All of bfinn's Comments + Replies

Yes, still doing this every morning, and it still works same as ever!

Somewhat relatedly, about 10 years ago I heard someone on the radio predicting that a long-term effect of social media would be greater acceptance of others' flaws, particularly youthful indiscretions that previously would have damaged a policitican's career - e.g. that they had smoked marijuana at college.

Such indiscretions would now be permanently documented on say Facebook when they occur. So everyone would gradually get used to the idea that such things are widespread and almost-normal, and almost all future politicians would be found to have such flaw... (read more)

Re departmental historians: the UK's Foreign Office does have (or had) a small team of historians; someone I know was one of them for a year or two. Apparently they were writing up the history of the Foreign Office in chronological order, at slower than real time; hence were falling further and further behind. They had got up to 1947 or something, but would never catch up. When they completed the history of a year, it was published in an internal book (I assume not publicly available due to national security etc.), which went on a shelf and no-one ever rea... (read more)

Upvote, not least for my first ever sighting in the wild of the interrobang.

The thickness has units of something like [effect]/[work]

I.e. presumably benefit/cost (work being a cost, whether financial or not), = the benefit-cost ratio (BCR) used in cost-benefit analysis in economics.

Is the moral of this really that all decisions should be made so as to maximize the ultimate goal of happiness x longevity (of you or everyone), in utilitarian fashion; whereas maximizing for subgoals is sometimes/often a poor proxy?

Or is it impractical to do utilitarian calculus all the time, but calculations/heuristics with the thin and thick lines can clarify the role of the subgoals so they can be used as adequate proxies?

(It's partly unclear in my head as I didn't grok the exact meaning of the lines & their thicknesses. And it's too late at night for me to think about this!)

7Adam Zerner1y
Ultimately you want to optimize for the supergoal. For some the supergoal is utilitarian happiness x longevity, but not for others. The post is agnostic on this question. The best way to optimize for whatever the supergoal is might be to lean towards calculations, or it might be to lean towards heuristics. The post is agnostic on that question as well. I think the big point in this post is that when you lose sight of what the supergoal actually is, you often fall into some bad failure modes and do a bad job of optimizing for it. (And also that simply being alive is, err, a pretty important thing.)

Others would have to report whether they find it more useful than what they do now (eg Pomodoro), but the reason I think it may well be is indeed the fact it fixes the various Pomodoro problems.

Re new downsides Third Time itself introduces, the one I'm aware of is indeed its extra complexity - hence it is best implemented in an app. But if others find other downsides, I'd be interested to hear of them.

(Alas I haven't got round to finishing Part 2 yet - been busy with other things, notably analyzing the academic research into what the best ways to spend a break are, which I'll write up in due course.)

Indeed - or simply break after completing a task - which also counts as a reward, hence an incentive to complete it. With the downside that you may be less motivated to resume work than if you're breaking in the middle of something.

This is really good stuff.

A minor suggestion: the list in prompt 5 is so important, I suggest it should be in bullets rather than a single paragraph, and ideally people should spend at least a minute or two thinking about each one.

On a detail (!) there are mouth guards (‘sleep clench inhibitors’) that you wear in your sleep to train you not to clench/grind your teeth both at night & in the daytime. I’ve used one; my dentist got one custom made to fit my teeth. You wear them nightly for a week initially, then just once every week or two. Unpleasant the first couple of nights, but you soon get used to them. Worked for me!

3Adam Zerner2y
Interesting, thanks for pointing that out! I'll look into it.

Otter (a smartphone app) is very good. So I've started using it recently for taking notes. Haven't tried using it to write an extended post about anything, though it could be a useful way of getting a first draft.

I like this idea of getting others to help write up ideas. I find writing up ideas vastly more time-consuming and difficult than thinking them up, or explaining them verbally. Even Eliezer, an expert writer, seems to have taken years to get round to writing up his recent list of AI risks.

Is there a good software that would record your voice and convert it to text?

When I was halfway through this and read about the 4 stages, they immediately seemed to me to correspond to four types of news reporting:

  1. Accurate reporting
  2. Misleading reporting (i.e. distorting real events, and fooling many people)
  3. Fake news (i.e. completely made up, but still pretending to be news, and fooling some people)
  4. Obviously false or 'pure fiction' (i.e. not even pretending to be news, and fooling no-one). You do get this kind of thing in the crappiest tabloids like the UK's Sunday Sport or maybe the US's National Enquirer. A well-known example in th
... (read more)

The focus produced by caffeine is enhanced by theanine (or L-theanine), which also counteracts jitters/headaches caffeine can otherwise induce. You can buy theanine in capsule form. Take 1-2 times as much theanine as caffeine. So for a cup of coffee (either brewed, or containing 2 shots espresso), which contains roughly 150mg caffeine, take say 200mg theanine.

You probably shouldn't routinely have more than 1 cup of (caffeinated) coffee a day if you want to avoid becoming tolerant of it, which removes its effects. And don't drink it in the afternoon or even... (read more)

I heard recently that sleepio is now prescribed by the UK's National Health Service, so has presumably been clinically demonstrated to be very effective.

I suspect AI like GPT-3 is good enough now to identify bad arguments quite well, maybe also things like cognitive biases

Mark Forster, the best productivity author I've read (even though I don't agree with all of his ideas), makes a similar point in one of his books. Though he doesn't frame it as finding the weakest link in the negative chain. Rather, as identifying the break in the positive chain that normally makes you act correctly, in cases where you usually get it right, e.g. if usually show up on time for a meeting, but sometimes fail to. You have to think through what actually happened this time that made it go wrong - exactly where did the chain break, and why, and what can you do to avert it? E.g. the traffic is usually OK but this time was bad, so you should always check the traffic in advance, or aim to arrive early.

Indeed. In fact IIRC the fact nuclear missiles were based there was secret at the time and long afterwards

Third Time means '1/3 of the time' (referring to break time = 1/3 of work time) and also 'the 3rd occasion'. It's only half a pun because 'the 3rd occasion' doesn't refer to anything here, but it's a common phrase like first time, second time etc. (E.g. 'the first time I ate caviar I didn't like it, nor the second time, but the third time I enjoyed it'.)

As for puns in the other names suggestions, there are too many to explain, I'm afraid!

Re the Malmstrom incident, there have been various reports over the decades of UFOs appearing at nuclear missile sites and even apparently interfering with (eg disabling) the missiles. Eg the Rendlesham Forest incident in 1980 at a USAF nuclear base in the UK, in which deputy base commander Lt Col Charles Halt and many other personnel spent hours observing (and filming, photographing etc.) UFOs over two nights.

(I'd link to the Wikipedia article, but last time I checked a while back it was being gatekept by ultra-skeptics who reverted any changes. I have ho... (read more)

UFO's aside, there are strong incentives not to disclose potential evidence of interference with a state's strategic nuclear deterrence.

Great post.

'Fabricated' doesn't seem quite the right adjective, as it implies deliberate deception, whereas your examples suggest it's usually unintentional. Indeed I initially assumed your post was about some kind of rhetorical trick rather than a mistake. So, how about something more along the lines of 'incoherent'? (Or see related terms below.)

In any case, I'm a bit wary of the introduction of new terms for apparently-new concepts, because they are often already quite well-known and built into English via established phrases, which to save brain space s... (read more)

An issue I find with debugging a complex program is that when you write tests (which put inputs into part of the program and then check whether the expected output is produced), your tests can themselves contain bugs, and often do (if they're not trivially simple). That is, your experiments isolating a small set of variables can produce confusing results due to the experimental design, not just unpredictability in what they're trying to test. Eg maybe your way of measuring the slope angle or sled weight is flawed. (Cf assumptions about the speed/straightness of light or a steady-state universe messing up your astronomical observations). As philosophers of science say, all observation is theory-laden.

I don't know anyone enlightened, so I'm not making a claim either way. Just that if this is roughly what enlightened is meant to mean (I surmise via the drawing analogy), then this might be an expected consequence, hence test, of it.

If Looking is something like bypassing one’s own mental machinery that interprets & somewhat distorts reality, then maybe someone who’s enlightened has few cognitive biases, or at least is capable of noticing & maybe bypassing them when they arise?

I would expect someone who's really excellent at Looking would be better at tracking what actually matters. Sidestepping cognitive biases requires something else. Like, you can be tracking what actually matters instead of what you think matters, but still get surprised by your systematically overoptimistic estimates about how long things will take. If you notice that affects what you care about in a way that matters, then you probably have to actually do something about your thinking. Things like survivorship bias also require noticing when something true and important is hidden. Tracking that is a matter of rejiggering the mind to notice where this bias arises and adding corrective factors. …but if a master Looker[1] does not care about these effects, they might in fact use examples that arise from (say) survivorship bias precisely because they make the point the Looker does care about. And they might dismiss and be disinterested in corrections to their thinking, basically for the same reason someone writing a text message might find it annoying for someone to correct their grammar & punctuation. So I daresay that really grokking Looking runs the risk of making one enact biases more. …because much of the time they just don't matter for what you actually care about. 1. ^ I'm avoiding terms like "enlightenment" or "enlightened" because of serious overloading here. I'm just talking about skill with Looking here.
3Said Achmiz2y
Bypassing one’s own mental machinery and instead using… what? Anyway, I have never noticed anyone “enlightened” having fewer cognitive biases than anyone else, or being more capable of anything in particular. If you claim otherwise, I should like to see the evidence.

Agreed. I found reading so many comments exhausting and, et, unenlightening, and after a while I just gave up

[EDITED] I recommend tea (black, green, oolong or white - i.e. white leaves, not with milk). As well as being rather lower in caffeine than coffee, tea includes L-theanine, which (a) increases the benefits of caffeine (viz. attention, accuracy, energy) and (b) counteracts jitters and headaches from caffeine.

Oolong tea usually has a bit more caffeine than the others, though green & white teas have a somewhat better ratio of L-theanine to caffeine. But caffeine and L-theanine contents are very variable anyway. (read more)

Seems like a good middle ground, especially depending on the person or as an alternative to quitting cold turkey. But the active ingredient is still basically the same, which has been in use for thousands of years (like alcohol) and probably started before humans were very good at self-evaluation and problem solving.

Thanks for this - interesting & quite useful. It reminded me of some similar observations:

George Orwell recommended somewhere that, when writing, you should try to visualize what you want to say for a while (i.e. non-propositionally) before making any attempt to put it into words, so as to be concrete rather than abstract/dry and falling back on cliches etc. that may not express it adequately.

At the end of Wittgenstein's Tractatus Logico-Philosophicus - the gist of which is that any meaningful statement can be expressed as logical propositions - he hin... (read more)

For elephant-and-rider, the equivalent more common metaphor is 'head and heart' (or, correspondingly, heart and head)

Thanks. Interesting that it gets the general idea of 'from behind' but the specifics garbled - eg bottom left the people should be sitting on the bench, not the other side of the table!

Thanks very much - yes, that one is pretty remarkable, as are several of them.

On the close-up I see loaves, some kind of gadget left of centre, and is that the baby Jesus (with beard?) they're about to tuck into?! (I assume DALLE-2 is not always sure how to show people from this perspective.)

Some prompts:

The Last Supper by Leonardo Da Vinci, but painted from behind.

The Last Supper by Leonardo Da Vinci, but painted from above, looking straight downwards.

The Last Supper by Leonardo Da Vinci, as an X-ray image.

Relativity by Escher, as a high-resolution photograph.

Boris Johnson dressed as a clown and riding a unicycle along a tightrope, spray-painted onto a wall, in the style of Banksy.

"The Last Supper by Leonardo Da Vinci, as an X-ray image" It's trying! 

I especially like this one (close-up):

6Swimmer963 (Miranda Dixon-Luinenburg) 2y
"The Last Supper by Leonardo Da Vinci, but painted from behind". (Based on previous playing around, I think that DALL-E does not have a super strong conception of "The Last Supper" in general, and sort of defaults to a generic supper table.) 

An obvious point just about worth mentioning, is that of course you don't need to spend a whole day over Sabbath. A half- or third-day per week is a lot better than none. Albeit perhaps increasing the chance that you'll start thinking about post-Sabbath plans during Sabbath (if it ends before night).

Relatedly, if Sabbath is to be a whole day perhaps it's better lasting from waking up until bedtime (or notionally say midnight to midnight), so it is bookended by the natural boundaries of sleep. (Albeit that reduces the point of the candle-lighting ritual.)

You may or may not know there's a blog called the Ancient Wisdom Project which tries out practices from various religions for a month at a time.

One issue I have with the regular meal with friends/family bit is that (aside from those in your household, who you would see anyway) this potentially sets up a regular commitment which could well become onerous. In that, if you establish a pattern of seeing the same people week after week, you may after a while start to get bored of it/them (even close friends can pall if seen too often), and want to see other people, or no-one, for the Sabbath meal. Which starts to make Sabbath tedious/stressful if not dealt with, and even if dealt with delicately can cr... (read more)

They prevent accumulation of personal-and-home-related work debt. A chaotic house is not restful. Postponed chores weigh on youThe deadline forces handling them in advance.

Cf a friend of mine once observed that taking a vacation has the benefit of forcing you to clear your desk of things beforehand - completing, ditching or delegating tasks, clearing your email backlog, etc. Often in a more decisive and forceful manner than you otherwise would, e.g. declining to do things you've been asked to do, finishing mini-projects in quick-and-probably-good-enough ways rather than spending too long over them, etc.

Re the last point, yes indeed. Mark Forster (an excellent productivity author) recommends training yourself to comply with your own resolutions this by, each evening (for a while), deciding on something big or small that you will do (or not do) the following day without fail, no matter what. Then do so, and repeat.

Comparing peak to sustainable running speeds: the world marathon record's average speed is 55% of the world 100m sprint record. And for the Olympic men's qualifying times, the ratio is 54%. Both quite close to 60%

Or if you’re paid by results not the hour, as a contractor, you can earn the same in less time. Or even as an employee, you can just be paid to waste most of your time (though this is fairly unsatisfactory). Eg a friend of mine worked with an excellent programmer who would do nothing for months - literally spend most of the time in the pub or messing around with things that interested him - and occasionally spend a weekend programming furiously to produce what was presented to the (crappy) management as what the entire team had been working on for months.

Re ‘Slack is life’, cf the phrase ‘work-life balance’, where life implicitly means non-working time, i.e. (roughly speaking) slack.

(PS heading ‘The Slackless Like of Maya Millennial’ presumably should read ‘The Slackless Life…)

Last time I looked into all this a while back, eg on - an excellent site which analyses research on supplements - Healthspan Elite Zinc Defence Lozenges seemed to be the only suitable zinc lozenge available in the UK for treating colds once you have them. Doesn’t come with instructions (!) but as soon as cold symptoms start, dissolve slowly in the mouth and avoid eating/drinking for say 15 min afterwards, and avoid citric acid for a while before too. You need to take 9 lozenges (1 sheet) per day. also suggests black elderberry, eg Sa... (read more)

A good, plausible movie about AI risk might be an indirect way to raise the chance of government action via public awareness (if that would be useful). Have there been such AI risk movies that I've missed? If not, how could people get one to happen?

Cf I get the impression governments take asteroid risks a bit more seriously thanks to various asteroid-strike movies over the years. Not least the recent Don't Look Up on Netflix, which despite being a comedy was probably quite thought-provoking to many viewers (e.g. showing how there could well be government &... (read more)

A movie version of would be cool!
While not for activist reasons, more for entertainment, I have a general outline in my head about a realistic thriller movie about a quantitative finance company that accidentally comes up with AGI. The plot follows a security consultant who is called in; they don't really know what's happening at first and they think it's a hacker, but it also could be a bug, they're not sure... Kind of writes itself really.

Maybe users who police bad comments/articles (like that one) should be rewarded for downvoting them by earning the $1 saved, thus maintaining financial equilibrium

Strong-downvoted, earned $6 by doing so. (I did not actually do this)
1Thomas Kwa2y
Then who decides whether the comment is actually bad?

I genuinely can’t tell how much of this is an April Fool joke. If all of it, it’s gone on too long now

3Yoav Ravid2y
It's not a joke (though it is a pun), It's a playful experiment with real money.

Cf The Dice Man (a good idea for a mediocre book), in which the protagonist decides to make all decisions large & small by rolling a die.

But (and I don’t recall if the book discusses this) much then comes down to what 6 options to choose from - a decision made entirely by the protagonist. Eg if you fall out with Fred, do you include ‘punch Fred’ or ‘kill Fred’, or merely ‘criticise/ignore/undermine/forgive Fred’? And what proportion of options should be nasty vs nice?

And perhaps that decision (deciding the options) is more instructive than just making a decision cold.

(1/3 feels like a bit too much though, e.g. 8 minute break for each 25 minute pomodoro. 1/5 to 1/4 would be closer to classic 25-5)

Actually if you take into account the longer break every fourth Pomodoro it works out close to 1/3. (Office workers not following a system effectively use fractions more like 1/5, the research shows, but most of them don't work in intense bursts.)

If you're in the flow, do you take the break anyway or skip it? If you skip it, it might be best to save the break up for later, so you get a longer break after longer work (somewhat like Third Time). As the academic research suggests, not surprisingly, that people need longer breaks after working longer.

I think making pomodoros longer when you're in flow makes sense, as does making them shorter if you're distracted because your attention span is short today. Though being distracted or interrupted by something external outside your control doesn't seem a r... (read more)

I skip it (because in the flow I don't want to be interrupted by a break) and start immediately the next pomodoro with longer interval. Yes, sounds reasonable, I will probably make the breaks proportional to pomodoro length. (1/3 feels like a bit too much though, e.g. 8 minute break for each 25 minute pomodoro. 1/5 to 1/4 would be closer to classic 25-5) Yes, in fact it's not automatic shortening/extending of the interval in my app, I just have convenient shortcuts for "start pomodoro 5 min longer / 5 min shorter / the same length", so I sometimes decide to keep current interval, too.

I don't know if you touch on this elsewhere, but I wonder what the relationship between the mafioso nature and psychopathy is. They seem fairly similar, but e.g. psychopathy is often hidden by psychopathic charm, used to manipulate people 'nicely', whereas it seems the mafioso nature's effectiveness comes from others knowing you have it and believing your threats.

If psychopathy is an evolutionary strategy, keeping it hidden presumably means people don't avoid/kill you. Or maybe psychopaths only keep it hidden until they get powerful enough that they are fairly safe, then reveal it in order to issue credible threats like a mafioso. (E.g. some dictators.)

Some years ago I did read research showing that improving sleep quality was the easiest intervention most people could make to improve their happiness. (I don't recall where, but googleable no doubt.)

Re sleep deprivation inducing mania, I sometimes fast (either on 100kcal or 700kcal per day for 5-6 days), and find (as others do) that after a day or so I go into a hyperactive state - very alert, clear-minded, energetic, probably not far from mild mania. (And unlike as described in the OP, I don't get hungry at all - or rather, only did the first couple of times I fasted. So it presumably it just takes getting used to.)

So I wonder if the analogies between sleep and eating are even closer than suggested.

Load More