All of Baughn's Comments + Replies

Open thread, July 31 - August 6, 2017

Dyson opens up another interesting question with this. Is it better to survive forever with a finite subjective time T, or it is better to consume 2*T experience in a finite amount of calendar time?

Isn't 2*T obviously better? Maybe I'm missing something here...

0Thomas4yWe are on the same page here. But a lot of people want to survive as long as possible. Not as much as possible, but as long as possible.
Open thread, Sep. 12 - Sep. 18, 2016

Data recovery is a last-ditch effort that often as enough fails, and if it succeeds will only get you back kilobytes or megabytes of your most critical material. (Unless you're lucky enough that it's actually a controller failure.)

If you want to avoid disk failures, invest instead in backups.

Open Thread, Aug. 22 - 28, 2016

When it comes to writing styles? Absolutely. There's a ton of skills involved, and deciding exactly which thoughts you want to convey is only a small part of it.

Suggestion: jog instead of walk

If jogging isn't convenient, then you can still walk faster.

Whatever your 'natural' pace is, you can easily up it if you consciously try. I guarantee it. Downside is, once you get used to it you'll start to get annoyed at all the slowpokes who seem to think they'll live forever.

February 2016 Media Thread

Garakowa: Restore the World

It's a movie about a suspiciously humanlike pair of anti-virus AIs, and their attempt to protect a backup archive of human history from destruction at the hands of sapient viruses. It's compressed a bit more than would be ideal, but otherwise well done.

Practically anything else I could say about it would be a spoiler, so I won't. What happened to humanity? Why are they there? Where do the viruses come from? You'll need to watch it to find out.

The story is also interesting due to the central Friendliness failure being novel, fairl... (read more)

AI safety in the age of neural networks and Stanislaw Lem 1959 prediction

But here's the thing: most of the suggested ways to solve this problem (including the one I use) involve keeping the massive weak constraint relaxation, throwing away all irrelevant assumptions, and introducing new features to get the structured symbolic stuff. And that revision process generally leaves you with hybrid systems in which all the important stuff is NO LONGER particularly opaque. The weak constraint aspects can be done without forcing (too much) opaqueness into the system.

Where can I read about this?

AI Fiction - Crystal Society

We already have a media thread. An average of several dozen media links are posted every month[0], and if they were all top-level posts the site would quickly become useless. While the story is interesting, I don't think it's interesting enough to overcome that.

[0]: Have not actually counted.

Open thread, Jan. 18 - Jan. 24, 2016

It's circular, and square.

That's literally all there is. I can't imagine it visually, the way I usually would. Wonder why. :P

2Gunnar_Zarncke6y― Lewis Carroll See also this article [] discussing the usefulness of believing impossible things.
0roystgnr6yI can imagine it. You just have to embed it in a non-Euclidean geometry. A great circle can be constructed from 4 straight lines, and thus is a square, and it still has every point at a fixed distance from a common center (okay, 2 common centers), and thus is a circle.
Open thread, Jan. 18 - Jan. 24, 2016

I also asked for a square circle which got the immediate reply that it is logically impossible.

I am now imagining a square circle. That's interesting.

2Gunnar_Zarncke6yCan you describe it?
Open Thread, January 4-10, 2016

Google for 'crashplan data loss', and you'll find a few anecdotes. The plural of which isn't "data", but it's enough to ensure that I wouldn't use it for my own important data if I wasn't running two backup servers of my own for it. Even then, I'm also replicating with Unison to a ZFS filesystem that has auto-snapshots enabled. In fact, my Crashplan backups are on the same ZFS setup (two machines, two different countries), so I should be covered against corruption there as well.

Suffice to say, I've been burnt in the past. That seems to be the onl... (read more)

0Lumifer6yI'm using Crashplan as the offsite backup, I have another backup in-house. The few anecdotes seem to be from Crashplan's early days. But yeah, maybe I should do a complete dump to an external hard drive once in a while and just keep it offline somewhere...
Open Thread, January 4-10, 2016

Use a backup system that automatically backs up your data, and then nags at you if the backup fails. Test to make sure that it works.

For people who don't want / can't run their own, I've found that Crashplan is a decent one. It's free, if you only back up to other computers you own (or other peoples' computers); in my case I've got one server in Norway and one in Ireland. There have, however, been some doubts about Crashplan's correctness in the past.

There are also about half a dozen other good ones.

1Lumifer6yLinks? I use Crashplan and would be interested in learning about its bugs.
Open Thread, January 4-10, 2016

I would be very interested in trying one of those. In particular, I frequently change up my writing style (deliberately), and it might be able to tell me what I'm not changing.

has anyone actually gotten smarter, in the past few years, by studying rationality?

I feel like spending time learning "domain specific knowledge" is much more effective than "general rationality techniques".

However, reading Lesswrong is what prodded me towards getting better at spending my time effectively, really getting into a growth mindset. My only problem nowadays is that there are too many things I want to learn, and that's a much better problem to have; I know I can, I just have to pick and choose. I'm getting better at that, too.

Maybe the same would have happened anyway, but I don't think it would have happened quite as fast.

Open thread, Nov. 16 - Nov. 22, 2015

Ah, no. I do agree that uploading is probably the best path, but B doesn't follow from A.

Just because I think it's the best option, doesn't mean I think it'll be easy.

Stupid questions thread, October 2015


Determining which possibilities this is false for in our particular universe would take some time, and depends on the exact form of the laws of physics (which we don't know), so let's use a simplified example.

Take the Game of Life. While simple, it is in fact turing-complete; this was demonstrated by implementing a turing machine on it, which is the best way to demonstrate that sort of thing. (It's fun to put one cell out of place and watch it disintegrate.)

Take an infinitely large game of life. Start it in a random state, and leave it to evolve for ... (read more)

Mean of quantiles

(E.g., the trimean will have a better breakdown point but be less efficient than the mean; a worse breakdown point but more efficient than the median.)

What does "efficient" mean, in this context? Time to calculate would be my first guess, but the median should be faster to calculate than the trimean.

4gjm6ySee here []. [EDITED to add:] Sorry, that's a bit rude; I should also give a brief explanation here. Any estimator will be noisy. All else being equal, you would prefer one with less noise. There is a thing called the Cramér-Rao inequality that gives a lower bound on how noisy an estimator can be, as measured by its variance. (But see the note below.) The efficiency of an estimator is the ratio between its variance and the bound given by Cramér-Rao. An estimator whose efficiency is 1 has as little variance as any estimator can have. (Such estimators need not exist.) Noisier estimators have lower efficiency. Efficiency depends on the underlying distribution. I shouldn't really have said that the mean will be more efficient than the median; if the underlying distribution is thin-tailed enough then it will be; e.g., for large samples of normally distributed data the mean has efficiency 1 while the median has efficiency about 0.64. But if the actual distribution is fat-tailed, the median may actually be a more efficient estimator than the mean. (Note: You might notice that I said something obviously false above. It's trivial to make a completely un-noisy, zero-variance estimator of any parameter: just always estimate zero. But this will be a biased estimator; i.e., its expectation will not equal the underlying value it's meant to be estimating. The Cramér-Rao inequality only applies to unbiased estimators. In some cases, for some applications, the "best" estimator may actually be a biased but less noisy one. For instance, suppose you have some samples of a normally-distributed random variable and you want to estimate its variance. The "obvious" thing to do is to compute 1/n sum(x-xbar)^2. That gives you an unbiased estimator but, famously, you can get rid of the bias by computing 1/(n-1) sum (x-xbar)^2 instead. But if you want to minimize your mean squared error -- i.e., minimize the expectation of (est. vari
Bragging thread September 2015

Did my first serious bit of Minecraft modding, and learned how to use Blender in the process. It's not as impressive as the things I do at work, but it's fun.

Power poles

Stupid Questions September 2015

There's a good chance you'll have a second elephant failure while the first one is giving birth to a replacement, so at least use RAID6.

Or ZFS RAIDZ2. That's also great.

Open Thread, Jul. 27 - Aug 02, 2015

Where, exactly? All I've noticed is that there's less interesting material to read, and I don't know where to go for more.

Okay, SSC. That's about it.
5Vaniver6yHere's [] one discussion. One thing that came out of it is the RationalistDiaspora [] subreddit.
-1[anonymous]6yTumblr is the new place.
0[anonymous]6y [] , [] , offline.
Open Thread, Jul. 13 - Jul. 19, 2015

Nothing terrible was going to happen. As has been pointed out, collisions that energetic or more happen all the time in the upper atmosphere.

0Thomas6yEnergetic perhaps. But as dense also?
The Waker - new mode of existence

But you're always stuck in one reality.

Let's take a step back, and ask ourselves what's really going on here. It's an interesting idea, for which I thank you; I might use it in a story. But...

By living your life in this way, you'd be divorcing yourself from reality. There is a real world, and if you're interacting solely with these artificial worlds you're not interacting with it. That's what sets off my "no way, no how" alert, in part because it seems remarkably dangerous; anything might happen, your computing infrastructure might get stolen from underneath you, and you wouldn't necessarily know.

0Jan_Rzymkowski6yDisclaimer: This comment may sound very crackpottish. I promise the ideas in it aren't as wonky as they seem, but it would be to hard to explain them properly in such short time. Here comes the notion that in posthumanism there is no definite reality. Reality is a product of experiences and how your choices influence those experiences. In posthumanism however you can modify it freely. What we call reality is a very local phenomenon. Anyhow, it's not the case that your computing infrastructure would be in danger - it would be either protected by some powerful AI, much better suited to protecting your infrastructure then you or there would be other copies of you keeping the maintenance in "meatspace" (Again, I strongly believe that it's only our contemporary perspective that makes us feel that reality in which computations are performed is more real then virtual reality). What's more, a Waker can be perfectly aware that there is a world beyond her experiencing and may occasionally leave her reality.
July 2015 Media Thread

Do you by any chance have those as MP3 or FLAC?

2gwern6yAfraid not. I convert everything to OGG and I don't have the original download link handy. You can probably find a working download link searching for the original album title & keywords 'さんぼん堂 幻想少女大戦永OriginalSoundTrack'; pay particular attention to Chinese links, they seem to last longer than English ones.
A Roadmap: How to Survive the End of the Universe

"not impossible" == "possible". And this article doesn't show either one.

Open Thread, Jun. 22 - Jun. 28, 2015

So, some Inside View reasons to think this time might be different:

  • The results look better, and in particular, some of Google's projects are reproducing high-level quirks of the human visual cortex.

  • The methods can absorb far larger amounts of computing power. Previous approaches could not, which makes sense as we didn't have the computing power for them to absorb at the time, but the human brain does appear to be almost absurdly computation-heavy. Moore's Law is producing a difference in kind.

That said, I (and most AI researchers, I believe) would a... (read more)

1RobFack6yWhile the work that the visual cortex does is complex and hard to crack (from where we are now), it doesn't seem like being able to replicate that leads to AGI. Is there a reason I should think otherwise?
0jsteinhardt6yThis seems an odd thing to say. I would say that representation learning (the thing that neural nets do) and compositionality (the thing that symbolic logic does) are likely both part of the puzzle?
[Link]: The Unreasonable Effectiveness of Recurrent Neural Networks

If you're being generous, you might take the apparent wide applicability of simple techniques and moderate-to-massive computing power as a sign (given that it's the exact opposite of old-style approaches) that AGI might not be as hard as we think. It does match better with how brains work.

But this particular result is in no way a step towards AI, no. It's one guy playing around with well-known techniques, that are being used vastly more effectively with e.g. Google's image labelling. This article should only push your posteriors around if you were unaware of previous work.

[Link]: The Unreasonable Effectiveness of Recurrent Neural Networks

The craziness it produced was not code, it merely looked like code. It's a neat example, but in that particular case not much better than an N-gram markov chain.

How much understanding should we expect from even a powerful AI, though? All it's being fed is a long stream of C text, with no other information than that - it gets no runtime output, no binary equivalents, no library definitions, no feedback on its own compression output... I'm not sure what a human with no knowledge of programming would learn in this context either other than to write C-looking gibberish (which, unlike generated images or music, we are not much interested in the esthetics of). The RNN might be doing extremely well, it's hard to say.

It w... (read more)

8Antisuji6ySyntactically it's quite a bit better than an N-gram markov chain: it gets indentation exactly right, it balances parentheses, braces, and comment start/end markers, delimits strings with quotation marks, and so on. You're right that it's no better than a markov chain at understanding the "code" it's producing, at least at the level a human programmer does.
June 2015 Media Thread

Log Horizon

If you liked the anime, you will likely find that this is better. If you felt that the anime was flawed, you may well find that the book is not, or not in the same way.

The story is slow, with a great deal of explanations and musings, especially in the beginning; it's trying to paint an entire world, and that shows. It is the sort of thing that is very difficult to adapt to an animated format. The book, however, was well worth the read.

Only the first volume is out yet, the second to come in July.

June 2015 Media Thread

nostalgebraist has started work on a new novel, The Northern Caves. It's off to a slow start, but looks interesting so far.

Devil's Offers

In addition to what James said, I'm reminded of the mechanism to change screen resolution in Windows XP: It automatically resets to its original resolution in X seconds, in case you can't see the screen. This is so people can't break their computers in one moment of weakness.

But you are absolutely allowed to break your computer in "one moment of weakness"; it isn't even hard. The reason for that dialog is because the computer honestly, genuinely can't predict if the new screen mode will work.

Open Thread, May 25 - May 31, 2015

I don't believe that it's mainstream transhumanist thought, in part because most people who'd call themselves transhumanists have not been exposed to the relevant arguments.

Does that help? No?

The problem with this vision of the future is that it's nearly basilisk-like in its horror. As you said, you had a panic attack; others will reject it out of pure denial that things can be this bad, or perform motivated cognition to find reasons why it won't actually happen. What I've never seen is a good rebuttal.

If it's any consolation, I don't think the possibility... (read more)

0Fivehundred6yThe writer did seem to think that it was very likely. But he dismisses the idea of FAI being a singleton.
Experience of typical mind fallacy.

I distinctly remember, at some point in my teens, realizing that other people sometimes thought like me and I could model their reactions as something more than inscrutable environmental hazards. So there's that.

April 2015 Media Thread

On the flip side there's Luv and Hate, which is an (incomplete! still good) rewrite of the Muv-Luv Alternate story with a guest protagonist from... Supreme Commander. Including the ACU.

It's well-written, mainly character-focused with a few amusing combat interludes, and oh so gratifying after attempting to read the grimdark original.

It's also a quest. If this doesn't mean anything to you folks... don't worry about it, you can treat it as an ordinary story if you wish.

Open thread, Apr. 01 - Apr. 05, 2015

We won't run out of coal anytime soon. It has other issues, but I think that invalidates his conclusion—coal power plants are pretty cheap, and are already being built.

I'm also more optimistic about politicians. Ten years may be beyond their reelection horizon, but it's not beyond their "This place is going to hell"-horizon.

Seeking Estimates for P(Hell)

Okay. I'm sure you've seen this question before, but I'm going to ask it anyway.

Given a choice between

  • A world with seven billion mildly happy people, or
  • A world with seven billion minus one really happy people, and one person who just got a papercut

Are you really going to choose the former? What's your reasoning?

1qwerte6yThe second option is a world with seven billion -1 really happy people and one person who is a tiny bit less than mildly happy? My reason to choose the former would be that all of those lives are experienced by only one person and everyone experiences only one life. In the former case, no subjective experience is worse than mildly happy. In the latter case, a subjective experience is worse than that. It doesn't matter how much happiness or pain a number of people will cumulatively experience because no one actually experiences the cumulative experience. All that matters is improving the worst life at any given moment. I won't be surprised if my reasoning is bullshit, but I'm not seeing it.
3Brian_Tomasik6yFrom a practical perspective, accepting the papercut is the obvious choice because it's good to be nice to other value systems [] . Even if I'm only considering my own values, I give some intrinsic weight [] to what other people care about. ("NU" is just an approximation of my intrinsic values.) So I'd still accept the papercut. I also don't really care about mild suffering -- mostly just torture-level suffering. If it were 7 billion really happy people plus 1 person tortured, that would be a much harder dilemma []. In practice, the ratio of expected heaven to expected hell in the future is much smaller than 7 billion to 1, so even if someone is just a "negative-leaning utilitarian" who cares orders of magnitude more about suffering than happiness, s/he'll tend to act like a pure NU on any actual policy question.
[LINK] Terry Pratchett is dead

If you have Alzheimer's, and you want to use cryonics, you should do your very best to get frozen well before you die of the disease.

This is problematic in all jurisdictions I can think of. Even where euthanasia is legal, I don't know of any cryonics organisations taking advantage, and there might be problems for them if they do. I'd very much like to be proven wrong in this.

0listic6yI believe it's probably only because of the woefully under-developed state of cryonics itself that the practice of voluntary death through cryopreservation ( cryothanasia [] ) haven't been seriously researched: rather counter-intuitively, cryonics companies are too few and mostly have enough trouble on their hands to bother disrupting the status quo. Getting frozen before you die can well be problematic, but not necessarily impossible in all jurisdictions. I believe it's just not well researched. Cryonics has low demand as it is, and cryothanasia requires even greater mental effort to voluntarily choose death, before dying the 'natural' way, and make all the necessary research and preparations yourself, so I wouldn't be surprised to learn that noone bothered yet, or noone who made their efforts public, at the very least. Which doesn't mean that this is impossible. Suicide tourism [] is a thing, after all. I would recommend directly contacting Danila Medvedev [] ( []) from Russian company CryoRus [], if you are really practically interested in the prospect of cryothanasia - it is likely that it is possible, but noone is going to offer it as a product, so far, you will have to research it for yourself; they are at least entertaining the idea.
1pcm6yIt is sometimes possible to die by refusing to eat/drink. Ben Best has some conflicting claims about how feasible that is with Alzhiemer's here [] and here [].
[LINK] Terry Pratchett is dead

It's a suspiciously pleasant way to go, but I see no reason to look more closely at this. Let's just be happy he got the end he wanted.

3Toggle6yPratchett himself stated an intention to commit suicide before his disease progressed past a certain point. "To jump before I am pushed", I believe was the phrase he used at one point. [] The BBC claims that he didn't take his own life, and given his advocacy I think that his family would have been honest about his suicide if it were one, but it's a reason to look more closely at least.
March 2015 Media Thread

Saenai Heroine no Sodatekata.

It's an anime about... making a game... that appears fully congruent with the contents of the anime...

In short, it seems to be a metacircular anime. It's worth watching because of the way it plays with tropes, and the origin of those tropes; it's marginally annoying in that many of the tropes it plays with are of the harem genre. There may be something more going on in the background, but I haven't watched enough to tell. It may be especially interesting to people who have long experience with japanese animation.

The first episode is fully representative, so I'd recommend having a look if the above appeals.

March 2015 Media Thread

The only other one that springs to mind is the one with the Nine-Brained Kyuubi.

Got any more?

1Leonhart6yRathanel's The Empty Cage (previously recommended [] on LW) and OmgImPwned's In Fire Forged []. Can't remember if the first is finished, the second certainly isn't. Waves Arisen is in a class by itself as regards sweet sweet ingroup jargon, however :)
Harry Potter and the Methods of Rationality discussion thread, March 2015, chapter 116

That's a literal Idiot Ball reference, not to mention canon. I don't think we'll see it.

2Luke_A_Somers6yBut it's not an idiot ball if there's an actual legitimate reason for being stupid. Like, in The Naked Time and The Naked Now, the initial infections were pure idiot ball. The subsequent infections, and the crew doing spectacularly stupid and even suicidal things, were not.
Best of Rationality Quotes, 2014 Edition

That worked. Running "strfile rationality" on my system produces a different rationality.dat than the one you have checked into your repository, but "fortune rationality" works now.

(It doesn't need to be copied to a system path.)

Best of Rationality Quotes, 2014 Edition

How do I use it? Fortune is being obstructive.

4Huluk6yFirst, you have to find the path where fortune files are stored. In the man file of fortune, there is a chapter "FILES" in which you can find the default path for fortune files. Your path may be different, maybe you have to search for it. Put the files "rationality" and "rationality.dat" into this path and test using the command fortune rationality. If it doesn't work, you can try to generate your own .dat file using the command strfile rationality. I'd be interested to know whether this worked for you and what steps where necessary, since I didn't try this with different fortune installations.
Imagining Scarcity

Read some blogs. I'd like to read the good blogs, but I have no idea where to find them; any pointers?

2[anonymous]6yQuite like myself....
0Xerographica6y"Good" is, of course, super subjective. But here are some that I'd recommend... More liberal perspective... Miles Kimball []John Quiggin []Noah Smith [] More market perspective... CafeHayek []Coordination Problem []EconLog [] AskBlog [] For both a market perspective (Tabarrok) and a liberal perspective (Cowen)... MarginalRevolution [] And for a pragmatarian perspective... my own blog... Pragmatarianism []
Harry Potter and the Methods of Rationality discussion thread, February 2015, chapter 111

I wonder if that was all acting. If it wasn't, then Hermione may now be very hard to kill indeed.

This can only be a good thing.

1Gondolinian6yExcept by AK and fiendfyre, both of which come easily to Voldemort, and the former can be cast by many experienced fighters.
1MarkusRamikin6yYou clearly haven't read some of the same manga as I have. (Though I won't be more specific because, well, infohazard.)
Harry Potter and the Methods of Rationality discussion thread, February 2015, chapter 108


I've been wondering about this. Voldemort knows this prophecy, and I expect he realises it's not about him. So, how can he expect to be able to kill Harry, in a self-consistent universe?

Perhaps Harry's reading of that plan is all wrong.

2lerjj6yActually, I'm no longer that sure about this. Does anyone have a source for prophecies being unavoidable? They certainly seem to have a knack of getting themselves fulfilled (see October 31st) but are they literally unbreakable? If Quirrell thinks that he can prevent a prophecy from coming true, he's committed this time around to do it properly and not to try to creatively fulfil it. He doesn't appear to think this is futile.
Harry Potter and the Methods of Rationality discussion thread, February 2015, chapter 108

Moreover, it ignores Merlin's interdict. Those children had not learned how to disillusion themselves.

I think you're on to something here.

8Transfuturist6yMerlin's Interdict is unrelated to wands. It simply prevents certain magical knowledge from being passed down in any way other than through oral instruction. It doesn't prevent spontaneous rediscovery of that magical knowledge; the magic was there waiting to be discovered.
Harry Potter and the Methods of Rationality discussion thread, February 2015, chapter 104

With professional players, whose job it was to play Quidditch.

It won't survive children doing the same. More to the point, the snakes and ravens are deliberately and obviously exploiting the current rules, which will trigger all sorts of fair play instincts.

0ReevesAnd6yAlthough the easier solution to this problem is to stop adding Quidditch points to House points. That's dumb to begin with. Maybe just add some points for winning the match.
I notice that I am confused about Identity and Resurrection

A rather large fraction of my discussions happen via IRC; I log every bit of it, and carefully back the logs up.

Occasionally, I go back and read some random fraction of the logs. It is usually a valuable experience. I am doing so right now, albeit without IRC.

Load More