D&D.Sci Pathfinder: Return of the Gray Swan Evaluation & Ruleset 2021-09-09T14:03:56.859Z
D&D.Sci Pathfinder: Return of the Gray Swan 2021-09-01T17:43:38.128Z
How poor is US vaccine response by comparison to other countries? 2021-02-17T02:57:11.116Z
Limits of Current US Prediction Markets (PredictIt Case Study) 2020-07-14T07:24:23.421Z


Comment by aphyer on Where do your eyes go? · 2021-09-20T17:54:43.866Z · LW · GW

I've played the Touhou bullet hell series (credentials: beat Perfect Cherry Blossom on Phantasm but haven't played the more recent ones as much) and viewpoint on that:

When you're first learning, your eyes are on your character. As you get better, your eyes are actually on a broader area that isn't focused on your character, most commonly somewhat above them (since most bullets come from above).

While you're learning on low difficulties, it suffices to keep your eyes on your character, see every bullet that gets close enough to be a threat, and reactively dodge it. On higher difficulties, you need to be thinking ahead, dodging towards areas of low incoming bullet density, reacting to high-level patterns, or manipulating shot patterns to not get walled off, and you can't do this if you are tunnel-visioned on the small bit of screen around your character. Most commonly your eyes will be a few inches above your character watching incoming bullets, and you'll be executing dodges based on a kind of peripheral vision and memory of where the bullets that passed through your focus one second ago were going. (It's hard to describe).

Comment by aphyer on D&D.Sci Pathfinder: Return of the Gray Swan Evaluation & Ruleset · 2021-09-10T01:12:22.581Z · LW · GW

Done, thank you!

Comment by aphyer on D&D.Sci Pathfinder: Return of the Gray Swan Evaluation & Ruleset · 2021-09-09T21:48:00.713Z · LW · GW

If I cannot manage to serve as a shining example, at least I can serve as a cautionary tale.

Comment by aphyer on D&D.Sci Pathfinder: Return of the Gray Swan Evaluation & Ruleset · 2021-09-09T17:06:18.950Z · LW · GW

Good catch, will fix.

Comment by aphyer on D&D.Sci Pathfinder: Return of the Gray Swan · 2021-09-08T12:20:05.881Z · LW · GW

If anyone is still working on this (either current answerers continuing their analysis or latecomers wanting to start), please let me know. In the absence of replies on this, I will publish the answer key tomorrow.

Comment by aphyer on D&D.Sci Pathfinder: Return of the Gray Swan · 2021-09-07T02:56:38.154Z · LW · GW

Is there a typo in your first specified route?

Comment by aphyer on D&D.Sci Pathfinder: Return of the Gray Swan · 2021-09-03T14:05:05.069Z · LW · GW

You don't know. That might be something you could figure out by looking at the Admiralty logs though.

Comment by aphyer on D&D.Sci Pathfinder: Return of the Gray Swan · 2021-09-03T10:47:54.207Z · LW · GW

That is definitely what the Admiralty data says.

(Yeah, usually I think freeware Office stuff is pretty much on par with Microsoft Office, but Excel specifically feels genuinely better to me than any of the freeware programs I've found.  Does anyone have recommendations?)

Comment by aphyer on D&D.Sci Pathfinder: Return of the Gray Swan · 2021-09-02T17:55:30.120Z · LW · GW

Names are not unique identifiers.

The Bloody Diamond and the Mopey Hind are captained by two different Captains named Angus MacDougal. If you have a problem with the MacDougal Clan's traditional naming practices, all eight-thousand members of it would like a word with you.

Comment by aphyer on D&D.Sci Pathfinder: Return of the Gray Swan · 2021-09-02T15:58:23.279Z · LW · GW

Fixed, thank you.

Comment by aphyer on Death by a Single Cut · 2021-08-29T08:31:00.916Z · LW · GW

In some cases, real-world reasons may be multifaceted.  If you present to me a plan that has three-hundred separate issues that might foil it, each of which has an independent 5% probability of occurring, it may be hard for me to state a single reason why I don't think your plan will work.

Comment by aphyer on Narrative truth · 2021-08-26T21:48:14.542Z · LW · GW

This seems to be based in the idea of manipulating the reader.

'Manipulating' is perhaps too harsh a word. In some contexts (fiction most obviously), the reader is essentially signing up to be manipulated. But in general, if you find yourself saying 'my readers will be insufficiently moved by the truth, so I must create a falsehood that will cause them to feel what I want them to feel, and think what I want them to think', it seems pretty accurate to describe that as manipulative.

If you were offended by something I said, and punched me in the face, the correct answer to 'Did you punch him in the face?' is 'Yes.' If you say 'Well, saying yes might lead people to conclude that I am a violent and dangerous person, and I do not think that I am a violent and dangerous person, so it would be more narratively accurate to answer No,' then you're just lying.

This is obviously an extreme example, but it doesn't seem qualitatively different.

(I'm also reminded of a quote by Bertrand Russell: "The mark of a civilized man is the ability to read a column of numbers and weep".)

Comment by aphyer on Is top-down veganism unethical? · 2021-08-22T17:37:48.466Z · LW · GW

(Disclaimer: am a carnivore)

I'll agree that lab-grown meat seems more attainable than genetically engineering vegan animals, but I think you're overstating the quality of Beyond Meat.

The only Beyond Meat product I've had is a 'Beyond sausage', which was really impressively unlike meat.  I'd imagine that sausage would be the easiest thing to replicate, but the Beyond Meat one tasted like...uh...I can't find good words to describe what it tasted like, but it tasted neither like sausage nor good.  Even the 'sausage' in McMuffins or in Dunkin' Donuts' breakfast bagels is in my opinion substantially better, and the 'low quality meat from a grocery store' is substantially beyond that.

I haven't actually tried Impossible Burger, but if it's anything like the sausage I'd say it'll be substantially worse than fast food burgers.

Comment by aphyer on Covid 8/19: Cracking the Booster · 2021-08-19T15:58:28.009Z · LW · GW

Your seven hypotheses alternate between being numbered 1-7 and being referred to as A-G, which is rather confusing.

Comment by aphyer on D&D.Sci August 2021: The Oracle and the Monk · 2021-08-14T00:21:26.216Z · LW · GW

My current plan and a few intermediate discoveries:

  • Energy types are not created equal.  Ash, say, is almost invariably single-digit, and is basically useless.
  • Several types of energy correlate: Solar and Lunar correlate very positively (around a ~30-day cycle that makes me suspect solar energy is a misclassified type of lunar energy).  Earth and Ocean correlate extraordinarily negatively.  Flame and Ash correlate very positively (though this is meaningless, since Ash is not useful to us).
  • Solar+Lunar considered together spends just over half its time above 70.  We're currently above 70, but we are trending down, and probably won't be in 10 days.  Additionally, there was a recent anomaly in Solar energy (the supernova?) which makes me leery of using it.
  • Earth+Ocean considered together varies very stably and predictably (on a 22-day cycle).  This cycle spends just under half of its time over 70...we are currently in one of its low points, but 10 days from now we should be in a high point.  If you plot (Earth+Ocean Energy) against (Day Number mod 22), 10 days from now (Day Number mod 22) will be 10, and there has never been such a day when Earth+Ocean was <70.
  • Therefore, I plan to recommend the monk use Earth and Ocean energy.
Comment by aphyer on D&D.Sci August 2021: The Oracle and the Monk · 2021-08-13T23:54:51.900Z · LW · GW

Do we know when exactly the supernova occurred? Is the whole dataset post-supernova, or did it happen partway through the dataset?

Comment by aphyer on What is the name of this fallacy? · 2021-08-11T02:53:26.573Z · LW · GW

Not an answer to your question, but a more general point:

If your question must have political examples, you are much more likely to foster good conversation if you can make your examples be from a variety of positions rather than all from the same one. It demonstrates that you're not trying to score political points on an issue, shows that you can consider multiple sides of an issue, and avoids triggering a bunch of defensive comments from people who disagree with you. I'd recommend substituting an example like this one for one of yours:

'So you say capitalism is horrible and must be destroyed - on the iPhone made by a corporation.'

Comment by aphyer on Stupidity as a mental illness · 2021-08-08T19:11:14.511Z · LW · GW

Let's say I describe a person as 'fighty'.  There are two different things I could mean by that:

  1. This person is good at fighting.  Perhaps they are large and strong...or skilled at martial arts...or carry weapons...or some combination thereof.
  2. This person is prone to fighting.  They tend to resolve problems by fighting.  When they dislike something a person has done, they will start a fight.

The first of these I think is...pretty much morally neutral?  

The second of these I think is fairly clearly morally bad.

These things are definitely correlated with one another.  People who are good at fighting are probably also more prone to fighting (since when you expect to win a fight you might be more likely to start one).  And people who are prone to fighting are probably also better at it (since when you fight a lot you get good at it).

But they're clearly not the same thing.  My friend's Chihuahua is highly prone to fighting, but not very good at it.  And a retired Marine veteran might be very good at fighting but not very prone to it.

When we are talking about fighting, though, we recognize these two as different things.  We can describe someone as 'violent', which pretty much means #2.  We can call someone 'a gentle giant' to imply #1 but not #2.  And so on.

When we discuss intelligence, these two different things get muddled together:

  1. This person is good at thinking.  They are naturally intelligent, and find problems easier to solve than other people do.
  2. This person is prone to thinking.  When confronted with a problem, they attempt to solve it by thinking.  They do not simply take a gut reaction and insist on it in defiance of reason.

Sadly we don't really have different words for #1 and #2 here.  We use 'smart' and 'stupid' to refer to both of them.  And this is something of a problem because, while the two are again correlated, they are not the same thing, and the correct response to the two is different.

Being 'stupid' in sense #1 is...I mostly agree with the pro-stupidity people here?  Being smart is better than being stupid, but holding natural stupidity against people who cannot help it is a bad thing to do.

Being 'stupid' in sense #2 I think is bad, and I think can legitimately be held against people. be frank, I feel like most stupidity in our current world is of type #2?  I remember going to high school.  If you consider these two sets of people:

  1. People who failed high school math because they were genuinely not intelligent enough to do it.
  2. People who failed high school math because they put no effort into it.

I think it's reasonably clear that group #2 is orders of magnitude larger.

Comment by aphyer on What does GPT-3 understand? Symbol grounding and Chinese rooms · 2021-08-03T16:00:27.788Z · LW · GW

I am a human, living on a ball of rock third out from a star in a galaxy I call 'Milky Way'... This is an additional group of words, group two, out of my quick try at writing what you ask for... Yakka foob mog grub puggawup zink wattoom gazork chumbul spuzz tirv yo klii voy tryvial toy nikto savaran to dillino...

It is hard to avoid a symbol that is common in most writing, but so far I am avoiding it...

Although following such constraints is annoying, I can accomplish such a task, which GPT cannot grasp at all... Svirak nilliak toynodil r saasak motooil bordonak xi toy vir nolonol darak, silvi sansak so, tirin vak go na kilian ay BOY no vak...

If artificial minds cannot grasp constraints such as this, it is hard to claim that such minds truly 'grok' what is output... By contrast, I find this a fairly straightforward task, though slightly annoying, and will find a bit of joy in finishing it shortly...


Comment by aphyer on Covid 7/15: Rates of Change · 2021-08-02T15:50:37.604Z · LW · GW

A better way of phrasing it might be "don't applaud a growing potential tyrant who seizes additional power just because they happen to do something you want with it."  I do think the message is clear even without that change.

Comment by aphyer on Covid 7/22: Error Correction · 2021-07-23T02:54:40.734Z · LW · GW

Who do you want to be in charge of deciding which lies were 'plainly a deliberate creation' and have 'caused harm', vs which lies were 'probably genuinely believed as opposed to deliberate lies' or 'did not cause harm because, while lies, they misled people in good directions rather than harmful ones'?

Comment by aphyer on Punishing the good · 2021-07-21T02:33:41.743Z · LW · GW

This probably depends heavily on whether people are capturing all the benefit of the good they do.

If Bob inflicts $100 of harm to do $1000 of good, and Bob was rewarded by $1000 for the good he did, taking $100 of that away for the bad seems reasonable. If he was not rewarded for the good he did, punishing him for the bad seems very strange.

Comment by aphyer on Fixing the arbitrariness of game depth · 2021-07-17T20:00:43.401Z · LW · GW

Is there some reason to think that the depth of a game by this sort of definition is a good measure to be using in the first place? Presumably the 'deepest' game available is 'test your height and whoever is taller wins', which with accurate enough measuring instruments has a depth of 7 billion?

Comment by aphyer on Essentialness of Data · 2021-07-15T22:52:30.781Z · LW · GW

This viewpoint on how media has changed over time seems really strange to me. My history would have been almost the reverse - a history of going from everything being expensive (pay per view TV?) and usually low quality as well to almost everything being cheap and many things being literally free, including a lot of high-quality things.

Comment by aphyer on Essentialness of Data · 2021-07-15T16:18:55.763Z · LW · GW

It seems strange to say that you do not consent to people gathering data about you when you are providing it to them yourself.

If you don't want me to see your baby pictures, the easy approach is for you to not send them to me. Instead, the more common approach seems to be to send me your baby pictures, then claim that you 'do not consent' to me looking at them.

Comment by aphyer on Ideal College Education for an Aspiring Rationalist? · 2021-07-13T04:05:59.326Z · LW · GW

If you don't feel like you 'get' reductionism on a gut level, the course that nailed that down for me was called 'Computation Structures' - name may vary on your end, but it's a class about how transistors and logic gates are built up through several layers of abstraction until you have a recognizably programmable computer.

As a side note, if you're willing to adopt a more pragmatic attitude towards what constitutes a 'rational' college class, may I suggest searching for information on which majors are highest-paid? :)

Comment by aphyer on What will be the aftermath of the US intelligence lab leak report? · 2021-06-27T16:07:57.233Z · LW · GW

An alternative story:

The report will be ~500 pages long, and shrouded in an impenetrable dialect of bureaucratese.  No-one in the world will read it in its entirety.  A lot of people will cherry-pick bits that sound like they support their previous conclusion.   Some people will declare that it proves the coronavirus was a deliberate Chinese bioweapon released to attack America and we should retaliate by nuking them.  Some people will declare that it proves that discussion of the lab leak hypothesis has always been unfounded anti-Asian discriminatory hate speech.  No-one will change their mind about anything or do anything different.

In the few cases where the report presents clear statements of fact, it will be pretended that they were already the viewpoint of experts.  Any media outlets that made earlier statements incompatible with the report will quietly edit or simply ignore their past statements.   Only a handful of people will notice or care about this.

Comment by aphyer on Scaling Networks of Trust · 2021-06-22T16:25:02.162Z · LW · GW

Attempted expansion:

If the person you are dealing with is willing to do the right thing even if it costs them something, a lot of different systems will work.

If you do not trust the person you are dealing with to do the right thing in the absence of incentives, you want a system that will impose incentives that ensure it is in their interest to do the right thing. I'm calling this 'game-theoretically sound' to mean 'the system accomplishes the intended goal even when one or both parties are the kinds of sociopath that prevail in game theory problems'.

This could be formal law (do your end of the contract or the government will punish you).

It could be informal threats (uphold the bargain or me and my mates will clobber you).

In some circumstances it could be reputation (honor your note or no one will deal with you again? refund my purchase or I will review you badly online?).

However, for reputation to actually work as a system that is protected against adversaries/sociopaths (as distinct from working as a system for nice people who are already pretty much trustworthy), you need the damages caused by reputation to exceed the benefits gained by defecting. This is plausibly true for ancient merchants living in small societies, or for companies that have large numbers of customers. I don't think it's true for your networks of trust.

That is not necessarily a problem for you! If you're just dealing with your neighbor, you don't need your system to be defensible against sociopaths. But if you want to scale up your system, it will become more and more relevant.

Hope this is clearer, apologies for length.

Comment by aphyer on Scaling Networks of Trust · 2021-06-22T15:30:20.718Z · LW · GW

One obvious reason this might not work as well for you as for Islamic merchants is the scale of the threatened sanction.

Even assuming that the merchant would not face legal penalties for reneging on his note, and even assuming a lack of extrajudicial violence against him, he is still potentially subject to extreme levels of social pressure not available to you today. You cannot prevent a grocery store from selling him food, or a job from employing him. Social networks in a smaller society could potentially do that. You can...threaten to say mean things about him on the internet?

I'm glad this approach has worked for you, but I don't think it's game-theoretically sound, I think it is relying on good faith.

As an aside, I'm somewhat skeptical of the story of the merchants' notes being traded so far afield. How do you avoid forgery? If a note is given and never traded, the merchant can remember to who a favor is owed - if you stay in the same area, people can see who is owed and keep him honest. How can a trader in Zanzibar know whether this is a genuine favor note from a trader in Algiers?

Comment by aphyer on Open problem: how can we quantify player alignment in 2x2 normal-form games? · 2021-06-16T15:44:27.613Z · LW · GW

Correlation between player payouts? In a zero sum game it is -1, when payouts are perfectly aligned it is +1, if payouts are independent it is 0.

Comment by aphyer on Are PS5 scalpers actually bad? · 2021-05-18T14:54:50.565Z · LW · GW

A stupid question: even if there are pressures that keep producers from selling at the true market price, why don't they produce enough to meet demand at the lower price they do sell at? A deliberate strategy to build hype by making consoles scarce? Something else?

Comment by aphyer on Against Dog Ownership · 2021-05-11T04:35:08.118Z · LW · GW

(Disclosure: am dog owner)

I found this article somewhat baffling.

At essentially no point does the article consider what the dog's next-best alternative is, or suggest one.  The two most likely answers are 'nonexistence' (for dogs you get from a breeder) or 'prison' (for dogs you get from a shelter).  I'm open to the possibility that some dogs (abused ones, possibly also tiny breeds with genetic breathing problems) have lives worse than nonexistence.  I don't think you can credibly claim most do, and I don't think you can credibly claim that any meaningful proportion of owned dogs have worse lives than they would in shelters. 

The article seems to assume that dogs have a full set of human drives, and are disappointed not to be given the opportunity to write Doggie Shakespeare or something.  I envision the author watching some videos of lions and then saying 'they have no lives beyond waking, eating, and sleeping.  Clearly their life is a misery.'  

The article does an impressive dive into Bulverist psychoanalysis of 'here's why dog owners are evil perverts for having a dog':

'What does it say about a human who enjoys this emotional transaction? It says that on some level they like the idea of having dominance over another being. And, they want that dominance to be a feature of their daily life.' 

On reflection I think the most accurate summary of this article is 'willful outrage-bait'.

Comment by aphyer on Why I Work on Ads · 2021-05-04T14:34:01.912Z · LW · GW

Fair enough (and apologies for the rudeness).  I do think I'd draw a pretty sharp distinction between 'ads dropped in public spaces where you cannot avoid seeing them' vs. 'ads on webpages that you watch in lieu of paying for things' - the latter seems much easier to avoid and much less likely to be harmful.

(And as I understand things OP seems to be mostly working on the latter?)

Comment by aphyer on Why I Work on Ads · 2021-05-04T02:31:33.534Z · LW · GW

So, this viewpoint is very harsh and I don't know how fully I endorse it, but my gut reaction is something like this:

If you can't benefit from positive-sum informative advertising because you are incapable of watching a 15-second ad without succumbing to mind control, this is a problem with you rather than a problem with ads.  The correct response is for you to avoid ads personally (and in fact  many websites that use internet advertisements give you the option to pay instead, e.g. Youtube Premium), just as a child who cannot prepare food without cutting themselves should not be given a set of steak knives.   It sounds like you are doing that already, so good for you!

The correct response is not for you to try to prevent a positive-sum thing from existing for others, just as the correct response to a child getting their hands on a steak knife and cutting themselves should not be to try to ban steak knives for everyone else.

Attempts to restrict advertisements on those grounds seem isomorphic to e.g. New York Mayor Bloomberg's infamous attempted ban on large sodas.  The justification there appeared to be 'I am incapable of existing in a world with large sodas without drinking too much soda and getting fat, therefore other people should be banned from positive-sum trade to protect me from my weakness without me needing to exert any effort.'  The argument against soda seems to me a substantially stronger argument than the equivalent argument against ads: first, I think the harms of obesity are substantially larger than the harms of advertisements; and second, I think it is easier to personally avoid exposure to internet advertisements than it is to personally avoid exposure to large sodas.

Comment by aphyer on Why I Work on Ads · 2021-05-04T00:02:40.371Z · LW · GW

While I think that ads can be (and, at least in my personal case, are) positive-sum, I don't think this article expresses the reason why.  I would phrase it as follows:

  • There are many things I would buy if I knew of their existence.  Certain kinds of music, games, etc.
  • Unfortunately, I don't know of their existence.  In a world with billions of people, figuring out what products I might want takes time and effort.
  • You could let anyone who wanted to tell me 'you should buy my product'.  Unfortunately, lots of people will waste lots of my time.
  • Instead, we set up an ecosystem where:
    • A website (let's say Youtube) shows me content I want to see.
    • Rather than charging me $20/month or whatever, they make me watch brief ads.
    • Companies that think they have a product I want to buy pay Youtube to advertise on it.
  • If the companies are right:
    • I buy the product.
    • I am very happy!  I got to see content I wanted to see anyway, without having to pay for it, and I also got alerted to the existence of a product I'm interested in!  Yay!
    • Youtube is happy.  They got paid for providing content to me (albeit indirectly) and I'm happy and likely to keep coming back.
    • The company is happy.  They paid a bit for advertising, successfully reached the target market for their product, and made more money.
    • This is a large win-win where a lot of value/money/information/amusement is effectively materialized from thin air with no costs.
  • If the companies are wrong:
    • I don't buy the product.
    • I am mildly happy.  I got to see content I wanted to see, without having to pay for it, but I had to watch a silly ad for an economically illiterate insurance agency.  Meh.
    • Youtube is mildly happy.  They got paid for providing content to me (albeit indirectly) but I'm less happy and might look for other websites (EDIT: and the company is less happy and might do the same).
    • The company is unhappy.  They paid a bit for advertising, failed to reach the target market, and wasted it.
    • This isn't an amazing win-win like the scenario before.  Nevertheless, the only person who's really unhappy with this is the company...who is effectively paying for their poorly targeted ads.


This...seems...obviously good to me?  A couple representative things I've run into from ads:

  • The band Mono Inc. took out a Youtube ad that was just the music video for one of their songs.  I liked it, and have bought several of their songs as a direct result.
  • The game Hades got advertised to me back while it was in early access. I bought it very early, and liked it a lot.


When I'm upset about advertising, it's usually because it's not targeted enough.  For example, Buffalo Wild Wings has started showing me lots of ads for their new delivery service...which does not deliver to my zipcode.  This seems like a foolish waste of effort that could be avoided if only advertisers took my zipcode into account.  I assume this is the result of some sort of silly privacy law that forbids them from using that information in a moral panic over Big Data?


I acknowledge that my experience with ads may not be the typical one.  Nevertheless, I think that:

  • Ads can be, and frequently are, heavily positive-sum.
  • In some cases, people may harm themselves with ads by being foolish around them.
  • I still don't want positive-sum advertising banned or heavily restricted because some people are foolish about it.
Comment by aphyer on Death by Red Tape · 2021-05-02T14:01:41.704Z · LW · GW

While I agree with the general point, I think this part:

(The TSA spends 5.3 billion on airline security annually. It's difficult to put a price on 9/11, but quick googling says that total insurance payouts were $40 billion. So very roughly, the utilitarian question is whether the TSA stops a 9/11-scale attack every 8 years.)

is rather poorly phrased in a way that detracts from the overall message.  I don't think this accurately captures either the costs of the TSA (many of which come in the form of lost time waiting in security or of poorly-defined costs to privacy) or the costs of 9/11 (even if we accept that the insurance payouts adequately capture the cost of lives lost, injuries, disruption, etc. there are lots of extra...let's call them 'geopolitical costs').

Comment by aphyer on D&D.Sci April 2021: Voyages of the Gray Swan · 2021-04-13T00:06:06.017Z · LW · GW

Looks like you've put in more work than I did below and come up with very similar answers, the one diff I see is that you tried 

foam swords as a speculative guess that water elementals are non-linear and explain the extra sinkings/a way to reduce non-sinking-related damage

whereas I just went for

a third cannon for a bit of marginal equity against Nessie

Comment by aphyer on D&D.Sci April 2021: Voyages of the Gray Swan · 2021-04-12T23:46:15.266Z · LW · GW


Looking at the 99% and 98% damaged ships:

  • We see 46 99%-damaged ships that were attacked: 31 by demon whales, 6 by Nessie, 1 by crabmonsters, 8 by merpeople.
  • We see 44 98%-damaged ships that were attacked: 22 by demon whales, 6 by Nessie, 6 by crabmonsters, 10 by merpeople.

These look fairly similar, and suggest that those are the threats that are likely to be most dangerous (until of course some more serious threat is reliably unsurvivable so we don't find out about it, or some threat deals non-linear damage...but let's ignore that for now.  I was assured in Linear Algebra class that the real world never works like that.)

Given this: 

  • Oars, Cannons and Armed Carpenters (in descending order) counter the most common threats.
  • Tribute to the merpeople reduces the chance that they attack, though it increases the chance that something else does.
  • Rifles and Foam Cutlasses counter attacks that don't seem to be serious enough to matter anyway.
  • Shark repellent is actively harmful, since it prevents sharks from attacking (no shark attack has ever done more than 9% damage (EDIT: wrong, I sorted these as strings because I am a dullard, but they still never seem to do near-lethal damage), which makes sense are sharks going to get into your boat?) and makes something more dangerous attack instead.

Adding up Armed Carpenters (20) + Tribute to Merpeople (45) + 20 Oars (20) + 3 Cannons (30) makes 115gp, which we can't quite afford.  My best guesses:

  • If trying purely to survive, buy Tribute to Merpeople + 20 Oars + 3 Cannons for 95gp.
  • If trying to survive while also impressing the Admiralty with your budget-consciousness (and your refusal to pay Danegeld to those damned merfolk!), buy Armed Carpenters + 20 Oars + 3 Cannons for 70gp.
Comment by aphyer on Preventing overcharging by prosecutors · 2021-04-06T17:02:24.424Z · LW · GW

An alternative approach that feels less gameable:

Every time a plea bargain is reached, there is a small chance (1 in 1000?) that we randomly select the case for review.  The defendant is imprisoned as per the terms of the plea bargain, but then we also bring the case to trial (federally funding a defense).

If the review trial finds the defendant guilty of the charges they pled to, we say 'okay, seems like the plea bargain was fair, the defendant would have been found guilty anyway'.

If the review trial finds the defendant innocent, we release them and publicize this fact.  If multiple reviews under the same prosecutor exonerate the defendant, it starts to look like that prosecutor is threatening charges to coerce defendants into unfavorable plea bargains.

A 1 in 1000 chance means that the cost of this should not grow too excessive compared to existing costs of the legal system, but also means that a prosecutor who is habitually coercing defendants into taking unfavorable plea deals is very likely to get caught.

Comment by aphyer on The best frequently don't rise to the top · 2021-03-25T19:38:12.298Z · LW · GW

Without commenting on the rest of your point, this part:

If my quality is, say, a 7/10, my level of success should be somewhere in that ballpark. Maybe the market would be inefficient and I'd only reach a 5/10 or a 4/10. Or maybe I'd be lucky and reach an 8/10 or 9/10.

is something I don't think you should necessarily have expected from a market to begin with.  Assuming that your startup was in a scalable field (e.g. a software company as opposed to a restaurant) with an efficient market, what you expect to happen is that quality 7/10 should lead to success 0/10 - there is no reason that anyone would buy from you when they could instead buy from a 10/10.

In reality there are lots of reasons this isn't quite the case.  Maybe your product is 7/10 overall but genuinely the best in the world for certain specific users/use cases.  (C++ and Python both continue to exist).  Maybe the market just isn't very efficient and some people end up buying inferior products.  (PHP also continues to exist).  But on the whole expecting a 7/10 product to experience a 7/10 success is not actually what you should expect to see.

Comment by aphyer on Thirty-three randomly selected bioethics papers · 2021-03-23T18:46:02.386Z · LW · GW

Numerical cost-benefit analysis. 

Comment by aphyer on Thirty-three randomly selected bioethics papers · 2021-03-23T18:36:31.394Z · LW · GW

Your hyperlink currently includes a period at the end of the link URL and so doesn't work right.

Comment by aphyer on Thirty-three randomly selected bioethics papers · 2021-03-23T14:57:38.776Z · LW · GW

I do think that packing five separate questions into one paper is too much, but also going through those questions one at a time:

(a) What is the difference between practice and research? 

This question seems...obviously stupid?  It might be intended as a Socratic lead-in of some sort, I suppose.

(b) What is the relationship between research ethics and clinical ethics? 

This question seems extremely vague.  I can imagine related sub-questions that could be meaningful: e.g. 'does research need to use different ethical standards than clinical ethics' (like (d) below), 'does research need separate ethical regulations from clinical treatment, or can it use the same ones', 'should clinicians be worried about the ethics of researchers who give them treatments/vice versa', but in the absence of a more specific question I'm not clear on what this means or what an answer would be.

(c) What is the ethical relevance of the principle of clinical equipoise? 

I hadn't heard of this.  Per Christian's answer above it might be a reasonable question, although it seems a bit tautological asking for the ethical relevance of a principle if the principle itself is an ethical principle.  Still willing to accept it as a probably-okay question.

(d) Does participation in research require a higher standard of informed consent than the practice of medicine? 

This is a good question.  If it were the only question in the paper I would like it.

(e) What ethical principle should take precedence in medicine?

We take a kind of abrupt turn here into a very high-level meta-question.  It's weird to combine this with a bunch of lower-level questions, and even weirder to put it at the end - surely if you need to decide what ethical principle to use that needs to be the first thing you do?

So I think we've got either 2 or 3 reasonable questions muddled together into one paper along with some silly/poorly defined ones.   Then, looking at the ending, the main thrust of their conclusion appears to be 'our approach is useless for thinking about these questions':

It can be concluded that [our approach] has not provided us with conceptual instruments that would resolve the ethical debate between proponents of the segregation and integration models... 

...[our approach] also does nothing to resolve the problem of a researcher's clinical obligation. Rather, it creates a new source of moral obligation: a health‐care system. Next, [our approach] seems not to resolve the controversy over the concept of clinical equipoise.

I am overall not very impressed by this paper.  I don't think it sounds actively evil or anything, I just think it sounds like a waste of (a great deal of) paper.

Comment by aphyer on Thirty-three randomly selected bioethics papers · 2021-03-23T13:00:49.611Z · LW · GW

I'll agree that I worded it pretty harshly, but I do think I'll stand by it not being a useful paper.  Imagine a science paper that claimed to be about the following list of questions:

a) What is the difference between iron and xenon?  b) What is the relationship between solid matter and gaseous matter? c) What is the practical relevance of the principle of least action? d) Does investigating radioactive materials require different experimental procedures than investigating other physics? e) What kind of statistics should be used in physics papers?

Comment by aphyer on Thirty-three randomly selected bioethics papers · 2021-03-23T02:10:55.287Z · LW · GW

Many of these are rather hard to read.  The following are my even-more-slimmed down summaries based on skimming just those few paragraphs.  Some may be in error, if you disagree with any let me know:

1. Argues in favor of more bioenhancement and against refusing to do bioenhancement for reasons of 'egalitarianism', though in a somewhat wishy-washy way 'we endorse a cautious proposal'.

2. Not really related to bioethics - argues in favor of pseudonymous publishing.

3. Argues in favor of 'nudging' patients to obtain consent for treatments.

4. Argues that if you favor assisted suicide when doctors do it, you should also favor it when for-profit entities do it.

5. Argues that all newborns should be screened for critical congenital heart diseases, without exemption for parental religious beliefs.

6. Tells the story of a patient who demanded his doctors do something really stupid and refused to budge.  Unclear what moral, if any, they want to draw.

7. A weirdly meta paper that evaluates methods for evaluating ethics.  I have no idea what this means.

8. Frets about off-label use of ketamine being worrying, recommends formal research into it combined with stricter regulation of off-label uses.

9. I have no idea what this paper is about.  It tells the story of a kid called James, but I don't know what it wants to draw from it.

10. Argues against 'nudging' people to register as organ donors.

11. Argues that traditional consent is inadequate for a 'biobank' (which stores biological samples of people?) and that you should use something called a 'Participation Pact' instead.

12. Argues that bioethicists should pay attention to neuroscience?

13. Agrees with someone called Lee that bioethics and environmental ethics should be linked, but disagrees that his idea of 'public health ethics' is a good way of doing it.

14. Asks about how to allocate limited supply of a drug called nusinersen for spinal muscular atrophy patients.

15. Argues that consent to organ donation is 'vague' in harmful ways.

16. Talks about 'right-to-try' trials where terminally ill patients try untested drugs.  Some waffling, I'm unclear if they approve or not.

17. Argues that diversity in researchers will help make research better.

18. Talks about what to do if your patient's relatives try to fix him with prayer and magic.  Appears to argue that you should support the relatives as well as the patient.

19. Argues that conscientious objection to abortion should not keep you from being sued if your refusal causes injury.

20. Proposes a different standard for how to evaluate parent's decisions re. medical care for their children.  Unclear how it differs.

21. Argues that standards for resolving e.g. disagreement between doctors and family members on what to do with an incapacitated patient should be simpler.  

22. Argues that all children should get measles vaccination.  Claims that this paper's conclusions 'reframe the dialog on measles vaccination...from a framework of what is owed to parents...[to] the framework of what is owed to children'.  Unclear why that wasn't the framework we were in already.

23. References Gattaca, then argues (against some other bioethicist) that bioenhancement can be good if regulated.

24. Argues that 'professional guinea pigs' who participate in trials for money are being exploited by capitalism.

25. Argues that palliative care for children is currently in a very bad state, and that we should do research into it and make it better.

26. I have no idea what this paper is about.  I don't think the authors do either. (EDIT: second opinion is that I'm being uncharitable here, see Kaj below).

27. Declares that a 2010 study on premature babies called SUPPORT was unethical, arguing that it placed one of its treatment groups at higher risk than they would have been without the study.

28. Talks about how to handle decision-making for unrepresented patients.  Unclear what they think you should do.

29. Argues that the HEC-C program (a healthcare certification program of some kind) is good.

30. Argues that the HEC-C program (again!) should be more diverse - what they seem to mean by this is not the standard concept of 'diversity' but that it should cover a more diverse set of medical situations.

31. Argues that informed consent is 'nonsense' because people sometimes believe multiple different things.

32. Argues (against some other bioethicists) that you do not need consent from a patient to test whether they are brain-dead.

33. Book review of a book arguing that medical ethics 'cannot be regarded as an extension of common morality' and must be treated as a completely different thing.

Comment by aphyer on How poor is US vaccine response by comparison to other countries? · 2021-02-27T16:41:17.239Z · LW · GW

I agree that the export ban is not cooperative policy, but I'm unclear if it has an effect.  If the US is banning exports of vaccines, but also paying more for vaccines than the EU, it seems unlikely that vaccines will be exported from the US to the EU anyway?

Comment by aphyer on Covid 2/11: As Expected · 2021-02-27T16:36:14.028Z · LW · GW

Update: Metaculus published a document explaining their scoring rule, which seems to say something similar.  The key quote is probably this:

Our positive sum Scoring Rule — in which for a given question the total points awarded across users is generally positive — encourages broader participation on more questions by not penalizing forecasters who participate in more questions

Comment by aphyer on How poor is US vaccine response by comparison to other countries? · 2021-02-20T01:41:05.319Z · LW · GW

Presumably the relevant question for vaccine rollout is 'are these places running their own vaccine distribution program, or are they covered under the umbrella of the UK's program.'  My guess would be the latter, leaving a total of four distinct vaccination programs of which one is fairly large, two are medium-sized and one is tiny (the UK, Israel, the UAE and Seychelles) as ones that outperform the US.

Comment by aphyer on How poor is US vaccine response by comparison to other countries? · 2021-02-17T19:45:38.234Z · LW · GW

Interesting.  To be honest, I'd gotten almost the opposite impression regarding our throwing-money-at-it phase.  Here is one article (found recently with Google, but it seems fairly representative of the sort of things I have been reading about our vaccine acquisition):

Just as thousands of Brits were lining up to get Pfizer’s mRNA COVID-19 vaccine yesterday, a troubling question emerged in the U.S.: Did the United States government fail to lock in enough doses of the vaccine to ensure a broad and quick rollout here?

The short answer appears to be yes.

Pfizer struck a deal with the U.S. government for 100 million doses of its COVID vaccine over the summer, but when the company offered more, [the US] administration declined, according to anonymous sources who spoke to the New York Times.


Pfizer and BioNTech went on to lock up supply pacts with other countries. They struck a deal with the U.K. government for 30 million doses in July, and in November nabbed a contract with the European Union for 300 million doses, promising the first deliveries by the end of the year.

It sounds like maybe our throwing-money-at-it phase was #4 (we did badly but other people did even worse), which has led our distribution phase to be #5 (we're doing badly but we have enough of a head start that we're still doing better than most)?  

Comment by aphyer on Insights from Euclid's 'Elements' · 2021-02-15T01:59:38.386Z · LW · GW

A bit pedantic but still worth pointing out - unless I'm very mistaken or Euclid uses the term very differently from modern mathematicians, this bit:

Two triangles are similar when there exists such an affine transformation (i.e., you can scale as well).

is not the correct definition of 'affine transformation'.  If two triangles are similar, there will exist an affine transform between them, but the reverse is not necessarily true.  The category of affine transformations is broader than you think - an affine transformation allows you to skew things as well (e.g. convert a square into a parallelogram), or to scale things differently along different axes (e.g. convert a square into a rectangle).  The ratio between two distances in an affine transform must remain the same only if those distances are parallel.  In point of fact, I believe affine transformations can convert any triangle into any other triangle.  

(Checks online: Wolfram says yes).