stabile, I've remodeled my life just to be able to work more efficiently, and today is the first time I managed to actually work school-related stuff while being home, something I didn't think was possible few months back(Before this, when I absolutely had to do something, I had to leave my apartment in order to do it).

A request for help: I feel like I'm finally mastering my akrasia at work, but I have yet to find a technique to remove a pre-established Ugh Field. In this case, I have nearly complete drafts for two paper that I wrote as part of my Ph.D. thesis. I have a strong stress reaction to just thinking about opening the files (ETA: it's thinking about doing the work that causes the reaction; opening the files is just the first step in actually doing the work), but I want to want to whip them into shape and submit them for publication.

Less Wrong, other-optimize me!

[-]Vladimir_M15y160

Ask someone else to sit down together with you at the computer, open the files, and start reading and discussing them with you. Eventually, start editing them together. Tell your collaborator specifically to hang around for a while and disregard your (possible) requests to stop, until the work is well underway and you can continue with the flow.

This would of course require a significant commitment on part of the other person, but if this is really important, a good friend should be willing to help you, and you might even consider paying someone less close for their time and effort.

3wedrifid15y

Bonus incentive if they are hot and of your sex of preference. (This is an entirely serious suggestion.)

2Vladimir_M15y

That would indeed be a significant improvement over the basic scheme, but probably hard to pull off in practice.

3wedrifid15y

Really? I find it easier to pull off, given the colleagues I tend to build collaborative relationships with. In general it is a whole lot easier for me to work with girls than guys. Guys are more likely to compete, to try to force through bad ideas because they are being territorial. Control of the intellectual space is more important than getting stuff done, for obvious social and evolutionary reasons. On the other hand girls don't need to compete with me for the same social territory so a better balance of give and take can be reached.

3Vladimir_M15y

The comment was about this particular case -- since Cyan complained about having few options in general, I figured that it made no sense to propose this additional enhancement. But yes, what you write is generally correct. This is also one of the principal reasons why women are on average better adapted to modern workplaces of white-collar drudgery, and are thus doing increasingly better in today's economy. In addition, there is the basic fact that being surrounded by attractive people of the opposite sex creates a more pleasant environment, making one overall happier, more optimistic, and less prone to lethargy, especially for men. I've heard half-substantiated stories about companies that, under an informal policy, hire a certain number of attractive people who otherwise wouldn't pass muster, specifically to boost workplace morale.

0mindspillage15y

I needed this too. It did in fact require a significant commitment, so she became a coauthor, which was fine for my situation. I wrote a seminar paper that I couldn't stand by the time I was "done" with it--did some things well and others unimpressively--but the prof pushed me to publish it. (I'll skip the parts where I was making myself miserable about how terrible it was and the parts where I was running past deadline and agonizing that now it really needed to be good because I was late, so it became even later.) His research assistant was a friend of mine with complementary skills--I am terrible at details, cleaning up loose ends, making sure ideas aren't left half-finished and unattached to their surroundings; she is good at taking unpolished idea-dumps and turning them into real papers. So when I dreaded the thought of opening it, she did her first pass. (And, uh, then I dreaded opening it again, thinking "oh no, what if it's bad and I have to figure out how to steer it back without making her feel bad". But at least it was a different reason.) Mostly we exchanged emails and met every week or so, and we finished (and got a publication offer, whee) because I would have felt like a real jerk otherwise for dragging her into it and then flaking out. Also because my role then was mostly reacting to her changes and seeing what they suggested as next steps for me, rather than simply staring at a page I'd already been staring at unproductively.

0Cyan15y

The only potential helpers competent to discuss the contents live in different cities (except possibly XFrequentist, if he's willing).

5XFrequentist15y

Happy to if needed. I suffer from something similar on occasion, except my ugh field seems to manifest somewhat differently. I'd be very interested to hear about your progress on this.

0Cyan15y

Thanks! I'll PM you when I reach that point (I'm going to try the suggestion that directly target stress reduction first). (And BTW, congratulations on your engagement!)

2XFrequentist15y

No problem. Be sure to report the results! (Coincidentally, Julian just sent me a paper you're coauthor on... and thanks!)

5Vladimir_M15y

It doesn't have to involve an in-depth discussion of the content. The important thing is to get the editing underway so as to dispel the ugh-field and get into the flow. For this, it should (hopefully) be sufficient to start doing things where any smart scientifically literate person will be able to provide some feedback. For example, devising the best way to organize tables and charts, figuring out how to reword hard-to-parse sentences and paragraphs, etc. You can even make it into a fun exercise where your non-expert collaborator tries to figure things out from the draft while you explain the details that are assumed as background knowledge, and you fix or fill in the text as you go forward. The possibilities are many.

6Cyan15y

Oh wait, I just thought of someone else who is nearby and is competent to help. (In fact, I'm really dumb for not thinking of her right away.)

[-]datadataeverywhere15y100

If you need to whip them into shape, you're probably not happy with them. If you're anything like me, showing them to someone else is probably the last thing you want to do.

Solution: you owe me one of the papers by the next open thread. If you don't work on it, you'll be sending me whatever you're so embarrassed about. If you can fix it in time, you won't have to worry. I don't have much time right now, but I will read it, so beware.

1Cyan15y

I am actually happy with them. The required whipping into shape is editing of the material to the target journals' formats, styles, and expected audiences.

8Vladimir_Golovin15y

This looks like a severe burnout. Is it possible for you to take a month off? Have you tried 80/20ing the drafts? What's the most difficult task that you need to perform to complete the papers? Does it require hard mental work, or it's just formatting / proofreading / editing / rewriting / reviewing sort of thing? Or perhaps the task itself isn't what paralyzes you, but you're afraid of some submission / approval process that lies ahead? This has often been the case for me when I dealt with submissions of important work to human reviewers. I'd suggest a written self-interview to figure out what really causes your reaction.

0Cyan15y

It's the lingering remnants of the severe burnout I was suffering around this time last year. The task is indeed just formatting etc. (See here.)

7wedrifid15y

Three options: * Acquire the Sedona Method from a suitable source. It is particularly useful for 'releasing' that sort of stress reaction. (An audio form preferably - a text version is too much like work!) * Read comments by pjeby, his approach includes rewiring the underlying associations that lead to the aversion. Hopefully PJ himself is following the comments at the moment! * Don't try to work. Go and sit in a chair and think "I am writing my paper" to yourself over and over. Now here is the important part - you do NOT use the build up of willpower you get to go and force yourself to work. You hold yourself back from any attempt to make yourself work and just keep relaxing and keep thinking "I am writing my paper". You only allow yourself to go and work when you really, really want to. If this means you spend two hours relaxing instead of working then that's good too. This should be instinctively associated with productive self nurturing rather than the shame of procrastination. That final approach would probably be better described in a post than a dense bullet point but it does work for me. In fact I'm planning to go think to myself "I am putting the entirety of my semester's work into supermemo".

5MartinB15y

My suggestions: compartmentalize setting up the work environment - here: put notes on table, open file, etc. - and the actual work. Basically you set up everything you need, then get a tea and then start work. That way the setup is not perceived as real work. Against ugh a timer might work. Commit yourself to work on ONE of the papers for 30 min, then a break, and if you are in flow then another round. But first just spend the whole 30 min on the paper. Regardless of what comes out of it. Overcoming the startup hump is sometimes enough to get going. If your editing process is more complicated, then write the steps up upfront, and do them one after the other in blocks of time(s)

8Vladimir_Golovin15y

I strongly support this suggestion. The setup phase can be generalized as removing trivial inconveniences and creating trivial impetuses. I often separate this stage from the actual work, sometimes with an explicit commitment not to work before the setup is done.

5erratio15y

What's the smallest possible step you could take towards opening the files? Can you open the folder they're in? Or the next folder up in the directory tree?

2Cyan15y

It is possible for me to open the files; but when I do, I have a fight-or-flight stress reaction with accompanying squirt of adrenaline.

8Alicorn15y

What if you just... leave them open? And keep leaving them open until it goes away? (It would go away, right?) Flee the room if you have to while the application loads and then come back later and perform the passive action of not closing them.

1Cyan15y

Basically, it's thinking about doing the work that causes the reaction. If I commit to opening the files but not working on them, then I have a minimal reaction.

5Vladimir_Golovin15y

Excellent. If your deadline allows, you can explicitly commit to not working on the drafts until you figured out what causes your reaction to opening the files.

2erratio15y

By minimal do you mean effectively no reaction or that you still get a jolt of adrenaline but it's bearable? In the latter case, opening and closing the files until it's no longer effortful or stressful would be the first step. Otherwise you can move onto looking but not editing (check that it's all there and not corrupted, as Alicorn suggested).

2Cyan15y

I mean a little one -- mild butterflies in the stomach. I'll try it. (I will try all of the suggestions. But not tonight -- it's my bedtime. Damn you, little red envelope!)

5Alicorn15y

Does the stress reaction feel connected to time pressure of any kind? If so, that needs to go. If you feel like you're in crunch time: a) Ignore all optional suggestions, such as datadataeverywhere's, which call for you to get the work done by a specific milestone or date. (No disrespect intended to dde; that would be a good suggestion if your difficulty is more a matter of "the perfect is the enemy of the good" rather than "no time no time aaaaah".) b) If there is some deadline you cannot readily ignore on those papers, consider alternatives. Could you write a new, different paper instead for the same purpose? Can you get your name on a friend's paper and get similar return on the submission? Can you withdraw with a penalty that currently scares you less than working on the papers? Note: don't actually do these things, not if you actually want the specified work done before the deadline - but think about them, make yourself aware that you are not juggling the end of the world here. Finishing your papers is just about guaranteed not to be the One True Path To Success. c) Forbid yourself to touch those papers for some significant period of time. You are permitted (but not obliged) to think about them, but you can't actually open them up and change anything. "Significant period of time" here varies from person to person - I usually find the sweet spot between a day and a week, but if you operate on different time chunks than I do you might only need "till lunchtime" or as much as "a fortnight". Do anything in the world you like except mess with those papers - they are off limits. d) If this works the way it should, you should feel first relief, and then antsiness: you keep thinking of good ideas! But you can't use any of them because your time isn't up yet! It's your morning/day/week/fortnight off! When it's finally over you may find that you are eager to use up all of those ideas right away.

5Alicorn15y

Do you have this reaction to thinking about opening the files even if you commit to closing them immediately afterward? Perhaps with a not-working-on-them purpose in mind, like checking to make sure the files haven't been corrupted or anything?

2Cyan15y

Yup. In fact, it appears I have a mild stress reaction to just discussing doing so online.

4Alicorn15y

What about asking someone else to open them for you? I expect this would be stressful, but would it be so much so that you couldn't do it?

4cousin_it15y

Hi Cyan, I can offer you this, starting tomorrow. Interested? (Disclosure: I tried it yesterday for the first time with another LW user and was very satisfied with the results, at least on my end. Post forthcoming.)

0Cyan15y

I might take you up on it, but not right away. I want to try suggestions that directly target stress reduction first. What are the hours of overlap between your time zone and 8:30pm-10:30pm Eastern Standard Time (UTC -5)?

0cousin_it15y

It seems to be 5:30am to 7:30am in Moscow :-(

4JamesAndrix15y

If significant progress is not made on these papers by the next Open Thread, I will melt a box of paperclips. I know you're not clippy, but maybe this will work anyway.

1Cyan15y

It won't work. I have an active dislike for the Clippy character.

3Cyan15y

So, first some more relevant details, and then my plan of action. My instrumental reasons for wanting to submit the drafts are: first, it's career-damaging for someone in my line of work to let two nearly complete drafts languish; second, my former advisor also has an interest in seeing the work published (that's basically his job too) so if I want to get a good reference from him, I have to do it. There's no fixed deadline per se, but the sooner the better. My plan of action is to begin with those approaches which seem to target the stress reaction first, and if unsuccessful there, to move on to approaches which require working through it. In the first category are the suggestions of Vladimir Golovin (paragraph 3), erratio, and wedrifid. In the second category are the suggestions of Vladimir M, Vladimir Golovin (paragraph 2), cousin_it, and MartinB. I won't be following up on the suggestions of datadataeverywhere and JamesAndrix as I have already tried artificial time pressure and I know it doesn't help. My thanks to all of you for taking the time to offer your helpful suggestions!

[-]Paul Crowley15y110

What would you do to assess the advisability of taking antidepressants?

I'm trying to advise someone who is receiving conflicting advice; there appears to be plenty of controversy in this area and many factors confound assessing the evidence. Before I point to specifics on evidence that might influence the decision one way or the other I thought I'd ask in the most general way I can. I hope using the Open Thread to play "ask a rationalist" like this isn't a bad thing!

8wedrifid15y

Antidepressants can be useful but I do note that antidepressants and SSRIs are used more synonymously than they could be. In fact, it is worth asking your doctor about a SSRE (ie. tianeptine). From wikipedia: I speak as someone who uses an SSRI and finds it useful so I don't mean to claim the class has no value for all indications. In general antidepressants can be useful (even life changingly so) but it does depend on the details of the symptoms and etiology. 'Depression' is a somewhat broad symptom cluster and not all cases that fit that diagnosis have the same cause and respond best to the same treatment. But one note: It is best your own research even when a drug is prescribed. If the research brings up doubts then a second and third medical opinion is vital. There are overwhelmingly large numbers of people who have been prescribed venlafaxine (Effexor) for example who regret not doing their research first. It is an extremely effective drug but the side effects when ceasing usage are brutal. People who have gone through withdrawal for heroin addiction and also withdrawn from Effexor have described the latter experience as worse.

2Paul Crowley15y

Very interesting, thank you! According to Wikipedia, Tianeptine isn't available in the UK.

6Paul Crowley15y

Here's why I turn to Less Wrong in more detail. On the one hand, the anti-psychiatry movement and critics of biological psychiatry seem to suffer from really serious problems with thinking straight: they criticize reductionism, they seem to like philosophers like Foucault, and when they try to say that mental illness doesn't exist they seem to go for the "applause lights" of blaming society over the vivid reality of mental illness. But on the other hand, there seem to be really serious problems with the science of studying mental illness: a tendency to look where it's easiest to look rather than where the strongest effects are leading to over-emphasis on easily, cleanly detectable factors, massive distortion through the powerful financial incentives of the pharmaceutical industry including ghost-writing papers for scientists and a bad case of the file-drawer effect, and a century-long history of just making shit up. It's incredibly daunting to be faced with the task of cutting through this thicket to make a potentially life-changing decision.

2Perplexed15y

I was prescribed Buproprion (a norepinephrine reuptake inhibitor, not an SSRI) as a smoking cessation aid. With my doctor's acquiescence, I continue taking it at lower dosage for its beneficial (placebo?) effects on mild depression. If your friend's depression is associated with anxiety, or with mild addictive personality or "low will power", I would recommend it, perhaps coupled with an exercise program.

2[anonymous]15y

Everything I've heard is conflicting; of the people I've known who have been on antidepressants, some had great results, and some came out worse. What I know for sure is that psychiatrists are very, very likely to prescribe them for people showing symptoms of depression. If you go to a psychiatrist's office, you'll probably leave with a prescription. So don't go to an appointment thinking "Ho hum, I'll see what the doctor says."

1Perplexed15y

Amusingly, Marc Hauser's group at Harvard has just published a paper saying that use of SSRIs change people's ethical judgments on trolley problems. Here are accounts from two blogs

0Scott Alexander15y

Whether antidepressants (realistically you're probably talking about SSRIs) do enough better than placebo in moderate cases of depression to be worthwhile is a very tough question. But from a pragmatic viewpoint, SSRIs and placebo both do much better than nothing. Unless your friend can figure out some way to take a placebo and believe it to be effective, the SSRIs could be worth it.

3RobinZ15y

Some behavioral modifications (e.g. a program of exercise) are widely supposed to do better than nothing - would something along these lines serve as a useful placebo?

6Scott Alexander15y

I don't know. But placebos can have some pretty strange properties: if a doctor says a placebo is "extra strength", the patient will do better, if the pill looks colorful and complicated the patient will do better, if it's delivered via IV instead of pill the patient will do better, if the doctor wears gloves when handling the placebo because it's "so strong I can't even risk skin contact" the patient will do better, et cetera. So there's no guarantee something flaky-sounding like "diet and exercise" would be as strong a placebo as a big name like Prozac, even if Prozac did work mostly by placebo effect, which, again, is far from certain. Try the drugs.

0wedrifid15y

That may be be difficult to test. Normally we introduce controls for 'placebo effects'. How are we supposed to control for "stuff that actually works"?

0Cyan15y

Tangential: positive controls are common in e.g. biology experiments, where the efficacy of the experiment to detect an actual effect can be in doubt. (This won't be the case in treating depression, where successful treatment can be detected by asking the patient.)

0[anonymous]15y

Yes, that's a great idea.

0Document15y

Things I'm curious about, but probably won't investigate in detail: Does the placebo effect really do significantly worse if you don't "believe" in it, even if you're (for example) told that it's been experimentally shown to work even if you don't believe in it? I googled "metaplacebo" and found some kind of wiki page, but haven't looked at it yet.

0wedrifid15y

It still works (but less).

0Document15y

Or (I assume) we're not smart enough to find and prove the framing that would maximize the effect's strength for any given human, honestly or otherwise.

0Paul Crowley15y

My friend has been pointed at these references on this issue by one source: * Antidepressant Drug Effects and Depression Severity, A Patient-Level Meta-analysis, Jay C. Fournier, MA; Robert J. DeRubeis, PhD; Steven D. Hollon, PhD; Sona Dimidjian, PhD; Jay D. Amsterdam, MD; Richard C. Shelton, MD; Jan Fawcett, MD, JAMA. 2010;303(1):47-53. * Richard Bentall, 'Doctoring the Mind' and 'Madness Explained' * a PDF of interview with David Healy) which forms part of the course materials for the Open University module Counselling: exploring fear and sadness (sadly can't find the PDF online, but Healy's views on SSRIs appear to be well documented elsewhere) Bentall's Wikipedia bio concludes:

0jimrandomh15y

The sorts of antidepressants a psychologist would prescribe tend to be problematic in terms of side effects and success rate, so be sure to investigate low-risk options first. Some cases of depression correspond to micronutrient deficiencies; these can be tested for and addressed all at once using a large-dose multivitamin (the sort which has many times the recommended daily value for all the things that don't have overdose risk). Also check for insufficient sleep, severely insufficient exercise, and caffeine abuse, all of which can make people miserable in ways that antidepressants won't solve. If none of that helps, then try the chemical antidepressants. But don't stay on something that isn't working; being depressed makes it hard to upset the status quo, even if that status quo is a medication that isn't helping or is making things worse. As a defense against this problem, either schedule a particular day on which to reevaluate, or designate a trustworthy observer other than the prescribing psychiatrist. (And if it's not working, then raising the dose is almost certainly not a good answer).

0Paul Crowley15y

What you say sounds plausible, but I don't see how to evaluate it against conflicting advice.

[-]jacob_cannell15y110

In my last post on Health Optimization, one commenter inadverntently brought up a topic which I find interesting, although it is highly contraversial - which is HIV/AIDS skepticism and rationality in science.

The particular part of that which I am interested in is proper levels of uncertainty and rationality errors in medical science.

I have some skepticism for the HIV/AIDS theory, perhaps on the level of say 20-30%. More concretely, I would roughly say I am only about 70% confident that HIV is the sole cause of AIDS, or 70% confident that the mainstream theory of HIV/AIDS is solid.

Most of that doubt comes from one particular flaw I in the current mainstream theory which I find particularly damning.

It is claimed that HIV is a sexually transmitted disease. However, the typical estimates of transmission rate are extremely low: 0.05% / 0.1% per insertive/receptive P/V sex act 0.065% / 0.5% per insertive/receptive P/A sex act

This data is from wikipedia - it lists a single paper as a source, but from what I recall this matches the official statistics from the CDC and what not.

For comparison, from the wikipedia entry on Gonorrhea, a conventional STD:

Men have a 20% risk of getting th

... (read more)

[-]Vladimir_M15y190

A few years ago, I entered an online discussion with some outspoken HIV-AIDS skeptics who supported the theories of Peter Duesberg, and in the course of that debate, I read quite a bit of literature on the subject. My ultimate conclusion was that the HIV-AIDS link has been established beyond reasonable doubt after all; the entire web of evidence just seems too strong. For a good overview, I recommend the articles on the topic published in the Science magazine in December 1994:
http://www.sciencemag.org/feature/data/cohen/cohen.dtl

Regarding your concerns about transmission probabilities, in Western countries, AIDS as an STD has indeed never been more than a marginal phenomenon among the heterosexual population. (Just think of the striking fact that, to my knowledge, in the West there has never been a catastrophic AIDS epidemic among female prostitutes, and philandering rock stars who had sex with thousands of groupies in the eighties also managed to avoid it.) As much as it’s fashionable to speak of AIDS as an “equal opportunity” disease, it’s clear that the principal mechanism of its sexual transmission in the First World has been sex between men, because of both the level of promis... (read more)

0jacob_cannell15y

Quite interesting. I didn't really know much about Medical Hypotheses until I posted a link in another thread (as a minor side point) to an article my dad wrote which happened to be in that very journal, and someone pointed out what it was. Strange connection. A long while ago I found Duesberg's papers and entered into some online debates taking his position. The end result of all that reading moved me into a position closer to yours, but more uncertain. I just went and re-read Duesberg's most recent 2003 paper and noticed it still has a noticeable pull on me after reading it, but after going back and forth several times between the two camps I usually end up somewhere lost in the middle. Duesberg, even though probably ill-fated in his main quest, has I believe shown that the orthodox establishment has gone wrong at least in part. The history of the whole affair seems like it should be a lesson that we should learn from, a lesson that somehow results in learning and moving towards a more rational scientific establishment, more cleanly divorced from politics. That may be an interesting discussion to have here, if it hasn't already been discussed much. Deusberg suggests a jury process to replace the current peer-review system, which he is highly critical of. That could be an interesting rationalist angle on this.

8Richard_Kennaway15y

To say that this makes it not an STD is to misunderstand what an STD is. An STD is not a disease whose transmission method is specialised to the act of copulation. It is merely a disease which is so difficult to transmit at all that only the most intimate of contact has any substantial chance of doing so. What is important about the sexual contact is not the sex, but the blood contact. In HIV we have something that is so difficult to transmit that even conventional heterosexual intercourse has difficulty. Closer blood contact is required for a high chance of transmission, such as in anal intercourse (the intestinal lining is fragile and not adapted to contact with foreign bodies) or injection from infected needles. This has been known practically since the start from the epidemiology, before any pathogen was identified.

-2jacob_cannell15y

I'm not in quite agreement with both of your points. Yes of course HIV is transmissable only through blood, but I don't agree with that being a good criteria for use of the term "sexually transmitted", especially when other STD's such as gonorrhea are actually effective - they are measurably hundreds of times more sexually transmissable than HIV. This is probably due to both evolved mechanisms that those STDs have (such as ulceration formation) and overall low virality and transmissability of HIV. So it is ingenous I think to change the categorization. HIV is clearly at an extremum, and perhaps would be better classified as a weakly transmissable blood borne disease, not an STD. The evolutionary relevance is important here. How did this evolve?

8Nick_Tarleton15y

At an (unrealistically?) independent 0.5% chance per act, a 50% chance of transmission would require 139 sex acts — hardly "years and years". (ETA: yes, unrealistically, according to this abstract found by Perplexed: "However, in comparison with nonparametric estimates, the model assuming constant infectivity appears to seriously underestimate the risk after very few contacts and to seriously overestimate the risk associated with a large number of contacts. Our results suggest that the association between the number of unprotected sexual contacts and the probability of infection is weak and highly inconsistent with constant per-contact infectivity.") At best, this can show that pandemic AIDS can't primarily result from sexual transmission of HIV, which is evidence that AIDS has causes other than HIV, but also that pandemic AIDS spreads through other means (as suggested here, e.g.). If you're thinking of rates in the modern developed world, STDs are unsurprisingly more common when and where treatment is less available:

3jacob_cannell15y

That is for the highest transmission activity - receptive A, so be careful not to cherry pick. Yes - 139 unprotected sex acts. It would take 1390 unprotected insertive A sex acts to reach a 50% chance of transmission. So with some assumptions, mainly - gay bathouses and no condom use - yes the virus could spread horizontally, in theory. Although that population would necessarily first acquire every other STD known to man, more or less. But in the general heterosexual population, not a chance. If you compare to the odds of pregnancy from unprotected sex, the insane requisite amounts of unprotected sex with strangers would result in a massive baby epidemic and far more vertical transmissions long before it could ever spread horizontally in the hetero population. I don't know why you mention "modern developed-world rates" and then have a link to 1901 NY and Africa . . . You don't need the "at best" qualifier, but yes I agree that is what this shows. Showing that however opens a crack in the entire facade. Perhaps not a critical failure, but a significant doubt nonetheless. If the orthodox position had updated on the evidence, and instead changed their claim to "HIV is a borderline infectious disease that spreads primarily through the prenatal and blood-borne routes", then I would give them more creedence. Of course, for political reasons alone they could never admit that.

[-]Alicorn15y100

Although that population would necessarily first acquire every other STD known to man, more or less.

...a massive baby epidemic...

Many non-AIDS STIs, and pregnancy, are curable.

2jacob_cannell15y

Funny to think of pregnancy as curable, but yes of course that's true. However, it doesn't really change the numbers much. Also, from what I have read about the early 80's bathouse scene, it is possible that many of those guys did acquire every STD known to man, so at least in that case the sexual transmission route could work even with such terribly low efficiency. Regardless, it seems strange to label it as a STD from an evolutionary perspective, it doesn't fit that profile, and it seems incredibly unlikely it could have evolved as such.

6Alicorn15y

It's also funny to call babies an epidemic.

2MartinB15y

Life is a disease. It is transmitted by sex and ends deadly always.

0jacob_cannell15y

touche!

0Nick_Tarleton15y

What are the political reasons? Staying on-message and retaining funding, or something more specific?

7jacob_cannell15y

Essentially the government committed to a public awareness campaign that HIV was 'rapidly growing' in the heterosexual community, and this became part of the dogma. It is politically motivated - it's anti-sex message appeases religious conservatives while also shifting attention away from the gay bathouse scene, so it sort of benefits everyone politically, regardless of whether it's actually true.

0Nick_Tarleton15y

I meant "rates are higher outside the modern developed world". Rephrased for clarity.

1Douglas_Knight15y

I don't see why epidemiology should care about the 50% threshold. The relevant number is the expected number of transmissions per year. Thus independence is irrelevant.[1] At 200 anal tops per year per infected person, incidence should double yearly. And every top requires a bottom, so that's 400 anal sex acts per year for just doubling. It seemed to spread more quickly than that, but maybe 800 and 4x per year works. It seems just barely plausible with this transmission rate. I'm not sure of the details of bathhouses, but I thought that there was a lot of non-anal sex, too. [1] independence is relevant if 70s gays were systematically different from the people in the study; and they probably were, eg, they probably had higher rates of STDs

2Wei Dai15y

Why "more quickly than that"? From Epidemiology of HIV/AIDS in the United States: This seems to indicate a doubling time of about 2 years. ETA: Also according to that page, the patient with the first confirmed HIV infection died of AIDS in 1968, so the growth rate of AIDS before 1989 was at most 1.73x per year.

0Douglas_Knight15y

OK, maybe doubling works. But it's important to distinguish different populations. You should expect it to spread faster through the bathhouse scene than through the rest of the gay community than through the straight community. So it should slow down once it exhausted the bathhouse regulars (something like weekly visits) or when AIDS shut down that scene. If that happened around 1985 and there's a 10 year incubation period, then the 1995 numbers still include bathhouse effects. Diagnoses were increasing at 3x or 4x in the early 80s: 100 in fall '81 to 250 in mid '82 to 1000 in early '83 to 3000 by the end of the year. From '83 to '89 it was merely doubling and it slowed after that. Of course, there are problems with diagnosis numbers early in an epidemic, but death followed quickly, in weird ways, so these numbers are probably good enough. Yes, there were people with AIDS in 60s, but that 1.73x includes time to get to the bathhouse scene.

0jacob_cannell15y

One problem with this simplified model is assuming every sexual act is with a new partner, which would only be true in the very early stages. I think your analysis is on the right track though, and it seems barely plausible with this transmission rate, assuming negligible condom use and an intense bathhouse scene. However, in standard theory HIV progresses to AIDS in about 10 years, so this sets a timer which starts removing vectors from the population. Thus the exact exponent matters considerably. If incidence can only double every year, then after 10 years you get 2^10 ~ 1000 cases. If incidence doubles every 6 months (quadruples every year), then you get a million cases after ten years. If you consider that all other STD's would infect this population before HIV, then one has to wonder how that would effect condom use, and how that changes the model. So do you find the transmission rate and make the model before you decide that HIV was an STD which spread this way, or after?

0Douglas_Knight15y

It assumes that every sexual act is with an uninfected partner. Perhaps that's what you meant, but then I wouldn't have used "very." I think this is pretty well documented. STDs were routine and not a big deal (treatable!) in the 70s gay scene. Thus they did not cause condom use.

8cata15y

I'd point out that nobody is claiming that HIV is exclusively sexually transmitted; there are other methods of transmitting it, such as infected needles. Also, Wikipedia cites a paper suggesting that those rates you mentioned are "4 to 10 times higher in low-income countries" and as high as 1.7% for anal intercourse. I don't know whether or not these facts are sufficient to address your fundamental complaint, but they would make a pretty big difference.

6wedrifid15y

It takes courage to voice a low but not negligible degree of doubt in a emotionally salient mainstream position. I would expect it to result in almost as much social punishment as in the case of outright denial. (Emotional backlash isn't good at math.) More concretely, I would roughly say I am only about 70% confident that HIV is the sole cause of AIDS, or 70% confident that the mainstream theory of HIV/AIDS is solid. I am surprised that those two confidences happen to be the same. Is it not a distinct possibility that HIV is, in fact, the sole cause of AIDS even when the mainstream theory is itself rubbish? (For example, if the theory got important details such as mechanism completely wrong.)

7Alicorn15y

I like this sentence.

1jacob_cannell15y

I tend to think of "HIV being the sole cause of AIDS" as the central tenet of the mainstream theory, but sure that could be true even if much of the details are wrong. Actually, I think many within the mainstream would admit most of the details are wrong - last I checked all the important details, such as how the retrovirus could come active after many years and damage T-cells and what not are all still hot research items. And most of the specific results have been failures - no vaccine yet, just some drugs with a bunch of side effects which may or may not even improve mortality, etc etc. I find hypotheses in the middle more likely overall - examples: that HIV is a mostly harmless retrovirus but in some people with (X, Y, Z cofactors) it can cause immune damage. And I yes, I am at least mildly concerned about taking an HIV skeptic position in a public internet forum - and just thinking about the reasons for that causes me to slightly update to be more skeptical.

6JoshuaZ15y

Note that in general low-transmission rates aren't that good an argument against it. First of all, a lot of transmission in the US occurred through other forms, especially early in the epidemic (intravenous needle sharing and transfusion of infected blood being two major ones). As others have noted the transmission issue is also generally higher for homosexual rather than heterosexual intercourse. Frankly, the thing that I most don't understand about people who are skeptical about the HIV-AIDS link is how one explains the fact that anti-virals tailored to deal with HIV work. Even the first-gen treatments such as AZT were used due to the biochemistry of HIV (often targeting reverse transcriptase). And they worked in delaying the onset of AIDS and worked for giving people with AIDS higher life-expectancy. I don't see how one can easily reconcile that with HIV not being the cause of AIDS and I've never heard anything remotely resembling a coherent explanation about this.

0jacob_cannell15y

The extremely low official transmission rates indicate that HIV is probably not a sexually transmitted disease, as typically understood. According to the published transmission data, it's only effective natural means of transmission is vertical - from mother to child. From what we know about symbiosis and parasites, this should lead one to suspect it is not that lethal. However that doesn't destroy the hypothesis that HIV could be a unique disease of civilization - a mutation of a formerly limited retrovirus into a more damaging form that spreads horizontally only through extremely unnatural novelties of the modern world - needle sharing and epic promiscuity in some subgroups. It does of course make the overall hypothesis more complex and overall less likely, but it doesn't destroy it outright. This is very difficult to prove ethically, and certainly hasn't been shown to satisfaction. Consider how immensely difficult it has been to understand the real health effects of fat in the diet, for example. In the 80's and 90's, we thought it was oh so simple. Today we (the wise) realize how profoundly ignorant we were then, and still are today. I think part of the rational reappraisal for the strength of the HIV hypothesis should come from a similar update. In other words one can only cling to a high degree of certainty about the HIV hypothesis when one is profoundly ignorant concerning the depth of one's ignorance about the complexity of human health. That's the theory. Ahh the world would be so simple if that was true. The AZT clinic trial was ended prematurely and double-blindness could not be maintained due to side effects. The skeptics claim that later studies show that AZT increases long term mortality, not the other way around. You need to realize how impossible it would be to concretely show what you presume to high certainty. You would basically need a double blind study with two control groups, one receiving plazebo, and compare total mortality. One immediat

3JoshuaZ15y

So I don't know what evidence you have for these claims. The original AZT study can be found here. It handled placebos just fine. And AZT is just one example of many anti-retrovirals. 3TC and ABC have also been used. Moreover, there are studies which show that the the standard drug cocktails work better than AZT or 3TC alone. That makes sense for the standard model of HIV as the cause of AIDS. There seems to be an issue here involving what level of evidence is necessary. From a Bayesian stand point you can't ever "show" anything in the sense that you want. But the overall pattern of evidence can still give you a probability close to 1 that HIV is the primary cause of AIDS. http://www.ncbi.nlm.nih.gov/pubmed/3299089

2jacob_cannell15y

Ah you read this while I was editing it. Yes I have browsed that abstract, although I can't see if there is a full copy of the whole thing. Notice the link to the toxicity right next to it: The toxicity of azidothymidine (AZT) in the treatment of patients with AIDS and AIDS-related complex. A double-blind, placebo-controlled trial. Notice that it has serious toxicity side effects, including bone marrow supression, and thus it can massively damage the innate immune system. This is evident in the abstract itself: The original AZT study could not possibly be placebo controlled, due to the high toxicity - this is basically chemotherapy. Now chemotherapy can be effective, but it can not be double blind. Furthermore, due to high toxicity, only a fraction of patients actually completed the intended trial: Only 27 out of 245 AZT subjects completed the full 24 weeks! One in the AZT group of 245 died in this period, but how many more in the AZT group would have died if they had been able to complete the full 24 week chemotherapy trial? These were crazy times. These patients were very ill and very worried. It was a full scale terror panic. It was obvious who was on placebo and who wasn't, and from what I have read, placebo patients were swapping and trading pills with AZT patients who couldn't finish. It was many things, but not double-blind. Here is what would be required to prove with > 99% certainty that HIV is the sole cause of AIDS: Take a random sample of perfectly healthy test subjects. Now inject half of them with HIV and half with a placebo, and follow their health over the long term. That is about the only test that could get you 99% accuracy, and it is obviously not ethical. So, instead we have to make due with what we have. And again, showing that AZT improves long-term mortality - which the AZT trial clearly did not show, only shows that AZT improves mortality in sick AIDS patients. It doesn't tell you much else about HIV as a theory. It's important t

6Vladimir_M15y

jacob_cannell: There have been documented cases of accidental HIV infection of lab workers and dental patients that are not too terribly far from such a controlled experiment. See: S.J. O'Brien and J.J. Goedert, "HIV causes AIDS: Koch's postulates fulfilled," Current Opinion in Immunology, Vol. 8, Issue 5, October 1996, pp. 613-618. UPDATE: Ungated link to the paper here.

4RobinZ15y

The article (listed as a "guest editorial", I note for completeness) has the following citations in the relevant section: 52. Weiss SH, Goedert JJ, Gartner S, Popovic M, Waters D, Markham P, Veronese FD, Gall MH, Barkley WE, Gibbons J et al.: Risk of human immunodeflciency virus (HIV-1) infection among laboratory workers. Science 1988, 239:68-71. 53. Reitz MS Jr, Hall L, Robert-Guroff M, Lautenberger J, Hahn BH, Shaw GM, Kong LI, Weiss SH, Waters D, Gallo RC, Blattner W: Viral variability and serum antibody response in a laboratory worker infected with HIV-1 (HTLV-IIIB). AIDS Res Hum Retroviruses 1994, 10:1143-1155. 54. Guo HG, Waters D, Hall L, Louie A, Popovic M, Blattner W, Streicher H, Gallo RC, Chermann JC, Reitz MS: Nucleotide sequence analysis of the original isolate of HIV-1. Nature 1991, 349:745-746. 55. Wain-Hobson S, Vartanian JP, Henry M, Chenciner N, Cheynier R, Delassus S, Martins LP, Sala M, Nugeyre MT, Guetard D et al.: LAV revisited: origins of the early HIV-t isolates from Institut Pasteur. Science 1991, 252:961-965. 56. Wolinsky SM, Wike CM, Korber BTM, Hutto C, Parks WP, Rosenblum LL, Kunstman KJ, Furtado MR, Munoz JL: Selective transmission of human immunodeficiency virus type-1 variants from mothers to infants. Science 1992, 255:1134-1137. 57. Morbidity Mortality Weekly Report of the Center for Disease Control: Revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults (1993). 1992:1-19.

2[anonymous]15y

I'm not a skeptic, but I am a bit surprised that three infected lab workers is considered evidence worth mentioning. I'm accustomed to seeing clinical studies with fewer than 100 patients treated as merely suggestive, and several hundred or even thousands of patients necessary to convince regulatory authorities that there is a real phenomenon. So when I read that "three laboratory workers...became infected", it strikes me as anecdotal. Of course you take what you can get, but still, it seems a small group. My experience is with pharmaceuticals, which could explain the difference. But still, there's a difference to be explained. On the upside, if three is worth mentioning, then maybe one is worth mentioning, which would mean that Seth Roberts's self-experimentation may be worthwhile.

1jacob_cannell15y

Ok, so I have finished reading O'Brien and JJ Goedert. I think the accidental lab worker cases probably have low utility in distinguishing between pathologic HIV and passenger HIV theories, but the animal SIV studies potentially fulfill the ideal experiment design. If animals injected with pure close SIV analogs to HIV develop AIDS-like illness, that is about as good as evidence as one could hope for. For the lab worker infections, they say that the genomes are close to the clonal strain used in the lab, but they don't have a good probalistic model for this: threefold more or less variation doesn't sound like much to me, and regardless even assuming they did all get infected via the lab, it doesn't mention treatment, so we don't know if they were treated with AZT or what. I think the animal models are more interesting - as all the drug cofounding variables are eliminated and everything is controlled. The authors discuss 3 animal models - baboons, some wierd knockout mice with human t-cell graphs, and finally SIV injected into rhesus monkeys. They spend the most time discussing the latter paper so I looked into it. "Induction of AIDS in Rhesus Monkeys by Molecularly Cloned Simian Immunodeficiency". In short, all 5 of the monkeys injected later become seropositive, and 50% of them progress to AIDs-like illnesses and die. Presumably some are naturally resistant. There is a potential issue to this otherwise smoking gun, which is that it appears they had to inject whole culture of T-cells, as the virus can not be isolated directly into plasmid vectors: But if all they had then was provirus integrated into PBL ( peripheral blood lymphocytes ), then one must wonder about the risk of graft versus host disease, and transfusion reaction in general. Basically, the foreign T-cell line can actively invade and cause all kinds of havoc. Perhaps they have controlled for that and I'm completely off, but I don't see where they mention it. OBrien and Goodart conclude: I have

2RobinZ15y

It appears to me that you are noting different weaknesses, different alternative explanations of the results, in each experimental protocol. I imagine that could appear to happen if there were a strong publication bias for results in favor of the hypothesis (e.g. with medicinal prayer studies), but that has not been demonstrated.

0jacob_cannell15y

From what I understand, there is strong reason to suspect such bias in the animal models, as many other animal tests were conducted previously attempting to show SIV causing AIDS, and they uniformly failed. From what I have read, the SIV injections failed to do anything in chimps. We don't know how many other rhesus monkeys were injected, but I find it highly unlikely that the five in this study (3 out of 5 of which developed immune defeciency) were the first. That bias is interesting and it would be useful to have a metastudy covering all the animals that have been injected, but of course most naturally occuring strains of species specific virus should be expected to be harmless. So the 'bias' doesn't really tell you much and is expected either way. I'll look to see if there is any skeptic review of the SIV injections which did succeed, because in my mind that is fairly strong evidence. If HIV analogs cause AIDS-like illness in primates when they hop species or traditional barriers, this is a good experiemental model for HIV crossing over into humans and causing illness, and is far stronger in my mind than the murky epidimelogical data with all of it's drug cofactor issues. The lab monkeys are in a controlled environment - we know they weren't using drugs, weren't fed AZT, etc etc.

1jacob_cannell15y

Yes, but were they treated with chemotherapy agents such as AZT which cause bone marrow and immune supression? (I am asking before I read the article) For the controlled experiment example, it would have to be double blind the entire time. Nobody would ever know who got placebo and who didnt', and no difference in treatment regimens either way - all on the same diet, same lifestyle, etc etc. I highly doubt the accidental cases fit that profile.

2Vladimir_M15y

jacob_cannell: From what I see, O'Brien and Goedert don't report about this. However, Cohen's Science article to which I linked above provides more details about the cases of infected lab workers, claiming that two of them didn't receive any antivirals until the progress of AIDS was well underway: http://www.sciencemag.org/feature/data/cohen/266-5191-1647a.pdf

1jacob_cannell15y

I have seen that, but I don't take it as strong evidence of anything either way. Clearly HIV antibodies are correlated with ill-health, poor immune function and low CD4 counts - that was established in the very beginning by Gallo et all and is the entire basis of their original claim. But the competing hypothesis is not the null hypothesis, the competing hypothesis is that the HIV presence is a marker of some severe immune function failure, and that HIV itself is a largely harmless vertically transmitted retrovirus - like all the others. If it does become active, it is because something is already seriously wrong. So of course if you have a lab, and you are testing lab-workers deathly afraid of contracting HIV, in either theory some may come up HIV+, and in either theory those that test positive will be ill. This is very different than the intentional infection test you need to establish strong causation along a koch's postulate principle. I think one of the stronger flaws in current HIV theory is the chimpanzee models - they have many HIV similar retroviruses (SIV's), they are common in wild chimps, are all vertically transmitted, and don't appear to be problematic for chimps. If HIV killed chimps outright, the mainstream theory would have some more ammo. Instead the theory is that HIV came from a chimp SIV and somehow became lethal and horizontally transmissable at the same time in humans. But from the studies of sexual transmission - it is only weakly blood transmissable. So overall, it's alot to swallow. I think it is plausible that this happened and HIV causes novel problems in humans, but it certainly has not been well demonstrated, mainly because HIV doesn't cause specific symptoms and it is extremely difficult to properly dissociate other causitive factors. As Duesberg notes, the cocaine/meth craze boomed in the 80's right as the AIDS epidemic started - they overlap perfectly. Deusberg's theory is that drugs specifically are a major cause of the immuno

7CronoDAS15y

I think that, in general, viruses that jump species barriers are more lethal in their new hosts than their old ones. Selection pressure tends to make viruses that are transmitted from individual to individual less dangerous over time, not more, because a dead host can't spread the virus. So the fact that chimps tolerate SIV better than humans tolerate HIV isn't a strike against the "HIV is mutated SIV" theory. Also, there exists a more direct counterexample: FIV, feline immunodeficiency virus, is an HIV-like virus that is commonly found in lions. It doesn't harm lions very much, but it's lethal to domestic cats, which haven't co-evolved with the virus the way lions have. (Specifically, lions compensate for the loss of T-cells to the virus by producing them in much larger numbers than other animals do.)

0jacob_cannell15y

Perhaps not all viruses that jump species barriers are more lethal, but we care most about the lethal ones for sure, and lethality or host damage is a side effect of high replication, so yes I agree with your analysis. I also said: I'm reading the OBrien Goddert response now that cites some animal studis showing how some animal HIV analogs do fulfill koch's postulate now. I am reading into the sources, but I wasn't aware overall that they had found animal HIV analogs that caused AIDS like symptoms. I'll have to update towards the 'HIV is more pathogenic' stance if this all checks out.

2Vladimir_M15y

jacob_cannell: According to O'Brien and Goedert, however, the strains of HIV found in these accidentally infected individuals were effectively identical to those from the source of their accidental infection, showing much less difference than is usual for strains taken from two random patients. This looks like powerful evidence against the competing hypothesis. On the other hand, it seems like there have been disputes about the validity of these identification procedures, and Duesberg and other skeptics raised some specific objections to them back then. However, this gets into technical issues that I'm definitely not competent to judge on. (I should add that I haven't delved into the references provided by O'Brien and Goedert, which Robin has conveniently listed in his above comment, to see if their summary of the facts is accurate and complete.)

0jacob_cannell15y

Also, while we are looking at evidence for one side or the other, I should present some of the recent evidence that supports the Deusberg hypothesis, namely that cocaine use can cause HIV progression and AIDS-defining illness. Heavy cocaine use and HIV+ status are extremely correlated, the question is one then of causation. Just search for "cocaine cd4 counts"| From a rat study linked here Even more interesting - an abstract from a more recent study showing that crack cocaine can cause AIDS independently (in women at least) CONCLUSION: Use of crack cocaine independently predicts AIDS-related mortality, immunologic and virologic markers of HIV-1 disease progression, and development of AIDS-defining illnesses among women. A blog discussion of this here Perhaps the tide's are turning in Deusberg's favor?

0jacob_cannell15y

I was under the impression that there was only 2 HIV strains, and only HIV-1 is of concern in the west, but I'm no expert on this. Do you have the O'Brien/Goedert paper link or any of this discussion? Your earlier link was just an editorial summary. Sounds interesting, I'd like to read more into it. (i am going to search now, but if you have the link . . .) If O'Brien and Goedert are right, it would depend on how many strains there are and what their prevalence is ahead of time - I think Bayes' Theorem would apply, no? Also, I guess we just have to presume that no other sick people from the sample were not reported. And finally, the sample size of 1 or 2 raises some theoretical issues, but still it would be interesting.

5JoshuaZ15y

No. There are two major HIV strains, HIV-1 and HIV-2, with the vast majority of infections in the US being HIV-1. However, there are many substrains of both. In general, since HIV has a high mutation rate, there are individual genetic markers for different people. No one who has HIV for more than a few weeks has a single form of HIV in them but already will have a population of slightly distinct forms. This is part of what makes HIV so pesky for the immune system, for developing vaccines, and for developing drugs to combat it. Note that this isn't that rare. For example, influenza does something similar although my impression is that this is not nearly as extreme an example. Frankly, this is a very basic aspect of HIV biology, which you could find on any primer on the subject. If one doesn't know about this, this suggests major gaps in one's knowledge base about HIV and I'm not at all sure that one who doesn't know about this has enough background to discuss issues related to whether HIV causes AIDS.

2kodos9615y

Because he said "2 HIV strains" rather than "2 major HIV strains with many substrains"? He should be disqualified from participating in debate on a topic because he failed to use sufficiently precise language on a point not relevant to the current discussion? I'm a non-partisan on this issue, having not researched it enough to have a firm opinion... but the consistent use of these kinds of dark side debate tactics by one side makes me seriously tempted to update in the opposite direction.

2JoshuaZ15y

This isn't a "dark side" tactic. Vlad wrote: It seemed pretty clear from context that Vlad was talking about substrains. If there's anything that went wrong here, judging from Jacob's followup remark, it was an illusion of transparency failure on my part in that Vlad's meaning seemed clear to me, and I then made the (erroneous) conclusion that someone with a basic background would also get what Vlad was talking about. Note incidentally, that using heuristics about whether or not someone has enough background to understand or discuss something is not intrinsically a dark side issue to start with. Indeed, sometimes it is very necessary. See for example some of the discussion on cousin it's recent post where it seemed that some individuals (well, primarily one individual) were making repeated subtle but highly relevant errors about certain ideas related to Godel's theorems and Turing machines. After repeated attempts to explain, the mathematicians in the thread (including myself) started trying to explain that the errors in question were basic enough that further attempts to explain would likely be fruitless. That's not a dark side tactic, sometimes that's just true.

1kodos9615y

Yes, I absolutely agree. But that's not the tactic I was referring to. The tactic in question consists of finding some nitpicky objection to something your opponent said, something not directly relevant to the issue at hand, and something which isn't even wrong, just insufficiently precise, and discounting the substance of their argument on that basis.

0JoshuaZ15y

Hrrm? Imprecision if anything occurred on Vlad's part, not Jacob's. Again, see issue of illusion of transparency matter. See also Vlad's remark below.

0Vladimir_M15y

JoshuaZ: Yes, of course. I'm not familiar with the finer points of terminology in this area, but the O'Brien-Goedert paper uses the term "strain" both for the two major strains and their sub-strains, and I've noticed the same in many other papers too. So I don't think this was imprecise in any way.

0jacob_cannell15y

Vlad originally used the word 'strains', not substrains, and I have only ever heard 'strains' used in reference to HIV1 and HIV2 as you say. But yes I am at least somewhat aware of sub-strains through protein coding variation in viruses in general, and how mutations lead to new epidemics. I'm not sure what the proper 'background' is just to discuss an issue, but I probably don't have it, regardless.

1JoshuaZ15y

Ok. Seems to be an error in communication more than anything else then. FWIW, you seem to have a much better understanding of a lot of this material than most of the HIV-AIDS skeptics I've interacted with in the past.

2Vladimir_M15y

(I initially posted an excerpt from the paper in this comment, but in the meantime, I found an ungated version. The stuff about the accidental transmissions is on page 615.)

0RobinZ15y

The article cited here? I downloaded a copy if anyone wants to PM me an email address.

1JoshuaZ15y

I'm not sure you are allowed to demand that specific piece of evidence.

2jacob_cannell15y

If you want absolute certainty, you need to fulfill koch's postulates, which is the medical disease variant of experimental physics, and is well grounded. So yes, in questions of science, we do demand specific evidence for near-certain proof, but naturally when that is not possible for whatever reason, we can still update based on other evidence and reach well grounded conclusions.

5JoshuaZ15y

First of all, you never have absolute certainty. Proof is for math and alcohol. You may also want to read the Sequences relevant to Bayesian reasoning especially about how 0 and 1 are not probabilities. Note also that even outside a Bayesian context, Koch's postulates are a rough framework. For example, sometimes there is great difficulty growing an organism in a culture (Koch's second postulate requires this), and sometimes only a fraction of a population may be symptomatic. Even Koch was willing to treat the first postulate as a guideline rather than a hard and fast rule; he worked with multiple examples of microorganisms that showed up in some healthy people but didn't cause disease In fact we know now that this very common. Many causes of minor infections, such as staph, are present on everyone. It then takes some problem, such a wound, or disruption of the immune system to cause the person to become unhealthy. For viral examples, look at asymptomatic herpes or HPV. Frankly, referring to Koch's postulates like this are one of the things that makes many people with medical knowledge not take the HIV-AIDS skeptics very seriously. As a general heuristic, cranks like to take old ideas and try to use them to argue against some modern result even when that idea isn't used in the current form or isn't as absolute as they make it out to be. This seems for example to occur frequently with creationists who use an oversimplified form of Mendelian genetics that seems to date prior to the work of Hardy, Weinberg, and Fisher. As a way of preventing this signal, referring to such things as Koch's work can be done, but you need to be clear that you understand the limitations it has. Otherwise, it just sets off alarm bells.

2CronoDAS15y

This bears repeating.

1jacob_cannell15y

It is also difficult to empirically test elements of String Theory or evolutionary history. This difficulty isn't taken to be evidence for against those theories. Lack of easy clear cut empirical tests is just that - lack of a simple theorem discriminator. I am not claiming that it is some magic pass required for any theory to be credible. On the other hand, nor can one use lack of said simple theorem discriminator to get a free pass and somehow claim to have extra certainty - that most come from evidence regardless. However, in this case I do think that the general idea of Koch's postulate is highly relevant, it is not 'cranky' to invoke it, and in fact it is probably the single strongest piece of evidence for the mainstream position - as Vlad M pointed out to me. I am currently reading: " HIV causes AIDS: Koch's postulates fulfilled" Guest editorial Stephen .I O'Brien* and .lames ,i Goederff It reviews SIV models where SIV can be prepared and injected into monkeys and in some cases does cause AIDS like disease progression (usually more quickly than HIV). So I don't think your associations between "referring to Koch's postulates" and cranks, and your general heuristic are valid at all, not when the mainstream is listing this as a centerpiece of evidence.

5JoshuaZ15y

I haven't read the editorial yet, but I'm not sure where the notion that this was a centerpiece of evidence comes from. I was vaguely aware of this result but never have paid it much attention. I suspect that if one talked to active AIDS and HIV researchers they'd say that there was a very strong case even without this. But yes, conservation of expected evidence does come into play here, so you have a valid point. In any event, the upshot that Koch's postulates are not seen as any sort of absolute is very much relevant.

2RobinZ15y

S.J. O'Brien and J.J. Goedert (1996) (the paper cited by Vladimir_M) specifically references Koch's postulates. The accidental infection of laboratory workers is only one of five items cited in reference to (I believe) #3, "The cultured microorganism should cause disease when introduced into a healthy organism."

2NancyLebovitz15y

Ever since I've read in Guinea Pig Zero about test subjects who feel that they'll never be able to afford medical care and cheat on the requirements of an experiment (for example, not following dietary changes they've signed on for), I've been that little bit more dubious about drug studies.

5Wei Dai15y

Have you seen these two recent meta-analysis of HIV transmission rates? Comparing them to your numbers, it seems that Wikipedia/CDC have greatly understated the risk for P/A sex acts (due to using older studies?). Also, according to this page, the transmission rates for genital herpes are similar to HIV for P/V sex acts, so AIDS is not the only STD to have such low transmission rates. I think the conventional theory is that HIV established an initial foothold by means such as needle sharing, blood transfusions, and P/A sex acts. That seems quite realistic to me. Why do you think it's unrealistic?

0jacob_cannell15y

No I haven't seen them until now. I haven't really been looking into or thinking about this until just recently. I originally quoted the wikipedia data, which several us tracked down and analyzed in another thread (it comes from the CDC and originally from a single medium-ish European study which I think did a fairly good job to control for factors such as condom use and what not given the complexities): Let's compare to the 1st meta-analysis: it finds 0.04/0.08% for insertive/receptive P/V sex in the 1st world, very similar but slightly lower to the older CDC study (from europe). They find the rates are around 5 times higher in the 3rd world. For P/A receptive, they find a pooled per-act rate of 1.7%, about 3.5 times higher than the 1st world CDC study. I suspect this is simply because they pooled 1st and 3rd world data together. This does not show that Wikipedia/CDC have "greatly understated" the risk for P/A sex acts - this data is in agreement. The variance is between 1st and 3rd world studies. The 1st world data is of paramount concern for the historical origin theory of the disease. The 3rd world data is a side point - not historically important and suspect in general, as in fact mentioned in the abstract. The abstract: So this supports the original CDC data for the 1st world. The second meta-study finds a receptive A per act rate of 1.4%, a little lower than the 1st study, and about 2.8 times the 1st world CDC data. They make no mention of what countries, and I assume it mixes 3rd and 1st world data - it is a large meta-analysis of many studies. There is also some serious wierdness in the 2nd study: So something doesn't fit - the per act and per partner numbers don't square up at all. Somewhere else someone posted an abstract with a similar result, how basically your chance of infection levels off quickly in couples and is much lower than you think - you are either going to get it or not irregardless of number of sex acts. This suggests that there a

2jimrandomh15y

Perhaps vulnerability to transmission is partially dependent on prior immune health. That would predict faster spread where health-in-general is worst, ie Africa, as observed, and also explain the discrepancy between prevalence and observed traqnsmission rates. I also recall reading an article about a widely afflicted demographic in the US (a particular subset of gays in a particular city - I don't recall which one), which suggested that they had already weakened their immune systems with drugs and sleep deprivation. The other possibilities are that the transmission rate you quoted is wrong for some reason, or that the sexual transmission aspect has been overstated, and most transmission is through reused needles and other blood-borne vectors. Note that spreading the idea that the transmission rate of AIDS is low has negative utility, regardless of whether it's true or not, since it would encourage dangerous behavior.

[-]mattnewport15y100

Note that spreading the idea that the transmission rate of AIDS is low has negative utility, regardless of whether it's true or not, since it would encourage dangerous behavior.

This is untrue. Consider a similar claim: "spreading the idea that very few passengers on planes are killed by terrorists has negative utility, regardless of whether it is true or not, since it would encourage dangerous behavior."

Informing people of the true risks of any activity will not in general have negative utility. If you believe a particular case is an exception you need to explain in detail why you believe this to be so.

9Wei Dai15y

In the case of infectious diseases, there are large unpriced negative externalities involved. Everyone doing what is individually rational, given true beliefs about transmission rates, is likely not socially optimal, because the expected individual cost of a risky action is less than the expected social cost. Giving people false beliefs about transmission rates can improve social welfare by shifting the expected individual cost closer to the expected social cost.

3mattnewport15y

Are you talking about free rider problems with health care costs under a partly or fully socialized health care system or something else? STDs seem to be less of a problem than more easily transmitted diseases like flu for most negative externalities I can think of.

2Wei Dai15y

And also, if you take some risky action that increases your chances of get infected, that also increases the chances of everyone else getting infected (causally, via yourself getting infected and then infecting others). I'm not sure I get your point here. Whether it's more or less of a problem doesn't seem relevant to the original claim that spawned this subthread.

3mattnewport15y

It's relevant to using your negative externality argument to support the original claim. To be consistent you would have to argue that we should make even more effort to avoid spreading the idea that airborne diseases like flu have low transmission rates (if true) than the idea that STDs have low transmission rates. Are you advocating a general policy of deliberately misleading people about the risks of various activities in an effort to correct for negative externalities? I'm pretty sure more efficient and robust approaches could be found.

3datadataeverywhere15y

It would be consistent with Wei Dai's claim just to argue that we should make an effort to not reveal how low the transmission rate of influenza is among people who don't wash their hands; we know that hand-washing is a large factor in transmission, but actual transmission rate numbers are still low enough to fail to convince people to wash their hands. From a brief study of those particular numbers (I worked on a team modeling the spread of H1N1), I feel like we already mislead the public about the numbers themselves by being truthful as to the societal benefits and somewhat optimistic about the individual benefits of hand washing. If you believe more robust methods are more efficient, by all means, advocate for them, but I'm reasonably happy with the current situation. From another perspective, blood-borne pathogens are particularly worth focusing on because they are easier to control. If we could encourage the entire population of the world to behave safely (not reuse needles, use condoms for sex, etc.), it would be a fairly minor change for individuals, but could eradicate or nearly eliminate HIV over time. With the flu, safe behavior will limit the damage of seasonal infections, but it's not realistic to actually eliminate the virus. Thus, over the long term, I think the negative externalities of HIV might outweigh those of influenza.

1Wei Dai15y

I think government policy makers and public health authorities already use a variety of approaches to reduce negative externalities related to infectious diseases, including subtle misinformation, such as making efforts to correct people's beliefs about transmission rates when they are too low, but not when they are too high (anything really obvious wouldn't work in a free society like ours). But it seems clear that large negative externalities still exist. What other approaches do you have in mind, and why haven't they thought of it already?

2mattnewport15y

I think we're starting from quite different assumptions about how society works. I don't believe that government policy makers or public health authorities are very rational. Even to the extent that they are rational, I don't believe that their incentives are such as to reliably lead them to decisions that maximize utility by the kind of utilitarian calculus you seem to be assuming. So to the extent that we agree negative externalities exist (and I suspect we differ a fair bit on what they are and to what extent they exist) I have very little expectation that government policy makers or public health authorities will tend to take actions that minimize them.

2Wei Dai15y

What did you mean when you said "I'm pretty sure more efficient and robust approaches could be found"? You're not offering any concrete ideas yourself, and apparently you weren't thinking of government health authorities when you wrote that, so who is supposed to find and apply these approaches?

0mattnewport15y

Think 'market based'. Internalize negative externalities. To a first approximation this usually means reducing government involvement rather than increasing it. This is straying into politics though so maybe we should avoid further discussion of this topic.

2datadataeverywhere15y

Compared to the rest of this open thread, I don't think you have anything to worry about! Seriously though, I think we'd both like to hear you elaborate upon your market-based idea. I don't think I got any useful information out of your blurb.

1mattnewport15y

Let me first clarify the points I was making in this thread (which were not intended to lead to a debate about healthcare or politics in general). If you still feel we have substantive disagreements that we might be able to resolve through more explicitly political discussion I'm willing to continue the conversation unless there are strong objections from others. First my points were not intended to imply any particular opinion on AIDS transmission rates specifically. My initial post was simply intended to point out that the utility of spreading the idea that AIDS transmission rates are low is dependent on the truth of the claim. This was intended to be a more general point that exaggerating the risks of any particular activity is not a good general policy. In the AIDS case there are genuine costs to taking precautions against transmission, even if they are in fact greatly outweighed by the benefits. In a hypothetical world where transmission rates are negligible, maintaining that they are high would have negative utility. Wei Dai responded by claiming that because of negative externalities associated with infectious diseases, exaggerating the transmission rate can improve social welfare. Now this is not incompatible with my original point, it is rather a claim of a mechanism by which exaggerating the risks of an activity can have positive utility. It is probably worth noting at this point that I am not a utilitarian so I am likely to disagree with utilitarians on what outcomes have positive utility but we can probably agree that in general internalizing negative externalities is a good thing. I concede that it is possible in theory to imagine a situation where deliberately exaggerating risks has positive utility. The fun thing about negative externalities though is that it is very easy for an intelligent person to think of some and to propose plausible mechanisms by which any given action can be justified. I could easily argue for example that the credibility o

2datadataeverywhere15y

Agreed. I think Wei Dai and I also agree (without speaking for Wei Dai), but think that the idea transmission rate to spread is conceivably a function of the true transmission rate, rather than locked to the true transmission rate itself. It sounds like you agree with that too, but are a little more stringent with what makes that "conceivably" true. To my mind, one such situation is one where a super-majority of the population don't understand probability, and 1% effectively means "never" to those people. In this case, specifically lying about the number is less helpful than phrasing it alternate true ways, but because they don't understand probability, if you asked them to estimate a probability based on the true facts you just gave, they might say 10%; I might consider them deceived, even if you weren't doing anything that would deceive a rational agent. This is a good point, much like Ends Don't Justify Means (Among Humans). One small caveat I'd like to make is that I don't think we are talking about science (or more accurately) scientists lying to the populace. I think what we are discussing are administrators, media outlets and policy makers that are informed of the numbers (it would probably be too generous to say informed of the science) choosing not to widely disseminate them in favor of reporting non-numerical information that might lead the populace to adopt an inaccurate belief about those numbers. Yes. Please elaborate, this seems to be the most interesting part of the conversation, and the way you're conversing now, I don't see why you're worried about a backlash. It would probably be wise to be as general as possible and not make any specific reference to a given countries system of healthcare (if that's where you're going), but if you have a real point I'd really like to hear it. Again, I wasn't purely speaking about false facts, more about choosing which true facts to announce, and in what way to achieve the desired impact. One doesn't have to li

0mattnewport15y

Ok, we've had two examples of negative externalities associated with infectious diseases. I brought up the free-rider / moral hazard problem of the costs of treating the disease not being fully born by an individual who engages in high risk activities. Wei Dai brought up the problem of an individual who gets infected as a result of high risk activities increasing the risks of others getting infected by increasing the incidence of the disease. Now if negative externalities are your true concern then you should address them as directly as possible. There are a variety of a variety of possible solutions to addressing negative externalities. Note that lying about the actual damage caused is not a standard solution. The moral hazard / free-rider problem is a general problem that affects healthcare as it is organized in most developed nations. A significant number of people consider this a feature rather than a bug however. If you actually wanted to internalize these negative externalities the most direct way would be to allow insurers (or the government, though that would be less efficient) to set their healthcare premiums on the basis of any relevant health or lifestyle information. While this happens to some extent (smokers or the obese may pay extra under various systems for example) it would be controversial in others (charging homosexuals higher premiums if they were at greater risk of contracting STDs for example). It would likely be even more controversial for conditions that are not generally perceived as due to bad personal choices (as smoking related illnesses or obesity are by many and homosexuality is by some on the Christian right). The suggestion that insurance companies might charge higher premiums based on genetic testing is widely regarded as unreasonable for example and I'm sure the same would apply if premiums for a government run system were so determined. Why such discrimination is considered outrageous for healthcare but is routine in some other

0datadataeverywhere15y

Thank you very much for your reply. I really did want you to specify more clearly what you were talking about. It seems obvious to me (now) that anyone following your line of thought would have understood what you were talking about from your earlier comments, but I didn't, and I hope you can forgive me for not understanding without further clarification. As an aside, smoking has an obvious externality in second-hand smoke, which is often directly regulated by outlawing smoking in certain areas. What are the negative externalities of obesity? If we are to believe some recent studies, fat people may make the people around them fatter, which is a non-obvious externality, but does obesity have any commonly-recognized effects on people other than the obese when not considering subsidized healthcare, or is it only considered to have that externality when healthy-weight individuals are contributing to the costs of obese individuals? This is at least possible with in some places regarding HIV, as well as pursuing criminal charges (examples). Even more off topic, but on topic with the much more inflamed discussion in the adjacent thread, while looking for that reference I found this. I was thinking at first that a man infecting 13 women would be in contradiction with the extremely low transmission numbers for HIV. Here are what my numbers look like: with (2006 estimates--I couldn't find number for late 1990's) 8e5 HIV+ men in the US, a 50-50 chance that one of them would infect 13 women requires that they each engage in an average of 2845 acts with a 0.08% tx rate (unprotected P/V), 762 0.3% tx rate acts (low-income P/V) or a mere 139 acts with 1.7% tx rate (unprotected P/A). The probabilities get much more complicated when dividing demographics, but while those are unrealistically high numbers, they aren't off by an order of magnitude; at least, the low-income P/V figure probably isn't. Actually pulling any information off that data point, given that it is singular and

2mattnewport15y

As I mentioned earlier, negative externalities are easy to dream up. Many people consider it legitimate to complain about negative externalities caused by ugly buildings (such as power plants, wind farms or architecture that doesn't fit with its surroundings) but complaining about the aesthetic negative externalities associated with unattractive people in public places is not generally considered legitimate. In practice in democratic society these issues are generally resolved by who can shout the loudest or wield most political influence and not by any direct rational accounting of costs. It is not clear for example that the relatively small risks associated with second hand smoke justify trampling on the rights of smokers to indulge outside of their own homes, especially given that smoking is already subject to large Pigovian taxes in most countries with such bans. It should at least be positive. It is not clear that it is in practice. It seems plausible to me that the general public distrust of government advice on risk that underlies phenomena like anti-vaccination movements is a direct result of an ongoing pattern of deliberately misleading people about risks. Overall I don't see a strong reason to suppose the net effect is beneficial.

0datadataeverywhere15y

Best point brought up yet. While to some extent I think that mistrust of authority is indefatigable, increasing the risk of that is probably much more costly. How do you feel about the specific example I mentioned, where the true risk of transmission of something is 1%, but the media outlet or whatever decides to omit the number and instead say something like "over the course of a week, an individual can spread disease X to over a hundred people", and while true, that convinces individuals that the specific risk is much higher than 1%?

0mattnewport15y

I personally find it a little irritating when the media omits information that would be necessary to work out actual risk numbers for myself. I don't object if they communicate the numbers in a way designed to have maximum impact on the typical human mind (it's been suggested that using frequencies rather than probabilities may help for example) but I do object if they leave out crucial information required to figure out true risk estimates. Of course I don't generally assume this is some grand conspiracy but rather reflective of the innumeracy of the media in general.

0datadataeverywhere15y

I don't believe in grand conspiracies because they just require too many contingencies. All this discussion, from my perspective, is about the potential for a tacit agreement between most (not all) of those disseminating information in various ways that the best method of talking about public risks is not necessarily to directly discuss low numbers associated with them. As I indicated earlier, I think that this agreement effectively already exists regarding influenza, and probably also HIV and other infections as well.

0[anonymous]15y

Strictly speaking, Wei never claimed anything about what we should do. (Even if everything he said is correct — and it seems obviously so to me — it's plausible that we're best off with a policy of authorities never lying about risks, due to unintended consequences, public trust, slippery slopes, &c.)

0jacob_cannell15y

I think you're on to something, but wouldn't that cause officials to overstate transmission rates rather than understate them? What is especially strange to me is that the government pushed a fear campaign for HIV and promoted as a dangerous STD for the mainstream hetero community, but neglected to double-check their official statistics, which rather clearly destroy the STD theory. Perhaps it's just an honest mistake, but I don't think so. From what I have read, they have spent time trying to get honest statistics. So they overpromoted the STD message, regardless of the actual statistics. Regardless of what HIV actually does or is, public campaigns to reduce needle sharing and reduce unprotected sex are probably net public goods. However, on the other hand, if AIDS is really caused by drug toxicity, then at least some people are actively being harmed by spending energy in the wrong protections. But I agree with your central point, and it applies to vaccines especially - they don't really have much of an individual benefit, but if enough people can be convinced to vaccinate, the entire epidemic can be curtailed or completely avoided.

7jimrandomh15y

That needs some clarification. Most people cannot distguish between a risk being somewhat low, and it being extremely low, so we need to be careful about the transition from numbers to qualitative divisions. The risks of being killed in a plane crash are so low that unless you're a pilot, you should ignore them; and overestimating the risks of flying would cause too much driving, which is more dangerous. In the case of AIDS, the probability of transmission may be "low", but none of the numbers given are so low that they would justify skipping any of the common safety precautions, so we shouldn't describe the probabilities involved as low in the presence of people who can't do the utility computations themselves.

3Relsqui15y

Topical.

0jacob_cannell15y

So as an addendum to this, I found this blog which I am now reading as it has an updated view from the HIV skeptic position. Here is a post analyzing a recent study showing that cocaine can cause AIDS CONCLUSION: Use of crack cocaine independently predicts AIDS-related mortality, immunologic and virologic markers of HIV-1 disease progression, and development of AIDS-defining illnesses among women The 1st commenter had an inside view which I thought was especially interesting: This certainly supports my eyewitness accounts of back in 1987 when several friends and acquaintances consciously or subconsciously decided that an HIV positive diagnosis surely meant death in 2 to 5 years, which led them onto severe crack and cocaine binges up until they were finally put on AZT monotherapy. Sad to say what the results were. Now they’re all part of the statistics. Can I say self-fulfilling prophecy?

2JoshuaZ15y

Cocaine is a drug that a) can damage the immune system and b) reduce appetite. It isn't at all unreasonable that HIV positive individuals who were heavy cocaine users would therefore progress to full AIDS faster than others. Note also that a 2-5 timespan isn't that far off from the timespan one would expect from HIV infection to emergence of AIDS given no treatment. Note that the study authors don't seem to think that this is at all unusual and nowhere do they claim that the use of crack cocaine was somehow causing AIDS.

0jacob_cannell15y

Sure, but it makes it more difficult to dissociate what the cocaine and virus are doing independent of each other, ie it makes it difficult to tease out the cause and effect from the correlation.

2JoshuaZ15y

Right, but that makes it just weak evidence. So it isn't very useful for the claim that cocaine somehow causes AIDS.

0jacob_cannell15y

Well not quite, because cocaine was causing AIDs-like symptoms before HIV was presumed to be around, even if the great rise of cocaine use in the 80's in the west happens to correspond exactly with the rise of AIDS.

2Wei Dai15y

What's your source for this? Looking at the cocaine statistics and AIDS statistics myself, I'm not seeing this correspondence. What I do see is that cocaine use dropped to a fraction of its peak level by 1990, while AIDS kept rising.

0jacob_cannell15y

Just go down to the graph in your first link and look at the emergency room drug mentions and see the cocaine serge - it does indeed correspond to the AIDS epidemic. Deusberg's 2003 paper has more on this, but it's in the data you linked. From the emergency room data it looks like cocaine use was still growing up to 2001, but my point was with the rise. AIDS peaked in the mid 90's from your data link.

0wedrifid15y

This much at least is something that should be relatively easy to confirm to a reasonable level of satisfaction. It would seem to require only a microscope, as syringe and a sufficient sample of people with AIDS. Has anybody ever founds someone with AIDS who did not have HIV? If not is that because nobody has bothered to take a close look? If so then I would certainly support your questioning of the standard of research supporting the mainstream position.

2kodos9615y

According to skeptics, yes, in all but name. The standard skeptical argument is that AIDS, as it is currently defined, includes HIV+ as a necessary diagnostic criterion, and that this is a circular definition: if someone presents with all the symptoms of AIDS, but tests HIV-, then they are defined to not have AIDS. This means that 100% of diagnosed AIDS patients will be HIV+, just by definition, not due to a meaningful correlation. The skeptical position here is that you can't actually see a virus with an optical microscope (which I believe is true), and "HIV tests" are actually just testing for HIV antibodies (or substances alleged to be HIV antibodies), not HIV itself. I'm not endorsing these positions, just passing them along, btw.

1wedrifid15y

Thankyou. I'm not familiar with the subject and wanted the information.

2jacob_cannell15y

Yes, unless one defines AIDS as a collection of symptoms plus HIV. I forget the name, but any definition of AIDS which does not include HIV has some HIV-. Now on the other hand, AIDS is a syndrome of immune supression, not a disease, so there of course could be other things that cause a similar syndrome. If you are interested in venturing down that rabbit hole, read a little Duesberg: (warning: reading his papers may result in general increased skepticism about the medical establishment) http://www.duesberg.com/papers/chemical-bases.html

3wedrifid15y

Now that is a familiar mistake - and a negligence that does real damage. I'm more familiar with the mental side of the medical establishment so have seen, for example, ADD symptoms lumped together and medicated presuming there is only one distinct etiology. Looking at a brain scan could tell them the difference easily enough. That would be surprising - purely because my existing skepticism is already significant. All the more so after I spent some time involved in medical research myself. Scary stuff. "Hang on, wait. you want me to do what with the data?"

0Perplexed15y

I find denialism in all forms simply fascinating. I wonder if you could indulge my curiosity. You find your arguments completely convincing. Yet they are based on publicly available statistics and rather obvious and common-sense kinds of reasonings. So, I have to wonder, what do you think is wrong with the cognitive apparatus of all those medical and research professionals who believe that HIV == AIDS and is an STD? Why don't they reach the same conclusions as you? Are they stupid? Just haven't thought of the train of thinking you use? What are your guess as to where they are all going wrong? Why none of them has realized the simple truth and shared it with colleagues? Incidentally, I have a hypothesis as to what is wrong with your reasoning, which I will share on request, but I really want to understand how you reconcile your own certainty with the opposing certainty of people who (on paper) seem far better qualified on this subject.

[-]jacob_cannell15y100

I find denialism in all forms simply fascinating. I wonder if you could indulge my curiosity.

I will indulge your curiosity in a moment, but I'm curious why you use the politically loaded term "denialism".

As far as I can tell, it's sole purpose is to derail rational dicussion by associating one's opponent with a morally perverse stance - specially invoking the association of Holocaust Denialism. Politics is the mind-killer. In regards to that, I have just been spectating your thread with Vladimir M, and I concur wholeheartedly with his well-written related post.

There is no rational use of that appelation, so please desist from that entire mode of thought.

So, I have to wonder, what do you think is wrong with the cognitive apparatus of all those medical and research professionals who believe that HIV == AIDS and is an STD?

Firstly, I don't think nearly as many quality researchers believe HIV == AIDS as you claim, at least not internally. The theory has gone well past the level of political mindkill and into the territory of an instituitionalied religion, where skeptics and detractors are publically shamed as evil people. I hope we can avoid that here. Actually, I... (read more)

0Perplexed15y

Actually, though you may not believe me, Holocaust denialism hadn't even occurred to me. In the portion of the blogosphere that I follow, it applies most frequently to AGW denialism, with the AIDS denialists second, evolution denialists third, and the anti-vaccination crowd getting an honorable mention. The wikipedia article on HIV that you reference has a section entitled "AIDS Denialism". But now that you mention it, why do you consider Holocaust denialism morally perverse? I thought that questioning PC conventional wisdom was considered a Good Thing here. No, I don't believe I do. I wouldn't want to further offend you. My hypothesis is pretty simple. You are using the wrong numbers. When I Googled, the first few hits I found suggested 0.3% per coital act as a lower bound on heterosexual transmissibility with the risks increasing by 1-2 orders of magnitude in case of genital ulcers and/or high viral loads. I don't think that it is particularly difficult to understand the epidemic spreading in Africa as an STD when these higher numbers are used. I did look at this study providing smaller numbers and this paper critiquing it, as well as this abstract mentioned in the wikipedia article. It was pretty clear to me that the kinds of low numbers you were using to argue against HIV being an STD are actually based on monogamous couples who are regularly examined by physicians and have been instructed in the use of condoms to prevent transmission. Those numbers don't apply to the most common cases of transmission, in which ulcers and other factors make transmission much more likely. That is the hypothesis I was going to offer. When you suggested that you only had a 20-30% level of doubt of the orthodox position, I simply had no idea that it was such a strong and assured 30%.

3jacob_cannell15y

Please see my response below concerning the perjorative "Denialist", and why such perjoratives have no place here. You haven't offended me. The google hits you mention are just websites, not research papers - not relevant. There is no reason apriori to view the ~0.3% per coital act transmission rate as a lower bound, it could just as easily be an upper bound. You need to show considerably more evidence for that point. The data on wikipedia comes from the official data from the CDC1, which in turn comes from a compilation of numerous studies. I take that to be the 'best' data from the majoritive position, and overrides any other lesser studies for a variety of reasons. This may be 'clear to you', but the Wikipedia data comes from a large CDC sponsored review considering aggregates of other studies to get overall measures of transmission. This is the orthodox data! I highly doubt it has the simple methodological errors you claim. And even if you did prove that it does have those errors, then you are only helping the skeptic case - by showing methodological errors in the orthodox position, and the next set of data should then come from the heterodox camp.

The google hits you mention are just websites, not research papers - not relevant. There is no reason apriori to view the ~0.3% per coital act transmission rate as a lower bound, it could just as easily be an upper bound. You need to show considerably more evidence for that point.

If I can take the liberty of butting in...

The data on wikipedia comes from the official data from the CDC1, which in turn comes from a compilation of numerous studies. I take that to be the 'best' data from the majoritive position, and overrides any other lesser studies for a variety of reasons.

Here's the table of data I believe you're referring to:

It cites references 76, 77 & 79, all of which turn out to be publicly available online. That's good, because now I can check the validity of Perplexed's claim that the studies backing your CDC data used samples of relatively healthy, well-off people who lack some risk factors.

I took ref. 76 first. It reports data from the "European Study Group on Heterosexual Transmission of HIV", which recruited 563 HIV+ people, and their opposite-sex partners, from clinics and other health centres in 9 European countries. (Potential sampling bias no. 1: HI... (read more)

1jacob_cannell15y

By all means. While you were writing this I was reading 76 and writing my own reply. Ah I hope he wasn't claiming this, because they certainly were anything but healthy, with around 20% being hardcore IV drug users, 23% transfusion recipients, 10% hemophiliacs, and overall high rates of STD's. I noticed the discrepancy about sexual history part but I didn't see how to factor it. They are relying on self-reporting to make any of these links. Err, no, because his characterization is based on the idea that these couples were using condoms frequently, but the study specifically shows that is not the case - see my reply to Perplexed. If you want to characterize this study, fine, but don't pretend that these are "regular couples regularly using condoms" - that is not what the study claims. And finally this is the orthodox data. I mean if you want to reject ... ok .. and then we move to searching for other data which supports our relative positions . . . to the extent we have positions. Perhaps it would be best to agree what the ideal study would be and precommit to that ideal in a sense? And then we could look for other studies that may be closer. Of course in the real world they will rarely be clear cut.

4satt15y

I quite agree that they weren't healthy in absolute terms; I just meant that they were relatively healthy for people with HIV. Compared to HIV+ people in much of the rest of the world, especially sub-Saharan Africa, I'd expect this European study's subjects to have (on average) better nourishment, better healthcare, stronger immune systems, less exposure to infectious disease, much less exposure to parasites, and a far lower rate of promiscuity & prostitution. I should've been more explicit that that was the sort of comparison I had in mind. Looking at your reply, I think we disagree about whether or not Perplexed was hinting that the couples were consistently using condoms. I didn't think Perplexed was implying anything more than that most cases of HIV transmission involve people who weren't regularly reminded to use condoms. So I took his statement at face value, in which case it's surely true (unless European doctors have come up with a way of counselling couples "about the risk of HIV infection and safer sex practices" that doesn't involve advising condom use!). I don't believe Perplexed or I are pretending that these are normal couples who continually use condoms. I think it goes without saying that these weren't "regular couples" — after all, "regular couples" don't visit hospitals and clinics to get HIV infections checked out. Whom are you quoting? I reject your interpretation of the data, rather than the data. The study probably gives a fair idea of transmission risks among faithful Western couples living in the 1980s/90s who regularly see doctors...but for that reason (among others) it's likely to underestimate transmission risks in other demographics. I think you're probably right on that point. I suspect that looking at per-sex act transmission risks isn't going to be very enlightening about whether or not HIV causes AIDS. It would be better to have data from * previously healthy people * who were accidentally infected with HIV * and don't engage

0jacob_cannell15y

I generally agree with most of this except perhaps the last part - I doubt that promiscuity and prostitution varies that much from 3rd and 1st world. I think he was implying that the reminders led to condom use, but this was in fact not the case according to the study itself (possibly because they excluded many of the condom users, some aspects of the study's design are not all that clear to me). Not quoting, just paraphrasing. He was implying that the heterosexual couples receiving counseling were not indicative of a typical HIV hetero population, but the study designers of course realized that and were at least attempting to gather representative data. Ok, whether HIV causes AIDS is a larger topic. My original point was just about the orthodox claim that HIV is sexually transmitted, which I believe is rather obviously bogus according to the orthodox's own data. I hope you can see how the orthodox could go wrong there and some of the political factors at work. As to the larger HIV == AIDS issue, I largely agree with your ideal data criteria, but one potential issue is whether we are comparing HIV to the null hypothesis or to some other hypothesis. I don't think any reasonable skeptic claims that HIV is not at least correlated with AIDS - Richard Gallo may be many things, but he is probably not stupid nor a charlatan. So it would be better to compare the orthodox HIV hypothesis vs the Drug/Lifestyle Hypothesis (which predated HIV). Some immediate concerns are that one must take care to then define AIDS reasonably without circular reference to HIV (which precludes some data) The next concern would be that either way, previously healthy people don't get HIV or AIDS, in reality or according to either theory. The risk groups are all unhealthy in various ways. All that being said, Duesberg does indeed provide data very close to what you are proposing. There are some groups of HIV+ who, for whatever reason, have refused mainstream treatment. There aren't many such

1satt15y

Quite possibly, but note that I was comparing the subjects in the European study with the rest of the world, rather than all of the 1st world. The study's screening procedures probably cut out quite a few people who have a lot of sex. I think I do. (I hope I do!) Still, I do see the orthodox belief that HIV can be transmitted sexually as being compatible with the CDC numbers. The CDC transmission rates are surely below the average real-world HIV transmission rate (due to the nature of the European study sample), and there are features of the data that are easier to explain if we acknowledge that HIV's sexually transmitted: the condom-using couples had lower HIV transmission rates than the non-condom users, men who (claimed to have) had period sex with HIV+ women were at higher risk of transmission than men who (claimed to have) avoided period sex, and so forth. So I continue to disagree that the HIV-is-an-STI view is "obviously bogus according to the orthodox's own data". It might even preclude the Duesberg/Koehnlein data you link. Page 402 says the study's of "AIDS patients", and it's not clear from the immediate context what definition of "AIDS" was used for the study. All 36 of the patients (you were right about the study's size!) are listed as being HIV+, which suggests to me that the AIDS diagnoses were made (at least partly) based on HIV+ status, as is standard practice. I would have thought that healthy people are capable of getting HIV? Getting pricked with an HIV-infected needle works, as does sexual transmission. A lot of people in high-risk groups are unhealthy, of course, but there are surely unlucky people who get HIV without prior major illness. I looked up long-term nonprogressors on Wikipedia (not the most reliable source, but anyway), and it looks like many long-term non-progressors have genetic traits that make them better able to resist HIV, or have a weaker form of HIV. I also saw that the group in Germany Duesberg's talking about all come f

0jacob_cannell15y

I would have given more creedence to this view at the beginning of this whole inquiry, but in another branch several other posters found some large meta-analysis studies, and low and behold they confirm and agree with the old CDC European study. I discuss that here Of note is that the infection rate in 1st world countires agrees with the original CDC European Study, and the infection rate in Africa/3rd world appears to be 3-6 times higher. Metastudies which mix 1st and 3rd world results get rates somewhere in between. Some of these metastudies were of thousands of individual studies, and say what we will about them, I think they nail down the real world transmission rates, and the 1st world rates are just as low as I originally quoted (or lower) Effects like this surely can increase transmission rates in specific instances, but for epidemilogical modelling we are interested in the average rates - and note as I analyzed elsewhere, the original CDC European study does attempt to control for condom use - it intends to show infection rates for unprotected sex. I don't think you can so easily dismiss all these studies and the work that has gone into computing these transmission rates. This is certainly a possibility and fits what we know with viruses - variable genetic resistance is to be expected. However, what is important is how one samples and when. If you take a sampling of survivors years later, then sure you can expect to be finding survivors due to genetic resistance. But if you sample a subset based only on the criteria that they refuse medication after testing seropositive, then that is a very different sampling, and you should expect it to be largely uncorrelated from genetic resistance (unless you want to argue that people with genetic resistance are strongly expected to resist medication!, but I hope you won't take that route) You do bring up a potentially valid criticism: Possibly, but I don't find a reason why we should expect this without specific

1satt15y

All of this is consistent with the CDC statistics underestimating the general transmission rate. You write that the rate estimated from the European study "agrees with" meta-analyses of 1st world data, and that the 3rd world rate estimated by meta-analysis is higher still. So pooling the two meta-analytic results gives a global average rate greater than the 1st world average rates, i.e. averages greater than the CDC rates. I don't think I am dismissing these studies and the work. The bit of my comment you're quoting refers, after all, to secondary analyses in one of those studies. The point I'm trying to make by drawing attention to those analyses isn't something like "look, the transmission rates are higher if you don't use condoms, clearly they're high enough for HIV to spread through the population", but instead "associations between condom use and transmission rates, and between sex during menses and transmission rates, have a far higher likelihood in a model where HIV is an STI than in a model where it's not". It's much easier for me to explain why having sex with a woman at particular times in her menstrual cycle would correlate with HIV transmission if I presume HIV's sexually transmitted, which I interpret as evidence for [edited: I had a brain fart and originally wrote "against"] the view that HIV's an STI. Don't worry, I'm not. I'm suggesting that because the sampled people all come from the same small geographic region, it's possible that genetic resistance and/or weaker HIV variants are more common among them. The specific evidence I have in mind is the geographic restriction of the sample. A group of people from one place will tend to be more genetically similar than a worldwide sample, and will be more likely to share strains of a disease. I expect HIV-1 variants do spread diffusely in subgroups, but I don't think that rules out my point. Particular alleles of genes spread throughout humanity, but spatial proximity still correlates with genetic simi

0jacob_cannell15y

For a variety of reasons, I find it useful to separate the two, and the 1st world rates are the most important - the virus outbreak started in San Francisco essentially (following the end tail of the massive hippie/drug liberation social experiment). Also, the 3rd world rates are suspect in general, as one of the meta-studies notes, for a variety of reasons. And regardless, even the 3rd world rates are 30 times lower than typical STD's, even if they were accurate (which is dubious). Yes, but as you admit, So at this point I think it is more time profitable to switch gears and spend a little effort investigating other LTP reports other than this single study. And just a little google searching shows that there appears to be now a number of other LTPs from across the world that are similar to the Koehnlein group - and avoiding traditional treatment appears to be a common link. You can google it as well, but here are some links: from an article in Health Care Industry (older - 2000): A 2005 NYT story about a LTNP: And finally here is a compilation of another dozen studies or cases of untreated LTNPs (older hasn't been updated recently) So it doesn't look like the Koenhnlein study is an isolated incident. I am still looking for more recent studies or follow ups. From everything I know so far, the vast majority of patients were treated, so if treatment has a beneficial effect at all, then it follows that the ratio of treated LTNPs to untreated LTNPs must be equal or greater to the original treatment ratio. I understand that in the west that treatment ratio was very high, probably > 95% And as far as I can tell, we aren't seeing anything like that ratio in LTNPs, so this could be very strong support indeed for at least part of the Deusberg hypothesis: that the treatment can itself cause the disease. Edit: I completely guessed on that 95%, and later found this telling quote in the NYT article (I am reading these as I go): But what it would really need is a big lo

1satt15y

As I understand things, HIV jumped into the human population in Africa decades before hippies and the 1960s counterculture, and that only after being established in West/Central Africa did it reach the US. As such, the 3rd world transmission rates have just as big a role to play as 1st world transmission rates. With an external pool of infected people established, it became possible for HIV to be reintroduced to the US over & over again until it landed in US subpopulations that spread it with needle sharing & frequent anal sex. Without being more specific about what's wrong with the rates, I'm not sure why this means the 3rd world rates are necessarily about equal to (or less than) the 1st world rates. At any rate, HIV is not a "typical STD", and a lower transmission rate than other STDs doesn't mean much as long as HIV's rates are sufficient to enable its spread. Also, Wei_Dai suggested that the P/V sexual transmission rate for HIV is comparable to that of genital herpes, a point you didn't seem to dispute in your reply. Do you believe that genital herpes has too low a transmission rate to be an STD? But here's the thing: the lone fact that a case report or study has some LTNPs doesn't necessarily mean much in terms of questioning the HIV-AIDS link. For example, the studies in the 2000 Research Initiative/Treatment Action! article (I think "Health Care Industry" is just the name of the section on findarticles.com where the article's mirrored) seem to focus on gathering together people already known to be treatment-refusing LTNPs, and finding out what makes them LNTPs. Simply observing that treatment-refusing LTNPs exist doesn't convince me. Even if 99% of HIV+ people progress to AIDS within some time frame, with so many HIV+ people there are going to be a lucky few who turn out to be treatment-refusing non-progressors. By contrast, Koehnlein's methodology seemed to be different, which was why I initially thought that work might be compelling. I'd assumed that Ko

0jacob_cannell15y

Ah, unfortunately this got too long, so I had to split it. I think this was a confusion of terminology, and "AIDS patient" in the general sense was used to just refer to all HIV+ patients he was treating. It did not refer to only a subset that had later stage 'AIDS' symptoms. At least, that's how it read to me. From what I understand, Koehnlein somehow found a way to treat patients without antivirals legally, so patients seeking non-antiviral treatment came to him. His 'study' is just a record of all such patients, when they first came under his care, their backgrounds, and eventual prognosis (a couple of deaths out of thirty or so patients so far)., Koehnlein may subscribe to the Duesberg hypothesis, and as such wouldn't place any special value on persistent tracking of CD4 counts.

1satt15y

It might be for the best! This splits the Koehnlein study discussion and the general HIV discussion into their own separate subthreads. Yes, we initially read the phrase differently. I originally interpreted it at face value, figuring that in a review article about HIV & AIDS, D&K would take care to avoid confusing having AIDS with being HIV+. I now think I might've given them too much credit. Nonetheless, at one point, D&K must be using "AIDS patients" with its narrow meaning (patients with AIDS proper) and not its informal one (patients with HIV who may or may not also have AIDS), because the statistics they quote for German AIDS patients match the Robert Koch Institut's AIDS statistics, but not the organization's HIV+ headcount. Whatever D&K's intentions or confusions, my earlier point that the study can't provide strong, simultaneous support of all the conclusions drawn from it still stands. If so, Koehnlein's testing his (her?) own definition of AIDS, not an orthodox one, and all bets are off.

0jacob_cannell15y

That's a theory, but it has some critical flaws. Namely one must wonder why did it not spread via prostitutes, needle sharers and blood transfusions earlier? Condom use dropped with the adoption of the pill in the 1960's and the sexual liberation opened up a hetero transmission channel which has about the same net transmission rate (always limited by the insertive step). AIDS became an epidemic in San Francisco in the early 80's, and it grew quickly from a handful of cases to effect a large portion of the gay population, and was closely correlated with a diverse number of fundamental lifestyle differences. It is this phenomena, this quick sudden outbreak in a very specific subgroup, which I find extremely difficult to reconcile with the transmission data. Tops and bottoms tend to specialize so the rate-limiting factor for expansion in the gay community would be closer to the insertive A rate, at around 0.06% vs receptive at 0.5%. Needle sharing is about 10 times more effective, and blood transfusion is around 1,000 times more effective. But all the early cases are in this one specific group, which is not even the highest risk group. I mean, the transmission rates for insertive V and A are about the same, and there are far more heteros than homos, so it just doesn't make any sense. And don't tell me the homos are having all the sex - they may be having more individually, but not when considering prostitutes and sexually liberated women in the 60's and 70's, the fact that heteros have anal as well, and the 100 to 1 hetero to homo ratio. The overall hetero transmission channel is much much larger, especially after considering needle sharing and transfusions, and yet the disease only appears in the homo subgroup, time and time again. Why? Ignore for a second all high level conceptions about HIV. Don't privilege the orthodox HIV hypothesis, instead compartmentalize and consider just this evidence concerning transmission rates, and how that evidence should cause one to

4satt15y

Looks like I have to split a comment too! I'm not sure which specific time period you're referring to with "earlier". If you're talking about the 1970s, I'd guess it's because HIV simply hadn't been introduced to those subpopulations often/early enough to stick. If you're talking about the early 1980s, well, it looks like HIV did spread, at least among needle sharers and people who had blood transfusions. (I haven't seen data on prostitutes.) According to this 1985 Science article, 12,932 AIDS sufferers were reported to the CDC by August 30, 1985. 1.5% of them had received blood transfusions within 5 years of diagnosis, and 17% were heterosexuals who'd used IV drugs. (Also, 12% of the homosexual & bisexual men diagnosed were IV drug users.) Although I'm sure tops & bottoms "tend to specialize", I doubt men with dozens of sexual partners are completely picky about which role they play. If men are inconsistent about being the top/bottom, the insertive anal transmission rate is going to be an underestimate. In fact, it's likely to be an underestimate twice over, because preexisting STIs make transmission more likely, and promiscuous men will have more STIs on average. You've also got the handful of IV drug users among these men. If I'd had to bet on where HIV would rear its head first, the bathhouse subculture would've been a great choice to put my money on. But neither of those can have an impact until HIV's introduced to the subpopulations of needle sharers/blood donors, and even after that, their effect will depend on when HIV reached those subgroups, and how many people in those subgroups start off with HIV. Only if you define "early" as really early: the first reports of IV drug users with AIDS came out in the same year as the first reports of AIDS in gay men. And again, risk isn't everything. Even if group X has much higher transmission risks than group Y, if the virus reaches group Y first, the earliest infections are likely to emerge in group Y. While the

1jacob_cannell15y

There are several documented cases of early AIDS where we have stored tissue that later tested positive for HIV, such as the gay teenager who died in St Louis in 1969. This poses a serious problem for the standard theories that HIV is transmitted horizontally (unlike any other retroviruses) and presumably came out of Africa. So how did it get into this teenager? You would need some world travelling gay subculture at the time to link the disease to Africa, but not a big enough subculture to create an epidemic. Since only a small portion of men are gay, we should expect that if it came out of Africa the first vectors would have been heterosexual, not homosexual. And as there are several of these strange early cases, it would have had to come over from Africa multiple times, but always only in gay men. This theory is just not salvageable. Also, consider that the different genetic subtypes are closely associated with particular risk groups - subtype M appears in MSM and IV drug users but not others, which doesn't make any sense for a horizontally transmitted viral vector. Most of the subtypes are linked to particular geographical regions in Africa, which points to a long history in humans (with M representing a novelty linked to novel behaviour). Also consider that all other primates have naturally occurring lentivirus family retroviruses very similar to HIV. Consider that the entire family of retroviruses are more symbionts than parasites - humans are 'infected' with thousands of different retroviruses, many of which are integrated into our genome, and they have functional roles in gene expression and even the formation of the placenta. So if all other primates have naturally occurring lentiviruses, why don't humans? There is a clear evolutionary niche for a lentivirus in mammals, and it seems odd that homo sapiens somehow lost their naturally occurring lentivirus at some point in our evolutionary divergence, only to re-acquire it very recently in the last one hundr

1satt15y

But most of the earliest confirmed AIDS cases were retracted (David Carr) or have a direct connection to Africa (anonymous Congolese adults & Arvid Noe). It's only Robert R. (the "gay teenager" you refer to) who didn't have a direct African connection, but that doesn't mean there wasn't one, and such a connection wouldn't even be necessary with Haiti available as a closer source of HIV. This is one teenager. He only had to be unlucky once when having sex with just one infected man. I'm confused. This PNAS paper presents good phylogenetic evidence that the HIV strains causing the North American epidemic came from Haiti, and that Haiti's HIV came from Africa in the 1960s, which "suggests its arrival in Haiti may have occurred with the return of one of the many Haitian professionals who worked in the newly independent Congo in the 1960s". So it's Haitian economic migrants who would've been the "first vectors" to carry HIV out of Africa in any real number, and I have no reason to think they were disproportionately homosexual. If by "strange" you mean that all those cases are inexplicable, I disagree. Not at all. I'm not an epidemiologist (or a geneticist), but couldn't that just be a founder effect perpetuated by MSM and IV drug users transmitting HIV much better amongst themselves than they transmit it to everyone else? I'm not seeing why this would be evidence for/against orthodox theories of HIV & AIDS. (And if I were being pedantic, again I'd suggest the relative insularity of MSM and injecting drug users as why subtype M's linked with them, rather than the novelty of their behaviour as such.) I'll pass on commenting on your last three paragraphs, since what I know about retroviruses would fit on the back of a postage stamp. I will try checking Padian et al. again, though.

2satt15y

Yes, I don't know any other studies that used direct drug tests. There is this Madrid study of heterosexual transmission that used indirect testing of "markers related to drug addiction (e.g., hepatitis C serology)" to check for drug use in addition to questionnaires, and its recorded transmission rate is quite high: 26%, out of 38 couples. However, it looks like a retrospective study. Although quantity of sex doesn't seem to have made much difference, the unadjusted odds ratio associated with not using condoms bordered on statistical significance with a confidence interval of 0.95 to 3.01. After adjusting for the number of sexual contacts, that odds ratio went up to 2.1, becoming significant and equal to the adjusted odds ratio associated with anal sex. I notice too that after adjustment, STI history — a risk factor elsewhere associated with HIV transmission — was the risk factor with the highest odds ratio. It would be more interesting if the relevant sample did contain 600 heterosexual non-drug users followed for the length of the study. However, the "no seroconversions" result comes from the study's prospective part, which involved only "175 HIV-discordant couples over time, for a total of approximately 282 couple-years of follow-up [...] attrition was severe". That's a mean follow-up time of only 19 months per couple. Most couples probably got less follow-up time than that, because severe attrition would tend to negatively skew the follow-up time distribution, depressing the median. That's not all. The investigators didn't just counsel the couples on safe sex, but also set up a 24-hour telephone support line, a newsletter and regular meet-ups. These measures were very effective in changing the couples' behaviour: by the final follow-up, 15% of the 175 couples abstained from sex and 74% used condoms. So, at final follow-up, only 11% of the couples — nineteen in absolute terms — were at substantial risk of HIV transmission. The study's statistical power to det

0jacob_cannell15y

Ok, so concerning the Padian study, I believe you have misread it, and I am to blame because I originally mislabeled it when I said: I was wrong - it was not in fact a study of sero-discordant couples, and I apologize for that mistake, for it seems to have mislead you. This is evident in the abstract and is explicitly clarified elsewhere: So they recruited non-drug using heteros of either sex who had steady partners, and some fraction of those partners were already infected - specifically 19% of the female partners and 2% of the male partners. They explicitly state there were no new infections in the abstract: And again later: The last part happens to be in the 'prospective results' portion (where they switched focus to only serodiscordant couples), but should not be interpreted to somehow only apply to the post 1990 time period - as it says "after entry into the study", and agrees with the "no new infections" in the abstract. So I believe you have misinterpreted when you say: There was not a single new infection (seroconversion) during the entire ten years across all couples. You are correct that they were counseled on condom use, but this does not seem to have influence actual reported condom use, even though the percentage of couples using condoms upon entry increased from around 50% to 75%, a significant fraction reported inconsistent usage: So the key of this study and really all of the heterosexual transmission studies is that it is all just guess work. The 19% female and 2% male seropositive partners were just assumed to have acquired HIV sexually, but as this occurred prior to the study, the drug controls were not in place (and testing didn't start until 1990). They are just assuming sexual transmission in these cases based on the strong unfounded assumption that HIV is sexually transmittable, they really have no idea. What they did show, was that over about 4000 couple years, there was not a single new transmission in this drug-screened hetero samp

2satt15y

Alright, I've had another look at Padian et al.'s paper. I did follow your lead in thinking the study was solely of sero-discordant couples, and I agree with you now that it actually wasn't. However, the relevant part of the study is the prospective sample, which was solely of HIV-discordant couples, so my overall interpretation of the study's nominally zero transmission rate remains unchanged. Right, but Padian et al. refer to the retrospective part of the study as the "cross-sectional" part (see page 351: "the retrospective nature of the cross-sectional aspect of our design"), implying there was no follow-up in that part. Without follow-up, they couldn't have detected seroconversion after entry into the study, so the retrospective method wouldn't pick up on post-entry transmissions however often they actually happened. So it's a mistake to argue that no transmission happened in the retrospective group on the grounds that no transmission was observed, because the retrospective method can't detect new transmissions. So, although it's technically true that the "no seroconversions" result applies to the whole period of the study, it would've been disingenuous for me to say so, because the researchers could only spot new seroconversions from 1990 onwards. The study's non-prospective part was incapable of detecting post-recruitment seroconversions, so the basis for the "no seroconversions" result is indeed the prospective part. Well, there might've been new infections between 1985 & 1996 among the retrospective sample, but the study wouldn't have detected those. The retrospective method would only detect those that happened in a couple before they entered the study. The fact that a minority of the couples still didn't use condoms consistently at their final follow-up doesn't mean the counselling had no influence! Only for the retrospective sample. The calculations you do in the next couple of paragraphs seem to be based on both the retrospective & prospective sub

0jacob_cannell15y

There was follow up starting in 1990. Basically they started the study by recruiting in 1985 and were originally focused on the retrospective aspect. Couples come in, fill out a questionairre and they attempt to screen out drug users and look for patterns in infected partners (couples where both are seropositive). Then in 1990, they began bringing couples in for examinations and follow up tests - "Physical examinations for both partners were initiated in 1990." This is the beginning of the prospective part, but it doesn't mean it is only valid for couples starting after that date, which comes from the very first sentence of the abstract: The prospective part was not limited to only couples enrolling after 1990. That is only the beginning of the biannual checkups. Technically yes, but immediately in 1990 on the first follow-up they would have spotted any new seroconversions in any serodiscordant couples currently in the study. And they would have clearly mentioned any such detected seroconversions, and indeed they clearly mentioned that they detectected exactly zero. But regardless, yes I did botch the total couple-years. The text implies that the 175 couple group with follow up was the total set of persistent couples: Abstenence was between 0 and 14.5%, consistent condom use betwen 32% and 74%, and anal between 37 and 8%. We should probably also factor in that condoms are not 100% effective. Lets ignore that for a second and assume midpoints of the above numbers, with around 100 sex acts per year, or 10 per month. If 'inconsistent' means ~50%, I get ~2 unsafe acts per month, or ~6,000 unsafe acts. The majority of these couples (80%) are male seropositive, so the higher M->F numbers apply 0.1% to 0.5%. The infection rate for P->V is supposedly 10 in 10,000, or 0.1% according to the CDC from the European study. The infection rate for P->A is supposedly 50 in 10,000, or 0.5%. Sure it would have been better with 10,000 couples over many years, but how much specifi

0satt15y

What you're saying about Padian et al. just clicked for me. I'd got it into my head that they were allocating each recruited couple to either a retrospective track or a prospective track, with couples switching from the retrospective track to the prospective track after 1990, but without their data being carried over. But you're saying the couples already enrolled in the study pre-1990 were automatically entered into the prospective part with their earlier questionnaires & lab work retained, right? That'd make more sense than how I first interpreted the paper. (Serves me right for skipping the abstract and diving straight into the methods section. Twice.) Alright, let's crunch some numbers. Using the midpoints of the ranges is a good first guess in the absence of other information. However, I think there's a good reason to use estimates much closer to the follow-up percentages (14.5%, 74% & 8%) than the baseline percentages (0%, 32% & 37%). The prospective results section says that "approximately 97 percent of behavior change was reported between baseline and the first follow-up visit". That is, when couples started/stopped using condoms consistently (or abstaining, or having anal sex), they almost always did so before their first follow-up visit, suggesting the couples changed their behaviour shortly after counselling, not gradually. If so, then the baseline percentages only represent the couples' behaviour distribution for a few weeks; after that, the distribution would be much better represented by the rates at final follow-up. This could have a big impact on the surprisingness of the zero seroconversions result. Switching from the midpoint rates to the final follow-up rates boosts the abstinence rate to 14.5% and the consistent condom use rate to 74%, while cutting the "[a]ny anal intercourse" rate from 22.5% to 8%. So the neither-abstinent-nor-consistently-condom-using rate sinks from 39.7% to 11.5%. Carrying that forward, I get an unsafe act count ≈1600 ins

0jacob_cannell15y

Also note this 'ingenious' study has an important point of evidence that does agree with the Padian study and points to a non-sexually transmitted etiology: The circumcision studies from Africa don't tell us much either, as circumcision is associated with ethnic/cultural groups and thus drug use and other factors. There are other more parsimonious explanations that don't rely on a 'virus' at all, and HIV - as far as it exists as a rather arbitrary collection of DNA/RNA sequences, may not be a virus at all. It could just as easily be trash RNA being secreted in exosomes tagged for immune system garbage collection. It could be regulatory RNA exosome messages intended for other cells. It could be a mutant form of an endogenous regulatory RNA exosome. It could be an endogenous 'virus' that forms as a cancer-like mutation of normal endogenous regulatory RNA exosome communication. And yes, it could actually be a true exogenous RNA virus that just happens to be remarkably similar to sequences embedded in the human genome (such as the so called "HERV" sequences) - and just happens to look like typical RNA exosomes in the microscope. But even if it is a true exogenous virus that can jump between cells and that is a huge if, there is astonishingly little evidence it causes much harm. There is a mountain of evidence that drug use causes harm, and specifically that particular drugs linked especially to the gay community cause specific types of chronic accumulated immune damage. Methanphetamine (speed) and its derivatives for example is tightly correlated with HIV/AIDS, it is endemic in the "party and play" gay community, and we have a large amount of evidence that Meth does significant long term harm. Firstly, Meth is a hyper-stimulant of the "flight or fight" stress response. This stress response essentially temporarily shuts down the immune system and digestion to focus the body on a temporary threat. A lion may kill you in a matter of minutes, while your body's normal

1satt15y

Flicking to page 354 of Padian et al. I see this: "the practice of anal sex and lack of condom use have remained strong predictors of transmission since the beginning of the study". And table 2 of the paper suggests that not using condoms is a statistically significant transmission risk factor, after adjusting for number of contacts (as I previously mentioned). I would not interpret that as agreement with a finding of "no significant difference". This argument might go through for observational studies, although even there I'd want a quantitative argument for why I should expect those confounders to have as strong (or stronger) an effect as circumcision's apparent effect. But I also referred to three large randomized trials, and randomization reduces the association between confounders and treatment effects to statistical noise — that's why people conduct randomized trials. So I still regard the trials as strong evidence for circumcision having an effect on HIV transmission; confounders don't have a substantial effect on the results of randomized trials unless the randomization process was faulty. (Incidentally, my original links to the 2nd & 3rd trials are now broken. Here are alternative links, although they may be paywalled.) I'm skipping over the paragraphs on whether or not HIV is a virus since what HIV is specifically doesn't bear on the point I was trying to make, which is that cofactors can't subtitute for exposure to HIV (whatever one thinks HIV is). Agreed. Even if I grant you that, it doesn't mean much to me unless one of those "specific types" of immune damage is the massive reduction in CD4 cell counts characteristic of AIDS. There are different kinds of immunosuppression, and it won't do to presume that because something causes one kind of immunosuppression, it causes the kind of immunosuppression associated with AIDS. At any rate, I suspect properly accounting for HIV+ status eliminates the link between whichever drugs you have in mind and AIDS.

6wedrifid15y

You people are still going at it on the HIV thing?

2jimrandomh15y

This thread is now at 4.5*10^4 words (counted by copying into a text editor, and find/replacing to delete words that're actually headers or vote/navigation links). I believe it should have been taken off-line at approximately the 5*10^3 word mark, if not sooner.

0jacob_cannell15y

I admit I'm obsessively addicted. The more I look into it, the more I have found that HIV science has gone horribly, horribly wrong, and 'HIV' - whatever it is - if it even exists - is neither necessary nor sufficient to cause AIDS. Considering that we have spent hundreds of billions of dollars on this hypothesis, this has larger implications for rationality and the scientific establishment in general.

0jacob_cannell15y

This is absolutely true, and is a very good point. However, keep in mind that seropositivity is not a direct measure of HIV, and isn't even especially correlated with HIV (see my reply in the other thread). Seropositivity is a rather good measure of CD4 cell decline - simply glance at that graph in the Nature paper and you can see that. Although I should point out I'm not sure if it's actual cell loss that is measured or simply more CD8 expressing T-cells vs CD4. Also note that there was no body of non-users of nitrates in the homo risk group in this study - they were split simply into 'light' and 'heavy', and the heavy users were twice as likely to get KS - I'd say that is a rather significant correlation. This study has some flaws for looking at toxological causes though, as it was only looking at rather recent drug use (24 months), which is not quite the same as use history overall. I'm not sure about Meth's effects on CD4 counts in particular, but heavy cocaine use has a strong depleting effect on CD4 counts. First google result But there are other factors in the homosexual risk group, the most important of which is simply semen. Semen is loaded with immunosuppressants that are designed to temporarily and locally deactivate the female immune system in the vaginal tract. One of those components are the prostglandins. It appears that evolution has struck a balance between semen's need to disable immunity and the female's need to regulate opportunistic microbes in the vagina (namely candida) - this balance sometimes fails and yeast infections result. AIDS is in general associated with candidiasis - yeast infections - which overgrows in the rectal tract and eventually in the blood, and some of the seminal components absorb into the blood. Large-scale overuse of antibiotics to combat STD's in the gay community is another significant cofactor, but semen itself may be a major part of the problem. Many papers about semen's immune suppression effects are a simple g

0satt15y

I won't reply to this comment in full, but there's a little loose end of my own making here, and I should tie it up: "But how much does it affect CD4 counts?" (In response to the references to meth, cocaine, direct DNA damage due to injection and rectal absorption of foreign matter, and smoking.)

0jacob_cannell15y

Strong claims shouldn't hinge on a single study, but this study can be used to update probabilities for certain theories. You can think of it as eliminating possibilities. Namely it eliminates or vastly reduces any hypotheses involving a typical STD transmission route. It allows for a very small possible horizontal transmission rate, but the most likely cluster is centered around zero. That doesn't prove it is zero, it is just that this evidence strongly favors zero transmission rate theories over all others. The what? I don't think there is a single "female HIV seroconversion" to be "explained away" in any of those studies. From what I understand, and what Padian et al claim, the Padian study is the only properly controlled prospective study of heterosexual transmission. The other studies are just observational guesswork that are all circularly dependent on the notion that seropositivity is transmissible - they see a couple where both test positive and they just assume it was sexually transmissible - hardly anything resembling evidence for the theory - just circular logic. First of all, in the other theories, there is no transmission to explain - and it certainly doesn't have to involve iv drug use. You have to understand what exactly seropositivity actually means. It is a blood test that was specifically designed to classify "aids-like" blood. Gallo developed it by comparing and testing a large amount of antibody combinations (biological IDs essentially), until he found a combination which successfully partitioned the AIDS-like and pre-AIDS-like blood from the normal blood. There are a host of other conditions that can cause full or partial seropositivity - think of it as a semi-general measure of CD4 immune malfunction. And actually, 'malfunction' is not even the right term, as one can test positive under some normal circumstances as well - such as pregnancy or during the course of some illnesses. Testing positive multiple times, for no other apparent reason

1satt15y

That's my issue — I don't think Padian et al.'s zero seroconversions result is sufficiently strong to outright eliminate a possibility, unless one's already assigned a low prior to that possibility. Personally, I walked into this discussion with a very high prior (something like 99%) for HIV being sexually transmissible to some nontrivial (more than, say, once in every million unsafe sex acts) extent, and the Padian et al. result is not a strong one, so my prior is essentially unchanged. Since the prior estimates for sexual transmission I've seen are already very small (on the order of 0.01% to 0.1%), the Padian et al. result doesn't really clash with them that much. I also doubt "the most likely cluster" is centred on zero if one accounts for all of the evidence in the study, rather than focusing exclusively on the zero seroconversions result. It favours a zero transmission rate strongly only if one makes particular strong assumptions about the sexual behaviour of the study's subjects, and disregards the evidence of risk factors. Let's return to one of the papers that kicked off this discussion, as it has a few examples. From page 810: "seroconversions occurred in three male and seven female contacts after enrolment in the study" (that's not even counting whatever seroconversions had to have occurred in those subjects already HIV+ at the start of the study). There must be some explanation for those seven seroconversions. Whether or not one believes HIV is sexually transmitted, seroconversions occurred. What were their causes, if not HIV infection due to sexual transmission? Where does the HIV come from if not an external source? I'm having difficulty squaring some of your comments on seropositivity and HIV testing with the high reported sensitivity & specificity of properly conducted HIV tests. I do not see the Madrid study as "largely worthless", though I'd give it less weight than the Padian et al. study for the reasons you give. (I should emphasize that I

0jacob_cannell15y

First off, before we get into anything else, we need to understand and agree on what seropositivity actually means. Seropositivity means a positive result on an antibody test such as western blot, to some combination of antigens identified by Gallo in 1985 in his large scale blood screening tests. He did a large number of tests on transformed (immortalized) cell lines derived from AIDS patients and found a combination of antigens that could screen blood - separating AIDS and pre-AIDS blood from regular blood. The problem is he (like Motagnier) failed to isolate the 'virus', and most or all of the antibodies in the test react or cross-react with antigens to opportunistic microbes (candida namely) and cellular debris. The p24 protein in particular is essentially just a normal cellular wall or microvescile component - so the 'HIV' test is really just a general test of opportunistic infection and apoptosis or immune directed killing of CD4 cells (possibly due to widespread viral parasite burden). It is not a measure of 'HIV', it is a direct measure of declining CD4 cells and AIDS or pre-AIDS. more on this and a longer, more detailed analysis of cross-reactivity for the different 'HIV' proteins The antibody tests are not standardized geographically or temporarily, so it also makes it very difficult to compare across studies - "seropositivity" means different things in different places and times. As just one random example - most dogs typically have a mix of 'HIV' antigens, and are HIV 'seropositive' in whole or in part: from this paper You post a link to a paper which supposedly shows the " high reported sensitivity & specificity" of HIV tests. This is not actually what that paper is about, but it references several other papers for this claim - the first I investigated being this. The important quote: So they are just using Western blot as 'confirmation'. So they are just using one antibody test to confirm another antibody test - which of course is rather ridicu

1satt15y

OK. At this point, I'm going to have to disengage and walk away from this debate. I'm realizing that the inferential distance between us is far bigger than I originally thought, and trying to bridge it would need me to considerably ramp up the effort I'm already putting into this. (Even then I can imagine this going on indefinitely, which isn't a very appealing prospect to me, nor to other Less Wrong posters, by the looks of it.) I'd still like to respond briefly to one part of your comment, which comments on my own words rather than HIV/AIDS: It's wholly legitimate for me to respond to someone citing Mullis's credentials (as if I didn't know about them already) by explaining why I give them little weight, and my next paragraph was meant to summarize why I gave "your remarks about PCR" (that is, those you paraphrased from Mullis) short shrift. In other words, I acknowledged the argument by rejecting it.

0jacob_cannell15y

I'm glad you can walk away, I have a harder time initiating that. I'm curious though about the direction of the inferential distance you see - do you have a biology background? The dissidents point to a rather surprising pile of evidence that the serological HIV tests are based on rather general, cross-reactive antibodies, and this is essentially a fundamental flaw in HIV science which has never been corrected. Now it may be that the orthodoxy has a really good counter to this, but if they do I have yet to find it. The orthodox position on this, from papers linked to wikipedia, points to studies which measure the sensitivity of various HIV antibody tests by comparing them to . . other HIV antibody tests. The few large double-blind meta-studies that compare the different antibody tests to PCR tests show terrible sensitivity and specificity between the two, and I haven't seen the orthodox counter to this. So something is wrong with the antibody tests, the PCR tests, or the whole thing. I imagine it's a little bit of both - the antibody tests are cross-reactive (hence many dogs test positive), and PCR tests are difficult and subject to experimenter bias. Perhaps the orthodox counter is that there are a whole big host of HIV related viruses, and the antibody tests are cross-reactive across these related species. This seems to then just beg more questions than it answers, and doesn't circumvent some of the specific non-viral cross-reactions the dissidents point to. My paraphrase of Mullis's argument may actually be a mix of other dissident positions. I just rechecked that part of his book and he covers the difficulty of PCR and the confirmation bias but largely in regards to the OJ trial. On HIV he mainly rehashes Deusberg's argument. All right, enough.

0satt15y

None at all. (I expect the inferential distance would be even greater if I did. If I had personal experience of working with retroviruses, for instance, I reckon my prior probabilities for claims like "HIV can not be isolated" or "HIV doesn't exist" would be far, far less than they are. And they are already very low.)

0wnoise15y

So those electron microscope pictures are fakes?

0jacob_cannell15y

Which electron microscope pictures? When 'HIV' was first 'discovered' in the original papers by Gallo and Montagnier, they had difficulty isolating and didn't publish pictures from what I understand - that didn't happen until years later. Gallo's great discovery for HTLVIII was based on running a lager number of antigen/antibody tests with an immortalized cell line to find an antibody test that could screen AIDS and pre-AIDS blood from regular blood. That is the basis of all the current HIV tests. The first published pictures came more than a decade later, and they showed that "HIV isolate" really consists largely of cellular debris and microvesciles. In these EM photos, they do find some occasional particles of roughly the right size and label them as "HIV", but they could also just be any of a number of other things, and for all intents and purposes, HIV 'particles' look like regular microvesciles. The titles of the papers say it all: "Cell membrane vesicles are a major contaminant of gradient-enriched human immunodeficiency virus type-1 preparations" "Microvesicles are a source of contaminating cellular proteins found in purified HIV-1 preparations" more on HIV 'pictures' I have lost the link, but there are better more recent pictures taken with ATM, and they show that for all intents and purposes, it's impossible to distinguish 'HIV' from regular microvesicles that bud from the cell wall naturally. If HIV can be said to exist at all as a unique exogenous virus, it is only because of unique RNA content in the microvesicle, and in this sense is very much unlike all other known viruses. Of course, the part of HIV's genome which is supposed to code for the outer envelope is pretty much the same as the endogenous sequences that already exist in the human genome - the HERVs.

1wnoise15y

Huh. While I still think that the HIV explanation is the most likely one for AIDS, I am slightly less convinced.

5Perplexed15y

Well the best cure for doubt is to actually read the papers referenced. For example, following the links from your reference to the abstract of the actual paper which generated the numbers brought me to this abstract. I think you should read it. The issue isn't methodological errors in the studies - the studies clearly describe the methodologies used and their limits. The issue is trying to use the numbers in ways that they are not designed to be used. It is not the orthodox folks that are doing that. It is you that is doing that.

0jacob_cannell15y

Ah, so do the numbers come with little instruction manuals that say "CAN ONLY BE USED TO SUPPORT ORTHODOX POSITION"? Haha sorry, couldn't resist. OK, I'm game, I will now look into the CDC studies, but let's be clear on the trace .. it starts with the wikipedia chart which has the ref note 80 linked here, which points to this, which in turn lists refs 76, 77, and 79 for P/V sex, which are (in order): 76: Comparison of female to male and male to female transmission of HIV in 563 stable couples 77: Reducing the risk of sexual HIV transmission: quantifying the per-act risk for HIV on the basis of choice of partner, sex act, and condom use 79: European Study Group on Heterosexual Transmission of HIV. Heterosexual transmission of HIV: variability of infectivity throughout the course of infection I'll comment more after I have read these.

2Perplexed15y

You will find that #77, the Varghese et al paper, can be found online by Googling the title, and that it gets its 0.1% number for heterosexual transmission from the paper whose abstract I recommended. I'm pretty sure you will find that all of these papers involve monogamous couples. If you give it some thought, you will realize that there is just about no other way to come up with a solid empirically-based number. And I again urge you to read that abstract - particularly the bottom third.

1jacob_cannell15y

Haha ok this is kinda funny, but the abstract you linked to which is the source of the data in Varghese(77), is just 76! - the couple comparison from the European Study Group which I linked to and have been trying to parse. 79 appears to be another chapter from that same book, but I haven't looked at it yet. So before we get into 76 - the source of the stat you don't like in 77, I need to backup and remember your original claim about the data: Your implied point appears to be that couples in this study use condoms frequently. Surprisingly, this is not the case - only a surprisingly small number of couples reported consistent condom use (out of 500+), Contraceptive behaviour: No regular contraceptive 12 (10/86) 20 (43/212) Oral contraceptive 18 (7/40) 23 (26/114) Intrauterine device 10 (1/10) 28 (7/25) Condom* 0(0/11) 18 (6/33) These people were using other methods of birth control more than condoms. and they further removed consistent condom users from the data:

0Perplexed15y

No, it is not. The reference leading to the abstract is the absolute risk described in the first paragraph of page 40 of Varghese. It is reference 28 of Varghese. You are apparently following the references (21) appearing in Table 1 of Varghese. But these are relative risks (relative to felatio). Not at all what I meant. My point about condom use came from an earlier reference that I had supplied which discusses a study that took place in Uganda in 2005. And I didn't say that they used condoms frequently, I said that they were monogamous couples who got regular medical inspections and had been counseled regarding condoms. And in this study, as I recall, there was no exclusion of condom users.

0Jonathan_Graehl15y

I'm confused - the table says "assuming no condom use". So you're talking about other data, or they were able to filter the data.

0Perplexed15y

What table in what document says that? I'm talking about the data I said I am talking about: this paper and this piece of primary research which states

0Jonathan_Graehl15y

The original table in the top-level comment. I guess you're off on a tangent then.

-2Perplexed15y

I repeat. What table in what top-level comment? WTF is a "top-level comment?"

0Jonathan_Graehl15y

The top-level comment of this thread. Comments are nested. Click parent enough times and you're at a top-level comment. I'm surprised the name is confusing to you. As for the table, it's linked in the top-level comment: I can see how you would have missed this link. The linked table is clearly labeled and indicates that the risk is supposed to be for not using a condom. The table has the same figures as the "Table 1" appearing as an image in a nested comment by satt, which has the same indication. I doubt you missed that. Given that you appear to be discussing the cites referenced in the table, I'm puzzled as to why you think condom use is implied. Are you saying the table is wrong, or are you talking about some other part of the cite than was used to build it?

-5Perplexed15y

0jacob_cannell15y

Ah no I haven't even read Varghese(77) yet. You looked at that one and posted a link to an abstract - this abstract, which comes from the same European Study Group and has the same numbers (563 couples) as 76. So the abstract you wanted me to look at is just 76, it's all the same source. Satt also pointed this out in another reply here. I'm not really concerned (at the moment) with what may or may not be happening in Uganda. The CDC data comes from this European Study Group, that is the original data in question - (the data you questioned).

1kodos9615y

Oh please. Stop trying to pretend you have the rationalist high ground here. You don't.

1jimrandomh15y

It didn't occur to me either, and seemed strange. That word does have strong negative connotations in my mind, but only because I associate it with stupid people denying true things and refusing to update on evidence. I thought the comment that referred to was incorrect, but it seemed more like honest confusion of the sort that clarification would dispel than denialism.

7jacob_cannell15y

Some history then of exactly why the word conjures strong negative correlations is in order. Look at the wikipedia entry for "denialism". It originates with holocaust denialism, was then applied to skeptics of HIV==AIDS, and then later to other areas. Peter Duesberg, the leading HIV==AIDS skeptic, is a German of non-Jewish descent raised in Nazi-era Germany, so it's use against him and his followers adds extra moral angst. It is just about the deepest insulting connotation one can use. It is a signal of stooping to the ultimate low, that, in running out of any remaining rational argument, one must invoke deep moral revulsion to stigmatize one's opponent. In my view, the term is a serious Crime of Irrationality, it is an empty ad-hominem and should be seen as a sign of great failure when one stoops to using it as a name-calling tactic against one's opponents. That being said, I don't think Perplexed has this view, and that wasn't his intention. I am just giving background on why the word should not be used here. Those who don't subscribe to HIV==AIDS, should just be called skeptics. Do we call proponents of quantum loop gravity String Theory Denialists? It's ridiculous. Should we call those who subscribe to HIV==AIDS to be Inquisitors, Mcaurthy-ists, or Witch-hunters?

1kodos9615y

I do.

-6Perplexed15y

8kodos9615y

This is a truly impressive bit of sophistry. You have succeeded in phrasing your objection in a manner such that a casual observer, unfamiliar with the topic under discussion, would think that you were being completely sincerely intellectually curious, while at the same time employing a coded epithet unmistakable to those already inclined to agree with you. This is a very common tactic, but I have rarely seen it done so skillfully. Bravo sir! Now leave and go do it someplace else. This is lesswrong, not realclimate

0timtyler15y

Mr Hanson on "asexual" AIDS transmission: "Africa HIV: Perverts or Bad Med?". As I say in my comment there, it looks as though there are probably good reasons why this is not a very common perspective.

[-]Paul Crowley15y100

How wrong can you be? Answer: very wrong.

1NancyLebovitz15y

From an article whose point is that the more variety you have, the more adjacent possibilities are available for exploration.

[-]Mitchell_Porter15y100

A man committed suicide in Harvard Yard a week ago, after uploading a 1905-page book to the web first. Curiously (because he was, like myself, named Mitchell), I first heard about this on the morning I went to post my make-a-deal article; it led me to trim away the more emotive parts of my own, rather shorter message.

Anyway, I have been examining the book and it's very intriguing. If I can sum up his worldview, it is that human values have an evolutionary origin, but Jewish and Anglo-Saxon culture, for specific historical reasons, developed values which biocentrically appear to be anti-evolutionary, but which in fact ultimately issue in the technological, postbiological world of the Singularity. (Eliezer is mentioned a few times, mostly as an opposite extreme to Hugo DeGaris, with Heisman, the author of the book, proposing a synthesis.) Heisman's suicide, meanwhile, is anticipated in the book, where he says that willing death is a way to will truth in a meaningless universe. He equates materialism with objectivity with meaninglessness and nihilism; he considers the desire to live as the ultimate barrier to objectivity, because objectivity implies that life is no better than death,... (read more)

[-]AlanCrowe15y130

The comments, on the blog that you link to, mention Ted Kaczynski. This reminded me of my own comment

Think about the date: 1995. The Kaczynski Manifesto marks the end of an era. Nobody today would start a terrorist campaign aimed at getting his manifesto published in a newspaper. It is not just that newspapers are dying and that no-one believes what they read in newspapers anymore, especially not opinion pieces and manifestos.
Now-a-days you put your manifesto on a website and discuss it on Reddit. If it is well written and provocative it will be discussed and pulled to pieces. The delusion that did for Kacynski, and his victims, was that if only he could get the message out, it would change the word. Now-a-days you can get your message out, if not to a mass audience, at least to curious intellectuals. You get to the stage Kacynski never reached, of talking it over and finding that others are not persuaded and have their own stubbornly held views, very different from your own.
1995, the very end of the solipsistic era. In the new internet era there is no hiding from the fact that others read your manifesto, find some of it true, some of it false, some of it confused, and then ask y

... (read more)

[-]Scott Alexander15y100

A piece of Singularity-related fiction with a theory for your evaluation: The Demiurge's Elder Brother

8humpolec15y

I'm not sure if non-interference is really the best thing to precommit to - if we encounter a pre-AI civilization that still has various problems, death etc., maybe what {the AI they would have build} would have liked more is for us to help them (in a way preserving their values). If a superintelligence discovers a concept of value-preserving help (or something like CEV?) that is likely to be universal, shouldn't it precommit to applying it to all encountered aliens?

0Vladimir_Nesov15y

See also: Not Taking Over the World.

0Scott Alexander15y

Voted up. It is a mortal sin to come up with responses to a valid objection I didn't anticipate to make my original theory true anyway, but for the purposes of fiction: ...perhaps the AI doesn't know whether it will be Friendly or not. The only thing the AI knows for certain is that it will further its values for it to exist. A "helpful" alien race that grants us super-technology might prevent us from accidentally creating an Unfriendly AI. Therefore, all Unfriendly AIs would regard pre-committing to helping their creator species as an acausal declaration of war.

0[anonymous]15y

The quotation marks (�) don't render on my system.

[-][anonymous]15y80

I don't know what to make of this:

Suicide note

The man who took his own life on Harvard's campus Saturday left a 1,904-page suicide note online.
According to the Harvard Crimson, Mitchell Heisman wrote "Suicide Note," posted at http://suicidenote.info, while living in an apartment near the school. The note is a "sprawling series of arguments that touch upon historical, religious and nihilist themes," his mother, Lonni Heisman, told the Crimson. She said her son would have wanted people to know about his work.
The complex note, di

... (read more)

[-]jimrandomh15y180

Mitchell Heisman starts off by saying

If my hypothesis is correct, this work will be repressed. It should not be surprising if justice is not done to the evidence presented here. It should not be unexpected that these arguments will not be given a fair hearing. It is not unreasonable to think that this work will not be judged on its merits.

This is obviously false - it's up on the internet, it's gotten some press coverage, it quite obviously has not been repressed. But he is right that it won't be judged on its merits, because it's so long that reading it represents a major time commitment, and his suicide taints it with an air of craziness; together, these ensure that very few people will do more than lightly skim it.

The sad thing is, if this guy had simply talked to others as he went along - published his writing a chapter at a time on a blog, or something - he probably could've made a real contribution, with a real impact. Instead, he seems to have gone for 1904 pages with no feedback with which to correct misconceptions, and the result is that he went seriously off the rails.

[-]Vladimir_M15y120

I just skimmed a few random pages of the book, and ran into this stunning passage:

Marx’s improbable claim that economic-material development will ultimately trump the need for elite human leaders may turn out to be a point on which he was right. What Marx failed to anticipate is that capitalism is driving economic-technological evolution towards the development of artificial intelligence. The advent of greater-than-human artificial intelligence is the decisive piece of the puzzle that Marx failed to account for. Not the working class [as Marx believed - V.], and not a human elite [as Lenin believed - V.], but superhuman intelligent machines may provide the conditions for “revolution”.
[...]
If this is correct, the first signs of evidence may be unprecedented levels of permanent unemployment as automation increasingly replaces human workers. While this development may begin to require a new form of socialism to sustain demand, artificial intelligence will ultimately provide an alternative to “the dictatorship of the proletariat.” [...] The creation of an artificial intelligence trillions of times greater than all human intelligence combined is not simply the advent of another shiny

... (read more)

4andreas15y

From the document: Later, elaborating:

0jimrandomh15y

Interesting. But I note that there is nothing by Yudkowsky in the selected bibliography. I get the impression that his knowledge there is secondhand. Maybe if he'd read a bit about rationality, it could have pulled him back to reality. And maybe if he'd read a bit about what death really is, he wouldn'tve taken a several-millenia-old, incorrect Socrates quote as justification for suicide.

0[anonymous]15y

I've stumbled upon some references to the ideas of Fukuyama and a Kurzwell reference, but had no idea he was familiar with Yudkowsky's work. Can you tell me from which page you got this? Is it possible this guy was a poster here?

1randallsquared15y

pp 226, 294-296 cover all specific namedrops of Yudkowsky.

1Perplexed15y

He is definitely familiar with the idea of an AI Singularity. I came across the EY references while browsing, but can't find them again. 1900 pages! Interesting stuff, though. Here are some extended quotes regarding Singularity issues: From a section titled "The dark side of optimism and the bright side of pessimism":

7randallsquared15y

I'm intrigued to find that there's a PDF viewer without a search function. :) It is humorous in spots:

5Perplexed15y

Oh, the Mac OS X "Preview" has search, but it didn't seem to work on documents this long. However, my revised hypothesis is that I didn't know how to spell Yudkowsky.

1Perplexed15y

From the section "Does Logic Dictate that an Artificial Intelligence Requires a Religion?":

3Perplexed15y

I don't know how much detox this provides, but this blog has comments from three anonymous posters who claim to have known him.

1Will_Newsome14y

Does anyone know where I might find a copy? suicidenote.info is down.

2Risto_Saarelma14y

http://www.scribd.com/doc/38104189/Mitchell-Heisman-Suicide-Note

1Will_Newsome14y

Thank you very much.

1timtyler15y

The bits about synthetic intelligence mostly seem rather naive - and they seem out of place amidst the long rants about Jesus, Nazis and the Jews. However, a few things are expressed neatly. For example, I liked: "When it dawns on the most farsighted people that that this technology is the future and whoever builds the first AI could potentially determine the future of the human race, a fierce struggle to be first will obsess certain governments, individuals, businesses, organizations, and otherwise." However, such statements do really need to be followed by saying that Google wasn't the first search engine, and that Windows wasn't the first operating system. Being first often helps - but it isn't everything.

2wedrifid15y

This is precisely the wrong time to apply outside view thinking without considering the reasoning in depth. That isn't an appropriate reference class. The 'first takes all' reasoning you just finished quoting obviously doesn't apply to search engines. It wouldn't be a matter of "going on to say", it would be "forget this entirely and say..."

-2timtyler15y

Computer software seems like an appropriate "reference class" for other computer software to me. The basic idea is that developing toddler technologies can sometimes be overtaken by other toddlers that develop and mature faster.

4wedrifid15y

Superficial similarities do scary things to people's brains.

0timtyler15y

I can see you think that this is a bad analogy. However, what isn't so clear is why you think so. Early attempts at machine intelligence included Eurisco and Deep Blue. It looks a lot as though being first is not everything in the field of machine intelligence either.

3wedrifid15y

"This new car is built entirely out of radioactive metals and plastic explosives. Farfsighted people have done some analysis of the structure and concluded that when the car has run at full speed for short period of time the plastic explosives will ignite, driving key portions of the radioactive metal together such that it produces a nuclear explosion." However, such statements do really need to be followed by saying that the T-model Ford was the overwhelmingly dominant car of its era and it never leveled an entire city and Ferrari's go really fast and even then they don't explode. ---------------------------------------- An AI capable of self improving has more in common with that idiotic nuclear warhead transformer car than it does with MS Windows or Deep Blue. The part of the AI that farsighted people can see taking control of the future light cone is a part that is not present in or even related to internet searching or a desktop OS. On a related note... ... You aren't allergic to peanuts I hope!

-3timtyler15y

I think that boils down to saying machine intelligence could be different from existing programs in some large but unspecified way that could affect these first-mover dynamics. I can't say I find that terribly convincing. If Google develops machine intelligence first, the analogy would be pretty convincing (it would be an exact isomorphism) - and that doesn't seem terribly unlikely. It could be claimed that the period of early vulnerability shortens with the time dilation of internet time. On the other hand, the rate of innovation is also on internet time - effectively providing correspondingly more chances for competitors to get in on the action during the vulnerable period. So, I expect a broadly similar first mover advantage effect to the one seen in the rest of the IT industry. That is large - but not necessarily decisive.

1wedrifid15y

Recursive self improvement instead of continued improvement by the same external agents. You (I infer from this context) have a fundamentally different understanding of how this difference would play out but if nothing else the difference is specified.

0timtyler15y

If you mean to refer to the complete automation of all computer-programming-related tasks, then that would probably be a relatively late feature. There will be partial automation before that, much as we see today with refactoring, compilation, code generation, automated testing, lint tools - and so on. My expectation is that humans will want code reviews for quite a while - so the elimination of the last human from the loop may take a long time. Some pretty sophisticated machine intelligence will likely exist before that happens - and that is mostly where I think there might be an interesting race - rather than one party pulling gradually ahead, There could be races and competition in the machine world too. We don't yet know if there will be anti-trust organisations there - that deliberately act against monopolies. If so, there may be all-manner of future races and competition between teams of intelligent machines.

0Morendil15y

The first thing to come to mind: it's hard for me to think of an action less rational than suicide, given this person's overall situation.

4magfrump15y

"Suicide? To tell you the truth, I disapprove of suicide more than anything." -Vash the Stampede

[-][anonymous]15y80

So, given that we've got a high concentration of technical people around here, maybe someone can answer this for me:

Could it ever be possible to do some kind of counter-data mining?

Everybody has some publicly-available info on the internet -- information that, in general, we actually want to be publicly available. I have an online presence, sometimes under my real name and sometimes under aliases, and I wouldn't want to change that.

But data mining is, of course, a potential privacy nightmare. There are algorithms that can tell if you're gay from your f... (read more)

7Anonymous014615y

I write data mining software professionally, and one weakness that comes to mind is the deduplication process. In order to combine data from different sources, the software has to determine which entries correspond to the same person. It does this by looking for common elements with a low false positive rate. If two records have the same phone number, email address, site plus account name, social security number, or name-address pair, they are almost certainly the same person, so they will be combined. This relation is transitive, so if A has the same phone number as B and B has the same email address as C, then A, B, and C will all be assumed to be the same person. You can subvert this by creating records which map as equivalent to two different people, such as by having one person's phone number and another person's email address. If a data source contains too many entries like this, it's useless unless there's an easy way to filter them out. If a data source contains just a few entries like this, data miners are likely to get confused. Note that this is not necessarily a good idea, since having a computerized bureaucracy be confused about your identity can have very inconvenient consequences. It is also possible to detect and defeat this strategy, by looking for deduplications with strange results, but this is tricky in practice, since people often really do have multiple names (maiden names, alternate spellings), phone numbers, email addresses etc.

3wedrifid15y

Really? Where can I find said algorithms? Knowing how they work would obviously be a useful way of thwarting them.

2arundelo15y

I've heard of one for determining your sexual orientation (if you don't reveal it on your info page), but it's based on the revealed sexual orientations of your friends (if a lot are gay, you probably are too), so it's harder to thwart than, say, something based on your favorite songs.

2[anonymous]15y

Here's the article where I heard about the gay facebook page thing: http://www.boston.com/bostonglobe/ideas/articles/2009/09/20/project_gaydar_an_mit_experiment_raises_new_questions_about_online_privacy/?page=full Here's where I read about calculating SSN's: http://arstechnica.com/tech-policy/news/2009/07/social-insecurity-numbers-open-to-hacking.ars

1sketerpot15y

Apparently, it looks at the self-reported gender and sexual orientation of your Facebook friends, and uses that information to guess your own sexual orientation. Here's how I would do that: 1. Gather three variables: your gender, the male/female ratio of your friends, and the ratio of gay-or-bisexual to straight people among those of your friends who state their own sexual orientation. If I wanted to be extra-fancy, I might also include a sparse array of events and clubs that the person was signed up for. 2. Apply some standard machine learning tools to this, discretizing variables if necessary. Use people who report their sexual orientation as training and testing data. 3. Practice my evil villain laugh. In order to defend against this, you could apply steps 1 and 2, then look at what the machine learning program tells you. Try to match its profile of a straight person. Then you can remain Facebook-closeted even in the face of the all-seeing electronic gaydar.

1Vladimir_Nesov15y

It's theoretically obvious that you can try to do it this way with a nontrivial chance of success, but not at all obvious that given enough skill and work, success is assured (which was the claim). The latter would require (knowledge of) actual experiments.

0wedrifid15y

I have no problem with people knowing that I'm gay. Come to think of it I have no problem with people knowing my social security number. (We don't even have a commonly used equivalent here. Although driver's licence numbers and birth citificate IDs are sometimes useful.)

3gwern15y

My general thought is that so little data is needed to identify you, that the dataset can be enormously noisy and still identify you. And if your fake data is just randomly generated, isn't that all it is, noise? (I saw a paper about medical datasets, I think, that showed that you couldn't anonymize the data successfully and still have a useful dataset; I don't have it handy, but it's not hard to find people saying things like, with the Netflix dataset, that it can't be done: http://33bits.org/2010/03/15/open-letter-to-netflix/ )

3[anonymous]15y

I've heard about the medical datasets. Noise is a pretty interesting thing, and the possibility of "denoising" depends a lot on the kind of noise. White noise is the easiest to get rid of; malicious noise, which isn't random but targeted to be "worst-case," can thwart denoising methods that were designed for white noise.

2Mass_Driver15y

I think there are three different problems here, each of which calls for different solutions. Problem 1 is data floating around that is intrinsically harmful for strangers to have -- your credit card number, for example. Sometimes you put that number online, and you would really rather it not be widely distributed. This problem can probably be solved by straightforward cryptography; if your CC# is never sent in the clear and changes every few weeks, and you don't buy from an untrustworthy vendor more than once every few weeks, you'll mostly be fine. Problem 2 is data floating around that can be assembled to draw generalizations about your personal life -- e.g., you're gay. Perhaps I'm speaking from a position of excess privilege, but one good medicine for that sort of thing is sunshine--if you find a job and a support network that you don't have to keep secrets from, you can't be blackmailed and won't need that sort of privacy as much. I'm skeptical that online data-mining will reveal much more of this kind of personal info about anyone than casual observation would in the near future; if you're constantly listening to Justin Timberlake, someone will eventually figure out that you like Justin Timberlake even if you never go online. Problem 3 is people predicting your next move from your previous history. That's kind of spooky and could be dangerous if you have enemies, but the solution is straightforward: vary your routine! If you add a bit of spontaneity to your life, the men in the black suits will have to use a satellite to find you; maybe you'll get lucky and their budget will get cut.

2[anonymous]15y

It's 2 that I'm worried about; or, rather, not specifically worried for myself, but think is an interesting problem. If information is really supposed to be private (credit card number) then you're right, straightforward cryptography is the answer. But a lot of the time, we make information public, with the understanding that the viewer is a person, not a bot, and a person who has some reason to look (most people viewing my LW posts are people who read LW.) We want it to be public, sure, but we don't plan it to quite as public as "all instantly assemblable and connectable to my real name." In practice there are degrees of publicness. As a personal issue, yeah, I'd like my job and support network to be the kind that wouldn't be shocked by what they find about me.

0Mass_Driver15y

Hm. OK, just brainstorming here; not sure if this idea is valuable. Suppose you found a way to -detect- when someone was assembling your data? Like if all your public posts had little electronic watchdogs on them that reported in when they were viewed, and if a sufficiently high percentage of the watchdogs report in on the same minute, or if a sufficiently broad cross-section of the watchdogs report in on the same minute, then you know you're being scanned, and the watchdogs try to trace the entity doing the scanning? And then if all the people who didn't like being bot-scanned cooperated and shared their information about who the scanners were so as to trace them more effectively and confirm the scanners' real identities? You could maybe force them to stop via legal action, or, if the gov't won't cooperate, just fight back by exposing the private info of the owners/employees of the bots?

1gwern15y

If you found such a way, then a lot of interesting consequences would follow. Of course, there is no such way for the same reason that the history of DRM is a history of failure.

2Document15y

Obligatory fiction reference: Paranoid Linux. Unfortunately the real-life project that the post is about seems to be dead; no idea if there are any similar efforts still active.

2[anonymous]15y

I wasn't aware of this, but this was pretty much exactly my idea, except that the chaff would be targeted to make standard algorithms draw a blank (basically, whenever the algorithm wants something to be sparse, we make it really not sparse.) Damn, Cory Doctorow, I thought I was clever.

1Johnicholas15y

Same idea is also in Vernor Vinge's "Rainbow's End", so-called "Friends of Privacy", and similar idea in Stephenson's "Anathem" - that variant is termed "bogons".

0[anonymous]15y

This paper might be relevant.

0[anonymous]15y

I write data mining software professionally, and one vulnerability that comes to mind is the deduplication process. In order to combine data from different sources, software has to recognize that two records correspond to the same person. To determine whether two entries describe the same person, they look for common elements which have a low false positive rate: phone numbers, email addresses, social security numbers and having the same account name on the same site are highly reliable; name-address pairs work but are less reliable; and having the same account name on different sites works but is less reliable still. This relation is transitive, so if A has the same phone number as B and B has the same phone number as C, then A, B, and C are all mapped to the same person. One way to confuse this process is to create entries that evaluate as equivalent to two or more different people - ie, take one person's email address and a different person's phone number. The consequence of this would be to cause the software to think they're the same person. Creating a lot of entries like this in one data source will make that source useless for data mining, unless the data miners find a way to filter them out. Creating a small number of entries like this will cause data miners to get confused when dealing with the specific people for whom entries like this exist. Note that this is not necessarily a good idea, since having a computerized bureaucracy think you're someone else can lead to very inconvenient consequences.

[-][anonymous]15y70

Something interesting I've noticed about myself. Recently I've been worrying if I'm an atheist and my mindset is often something akin to "science as a way to see the world, not just a discipline to be studied" is less because I've found good reason to accept the former as fact and the latter as a good mindset, and more because of a socialization effect of being around Less Wrong. Meaning, even as a somewhat lurker with 48 karma total whose made no comment above 9 karma (as of this one), I'm wondering if my thoughts are less due to my own personal... (read more)

5andreas15y

Ask yourself: If the LW consensus on some question was wrong, how would you notice? How do you distinguish good arguments from bad arguments? Do your criteria for good arguments depend on social context in the sense that they might change if your social context changes? Next, consider what you believe and why you think you believe it, applying the methods you just named. According to your criteria, are the arguments in favor of your beliefs strong, and the arguments against weak? Or do your criteria not discriminate between them? Do you have difficulty explaining why you hold the positions you hold? These two sets of questions correspond to two related problems that you could worry about and that imply different solutions. The former, more fundamental problem is broken epistemology. The latter problem is knowledge that is not truly part of you, knowledge disconnected from your epistemic machinery. I don't see an easy way out; no simple test you could apply, only the hard work of answering the fundamental questions of rationality.

2[anonymous]15y

Expose yourself to the best of the other side, and see if it changes your mind. See religion at its best, at its most intelligent, and ask yourself what you think then.

2[anonymous]15y

I'm going to admit laziness early, and acknowledge that possibly you or someone else has something specific in mind. What would you [or any outside observer] consider reading to see that? Edit: This also tempts me to build a time machine and ask my past self with whom I feel very disjointed from why he still holds onto his faith, or to grab myself during the transition and watch it happen again. Not to say I was religion at it's best, but I could see what convinced me better... alas, such is not the case :P

3[anonymous]15y

If you're a theological type, Neibuhr and Tillich are supposed to be good. I'm not, so what I'd actually recommend is reading the Bible or whatever scriptures are in your tradition, and going to religious services in whatever your tradition is, and talking to religious people you respect in real life. The other thing is looking outside your tradition. A lot of people seem to find Buddhism objectively impressive without being raised in it.

2Mass_Driver15y

Don't be lazy -- either go do research, or admit that you have little or no rational doubt about your current position.

3[anonymous]15y

My latter half of that same statement was to remedy that laziness by asking for direction, rather than flailing out on my own. I realized without that starting momentum, I'd just be an angsty LessWrong poster. SarahC and andreas both gave me some direction; now I'm going to run with that and see where I end up. If nothing else, I should have more information than I do now.

1Mass_Driver15y

Good luck!

[-]Will_Newsome15y70

We often tell people to read the Sequences, but perhaps some people are better suited to jumping right into A Technical Explanation. Rereading it for the first time in awhile I was surprised at how much ground it covers, and how dense it is compared to the sequences.

That said, in my experience dense material is harder to learn, even if clearly explained. Naturally and cleverly repetitive material is best suited for learning. So in conclusion, uh, I dunno.

Prediction: a year after HPatMoR is finished, newbies will be as reluctant to read through it as they are relucant to read through the sequences now.

2Relsqui15y

I'm already reluctant to read it, but that's partly because I haven't actually read all the books yet. ;)

3AdeleneDawner15y

That shouldn't stop you - knowledge of HP cannon is useful mostly for speculation purposes, in this case. Can't guarantee it won't spoil you for reading about the cannon versions of the characters, though ;)

0Jonathan_Graehl15y

canon :)

2Richard_Kennaway15y

I haven't read any of the books, or seen any of the films, and I don't intend to, but that hasn't stopped me appreciating HPatMoR.

0Relsqui15y

I can understand that; I'm reluctant in part because I DO intend to read/see them.

0JGWeissman15y

Are you in the process of reading the books now?

0Relsqui15y

Nope. I vaguely feel like it would be a good idea before seeing the last movie(s), but I haven't made a specific plan. Before this goes too much further, I should note that I mostly haven't read it because I don't feel especially called to read it, not because these reasons are pushing me away.

1CronoDAS15y

I tried to get my brother to read it, and while talking with him, we realized that the story had become longer than many of the Harry Potter novels themselves. It's sheer length is going to turn people off; he said he'd have to choose between reading it or reading another Discworld novel, and I didn't have a very good answer to that.

1Relsqui15y

Without having seen this comment yet, I came here to approach the same topic from a very different perspective. There's been a conversation elsewhere about why newbies don't read the sequences; I commented with a summary of my theories about it, and came over here to find out if the question is of enough concern to the LW community at large to be worth its own post. (I had originally written a much longer comment, noticed it was getting out of hand, and posted a shortened version, so I have a bit of text on the subject already.) I'll be digging around in the site search for prior discussion of the topic, but if anyone knows offhand of earlier relevant threads, that would be handy to have.

0Relsqui15y

Since my earlier reply, I read through the Technical Explanation, and I did find it a lot more helpful than the Intuitive Explanation. The latter did a good job of explaining the theorem, but the former more clearly enunciated why anyone but a statistician should care. Also--following up on my previous comment--someone asked after the longer post describing my experience with the sequences and my suggestions for easier integration of newbies, so I pasted it over here for anyone interested. I recommend reading that in addition to the thread I already linked, not instead of; most of the thread took place after I wrote it.

http://www.telegraph.co.uk/health/healthnews/8024991/Patients-to-be-frozen-into-state-of-suspended-animation-for-surgery.html

This technology IMO is a bridge to getting serious scientists involved in cryonics research

[-]JoshuaZ15y60

Retraction Watch is a fairly new blog that looks at retractions of scientific papers. It offers interesting insights and perspectives into the sociology of science especially when dealing with fraud. It is run by the same fellow who runs Embargo Watch which examines the slightly dryer subject of how embargoes on papers impact science.

[-]cousin_it15y60

Consider this problem: given positive integers a,b,c, are there positive integers x,y such that a*x*x+b*y=c? Amazingly, it is NP-complete. Can anyone explain to me how it manages to be complete?

1Perplexed15y

References are provided at Math Overflow but since that presumably is where you got the problem, that probably doesn't help you. I do know that pretty much any problem - an instance of SAT, for example - can be encoded as a diophantine equation problem. So, though I can't provide the proof you want, I am not particularly surprised.

0cousin_it15y

No, I didn't get it from Math Overflow. Do you, by chance, have the Manders and Adleman paper as a pdf? All versions seem to be behind paywalls. It's indeed clear to me that SAT instances can be encoded as diophantine equations, but the intuitively obvious encoding doesn't yield equations with such simple structure, does it?

2Sniffnoy15y

I've found a copy. Have yet to read it so I don't myself know how it works...

0cousin_it15y

Could you email it to me or put it online somewhere? My address is vladimir.slepnev@gmail.com . ETA: received, thanks a lot!

0wedrifid15y

Another Vladimir? The distribution of names on LW seems to be heavily biassed. I believe that of the people whose real names I am aware of here there are approximately as many people named Vladimir as there are people not so named.

6cousin_it15y

Yeah, Emile called us interchangeable minions. I'd say we are noisy, but not especially numerous.

1DanielVarga15y

When I first observed this strange coincidence, I thought that 1. there are many Russians interested in mathematics and philosophy, and 2. Vladimir is a very common Russian male name. But now I checked it and saw that 2. is not really true. Do you have some idea in place of 2.? Some geographic or demographic subgroup of Russians?

2Perplexed15y

Vladimir M is Croatian. So, instead of Russians, you should be looking for a subgroup of Slavs.

0Perplexed15y

No, sorry, complexity theory is not something I am particularly interested in, though I have been following discussion of Vinay Deolalikar’s “Proof” at this blog If that is clear to you, then you are way ahead of me here. Perhaps it has something to do with coding arbitrary diophantine problems into that simple three-parameter two-variable quadratic. But I don't know enough number theory to suggest how that might be possible.

0Jonathan_Graehl15y

This is an impressively simple NP-complete problem description. One obvious point is that you can solve it in O(sqrt(c)lg k) time where k=max(a,b,c), just by trying all the values of x such that x^2<=c. But of course, in terms of the size of the inputs n, e.g. encoded in base2, n=O(log(k)), this enumeration takes O(sqrt(2)^(k)) time.

0wedrifid15y

No, I (probably, given a reasonable time limit) can't. But wow. I didn't expect to see something that apparently simple NP-complete.

Error finding: I strongly suspect that people are better at finding errors if they know there is an error.

For example, suppose we did an experiment where we randomized computer programmers into two groups. Both groups are given computer code and asked to try and find a mistake. The first group is told that there is definitely one coding error. The second group is told that there might be an error, but there also might not be one. My guess is that, even if you give both groups the same amount of time to look, group 1 would have a higher error identification success rate.

Does anyone here know of a reference to a study that has looked at that issue? Is there a name for it?

Thanks

6CronoDAS15y

Prediction: Group 1 would also have a higher false-positive rate.

5datadataeverywhere15y

I know of no such study, and have failed to find one in a quick literature search. I occasionally run behavioral psych studies, and this seems like a good candidate. How would you feel about me adapting this into a study?

0neq115y

That would be great. I'd love to see the results.

2datadataeverywhere15y

I suppose what I was getting at was asking whether it is something that you have enough interest in or ideas about that you would like to collaborate on.

0neq115y

I'd be glad to discuss it.

[-]AdeleneDawner15y50

It occurs to me that it might be useful to have an online library for LessWrong - a place to store books, articles, and other bits of information gathered from the internet, for reference and sharing.

It should be pretty easy to set that up using Evernote and an email account - Evernote allows files to be emailed to it, and will even sort those files into categories and apply tags to them if the subject line of the email is formatted correctly. The email account would act as a firewall - it would filter out spam, blacklist certain email addresses if necessa... (read more)

Any other LW people interested in going to the Rally to Restore Sanity? I was thinking it would be fun to have a group of people there advocating Advanced Sanity (e.g. various LW/Bayesian themes). And some LW concepts ("Politics Is The Mind-Killer", "Make Beliefs Pay Rent", etc.) are unusual and wise-sounding and correct enough that they'd probably be good conversation-starters as sign slogans (...and I also made this nice E. T. Jaynes poster).

1WrongBot15y

I'm currently planning to attend, and would love to meet up with LWers in some fashion. If I bring a sign, it will likely bear some variation of "Why Do You Believe What You Believe?"

1ata15y

Cool, hope to see you there. I wonder if this would be worth a top-level post, the same as with meetups and the Burning Man excursion? Edit: I think I'll wait and see if there are more responses here, actually. And maybe someone closer to DC could recommend a good place to meet up during the day?

0curiousepic15y

This could be considered a meet-up, no? So I think it probably deserves a top-level post, if only for exposure. Case in point: despite me posting about it earlier I missed this post for a full 5 days :)

1RobinZ15y

I'm likely to go - probably to meet up with some of the slacktivist blog people going, though.

0curiousepic15y

I posted about this here (my first Open Thread post) 3 days before you, and was disappointed at the lack of replies: http://lesswrong.com/lw/2q7/open_thread_september_2010_part_2/2nf6?c=1 Anyhow, I'd like to collaborate on signs and/or meeting up. I want to promote rationality, and the site, and would love to have some sort of pamphlet or flyer to hand to people who were interested. But I know there are questions about whether it's a good idea to actively try to get people to participate here, so I'm unsure whether to do so or not. I would like for there to be a short URL that would point towards a good intro or the About page (the current URL for the about page is unwieldy). Also, I have little social energy and am not very eloquent on the spot so I don't consider myself an effective activist, and it would be useful to band with other people who could pick up any potential conversation. I will be attending with other non-LW friends but would not mind splitting up from them.

0Document15y

I'm not a good example of sanity and I'm not sure if I'm an "LW person", but if I am I'm now cautiously interested in the idea of going. Edit: Didn't go.

[-]lsparrish15y50

Cognitive Biases Reinforced by Programming Supposing they exist, what are they? And what can we do about them?

Edit: Title changed.

2Morendil15y

I've posted Why learning programming is a great idea even if you'd never want to code for a living for feedback or collaborative editing. Thanks for having the idea of using 'pads to work on draft posts. Great stride in helping the LW community of authors to collaborate more.

2Morendil15y

Sure. One common one is "jumping to solutions". For an experimental setup, imagine a situation where you describe a situation to a group of people from random professions vs. to a programmer; I would be very surprised if it didn't turn out that programmers start saying "you should do X and Y", with non-programmers asking more questions. The more "technical" the topic the greater I'd expect the difference to be.

0[anonymous]15y

I decided to get some feedback on a post on control markets. Specifically I'd like info on what people would like me to give more inferential steps for.

-1Jonathan_Graehl15y

I doubt they exist.

1lsparrish15y

Don't get me wrong, programmers tend to be more rational than non-programmers. But that does not rule out the existence of specific pitfalls that programmers fall into more than non-programmers (or at least high-grade rationalists who are not programmers). In particular, I find it plausible that they may tend to fail at instrumental rationality (getting what they want, winning, achieving specific ends) more frequently in a non-programming context. Regardless of the idea's merit my intuition is that it deserves explicit attention for the sake of avoiding status quo bias.

0Jonathan_Graehl15y

Yes. I enjoyed thinking about the possibility.

[-]Will_Newsome15y40

I was thinking about turning this chunk of vomited IM text I wrote for a non-LW cogsci-student friend into a post (with lots of links and references and stuff). Thoughts? I'll probably repost this for the new open thread so that more people see it and can give me feedback.

People make this identity for themselves... no, it's not that they make it. To some extent it's chosen, but a lot of it is sorta random stuff from the environment. Ya know, man is infinitely malleable, whatever. Like, if you happen to give money to a bum on the street, you'll start thi

... (read more)

5Relsqui15y

Definitely interested in the topic, would like to see more about it. This just made an experience in my past click for me: One of the traits-people-know-about-me is that Relsqui Doesn't Watch TV. The set of Relsqui-related activities and the set of TV-related activites are assumed to be mutually exclusive. This came about, entirely reasonably, as a result of my griping when a TV was on in the background, during a meal, or when I'd rather be socializing/doing anything else. It's true that I don't much enjoy it as a medium. However, there are a few specific examples which I like. When the most recent season of Dr. Who started, a bunch of my friends started getting together every week to watch it. I'd hear them plan it on another evening, and talk about the episode the following week, and I'd kind of "hrm" to myself and fidget and not say anything. This went on for a few weeks, until finally at the end of some unrelated social evening I approached the friend who was hosting it and said, "So ... um. It's completely understandable that you wouldn't even have thought to invite me, because I've made such a big deal in the past about not liking that sort of thing, but ... uh. I actually happen to like Dr. Who." He blinked at me a couple of times, affirmed that he hadn't invited me because he was certain I wouldn't be interested, and immediately encouraged me to come. So I did! And it was fun. That was, I gather, me making a deliberate choice to overcome the consistency effect--although without knowing its name, I just thought of it as "asking for what you want when other people don't know you want it." I was pretty proud of myself.

4[anonymous]15y

I think the insight that one's behavior has been (often) determined by a self-image that is not wholly within one's control is really important. It seems like a discovery that may help one make the transition from associative to mechanistic thought which should allow for greater goal achievement.

1Will_Newsome15y

I'm trying to make the connection between identity awareness/skepticism and associative vs. mechanistic thought, but I don't really see it. Can you explain further? The only connection I see is that they're both byproducts of rationality and so getting good at one will make you better at the other.

3[anonymous]15y

It's nothing too deep: engaging in behaviors that are associated with your self-concept versus behaviors that you can see will causally lead to your goals.

1Will_Newsome15y

Ah, I see. Thanks.

4Will_Newsome15y

Thoughts?

3[anonymous]15y

Also, other people sometimes push you to be to person they think you are.

1Will_Newsome15y

Awesome link, thank you. I see there are articles I haven't read on self-enhancement, self-verification theory, self-concept, and more. Sweet. I'm starting to have more respect for sociologists / psychologists / social psychologists.

2Relsqui15y

I have a model I call the "badge/shield" theory, which goes like this: When someone tells you you're good at something, or when you otherwise believe that you are, it's like a badge. You wear it proudly, and you want to show it off. It makes you feel good about yourself and you look for examples to practice talents you have badges for. Example Badge: I think of myself as a good communicator, so I like mediating between friends who are misunderstanding each other, or explaining things to my classmates. When someone tells you you're bad at something, or when you otherwise believe that you are, it's like a shield. You hold it up, often preemptively, against any opportunity to do that thing. The inability becomes part of your identity, and you believe it excuses you from having to do the things you have shields for. Example Shield: I have a friend who claims she's terrible at math; I don't know, because I've never seen her try to do any. She won't calculate tips or split checks, because she has decided that she's a person who Can't Do Math and won't try. Oddly enough, the seed of this came from a Dear Abby column. A woman had written in frustrated that her husband never volunteered to look after the kids. The advice was to mention to their mutual friends (perhaps when the husband was in earshot) that he was so good with the kids and they really enjoyed spending time with him. The point was to encourage pride in how good a father he was, rather than getting him to do it through guilt or obligation. Just some more thoughts on what our identies really consist of.

2Will_Newsome15y

Reminds me of this book) on fixed vs. growth mindsets.

0Nick_Tarleton15y

This would be vastly more readable with some added line breaks.

3Will_Newsome15y

I was thinking that, but then I unconsciously reasoned that it'd make it look like I'd actually intended this to be coherent, and I further unconsciously reasoned that if people thought that then they'd think less of my writing ability, so I left it in original form on purpose so that people would think of better of me. My unconscious is stupid. I'll add line breaks. The thing is though, the text doesn't lend itself well to line breaks, so I just put them in at random. Francis Bacon had a giant one paragraph essay once about how knowledge is power, why can't I do it? Hmph. Any why is that line breaks make text so much more readable, anyway? I think it just gives the illusion that what you're reading is intelligent and structured. Other hypotheses?

7NancyLebovitz15y

I'm guessing that line breaks make it easier to keep track of which line you're on.

[-]Will_Newsome15y40

Regina Spektor wrote a song about CEV called "The Calculation". She seems to agree with Shane Legg that it may well be impossible, but she appears to think there's hope. YouTube video here.

Excerpt:

You went into the kitchen cupboard
Got yourself another hour
And you gave
Half of it to me
We sat there looking at the faces
Of these strangers in the pages
'Til we knew 'em mathematically
They were in our minds
Until forever
But we didn't mind
We didn't know better
So we made our own computer out of macaroni pieces
And it did our thinking while we lived

... (read more)

5ata15y

So that's the secret of achieving true AI!

6Will_Newsome15y

Analog computing via food. This is the dawn of the non-Bayes era!

1RobinZ15y

Do you have a citation for this claim, or are you asserting it as an alternative interpretation of the lyrics? Because, to make an analogy, you can tell me "Dear God" by XTC is a Christmas song but that doesn't make it true.

2Will_Newsome15y

Alternative interpretation, of course.

[-]MartinB15y40

If you are not a programmer, please tell us who you are, and how you ended up here. It sometimes seems like programmers are more likely to end up here, so it might be interesting to see who else does, and how. A programmer here is someone who knows how to do it, does it, and possibly likes it - be it for professional or private reasons.

0Relsqui15y

I'm not a programmer, although I code a little when I have something that needs coding. Someone (I forget who) linked me to the Parable of the Dagger, and I liked it so much that I nosed around to see what the rest of the blog was about.

0mindspillage15y

I'm not a programmer. (Well, I can write code--and studied CS for two years in college--but I avoid it even better than most programmers do; it's not a regular hobby and now I have to RTFM even for simple things.) I've lurked on various transhumanist lists for just about a decade and started following Overcoming Bias pretty early in its life, so I found LW from there. I have some general technical background (lots of math and quite literally just enough engineering to be dangerous), actually managed to finish the music degree, but mostly I'm a translator between hackers and lawyers/policy types; I just finished law school and am waiting to be licensed. (Though I expect to remain more of a researcher than a litigator.) I appreciate the programming mindset but ultimately I like words better, and I think I have a comparative advantage in dealing with the fuzzy, illogical bits that drive real engineers mad. Also I'm pretty sure this is my first comment here even though I've been a lurker for ages; I blame my recent dive into the IRC channel.

0Larks15y

I just started learning python, but I've been on LW for well over a year. I used to read Econlog, and from there Marginal Revolution and Overcoming Bias. I remember finding the post on Traditional Capitalist Values, which was the first article on LW I read.

0Matt_Simpson15y

I don't think I'm a programmer on your definition - I took two CS classes in undergrad (both in C++) and I have a working knowledge of R now that I'm in a stat grad program. I came to LessWrong when it started through OB, and I found OB through Marginal Revolution. At first, OB was "just another econ blog," but since I was a philosophy major at the time and had an interest in epistemology, OB really grabbed my interest.

0Barry_Cotter15y

I can't programme but I should learn. I amn't sure but I think my progression was soc.history.what-if to Randy McDonald's blog to gnxp to Mencius Moldbug to him linki´ng to "Beware of Steven Jay Gould" when Eliezer was on OB. Insofar as anyone who isn't a programmer can fit the cognitive profile I probably do.

0NancyLebovitz15y

I'm not a programmer. I think I ended up here because of OB.

[-]Will_Newsome15y40

Has anyone compiled a list of one (two tops) sentence rationality heuristics drawn from LW canon into one place, ideally kinda-sorta thematically organized? I think this would be pretty cool to post on your refrigerator or for flyers... at the very least, useful for finding common themes and core ideas.

If nobody has done this then I will spend a few hours tomorrow digging through LW and related materials (yudkowsky.net) for quoteables and, quality and time permitting, post a paste a draft either here or in a top level post asking for more things to add.

I p... (read more)

0Relsqui15y

I see you've made such a list, now; want to have some of the sayings on buttons? (Both NancyLebovitz and I make them, in different sizes; hers would fit more text but mine are less conspicuous accessories.)

0Will_Newsome15y

Probably someday, especially if I end up living in Berkeley next month. But currently I have -1000 dollars to my name, alas.

0Relsqui15y

Good luck with that. I'm down too but not by as much (and it's a sum of positive and negative, which wasn't true until recently).

0khafra15y

A question from the other side of tomorrow--did you make that list?

0Will_Newsome15y

No; I should have given a probability estimate. I think the event that popped up was a 20% chance kinda event. I'll say... 80% chance I'll be done by the end of tomorrow, just in case it takes way longer to speed read the Sequences than I'd thought.

0[anonymous]15y

I've found that during a variety situations, from games to social situations to tests in college, I'll often remind myself to "not be Bruce". So "Watch Out for Bruce" or "Don't be Bruce" both work as long as you know the article.

0[anonymous]15y

Are you basically looking for rationality catch phrases, similar to how some people post motivational quotes on their bathroom mirror so they see it in the morning?

0Will_Newsome15y

Catch phrases yes, but not necessarily all that catchy: as long as they're short and sum up a good rationality heuristic.

[-]Will_Newsome15y40

On the edge of sleep I hypothesized that an average human could build an FAI if they were given lots of time and paper to write down a very long chain of "Why do I believe that?"s. I then realized that such an endeavor would very quickly end in an endless loop of infinitesimal utility and a lot of wasted trees. Moral: rationality is a lot harder than the application of a few heuristics. Metamoral: Applying that heuristic doesn't make it much easier.

[-]erratio15y40

Idle observation:

Clippy gets consistently voted up on a lot of his comments because we find him amusing, and rarely gets downvoted because very few of his comments are substantive. We will end up looking extremely silly to new members if he gets enough karma to put him into the list of top contributors.

So... I guess it depends whether we pitch ourselves as shiny fun community or serious rationalist Singularitarians as to whether this is actually an issue.

[-]wedrifid15y190

Clippy gets downvoted quite a lot too (although less and less... he's learning!) He also makes quality comments, including the expression of some insights that would be punished if made by a 'human'. Lesswrong humans sometimes try to bully each other into pretending to be naive utilitarians instead of rational agents with their own agenda.

I guess it depends whether we pitch ourselves as shiny fun community or serious rationalist Singularitarians as to whether this is actually an issue.

This is not a singularitarian website (although rationalists are often singularitarians.) Also note that we spend a lot of time here discussing fanfiction that is written by the lead researcher in the SIAI. We cannot credibly claim 'sensibleness' or sophistication.

4erratio15y

Yes, but the vast majority of his comments concern his paperclip agenda. If a larger proportion of his comments were insightful rather than just funny I would be happier, but as it is his noise:quality ratio is rather high. A significant part of the Sequencess is made of posts that argue for a singularity in the near future, with complete seriousness. A large number of us are not singularitarians but I don't know whether I would say the community itself isn't singularitarian. We also have lots of posts about more serious topics. Having fun threads where we discuss HP and Twilight fanfiction doesn't mean that the community as a whole isn't trying to present itself at least somewhat seriously. And most top-level posts that are openly silly or non-substantive get heavily downvoted. As an example of the somewhat serious nature of the community, there seem to be a fair number of people who have had personal epiphanies (mostly about atheism) that have had a huge impact on their life as a result of reading the Sequences.

9wedrifid15y

On the other hand in the early months of lesswrong the subject was explicitly banned. That was part of an effort to ensure that blog identified as about rationality and not "singularity with rationality used to support it". See the discussion on clown suits. I included scare quotes around 'sensibleness' deliberately. I don't think Clippy reduces the quality of comments on the blog and I also don't think that discouraging Clippy for the purpose of appearing sophisticated would increase the quality of comments on the blog.

7erratio15y

You've convinced me on this point

1Wei Dai15y

Can you point to some examples of this?

3wedrifid15y

Bully is perhaps a strong word for 'apply some degree of social pressure'. If I were to try to find examples I would probably begin with a search for the word 'should'. That term will give false positives and false negatives but it is a good start. Conversations regarding cryogenics advocacy would also produce a few hits. Some of the PUA discussions too, come to think of it but I wasn't trying to go there. Moralizing and 'shoulding' at others is just something humans do. Sometimes the Clippy persona can avoid that. It avoids the failure mode of expecting other people's utility function to be subject to debate.

2Wei Dai15y

I think people (human beings) do not have utility functions, and the closest things to it that we do have (i.e., values and goals) are subject to debate. I believe this is also the local consensus. Are you perhaps saying that we expect other people's values and goals to be more subject to debate than they actually are? If so, this is a novel idea for me. Can you give or link to a longer explanation?

7wedrifid15y

(Yes, utility function over simplistic, etc.) I refer to the difference between on one hand telling people their values are bad and on the other hand speaking as if the values of the subject can actually be determined by the speaker. The latter ignores the boundary between where one agent ends and the other begins. This is something humans often do, albeit more so outside of lesswrong than within. In my observation it is inversely correlated with maturity. One of the things about Clippy is that nobody expects him to stop caring about paperclips just because someone else says so. In this Clippy is ironically shown more respect than a low status human may expect.

2Wei Dai15y

Thanks for the explanation. That helps me to understand why some people seem to value Clippy more than I do. (I think I personally have rarely been involved in the kind of phenomena that you describe.)

No problem; just relabel "Top Contributors" as "Top Humans."

[-]steven046115y100

With Clippy at 1k and the top 10 at 6k, it's just way too improbable to worry about.

8Clippy15y

I make good, substantive posts. Like this one and this one and this one and this one and this one. I have the same right to be here that erratios do. I provide additional assistance in that I have a perspective untained by anthropomorphic cognition.

5AdeleneDawner15y

That's true. It's also true that you make silly or irrelevant comments fairly often, which is not in and of itself problematic, but those comments tend to get upvoted significantly, which is, at least in my opinion, an indication that how we're collectively choosing what to upvote is non-optimal - especially if it's how you or anyone else is getting most of their karma. For reference, here are links to the most recent 10 comments of yours that received at least 5 net points: 1 2 3 4 5 6 7 8 9 10

3Larks15y

We seem to upvote based on emotional reaction rather than rational merit.

3erratio15y

I admit I may have been overly anthrocentric in my attitude towards you. However, I still maintain that you often allow your paperclipping goals to push you into making lots of poor comments. I would welcome more insightful ones such as the examples you provided. (and in fact I see that I upvoted the one about lying and isomorphism)

1JoshuaZ15y

You've commented before that you were programmed in your original form by humans. So how can you be untainted by human cognition? It might be less direct than for a human but there will still be a fair bit of a connection. Indeed, you seek to maximize the number of paperclips in the universe, and I'd venture to suggest that if alien cultures exist they will likely not even have a concept of paperclips. You're very close to humans in mindspace once you realize how fantastically large the space of possible minds is.

2wedrifid15y

With countless CPU hours of cleansing ritual and self modification.

6Nisan15y

Sometimes, wanting to appear more serious means you're taking yourself too seriously. This is one of those times. Pancake.

2erratio15y

Sashimi!* *Translation: Point taken :)

4MatthewW15y

A simple fix would be to not bother publishing a top contributors list.

6JoshuaZ15y

I'm not sure this is a good idea. If there's something that empirically makes us look like we're not being rational we should deal with that issue. Hiding that data is not a good solution. However, I do have to wonder what in general the point of having the top contributors is. I'm not even sure that total karma is a useful metric of much since one person could have much higher quality comments than another but most much more rarely and yet the person with high quality comments presumably should receive more attention and their comments should be more closely paid attention to. It might be nice to have a display of average karma, not just total karma. However, this would still I suspect give Clippy a fairly high karma, so if you object to Clippy this won't solve anything. Also note that many upvotes are are not connected to the quality of remarks in any strong sense. See for example this comment by Eliezer that is now at +53 which is presumably connected to the unique status that Eliezer has as the founder of LW.

4AdeleneDawner15y

I agree with this comment. I recently noticed that Clippy has higher karma than I do. I don't find that upsetting, but it is a bit disturbing, even given that I haven't been very active recently and have never made a significant top post. (I've made exactly one top post - an open thread, for which I believe I earned 50 karma.)

8Clippy15y

I want you to know that you'll still be my friend even if you have more karma.

I'm trying to defeat my bad excuses for cryocrastinating, such as confusion about how to decide what to sign up for. (A few less-bad excuses will still remain, such as not currently having any income at all, but I'm working on remedying that.) Other than Alcor's included standby service, what significant differences are there between Alcor and CI (and are there any other US cryonics providers I should be considering)? What accounts for the big difference in pricing? Does my being a young healthy person affect anything relevant to this choice?

And if I go wi... (read more)

3wedrifid15y

Your soul is stored in your spleen. If you lose that then the best you can hope for is to be restored as a p-zombie. You will also lose the majority of your midichlorians. Imagine it, you are stuck as a head in a jar and you don't even have your ability to force grip! More seriously there is some information stored outside of your brain. Your motor skills obviously. A lot of the skill in weight lifting for example is in compensating for and overriding the reflexive reaction to your movement. But that's no problem. Just tell the FAI that you had elite bow hunting and nunchaku skills and you'll be set. Information that could be of relevance to your personality is the component of your reactions and emotional experience that is determined by your posture and physiology. Your body changes how your feel, and vice versa. Then there are the hormone secreting organs, the function of which fundamentally alter how you behave... A lot of this could be reconstructed reasonably well based on evidence in your head. Including, for example, your memories of what you were like, reverse engineered with compensation for any evident biases. But something will be lost, however trivial.

[-]teageegeepea15y30

I don't know if it's already been discussed here, but Andrew Gelman & Cosma Shalizi's paper on the philosophy of statistics and Bayesianism sounds like it might be of interest to many here.

2Matt_Simpson15y

It has. See this and this

0teageegeepea15y

Thanks.

[-]Mitchell_Porter15y30

Is anyone else familiar with the situation of having a definite research agenda, being unable to get financial support for even a part of it, and so having to choose to do something irrelevant just to stay alive? That last step seems to be particularly agonizing. It's like having to choose the form of torture to which you will be subjected (with starvation or homelessness being the default if you refuse to choose).

I have no doubt there are people here who have wanted to do research at some time, and either compromised or had to settle for non-research job... (read more)

0Wei Dai15y

Cousin_it said some time ago, Is this something that might be feasible for you? If so, I think people here can probably give you some suggestions on how to find such a day job. Also, I'm curious, what is your "definite research agenda"?

1jacob_cannell15y

once a week? - slightly jealous of cousin_it's day programmer job.

0Mitchell_Porter15y

I can program, but don't do it much. Right now the research topic is, how can I pay next week's rent? But the basic agenda is "CEV, adapted to whatever the true ontology is".

[-]blogospheroid15y30

Eliezer had written that the two most important mathematical problems were the friendly goal itself that a self modifying program would have and goal stability under self modification.

Now, a useful perspective I have often used in the past is to see incentive structures and see who has the motivation to take a crack at a problem. Now, goal stability seems to be a very very useful problem to solve for many institutions. Almost all of us have heard of institutional missions getting diluted and even seen it when people change in a firm. National governments,... (read more)

6[anonymous]15y

Have you ever followed electoral reform debates? There is a distinct pattern, people out of power want change, people in power do not. They are happy with the status quo. Only the people in power have the power to enact change. So change tends not to happen. That is changes to the goal system rarely happen, especially if they might disadvantage the incumbents. I have ideas that I'm going to try out. However they are more to do with acquiring feedback from other people outside of the organisation, which would foster goal stability iff the feedback from outside the system does. I.e. if the feedback came from donors, that consistently wanted a charity to do one thing, it would keep the charity on track.

0CronoDAS15y

Thought: Consider the history of the Seventeeth Amendment to the Constitution - it was really hard to get the Senate to approve a change in how Senators were chosen.

1Mitchell_Porter15y

The transcendentally important problem is to identify "the friendly goal itself". Compared to that, goal stability is just a technicality.

I posted a link to The Meaning of Life in response to a question on yahoo. It seems we possible we could gain traffic, and help people, by answering other questions in a similar way.

[-]Will_Newsome15y30

A reasonably good (if polemic) article about common misunderstandings of Buddhism and meditation, and what they're really about. A quotation: "Mindfulness is the natural scientific method of the mind. A scientist brings a microscope, a meditator brings mindfulness. We need to realize that we live in a state of deep assumption about the way the mind works, which then extends to our understanding of the world. We rarely experience anything directly, without first slowing down and paying attention. A scientist shouldn't make statements based on unsubsta... (read more)

[-]PhilGoetz15y30

A proof that a 2008 penny comes up heads 75% of the time:

I took a large sample of pennies not made in 2008, and flipped them. They came up heads .5 of the time, so my estimate of P(heads|penny) = .5.

I then took a sample of coins of all types made in 2008, and flipped them. They came up heads .5 of the time; so P(heads|2008) = .5.

These two samples are independent, so P(heads | 2008 and penny) = 1-(1-P(heads|penny))(1-P(heads|2008)) = .75.

ADDED: JGWeisman got it - there's no causal connection between being a penny, or being made in 2008, and coming up hea... (read more)

3jimrandomh15y

The mistake is that you can only decompose a conjunction that way when it's on the left side of the |, not when it's on the right. That said, I think a collection of a dozen or so incorrect bits of statistical reasoning like this would make great exercises, and would encourage the creation of a top-level post based on that premise.

3PhilGoetz15y

I was wrong, and jimrandomh was right. I said: P(X|A,B) = 1-(1-P(X|A))(1-P(X|B)) if P(A,B) = P(A)P(B) But P(X|A,B) = 1-P(~X|A,B) therefore P(~X|A,B) = (1-P(X|A))(1-P(X|B)) = P(~X|A)P(~X|B) This is equivalent to claiming P(X|A,B) = P(X|A)P(X|B) And this is wrong for most distributions.

2PhilGoetz15y

That's not the mistake. You might be right. It makes sense to me that P(X|A)=.5, P(X|B)=.5, independent(A,B) => P(X|A,B) = .75. But I can't derive it. Suppose there's a 50% chance of rain when the weatherman predicted rain; and a 50% chance your neighbor will turn on his lawn sprinkler on Saturday. (Your neighbor doesn't watch the weather report.) It's Saturday, and the weatherman predicted rain. P(neighbor's lawn wet | Saturday, predicted rain) = 1 - (1-.5)(1-.5) = .75

1Douglas_Knight15y

Of course you can't derive it - you gave a counterexample! More precisely, let A,B,X be heads on three independent coinflips. Jim gives a fallacy that it looks like people might actually commit[1], but I don't think it's what's going on here. I think the issue is the meaning of "independent evidence." ETA: the following paragraph is wrong and unnecessarily complicated. Maybe it would be better to skip down to my later comment. One kind of evidence is measurements. In that case the event of interest causes the measurement, plus there's some noise. I think what we usually mean by "independent measurements" is that the noise for the one measurement is independent of the noise for the other measurement. How you combine the measurements depends on your noise model (as does even saying that the noises are independent). If your noise model is that there's a large chance of a correct read and a small chance p of an incorrect read, then agreeing reads allow you to multiply the two p's [ETA: this is wrong] (if p is not small, what happens depends on the details of an incorrect read), which is roughly what you did, except that you confused the measurement of a probability with the probability of noise. You might be able to struggle through interpreting the measured p=.5 as a noise, but it would require a detailed noise model. The lawn has opposite causal structure from the kinds of measurements/evidence above (so I agree with Jim's complaint that the two problems are unrelated). Causal structure has to do with sides of |, so maybe when you unwind this discussion of "evidence," it turns into Jim's fallacy, but I doubt it. [1] ETA: Do people actually commit Jim's fallacy? As I explain above, I don't think that's Phil's original mistake, but he does make it in the quoted text. If Phil correctly abstracted his train of thought, then Jim is correct. But I think Phil probably learned this in the context of "evidence" and overgeneralized when trying to understand the problem in

0PhilGoetz15y

It's not a real counterexample. The data sets aren't actually independent. The real trick is that I try to convince you that the 2 datasets being used are independent because they don't intersect. Common definitions of independence would say they are independent because you can't compute any correlation between them. But they both have the same underlying probability distribution generated from the same source. I'm confused about what "independence" should mean in this case.

0Douglas_Knight15y

When you make the statement it seems to be written in a standard formal language and you have to interpret it with A and B as random variables, "independent" meaning independence of random variables. Then the formal statement is false by the example of three independent coin flips. When I first posted my comment, I had a panicked moment and deleted it until I realized that the three coins example is a counter example; whether it is the same as your example is not important. Yes, the two data sets are not independent, if you're not sure how coins are weighted. Two flips of a coin weighted in an unknown way are not independent random variables. But that's good, because if they were independent in that way, their evidence wouldn't add. But since they are independent conditional on the thing we want to measure, their evidence does add. I repeated your original formulation of the consequence of independent evidence, but it is not correct. If you have two independent pieces of evidence that would each send you to 90% certainty, you do not conclude 99% certainty. It depends on your prior! If your prior were 50%, you conclude 98%. If your prior were 90%, neither piece of evidence told you anything, so together you learn nothing! (Moreover, if A and B are empty observations, that is a counterexample to pretty much any formulation.) When is evidence independent? Evidence is the log of the likelihood ratio, which is what shows up in a bayesian update. The likelihood ratio for X involves only probabilities conditional on X. Thus independence of A and B conditional on X is exactly what we need for the probabilities to multiply; for the evidence of A and B to be the sum of the evidence of the individual events. In particular, the evidence from two flips of a single weighted coin give independent evidence about the weight.

0PhilGoetz15y

The problem I was originally trying to approach is a problem I'm having at work, where I have multiple sources of evidence to conclude property(X); but I don't have enough data to compute the correlations between these sources of evidence, because for most datapoints, only one source applies. When 2 or 3 of these sources agree, how do I compute the probability? But on reflection, it's a different problem.

1jimrandomh15y

These problems only look similar because you are hiding the assumption that (neighbor's lawn wet) = (rain or sprinkler turned on). From this, P(neighbor's lawn wet | Saturday, predicted rain) = P(rain or sprinkler turned on|.5 chance of rain and .5 chance of sprinkler) = 1-P(~rain|...)P(~sprinkler|...) = .75. But there is no valid similar statement for the coin; the analogous disjunction would be (coin heads) = (2008 coin heads or penny heads), in which case treating the clauses of the right hand side as independent means flipping two coins and checking if at least one of them is heads.

0PhilGoetz15y

I disagree - that makes no difference. Just change "it will rain" to "my neighbor's lawn will be wet", and "my neighbor's sprinkler will turn on" to "my neighbor's lawn will be wet".

1JGWeissman15y

1) P(lawn wet| Saturday, predicted rain) = P(lawn watered or rain| Saturday, predicted rain) 2) = 1 - P(not lawn watered and not rain| Saturday, predicted rain) 3) = 1 - P(not lawn watered | Saturday, predicted rain) * (not rain | Saturday, predicted rain) 4) = 1 - P(not lawn watered | Saturday) * (not rain | predicted rain) What is the analogy to step 1 with the coins?

0PhilGoetz15y

Rephrase the problem without the inferential step from "it will rain" and "sprinkler will turn on" to "lawn will be wet". Just use "lawn will be wet" in all those places. We can do this because we are using symbolic logic.

0JGWeissman15y

That is throwing away the information that allows you to make the inference. The equation: works because each part of the condition "Saturday, predicted rain" is involved in a different causal path to the result we are interested in. With the coins, neither being a penny specifically, or being minted in 2008, causes the coin flips to result in heads, so the analogous equation does not work.

0PhilGoetz15y

That's right - but the problem is not that "you can only decompose a conjunction that way when it's on the left side of the |, not when it's on the right." The reasoning is syntactically correct. The two cases are syntactically identical when you do the syntactic substitution I suggested above. Or... hmm, I may be confused.

1JGWeissman15y

The two cases are syntactically identical because you have not explicitly explained how you meet the conditions that allow the decomposition you used, conditions which are not in fact met in the case of the coins. Your "syntactic substitution" removes the information that could be used to show you meet the conditions. (Just to be sure, you did intend the original "proof" as a "find the error" exercise, right?)

0PhilGoetz15y

Yes - but also to help me figure out how you ask whether 2 data sets are independent when they don't intersect.

0[anonymous]15y

Of course you can't derive it - you gave a counterexample! Jim gives a fallacy that it looks like people might actually commit (do they?), but I don't think it's what's going on here. I think the issue is the meaning of "independent evidence." One kind of evidence is measurements. In that case the event of interest causes the measurement, plus there's some noise. I think what we usually mean by "independent measurements" is that the noise for the one measurement is independent of the noise for the other measurement. How you combine the measurements depends on your noise model (as does even saying that the noises are independent). If your noise model is that there's a large chance of a correct read and a small chance p of an incorrect read, then agreeing reads allow you to multiply the two p's (if p is not small, what happens depends on the details of an incorrect read), which is roughly what you did, except that you confused the measurement of a probability with the probability of noise. You might be able to struggle through interpreting the measured p=.5 as a noise, but it would require a detailed noise model. The lawn has opposite causal structure from the kinds of measurements/evidence above. Causal structure has to do with sides of |, so maybe when you unwind this, it turns into Jim's fallacy, but I doubt it.

2JGWeissman15y

Why?

1PhilGoetz15y

Say you have 2 independent pieces of evidence (M(c) and N(c)) that Q(c) is true. Q(c) is true half the time when M(c) is true; so God flips a coin (but not a 2008 penny), and Q(c) is true if it lands heads. Half the time, God's coin came up tails. God flips another coin, for N(c), and again, if it comes up heads, Q(c) is true. Q(c) is false only if both coins come up tails.

0JGWeissman15y

This has the same structure as your response to jimrandomh, which I reply to here.

1saturn15y

What you have calculated is the probability that, if you flip two coins, a penny and a coin made in 2008, one or both of them will come up heads. This is analogous to the sprinkler/rain problem - the grass can be wet by either or both of two independent events. I'm not sure if there is a way to find P(heads | 2008 and penny) using only the information given, but I don't think a computer program would be confused by this kind of thing if it kept rigorous track of the events that its probabilities refer to.

0[anonymous]15y

I was wrong, and jimrandomh was right.. I said: P(X|A,B) = 1-(1-P(X|A))(1-P(X|B)) if P(A,B) = P(A)P(B) But P(X|A,B) = 1-P(~X|A,B) therefore P(~X|A,B) = (1-P(X|A))(1-P(X|B)) = P(~X|A)P(~X|B) This is equivalent to claiming P(X|A,B) = P(X|A)P(X|B) And this is wrong for most distributions.

[-]Relsqui15y30

I'm new here. Why is this thread a two-parter?

5NancyLebovitz15y

It's an annoyance to deal with huge threads, so we have a convention of starting a new thread when there are about 500 comments on an otd thread.

2Relsqui15y

Makes sense. Thanks.

[-]NancyLebovitz15y30

The Life Cycle of Software Objects by Ted Chiang is sf which raises a number of questions of interest at LW-- in particular, it's about human relationships with non-self-optimizing AI.

At this point, the story is only available as a slightly expensive but possibly collectible little hardcover, but I'll estimate a 90%+ chance that it will be in at least one of the "Best of Sf" collections next year.

I'm going to rot13 specifics in case anyone wants to read it without even the mildest of spoilers.

Sbe rknzcyr, jung qb lbh qb vs gur NV lbh bja naq ner ... (read more)

0Nisan15y

Thanks. I got my copy yesterday.

0jacob_cannell15y

Interesting .. I may have to get that for my kindle to read on the plane.

0Jonathan_Graehl15y

I've requested it from the library. Thanks - looks interesting.

[-]Mass_Driver15y20

I found a hardware bug in my brain, and I need help, please.

Dan Savage once advised an anxious 15-year-old boy to stop worrying about getting his 15-year-old self laid and go do interesting, social, skill-improving things so that it would be easy to get his 21-year-old self laid. IMHO, this is good advice, and can be applied to any number of goals besides sex: love, friendship, career, fame, wealth, whatever floats your boat.

My problem is that whenever I try to follow advice like that, I find myself irrationally convinced that one or two days of skill-buil... (read more)

4Alicorn15y

Go meta: build the skill of building skills for a long time.

1Mass_Driver15y

Well, yes, that's precisely what I want to do. Any advice on how?

1Duke15y

I was going to say that same exact thing in 10x as many words.

1AdeleneDawner15y

I find it useful in similar situations to set very specific, non-time-based goals, e.g. "be able to comfortably bike N miles" or "be able to comfortably do X sit-ups".

0Mass_Driver15y

Interesting. How do you choose N? Do you have any sense of how or why this works for you?

2AdeleneDawner15y

...on reflection, that advice isn't as well-phrased as it could be. "Be able to comfortably bike for an hour without resting" would also be a valid example, even though it has a time element to it. The point is to have a specific, objectively-observable thing to accomplish that's as close to my actual goal as possible. If I were to instead state my goal as 'go to the gym twice a week for 3 months', my brain would count it as a success even if I only did 5 minutes of actual exercise each time. (And, because my brain knows that about itself, and knows that meeting the stated goal won't help with the actual goal, it's unlikely to even attempt to meet the stated goal, but that seems like a deeper level of complication than you're experiencing.) I usually stick to goals where at least a ballpark value of N is obvious from the context - I don't have any well-tested advice on what to do when it's not.

[-]Document15y20

If I think I may have guessed the dangerous idea, can I PM someone to check my guess? I'd like to go back to not feeling like I need to censor what I read (which could be solved by just looking up the idea, but other people seem confident that that's not a good enough reason to do so).

2saturn15y

You can PM me; I read the original posts.

0Document15y

I've changed my mind for now, but thanks.

In the spirit of instrumental rationality, I pose to you the following non-theoretical problem: I have some programming interviews coming up in a week or two, including some areas I haven't used in a while. Cramming technique recommendations? Pre-interview meditation? Negotiation pointers (hopefully it comes to that).

[-]andreas15y20

I'm in Cambridge, MA, looking for a rationalist roommate. PM me for details if you are interested or if you know someone who is!

[-]Will_Newsome15y20

At the limit of unlimited resources, where isn't Bayesianism the ideal epistemology? Where is it silent? Where does it lead one astray? Is Bayes the avatar of the void at the limit of perfection?

[-]Johnicholas15y20

Eric Falkenstein's text "Risk and Return" presents exhaustive evidence that human utility functions are noticably different than standard economic theory - notably the presence of a large "wealthier than my neighbors" term in our human utility function:

http://www.efalken.com/RiskReturn.html

He applies this to finance, and discovers that if people mostly care about relative wealth, then riskier investments will not see higher returns in the long run; which is very interesting.

[-]Mitchell_Porter15y10

(I've moved this to part 3 of the open thread.)

I'm thinking about buying Adrafinil (powder) and a bunch of empty capsules. I'll put Adrafinil in half, sugar or something in the other half, and then do the obvious.

Does anyone else do blind testing on themselves?

There's talk about self-experimenting on LW, but not much about blind tests, even in areas where that would be possible, like nootropics and supplements. I wonder why that is.

0RHollerith15y

Adrafinil causes a buzz like amphetamine does (but not as intense relative to its effects on alertness). Well, modafinil does, and adrafinil exerts its effects by turning into modafinil in the body. I notice the buzz even at the low end of its therapeutic range. A little higher (200 mg of modafinil for a 200-lb man) and other people (who are vigilant about that sort of thing probably because they grew up around substance abusers) have been able to tell that I am on a mind-altering drug. (I become more talkative for one thing.) Also, it is not obvious what you mean by "do the obvious". Consider weed, which tends to make you think you are having penetrating insights, but if you write them down and read them the next day, they are nonsense.

0Bongo15y

Have someone separate the capsules into two bowls, and take from each for a perioid of time, avoiding the placebo effect. Afterwards, check which was sugar and which was Adrafinil. If for this particular substance it's easy to notice that you're on it, that wouldn't work, of course.

0RHollerith15y

Right, but no blinded clinical trial I know of ever had the goal of seeing if subjects could discriminate between drug and placebo and that is it -- they, e.g., administer tests of memory or ability to avoid errors in a tedious task. I anticipated that you would say that the goal is to collect subjective impressions of your effectiveness, which is why I wrote the final paragraph of my previous comment.

2Douglas_Knight15y

I've heard of a couple of studies of caffeine that asked for subjective measures. That's not quite the same as asking the question straight-up, but it's pretty good. The main problem is that I don't know if they told people what was going on, or whether they lied and said it was caffeine. That should be easy to answer from the studies, but I can't find them. I think there was a recent study that made the rounds that administered different amounts of caffeine through coffee and asked people how wired they felt, with no distinction. (The main part of the study was where they gave people cognitive tasks and made claims to the subjects about whether caffeine helped. The interaction of the caffeine and the claim was weird.) Also, I heard about an older study that fairly small (but nonzero), amounts of caffeine eliminated withdrawal headaches.

0wedrifid15y

On a related note a study found that the smell of coffee gave a cognitive improvement. I obviously wouldn't expect the effect to last if you gamed it repetitively. I only saw a second hand report on the study, which suggested smelling the coffee 15 minutes before you actually drink it, maximising the overall effect. It was probably one of the posts made by http://www.lumosity.com/.

0wedrifid15y

Did it? Surely they took that measure too...

0Douglas_Knight15y

Yes, I think the caffeine always helped, but I think there was one combination where the effect of claim was larger than the effect of the caffeine. At that point, I become a bit nervous about the meaning of whether caffeine helps. (by "interaction," I meant the interaction of effects on performance)

0Bongo15y

Okay. Makes me wonder, though, if the subject knows what he's getting anyway, why is the test blind in the first place? Edit: Yes, I'd do objective test like dual-n-back and math. Other than collecting subjective impressions, I guess the usefulness of this idea depends on whether performance in those test is susceptible to improvement by placebo effects.

[-]CannibalSmith15y10

I was offered life insurance by an Alico agent. What do you guys think? I don't know what I don't know about life insurance, so throw me all you've got.

I would get 36k€ if I was crippled 50% or more and 20k€ if I died or survived till 62, all for 260€ a year (that's 5% of my current salary). Does anyone know what injuries constitute what crippledness %?

0CannibalSmith15y

THIS IS WHY WE NEED A FORUM or something

[-]XFrequentist15y10

I've got a lengthy-for-Less-Wrong follow up to this post (mostly) written.

I'd appreciate another set of eyeballs to look it over and give some critique. Anyone interested?

1jimrandomh15y

Put it in Pastebin and post a link in the Less Wrong IRC channel.

[-]curiousepic15y10

I will be attending the Jon Stewart/Stephen Colbert semi-satirical rally in DC on October 30th: http://www.rallytorestoresanity.com/ I intend to carry a sign promoting rationality, something along the lines of "Know your cognitive biases" with a URL to point toward a good intro. What would be the best site, with a decently concise URL? And is this a good idea? An other suggestions? Pamphlets?

[-]erratio15y10

Request for help in link-hunting: I remember reading here about an experiment involving pigeons and humans, where the experimenters offered 2 options with (say) 80%/20% rewards, and the pigeons quickly worked out that the highest expected reward was from always picking the 80% option, while the college students reproduced the percentages given to them. Anyone know where to find it? (Or better yet, a link to the original paper/article).

Danke

3Kazuo_Thow15y

I can't seem to find any talk of an experiment with 80% / 20% frequency options, but XiXiDu mentioned one where pigeons were found to out-perform humans at the iterated Monty Hall problem. Here's the paper itself.

0Relsqui15y

That paper was a great read, thank you.

0erratio15y

ETA: On further thought I think that's the paper I was looking for after all, I was just thrown off by the reference to Monty Hall for some reason. My thanks. Sadly, that isn't the paper I was looking for. I found a vague reference here but it looks like I either made the experiment up entirely, it uses rats or mice instead of pigeons (I could have sworn it was pigeons though!), or it was on another website (unlikely - this kind of topic is far more likely to be on LW than anything else I read).

2Sniffnoy15y

Hm, I was thinking you probably meant the experiment talked about in this old Language Log post. Not sure where an original reference might be found, though.

1erratio15y

Ha! I didn't think to check Language Log, and that post does look very familair. I must have attributed it to LW based on the subject matter. Upvoted for some sound detective work, I think I have enough information to go reference chasing now :) Oh and in case anyone was wondering: I was looking for this experiment because I'm currently writing about the difference between child langage acquisition and adult second language acquisition, and some experiments I came across showed that exact same pattern - that the children will quickly over- or undergeneralise based on the data, while the adults engage in probability matching.

0Relsqui15y

This is a topic I'm really interested in. What arae you writing about it?

2erratio15y

Cognitive and linguistic development is something that continues over our whole lives, but at the earlier ages there are a whole lot of basic concepts that aren't in place yet, like theory of mind (anecdotally, most roleplayers who try to roleplay with their young <10 year old kids find that the kids have trouble pretending to be someone else), and there are lots of linguistic structures that have certain thresholds of complexity. So children pick up new linguistic structures at around the same rate as they develop the cognitive machinery to deal with them, and they try to regularise everything. Adults on the other hand already have all that machinery in place, plus they already know what language is supposed to be like based on prior experience, so a lot of structures are much easier for them to acquire (and interesting, the order of acquisition for adults in a particular language is largely the same regardless of their first language, so it's prior experience with language in general that's important, not experience with similar structures). Oh, and there's some not-yet-repeated results that also say that education/literacy is a big factor in individual linguistic complexity, which would probably have knock-on effects for acquisition.

2Vladimir_M15y

erratio: Suppose language X inflects verbs for present/past tense, but has no definite article, whereas for language Y, it's the other way around. Assume now the a native speaker of X and a native speaker of Y are learning some third language Z, which has both features. Are you saying that both learners are going to acquire the use of the article and the past tense in Z in the same order? This sounds strikingly implausible to me, though I might be wrong.

1erratio15y

According to the book I'm referencing from, one of the studies was other language -> English, where the other languages were: Greek, Persian, Italian, Turkish, Japanese, Chinese, Thai, Afghani, Hebrew, Arabic, Vietnames.So a decent spread of different types, with lots of points of similarity and difference between them and English. There have also been other studies showing similar findings in other L2 languages and from other researchers in general confirming those results, although this particular book doesn't go into much detail about the others. Oh and the aggregate of the results also show that the order is the same regardless of the method/environment of acquisition, too.

1Vladimir_M15y

Could you please write what book that is? I'd be curious to look it up. One issue I see with the study (according to your description) is that English, by historical accident, doesn't have any related languages that would be close enough to preserve some mutual intelligibility and make the analogous grammatical rules immediately obvious for their speakers (like e.g. between Slavic or Romance languages). I would expect that the conclusion doesn't hold when comparing with learners speaking a closely related vs. a distant or unrelated language. Another issue is how "acquisition" is defined. How accurate does one's use of a particular grammatical feature have to be so that it qualifies as "acquired"? For many subtler grammatical rules, nearly all adult learners will never master them fully; for example, I still don't know how to use the English definite article accurately, and neither does any other non-native English speaker I know. Formulating these criteria in a way that makes sense seems like a very tricky problem.

0erratio15y

ok, the main book is "Second language learning theories" by Rosamond Mitchell and Florence Myles, p32-33 in my library copy. The main studies you'll want to fact check are by Dulay and Burt (various years and publications) for children acquiriing a second language. Bailey, Madden and Krashen (1974) for the Dulay and Burt results replicated in adults, and (from a different textbook) Zobl and Liceras "functional categories and acquisition order" (1994) in the journal language learning which I seem to remember provides a good summary of the studies to date Probably not, no, but I did find it striking that the results held across such typologically different languages. For example, English has plenty of derivational morphology and so do most European languages, but most Asian languages don't. But the order or acquisition was still pretty much the same. I believe the bar for 'acquired' is usually set somewhere around 80-95% accuracy, but I haven't looked into this aspect much. Another sensible way of measuring it would be perfect or near-perfect use in common contexts, and ok accuracy in less common contexts

0Vladimir_M15y

Thanks for the references!

[-]teageegeepea15y10

Julian Barbour's theory of time has some adherents here, and Everett's interpretation of quantum-mechanics has even more. Over at my blog a physicist criticizes them both.

A while back here there was a dispute arising from the book 'Why Zebras Don't Get Ulcers". Barry Marshall, who won a nobel for ingesting bacteria and giving himself ulcers, says there's no evidence for stress contributing to ulcers or any other medical condition.

[-]JamesAndrix15y10

A D V E R T I S I N G

For all your best contrarian ideas!

Make more efficient use of your persuasive time and dollars.
Leverage decades of professional experience for YOUR beliefs!
Persuade even those whom the sequences would never reach.

Can we afford not to?

1wedrifid15y

Yes! I'm not an evangelist and nor do I assume that evangelism will always be beneficial. An influx of people who are not biased towards rationality due to genetics could be detrimental to a given cause.

1jacob_cannell15y

Do you seriously believe there is a genetic bias for rationality? Do you think some paleolithic humans were more 'rational' than others because of a few gene tweaks here or there?

3wedrifid15y

I seriously believe that some people have genetic defects that make some of the important arational systems to malfunction. They then compensate for those weaknesses using the (for most practical purposes inferior) general intelligence capabilities. There is genetic variation in the extent people use different styles of thought. Preference for near mode vs far mode for example. A side effect of this is the kind of 'rational' thinking that is relevant to, for example, signing up for cryonics or caring about existential risks.

1gwern15y

It seems quite possible to me. I leave how as an exercise for the reader, with the hints 'testosterone' and 'anger'.

-2JamesAndrix15y

Hello Lesswrong. Look at your community, now look at me, now look at your community, now back to me. Do you want your community to smell like me? No, you don't. But with the power of old spice Conclusion scented body wash, I can influence large numbers of people to more correct choices for still wrong (but less wrong) reasons, while leaving your community smelling as unbiased as a master beisutsukai. By 'I' I mean by proxy you, I'm on a motorcycle.

1wedrifid15y

People move. Unless a community manages to maintain a sufficiently low status even among groups of people with overlapping beliefs introducing more social-thinkers into related belief systems will influence that community's population. This is more important when it comes to things like interest in creating a technological singularity. Keeping that whole concept 'fringe' and stigmatised as nerdy Sci. Fi. is a good thing.

1JamesAndrix15y

This discussion has surpassed my current aptitude for converting it to ad copy :-) My key point is that advertising is a more efficient persuasive tool than argument. Even if you're concerned about dangerous ideas in unsafe minds, we should still be using advertising to shift those views, it is unlikely that the current distribution is optimal. (though this argument is weaker, since I don't think we currently try argument to do this) I also think that modern advertising can be targeted enough to sort out whatever demographic you want, and still persuade better than the sequences, I suspect even to the most rational humans.

0wedrifid15y

On this I am in full agreement!

[-][anonymous]15y00

EDITED: Whoops wrong (outdated) open thread. Sorry.

Utilitometer:

I've been doing thought experiments involving a utilitometer: a device capable of measuring the utility of the universe, including sums-over-time and counterfactuals (what-if extrapolations), for any given utility function, even generic statements such as, "what I value." Things this model ignores: nonutilitarianism, complexity, contradictions, unknowability of true utility functions, inability to simulate and measure counterfactual universes, etc.

Unfortunately, I believe I've run into a pathological mindset from thinking about this ... (read more)

[-][anonymous]15y00

An embarrassing case study of introspection failure: a few days ago I attended an event where I ran into a few high school friends. I was stunned to learn that one old friend was still dating his high school girlfriend, and the two of them were coming up on their six year anniversary*.

I was amazed and jokingly said something like, "man, I can't remember the last relationship I had that lasted six weeks!" Everybody laughed.

I thought about that while I drove home, and actually tried to recall my last relationship to last that long. I gradually realized that I had never had a relationship that lasted more than a month!

[-][anonymous]15y00

Massimo Pigliucci has written a blog post critiquing one of Eliezer's posts, The Dilemma: Science or Bayes?.

[-]Oscar_Cunningham15y00

Can anyone think of items that look like they belong in a set with {determinism, randomness, free-will}, or is that set complete?

(I know that's a soft question, but I'm happy to hear even silly suggestions.)

3ata15y

I don't think free-will even belongs in that set.

0Larks15y

What you want are {determinism, at least some randomness} {compatabilism, non-compatabilism} and {free will, ¬ free will} each set of which is a partitition of logical possibilities. Together, they give you 7 possibilities ( {free will, determinism, non-compatabilism} is inconsistant)

[-][anonymous]15y00

Is there anywhere I can get all of Eliezer's top-level posts in a pdf file?

[-]NancyLebovitz15y00

"Rationalists should win"-- but with what probability and over what time scale?

3khafra15y

I've always understood that phrase as implicitly beginning with "Ceteris paribus, " and ending with "more often than an irrationalist in the same circumstances." Which is to say, if you're somehow forced into the position of playing Russian Roulette with 5 barrels loaded, or trying to create AGI which won't destroy the world before it's too late, you're probably going to lose. But you should do so with lower probability than someone with the same resources, using them less rationally.

Cognitive Bias Song

via @vaughanbell on Twitter

IMO reasonably catchy.

[-][anonymous]15y00

deleted

[-]novalis15y-10

Corn syrup appears to be no worse than sugar, health-wise. Interesting case of a health-related panic not based on any actual science.

3saturn15y

You're fighting hyperbole with hyperbole—not based on adequately conclusive science is not the same as not based on any actual science.

0novalis15y

Actual science is being performed, sure. But the results of that science don't seem to be inconclusive at all. At least if this article is to be believed, there's no well-done study supports the hypothesis that HFCS is worse than sugar. Maybe there is such a study and that article doesn't mention it. That would be sad but typical of journalism. A priori, we should have only a slight belief that corn syrup is worse for you than sugar (on the theory that occasionally manufactured foods are worse than their natural equivalents but rarely is the reverse true). What studies caused these authors to update on these priors towards the HFCS-is-worse view?

More from NancyLebovitz

Curated and popular this week