## Posts

My prediction for Covid-19 2020-05-31T23:25:17.814Z · score: 17 (6 votes)

Comment by themajor on My prediction for Covid-19 · 2020-06-04T07:30:57.150Z · score: 3 (2 votes) · LW · GW

I'm happy to hear that some of these changes have been unexpectedly positive for you! Personally I already did a bunch of these things (shop for one week's worth at a time, have days working from home, order online). To offer a bit of a peek at the flip side: I work in mathematics, which is uniquely suited for working at home (there's a modern joke that to do mathematics all you need is pen, paper, a bottle of water and a supercomputer), yet 2 of the 11 colleagues in my group have suffered burnouts since March from the added stress of having to look after their households. We are considering going back to 20% office capacity soon in staggered shifts, which while nice still means we won't have a chance to talk or work together in practice. Obviously this is exactly the point, but I want to note that this is a far cry from normal. My productivity at the moment is at an all time low, and several of my other friends have already heard that if this situation continues for much longer they will be let go from their jobs. In this sense I think this is unsustainable, or at the very least a serious hit to our global growth and productivity.

I have absolutely no quantitative guess what the impact of superspreaders is, and it would be amazing if we could stop them quickly. I think Zvi also pointed out that superspreaders get eliminated quickly in a pandemic, one way or another (and 'having the disease, surviving and then becoming immune' counts as the other).

I thought that the current spread of Covid-19 in warmer countries (Brazil, India) was evidence against the virus being very susceptible to temperatures, but there are a lot of confounders. If you know of any good summary of the current knowledge on this please let me know, I am very interested in this (and it would likely change my predictions massively).

Comment by themajor on My prediction for Covid-19 · 2020-06-03T09:37:01.166Z · score: 1 (1 votes) · LW · GW

Thank you for your comment. I don't think we disagree - Slovakia (and other European countries) have done extremely well by acting quickly. I also fully support washing hands, wearing face masks and being smart about social contact. I think you are responding to my sentence

[...] this suggests to me that 'The Dance' will look a lot more like 'a full quarantine, but with a few restrictions lifted' than like 'restoring social contact, but wash your hands and wear a face mask'.

I'm sorry for the confusion. The situation you describe sounds more like my scenario 3 than my scenario 2, and I think I explained poorly where I draw the line in my quoted statement above. Going by wikipedia, the most recent round of relaxations in Slovakia still sounds far from life as normal to me. The maximum occupancy set at shops is up to 1 person per 15 square meters, which is 25% of what it is in normal times. I imagine similar concerns apply to offices, but I can't find how many people are back to working at the office. The opening of schools happened only this Monday, so it really is too early to tell what the impact of that will be. Zvi voices some concerns.

Lastly, my entire goal was to try and talk about relative impact, and give perspective to the magnitude of the measures we can expect going forwards. I don't see where I have mistakenly given absolute statements on effectiveness of measures (in fact I only mentioned face masks once, without making any statement for or against them in the OP), but if you point them out I would be happy to change them.

Comment by themajor on My prediction for Covid-19 · 2020-06-01T13:04:01.115Z · score: 1 (1 votes) · LW · GW

I agree completely. However, I think the amount it has gone up is critical here. A lot of the countermeasures and increased preparation are linear countermeasures against an exponential threat - maybe a region that could previously only handle 1000 ICU patients can now take care of 2000, but if R0 is significantly above 1 (lets say 1.5) this only buys you about one and a half week. I think this topic deserves its own entire post at some point, and I didn't want to get bogged down in details in the section on "What doesn't change", but if the true rule is "if under X circumstances in March it was smart to go into lockdown, it is November smart to go into lockdown 2 weeks after seeing X" my conclusions are still the same.

I might write that full post sometime on this and more back-and-forth, if people are interested. I made serious concessions to brevity above.

Comment by themajor on My prediction for Covid-19 · 2020-06-01T09:03:32.958Z · score: 5 (2 votes) · LW · GW

Thank you for your comment! I'd like to briefly clarify that I don't consider the "lockdown versus kill the people" question a dilemma - I'm not trying to propose policy, just attempting to make realistic predictions for the next few months.

I think going for the approach you suggest ('option 2') is unrealistic in most western countries, and that the countries where this strategy worked had unique circumstances that worked in their favour.

As gbear605 points out, it is a lot more difficult to revert to this strategy when the disease is endemic, compared to when the number of active cases is in the several hundreds. In the post I link to an estimate of 1.1 million active cases of Covid-19 in the US today. One of the bigger weak links in the non-quarantine approach is that, when people test positive for cold/flu symptoms, there needs to be some plan or support in place to thoroughly isolate them for up to two weeks. At lower numbers this can plausibly handled by hospitals or nursing homes (which has been a big part of South Korea's method), but at up to a million cases this is unrealistic. In effect this would be akin to a lockdown with some extra steps.

Another advantage that South Korea, Taiwan and Singapore (as far as I know the biggest success stories of this method) had is that they are geographically isolated. I imagine this cuts down on the threat of re-infection from outside the country, and also why it is so much more difficult to prevent Covid-19 from becoming endemic in the US.

Widespread screening for elevated temperatures and cold/flu symptoms sounds like a very clever (and cost-effective!) method to combat the virus, if we also have a plan ready for people who test positive. But I don't think it can be the success story in western countries that it is in some Asian countries. We either missed our window and let Covid-19 grow too much, or we never had a window to begin with.

As for your later two points, I believe I addressed those in the original post. I think bringing R0 down to around 1 and keeping it there is prohibitively taxing, in multiple ways, and that even slight relaxation of our response will already have severe consequences. We should be thinking on the R0 scale, which runs from about 0.7 to about 2.5-3. Also I explicitly mention that I am fine with leaving the level at which we achieve herd immunity open for debate (I give 20% as a somewhat lower bound, based on estimates that this exposure level has already been reached in NYC).

Comment by themajor on Covid-19: My Current Model · 2020-05-31T20:38:56.417Z · score: 2 (7 votes) · LW · GW

Excellent post, thank you for writing this! I’ve been meaning to write up something similar, about my opinion on the public response to COVID-19 and a few future predictions. I think this comment section is a perfect place for this, and I’ll write my own thoughts as a response to yours. My experiences are not based on the situation in the USA.

As a disclaimer, my predictions are among the most pessimistic of all the ones I’m seeing people discuss. Consider it the bottom pit of the Overton window. Having expectations that COVID-19 will be worse than people are thinking so far can and does have negative effects on your mental health, so be careful (and stop reading). I’d love to be proven wrong.

Amazingly written, I did not give this enough thought. I should stop doing my own grocery shopping, and have it delivered instead.

Sacrifices To The Gods Are Demanded Everywhere

I agree fully. Your steelman hits quite close to the mark in my opinion. I want to add that if everybody has internalised that ‘pleasing the Gods’ is the way to respond (to anything, really), then this actually becomes the correct way for authorities to write guidelines – anything else will just add another layer of translation.

Governments Most Places Are Lying Liars With No Ability To Plan or Physically Reason. They Can’t Even Stop Interfering and Killing People

Yes.

Silence is Golden

I am completely unsure on the relative importance of this, compared to other risk factors. How much of the spread are you willing to attribute to this? Have there been studies on this?

Surfaces Are Mostly Harmless

I have been keeping all packages that arrived in a sealed plastic bag for 24 hours before opening (like a Good Boy), I guess that was stupid. Reading this paragraph changed my mind.

Food Is Mostly Harmless

Yeah, this all makes sense.

Outdoor Activity Is Relatively Harmless

As does this.

Masks Are Effective, And Even Cloth Ones Are Almost Enough

I totally disagree with this. Where are the numbers coming from? I consider cloth masks closer to a Sacrifice to the Gods than a real countermeasure.

Six Feet Is An Arbitrary Number, People Aren’t Treating It That Way, And That’s Terrible

You have this one completely backwards. People really really are not able to make risk assessments and benefit analyses. This world you describe, where the rule is taken as a best point estimate, and people interpret the ‘6 feet’ as anything else than a Boolean yes/no question, was never going to happen. I consider the current public awareness of the importance of keeping distance an amazing improvement on the status quo before this rule went public. You mention this (as the opening sentence, in fact). I disagree with any suggestion that we could plausibly do better collectively.

Herd Immunity Comes Well Before 75% Infected and Partial Herd Immunity Is Super Valuable

This is a very good point, but can you explain the numbers you give below? I have no intuition on what typical spread in personal infection risk is (sure, the range is large, but we also need the frequencies with which these high risks are taken).

Yes, We Know People Who Have Been Infected Are Immune

I agree fully.

Our Lack of Experimentation Is Still Completely Insane

Agreed.

We Should Be Spending Vastly More on Vaccines, Testing and Other Medical Solutions

I agree here, except with the last line. With all the regression to the mean the current medical system is experiencing, it is totally unclear to me how we would plausibly create a vaccine on a short (“a few months”) timescales.

R0 Under American-Style Lockdown Conditions Defaults To Just Under One, Which New York Escaped Via Partial Herd Immunity

Another great observation, and I’ll reply in more detail in a separate post..

The Default Infection Fatality Rate (IFR) Is 0.5%-1%, Depending on Conditions

There has been discussion of this on LessWrong before, and I agree.

Typically in America, 33% of Deaths and 90% of Infections Are Missed

This sounds like the correct order of magnitude for both, and also agrees with my estimates for the under-reporting in the official numbers where I live.

People Don’t Modify Behavior In Response To Re-Openings All That Much, When Given a Choice

This title is a very important (positive) remark, because it suggests (contrary to what I personally believe) that it might be possible to resume a lot of activities almost as normal while maintaining R0<1. I agree that the situation for schools is uniquely bad, and unfortunately I can tell you from personal experience this is one of the places completely infected with making Sacrifices to the Gods. This is a disaster waiting to happen.

It’s Out of Our Hands

When was this window for people in authority? Looking back I honestly think we were never capable of dodging this disaster (our society isn’t adequate at the level of saving people during a pandemic).

Comment by themajor on Corrigibility as outside view · 2020-05-09T16:15:26.574Z · score: 3 (2 votes) · LW · GW

Both outside view reasoning and corrigibility use the outcome of our own utility calculation/mental effort as input for making a decision, instead of output. Perhaps this should be interpreted as taking some gods-eye-view of the agent and their surroundings. When I invoke the outside view, I really am asking "in the past, in situations where my brain said X would happen, what really happened?". Looking at it like this I think not invoking the outside view is a weird form of duality, where we (willingly) ignore the fact that historically my brain has disproportionately suggested X in situations where Y actually happened. Of course in a world with ideal reasoners (or at least, where I am an ideal reasoner) the outside view will agree with the output of my mental progress.

To me this feels different (though still similar or possibly related, but not the same) to the corrigibility examples. Here the difference between corrigible or incorrigible is not a matter of expected future outcomes, but is decided by uncertainty about the desirability of the outcomes (in particular, the AI having false confidence that some bad future is actually good). We want our untrained AI to think "My real goal, no matter what I'm currently explicitly programmed to do, is to satisfy what the researchers around me want, which includes complying if they want to change my code." To me this sounds different than the outside view, where I 'merely' had to accept that for an ideal reasoner the outside view will produce the same conclusion as my inside view, so any differences between them are interesting facts about my own mental models and can be used to improve my ability to reason.

That being said, I am not sure the difference between uncertainty around future events and uncertainty about desirability of future states is something fundamental. Maybe the concept of probutility bridges this gap - I am positing that corrigibility and outside view reason on different levels, but as long as agents applying the outside view in a sufficiently thorough way are corrigible (or the other way around) the difference may not be physical.

Comment by themajor on Is this viable physics? · 2020-04-14T20:38:11.729Z · score: 6 (9 votes) · LW · GW

I've tried to read through the linked page, and swapped to academic reading' (checking the pictures, and sometimes the first and last line of paragraphs) halfway through. I think this is not viable.

There is a host of "theories of the universe" with a similar structure on a meta-level, consisting of some kind of emergent complexity. It is important to keep in mind the strength of a theory lies in what it forbids, not in what it permits. To date most theories of the universe fail this test hard, by being so vague and nonspecific that any scientific concept can be pattern-matched to some aspect of it. Judging by what I've read so far this is no exception (and in fact, I suspect that the reason Wolfram references so many big scientific theories is because large concepts are easier to pattern-match, whereas specific predictions are not as open to interpretation). Why will his patterns produce Einstein's equations (note that they currently do no such thing, he states we first need to "find the right universe"), and not Newton's, or Einstein's with double the speed of light?

As always with these nonspecific theories' it is very difficult to nail down one specific weakness. But currently all I'm seeing are red flags. I predict serious media attention and possibly some relevant discoveries in physics (some of the paragraphs sounded better than all other crackpot theories I've seen), but the majority of it seems wrong/worthless.

Comment by themajor on Coronavirus: Justified Key Insights Thread · 2020-04-14T20:09:16.887Z · score: 3 (2 votes) · LW · GW

I don't have a good model to give me any predictions on what reasonable numbers of asymptomatic cases would be, or how truncation influences these numbers. Could you explain why the inference is idiotic, and perhaps give a more reasonable one?

Comment by themajor on Coronavirus: Justified Key Insights Thread · 2020-04-14T19:24:44.839Z · score: 2 (2 votes) · LW · GW

Is there reason to believe the raw numbers are more accurate estimate of the rate than the model prediction? Also, what are the type-1 and type-2 errors of the tests used on the Diamond Princess? I heard some early reports that both of these might be significant, but then never heard anything about them again.

I checked that link above and followed their references to find other datasets, but two of them are in Japanese, one only deals with self-selected patients who showed symptoms, and the last two have small sample size (12 patients, two papers cover the same event).

Update: I have found https://www.eurosurveillance.org/content/10.2807/1560-7917.ES.2020.25.3.2000045, which benchmarks the real-time reverse transcription polymerase chain reaction (RT-PCT) tests. They state zero false positives in a trial with 297 non-COVID-19 samples, although they do retest 4 samples that showed "weak initial reactivity". Since the non real-time version of RT-PCT is supposed to be even more reliable, this means false positives are presumably not a big deal (even at a pessimistic 4/297 false positive this still means only 41 false positives out of 3063 tests done on the Diamond Princess).

Comment by themajor on Where should LessWrong go on COVID? · 2020-04-14T08:17:53.155Z · score: 21 (12 votes) · LW · GW

I think it is still very unclear what the situation a medium (~a few months) to long (~a year) amount of time from now will be like. I would love to see more discussion on this. On a related note, I think mental health and self-help are going to very important very soon, and while I am in a fine place personally I would still like to know a lot more about this, including how to help others. This strays a little bit from the other COVID discussion topics, but I do think LessWrong might have a comparative advantage here (especially compared to the crapshoot baseline that is the internet).

Comment by themajor on Coronavirus: Justified Key Insights Thread · 2020-04-14T07:49:14.051Z · score: 4 (3 votes) · LW · GW

Have you got a source for that 'about half the cases are asymptomatic'? I was under the impression that far more cases show symptoms eventually, and that the studies showing half of the infections are asymptomatic add the disclaimer 'so far', which means very little if the spread is growing exponentially with a doubling time of several days.

Comment by themajor on How strong is the evidence for hydroxychloroquine? · 2020-04-08T17:34:03.568Z · score: 5 (3 votes) · LW · GW

Thanks for sharing! I'm not a doctor, so I found this a tough read. This document is clearly a proposal (attempting to convince the reader) instead of a summary, but it still contains a lot of useful information. Nevertheless, there were some parts I found especially confusing.

On page 2 they mention there are currently 22 studies, of which one has completed, on the effect of hydroxychloroquine (HCQ) treatment on COVID-19 patients. Further down in the piece (in particular in the section "What about the studies that show no benefit from HCQ?" on page 11) they dismiss some studies showing little or no effect. Is there a place to find more discussion on which studies are being discounted, and for what reason? They link one study only, citing that "only 400mg daily for 5 days was used", although the suggested treatment in this document is "HCQ: 6.5-15mg/kg PO in divided loading dose followed by 400-800mg/day in divided doses for 4-9 days" (which encompasses 400mg daily for 5 days).

The recommended treatment is a combination treatment with four different components - an initial oral hydroxychloroquine administration and a daily treatment of hydroxychloroquine and two other medicines (zinc and Azithromycin). Furthermore the document states that this treatment is expected to work a lot better in early stages of the disease (this part is also unclear to me - again on page 11 they state that "[some studies] waited to initiate treatment until the disease was too far progressed to be effective" as grounds for dismissal). Does this mean this treatment is expected to have next to no effect in late stages? I'm worried about Bonferfoni-esque situations here; are 21 incomplete and 1 complete study strong enough to motivate this complicated treatment, especially if we allow ourselves to discount some papers with conflicting conclusions as well as restrict the time period over which the treatment is supposed to be effective?

Comment by themajor on How strong is the evidence for hydroxychloroquine? · 2020-04-06T08:42:29.978Z · score: 2 (2 votes) · LW · GW

I am very interested in discussion on hydroxychloroquine, but do not have a Facebook. Is there some other way to read the megathread?

Comment by themajor on A Kernel of Truth: Insights from 'A Friendly Approach to Functional Analysis' · 2020-04-04T09:16:35.445Z · score: 4 (3 votes) · LW · GW

Very nice! Two mistakes though:

• Technically the introductory part on derivatives on is incorrect, in two different ways.
• Firstly the derivative of a map is a map , that assigns to every point x a linear map sending direction y to a real value (namely the partial derivative of f at x in direction y). Thankfully the space of linear maps from to is isometrically isomorphic to through the inner product, recovering the expression you gave. Similarly the derivative of a map is a map .
• Secondly technically the domain of any derivative like the one above is not the vector space we are working with, but the set of directions at point x. This notion is formalised in Manifold theory and called the tangent space. Thankfully for any finite-dimenional vector space the tangent space at any point is canonically isomorphic to the vector space itself (any vector is a direction, that's what they were invented for). In infinite dimensions this still holds just fine except for the small detail that the notions of manifold and tangent space don't exist there. The same distinction is necessary in the range. So truly, formally, the derivative of a map is a map , with and similarly , with the condition that is simply on the first coordinate. This coincides with the map above: for every we get a linear map
• The above may seem very confusing for , since I claim that the derivative in that case is a map instead of simply a real-valued function. This is resolved by noting that each linear map from to can be represented with a number, similar to the top bullet point above (the inner product on is just multiplication). I think lecturers are quite justified in not exploring the details of this when first introducing derivatives or partial derivatives, but unfortunately in possibly infinite-dimensional abstract vector spaces the distinctions are necessary, if only to avoid type errors.
• In the definition of the partial derivative of M at f with respect to g (so with a range inside a vector space Y) we do not take the norm or absolute value of that expression, it should be the straight up limit . The claim that the limit exists does depend on the topology of Y and therefore on the norm, though.

Also there are a lot of discontinuous linear maps out there. A textbook example is considering the vector space of polynomials interpreted as functions on the closed interval , equipped with supremum norm. The derivative map is not continuous, and you can verify this directly by searching for a sequence of functions that converges to 0 whose image does not converge to 0.

Comment by themajor on A Kernel of Truth: Insights from 'A Friendly Approach to Functional Analysis' · 2020-04-04T08:54:17.816Z · score: 1 (1 votes) · LW · GW

Personally I did the exact opposite, and found that very refreshing. Whenever I ran into a snippet of applied functional analysis without knowing the formal background it just confused me.

Comment by themajor on What are the best online tools for meetups and meetings? · 2020-03-28T22:44:43.130Z · score: 4 (3 votes) · LW · GW

I have a bit of a negative answer, in the sense that I don't know what to recommend. A lot of people in my life (colleagues, friends, family) want to stay in touch (and I with them), but each person seems to have their own preferences about software, or at least sufficiently many subgroups of them do that no single tool will suffice. At this point I honestly think that it's not so much the quality of the tool, but more who are already using it, which determines which software is best. If you and your friends all used to meet on Minecraft every Tuesday, then that is probably the ideal way to keep doing things.

As of yet I'm using Skype, WhatsApp (+WhatsApp Web), personal email, two work emails, two Discord clients (one in browser, one as an app, with separate accounts), my phone, two Slack workspaces, weekly Zoom group meetings, Google Talk, MS Teams and the occasional Jitsi call. This is crazy, but all of them are sufficiently low traffic that I don't really mind.

All of them work fine, I do have mild personal preferences (for Slack, Discord and my phone over all the others) but like I said above it's far more relevant to stay in touch at all than to do it with the right tool.

Comment by themajor on The Hammer and the Dance · 2020-03-22T22:19:20.822Z · score: 2 (2 votes) · LW · GW

I have no idea what you mean, sorry.

Comment by themajor on The Hammer and the Dance · 2020-03-22T10:05:44.432Z · score: 5 (3 votes) · LW · GW

I am quite underwhelmed by this article. The first half neatly summarizes some statistics and plans that have been proposed, and the potential impact of these. I like this a lot, and the strong line (further down in the article)

If you’re a politician and you see that one option is to let hundreds of thousands or millions of people die with a mitigation strategy and the other is to stop the economy for five months before going through the same peak of cases and deaths, these don’t sound like compelling options.

accurately summarizes my feelings on the measures I've seen so far.

However, almost immediately after the first half the article changes tone dramatically, and I feel uncomfortable with most of the remainder. The entire argument hinges on being able to control 'the Dance'. Personally I think there is little to no chance of this working, due to the sheer amount of public cooperation and government coordination required for the measures you suggest. I'm willing to dismiss the examples of South Korea and Singapore as unique circumstances that do not readily apply to a lot of western countries (for example Singapore is tightly confined geographically and implemented their measures very early), an argument which is presented and immediately dismissed in the article without elaboration.

I would love to see a more quantitative analysis of what this dance would look like in a western country. The article states that the impact of contact tracing, travel restrictions, big crowds and plausibly other measures are vastly understated - what is the quantitative impact of these? After the excellent introduction I was expecting an equally high level summary of this new strategy, but instead it was only motivated by analogy and rhetoric.

Comment by themajor on Meetups in the era of COVID-19 · 2020-03-16T08:02:27.025Z · score: 1 (1 votes) · LW · GW

Yes, the bullet point list is for Discord. I only tried Jitsi for an hour or two, but the quality of the call was horrible and the session crashed when a second person joined (so really I could just see my own face, pixellated). When I tried to download the Ubuntu client I got dpkg errors and a failed installation. Maybe Jitsi is actually great(?), but I'm not very tempted to try again without outside help.

Comment by themajor on Meetups in the era of COVID-19 · 2020-03-15T18:07:27.760Z · score: 3 (2 votes) · LW · GW

I've been looking into ways of staying in touch with people while in quarantine, in particular Skype, Discord, Slack and now recently also Jitsi. Personally I have to unfortunately keep using all because of confusing Venn diagrams and people who are not tech savvy, but if you have the option to choose I recommend Discord over all the alternatives:

• I think it works on all these operating systems, I can personally confirm Linux, Windows and Android. I did have some sound quality issues with voice conference calls on Android, but I doubt any competing software does better.
• You all need to download the client to have video calls. You can have group voice calls and text messaging in-browser. [EDIT: for clarification, if some but not all of you have a client, those last few will not be able to send video. I don't know if they can still watch other people's video, they can definitely just join as a voice call.]
• Regardless of the choice above you do need to make an account.
• The quality is very decent. Setting the right sensitivity setting for your own microphone can be a bit finicky (the 'automatic' option is not always great). Other than that I have no complaints.
• If the organiser leaves everybody can continue (in fact, people can spontaneously make more events provided the organiser has set these permissions). The way I've been holding events with Discord there's really more of a 'moderator' (with the permissions to kick people form the server) than an 'organiser', although in practice these are often the same person.
• Yes, there is text chat. Split in channels at your own desire, with option to link to other channels/specific messages and more. You can also have text channels and voice channels open simultaneously, These are open at all times, completely independent of the voice channels.
Comment by themajor on Why would panic during this coronavirus pandemic be a bad thing? · 2020-03-09T08:27:36.300Z · score: 4 (2 votes) · LW · GW
However, my impression is that if P(>10% of the population is infected) is reasonably high — say, >25% — then the cost to the economy would be tremendous, and it would be worth paying a huge cost right now to avoid that possibility.

Are there any explicit approaches you're thinking of that can be taken? Truth be told I don't see how we would realistically stave off this scenario, other than the harsh quarantine measures that worked in China. This is (as far as I can tell) a main part of why so many people here are freaking out - we're headed straight for this scenario and governments are not seeing the smoke. As an example, consider Italy to see the lack of preparedness to take action (closing off a massive region now because it's too late to contain the Corona locally, 366 deaths total so far, leaked documents on containment causing people to move out of containment areas before containment set in).

For those questions you proceed to pose, my thought is that you have to just make your best guess and go with it. Best guesses may not be perfect, but I would expect them to be solid. In other words, none of those questions seem difficult enough where it would stop us in our tracks.

I totally disagree. I think "someone's best guess and go with it" is going to be horribly mismatched with what we actually want from stores and supplies, and will be actively harmful. I don't really know how to explain this in more detail, but I do not think most governments are adequate at the level needed to supply a whole country in an emergency.

I thought the CDC researches and plans for all of these scenarios? Isn't that their entire purpose? (Sorry for the sass; it isn't aimed at you :))

No problem ;). I don't know much about the CDC in particular, but I am currently seeing rather varying responses from health officials globally. This is one of the weaker points in my fears though, maybe our health officials have been preparing for an event like this outbreak for a long time, and have entire flow charts and calling lists and plans ready. I rather doubt it though, considering how (at least for me locally) they've been described as overwhelmed, and it's taking them rather longer to respond than I expected. Also I think most of these health officials have other tasks than specifically targeting novel epidemics, such as fighting the seasonal flu, informing the public and dealing with more 'mundane' but far more commonly occurring disease outbreaks. I would love to be wrong here.

20$/50$ per person for a corona test

I think this is simply naive. First of all, there is (again) a huge difference between one test bought for yourself, and purchasing tests for everybody in a country. My limited experience with healthcare systems and bureaucracy suggests that the cost goes up from the increased scale, instead of down. Plus, there are huge costs you're not including here. People need to administer the test (you think the public can do that individually? Maybe some can, but most? No way), they need to be distributed, and proper action needs to be taken afterwards on positive results. These tests first need to be checked for quality (do you happen to know the rate of type-1 and type-2 errors of that test you mention?), and we need a good way to deal with the millions of false positives that would come out of such a plan. All of this needs to be organised in as transparent and accountable a manner as possible, which is difficult and expensive. I think you're easily looking at costs an order of magnitude above your estimate (so $100B+ for the USA, based on your number), and plausibly way more. Plus, even if this would be economically beneficial, you still need someone to step up (ideally right now, instead of in three weeks) and say "The estimated costs are through the roof, let's take a certain hit of billions of dollars now to prevent a potential loss of way more in the future.". This is a very risky gamble, both from an economic point of view but also unfortunately as a career move. Also keep in mind the eventual costs to the economy are wide-spread, but the costs from triggering such a plan are highly localised. Comment by themajor on Why would panic during this coronavirus pandemic be a bad thing? · 2020-03-08T22:43:16.526Z · score: 4 (3 votes) · LW · GW Do you mind elaborating on that? Of course, although I do run into the problem that to me most of this is self-evident, which makes it hard to motivate. But I'll try to explain some of these claims in more detail. • Declaring a national emergency is a huge cost for a country in general and a government in particular. I think the best way to look at this is that the converse state, "everything is fine please continue operating as normal", is a very profitable and desirable state, and you're destroying that. At the very least this can disrupt economies and production chains, but also public trust (between members of the public, the public and industry, the public and government, the industry and industry, etc.) • Regulating supplies and stores is hard. How do you decide how much goes where, which stores need to stay open and which can close, how much to downsize your public transport and community spaces and services? How much extra money is this all allowed to cost? Below you mention to steve2152 that workers will keep working if you just increase their salary, but who pays for this (which department/ministry/bill in particular)? How high are the salaries supposed to be? How do new findings on the spread of the disease impact each of your answers, what are your tipping points for swapping to a globally different approach? Do you even have people at the right level of the chain of command to suggest these ideas (to be honest I've never heard of governments paying workers extra during national crisis to keep them working, outside of cleanup of nuclear meltdowns). On top of this I think governments have very little experience with epidemics like the one we are facing (globally) today. Like I said I think this is very hard, and if there wasn't any particular reason for my government to sort all of this out beforehand, I expect them to not have this sorted out at all. • (On testing the literal whole country) I don't know about the global situation, but at least where I live I know we're not doing door to door screening (in fact, general practitioners over here are already overburdened right now, with only the paranoid fraction of the population asking for tests. Which, mind you, are currently being denied unless the doctor deems it reasonably likely to give a positive result). I think the idea of "Eventually we'd be able to test the whole country" is an extremely weak link in any public health plan, and simply is too ambitious to work. • Providing food and healthcare packages is also a massive coordination and transportation problem. Out of all the points I mentioned this might actually be doable(?), but it would still require massive resources (the food/medical supplies themselves, trucks, workers distributing them, people filling the packages, and even people planning what should or shouldn't be in these packages, people coordinating internationally?). On top of that this plan incurs all the costs of the 'national emergency'. Plus it's not even clear to me how giving everybody a food and healthcare box will fix the epidemic, it sounds more like a stopgap measure to me (isolation in combination with this would be the real solution, but telling entire regions to self-quarantine is again incredibly expensive). Would it be such a bad thing if it took weeks? I would think that most people have enough food and hygiene products to last weeks. And for those that don't, I would expect most of those people to have family, friends or neighbors who would share. And if that all fails, I would think the government would be able to provide some sort of early before the more official support that would come later. I think the relevant scale to compare this on is not "life or death", but "normal life, just trying to save up for a nice vacation/new car/future expenses". The actual death rate of corona is not that high (I'm avoiding numbers on purpose, because I have no idea about the number of asymptomatic infections), and I expect most people to make it through this just fine on the scale of life to death. But I also expect entire stores and businesses to lose out on major profits, big events to be cancelled, supply chains to be closed off. This is a massive opportunity cost we are paying (in fact, some of my friends are already worried that the corona-recession might get them laid off). In a lot of instances it's even a real cost, not an opportunity cost, because people have already committed resources to society continuing to function as normal. Comment by themajor on Why would panic during this coronavirus pandemic be a bad thing? · 2020-03-08T09:37:32.987Z · score: 10 (4 votes) · LW · GW Several of your bullet points read "X goes wrong, until the government steps in". What makes you think the government is able to put out these fires at the same rate as people run into them? The government also just consists of people. Declaring a national emergency, regulating supplies and stores, testing the literal whole county and providing food and healthcare packages takes time, planning and frankly skill that I'm not sure governments have. That being said, I agree with quite a few of your points. But I think the negative impact of empty grocery stores, people hoarding hygiene products, shops closing because too many employees are staying indoors etc. will be very serious, and that it will take at least weeks before any centralised plan will be able to catch up with this. Comment by themajor on How did the Coronavirus Justified Practical Advice Thread Change Your Behavior, if at All? · 2020-03-06T08:03:03.858Z · score: 5 (3 votes) · LW · GW • I bought copper tape and put it on all door knobs, my phone and the fridge handles in my house. • I bought vitamin D and C supplements. • I stocked up on some food, enough to last me (but not my roommate, who is currently not home but will be back soon) ~10 days. I'll probably make another shopping trip tomorrow to increase that number a bit higher. • I bought hand moisturiser. • I discussed these purchases with friends and family, recommended they do the same, and discussed my expectations for the coming weeks and how to properly respond. I did not buy electrolytes, I found the discussion on ORS, even self-made, versus electrolyte powder not convincing if it's only for a few weeks tops. Comment by themajor on Covid-19: Things I'm Doing Differently · 2020-03-05T14:32:07.032Z · score: 1 (1 votes) · LW · GW Not much to add, just wanted to say thanks for sharing, and I have a lot of overlap with you too (washing hands often, copper on doorknobs/fridge handles/back of phone/other surfaces, not shaking hands in public and more). Comment by themajor on Coronavirus: Justified Practical Advice Thread · 2020-03-02T11:11:45.514Z · score: 3 (2 votes) · LW · GW Where does the 12x come from? The link mentions 1/2 a tablespoon of salt versus 2 tablespoons of sugar, which is a factor 4 in volume. A quick google says the densities only differ by 25% (and in the direction that makes the ratio closer, not further apart), so this is not mass percentage either. EDIT: Never mind, they mention 1/2 a teaspoon of salt. My mistake. Comment by themajor on Continuous Improvement: Insights from 'Topology' · 2020-02-23T00:13:49.497Z · score: 4 (3 votes) · LW · GW Very nice! Two small notes: • The two notions of continuity (sequential continuity and topological continuity) you present under "Multivariate continuity" are not equivalent. In a sense the topology around a point can be 'too large' to recover it from just convergence of sequences (in particular, these notions are equivalent for first countable spaces (I think? Second countability is definitely enough, but I think first countability also is) but not for general topological spaces). You can fix this by replacing the sequences with nets. • The compactifications (one-point and Stone-Cech) are very useful for classification and representation theorems, but personally I've hardly ever used them outside of that context. These compactifications are very deep mathematical results but also a bit niche. I remember back when I took my course on Introduction to Topology that we spent a lot of time introducing homotopies and equivalence classes, and later the fundamental group. And then all that hard work paid off in a matter of minutes when Brouwer's fixed point theorem (on the 2-dimensional disc) was proven with these fundamental groups, which is actually one of the shorter proofs of this theorem if you already have the topological tools available. Comment by themajor on Jan Bloch's Impossible War · 2020-02-18T09:38:11.805Z · score: 4 (2 votes) · LW · GW I like the post a lot! And I also have a silly question: is there a good way to navigate hivewired? Ideally I'd like to browse through the blogs in chronological order, but all I can find for that is the monthly archive pages at the bottom of the site. Unfortunately those contain the full posts in anti-chronological order, making it a chore to find the links to the actual posts. Is there a full list of posts sorted by date somewhere? Comment by themajor on Category Theory Without The Baggage · 2020-02-05T09:44:49.542Z · score: 13 (5 votes) · LW · GW I'll give it the good old college try, but I'm no expert on category theory by any means. I've always been told that a real-world application of category theory is kind of an in-joke; you're not allowed to say they don't exist, but nobody has any. The purpose of category theory is to replace (naive) set theory as a foundation of mathematics. It's therefore not really surprising that this is far removed from describing properties of a pot of water, for example. I think in light of this my previous bullet points weren't that horrible. If you ask something like "I'm walking down the street, suddenly I see a house on fire. How does category theory help me decide what to do next?" the answer is "it does not, not even the slightest bit". Instead it helps on a very abstract meta-level: it is there to structure the thoughts you have about fields of mathematics, which in turn will let you gain faster and deeper insight into those fields. Often (for example in algebra) even this will not translate to real-world applications, but you can then use those fields to better absorb something to make predictions about the real world. As an example I'm thinking of Category theory helping you learn Group theory which helps you learn Renormalization theory which gains you real-world insight into complicated many-particle systems, like a gas out of thermodynamic equilibrium. If you are looking for anything more direct than this I flat out don't have an answer for you. Finally a bit about the bullet points: • The study of non-invertible linear operators on infinite dimensional vector spaces led to notions of spectrum, and later the Gelfand-Naimark representation and study of Fredholm operators, which are some of the core deep ideas behind mathematical formulations of quantum mechanics (amongst other fields). Also the difference between [a linear map failing to be an isomorphism because it is not bijective] (related to the so-called Point spectrum) and [a linear, bijective map that is still not an isomorphism because its inverse map is not a morphism] (related to the so-called Essential spectrum) is of great importance when using numerical simulations to determine behaviour of differential equations. In particular, the point spectrum depends on the implementation and the essential spectrum does not. As far as I know this is a topic of intense debate in simulations of fluid/air flow and turbulence. • This is a fair point, and I don't have an answer for you. I do briefly want to remark that (to my knowledge) all of mathematics, both modern and old, falls under the umbrella of category theory somehow. So the claim "I might be looking for properties that this theory cannot capture" has a high burden of proof. • Personally I run into this regularly - statements like "the topology generated by this norm is equivalent to the product topology from these underlying spaces, therefore we now have the following (universal) properties". On the other hand I don't have a general theorem about universal constructions for you that says something very exciting. I think universal constructions are more about pre-caching knowledge, so that you may immediately use a range of (in your words fairly obvious) results after you verify some commonly occurring conditions. Comment by themajor on Category Theory Without The Baggage · 2020-02-04T14:27:17.001Z · score: 4 (3 votes) · LW · GW Category theory provides a theoretical structure that fits many (all) fields of mathematics. Some non-trivial insights from it are that: • The notion of 'invertible' depends on the context (specifically, it needs to let you recover the identity path, which need not be a set-theoretic identity function. I think this is what Eigil is saying above. As an example, consider measure theory where we introduce 'equal up to sets of measure 0'). • The properties of an object are encoded in the maps we allow to/from the object (as a weak example, a topology induced by a collection of maps. As a strong example: Yoneda's Lemma). • Some properties or constructions are universal in mathematics (direct products, direct limits, inverse limits, initial objects, final objects), and studying these in a general setting provides a lot of insight in actual applications. Also personally I've had great mileage out of noticing when my wild ideas could not be cast in category-theoretic form, which is a huge red flag and let me identify errors in my reasoning quickly and cleanly. Comment by themajor on Rule Thinkers In, Not Out · 2019-12-14T07:55:11.251Z · score: 1 (1 votes) · LW · GW The question is QM is local or not is precisely what is still up for debate. Comment by themajor on A Brief Intro to Domain Theory · 2019-11-21T08:47:26.203Z · score: 1 (1 votes) · LW · GW This looks really interesting! Your special notation for composing functions is making this a lot harder to read for me though, I've read the part up to 'Embeddings and Projections', but I really have to take a break here and set extra time aside to read it at a slower pace from that point. I'm happy the notation works for you but it's really jarring for me. Comment by themajor on How do you assess the quality / reliability of a scientific study? · 2019-10-30T11:49:11.687Z · score: 4 (3 votes) · LW · GW Already partially mentioned by others, including OP. I usually start with comparing the conclusion with my expectations (I'm painfully aware that this creates a confirmation bias, but what else am I supposed to compare it with). If they are sufficiently different I try to imagine how, using the method described by the authors, I would be able to get a positive result to their experiment conditional on my priors being true, i.e. their conclusion being false. This is basically the same as trying to figure out how I would run the experiment and which data would disprove my assumptions, and then seeing if the published results fall in that category. Usually the buck stops there, most published research use methods that are sufficiently flimsy that (again, conditional on my priors), it is very likely the result was a fluke. This approach is pretty much the same as your third bullet point, and also waveman's point number 5. I would like to stress though that it's almost never enough to have a checklist of "common flaws in method sections" (although again, you have to start somewhere). Unfortunately different strengths and types of results in different fields require different methods. A small Bayesian twist on the interpretation of this approach: when you're handed a paper (that doesn't match your expectations), that is evidence of something. I'm specifically looking at the chance that, conditional on my priors being accurate, the paper I'm given is still being published. Comment by themajor on Dual Wielding · 2019-08-28T07:58:39.947Z · score: 3 (3 votes) · LW · GW Most of the problems you mention deal with limited battery life. I have never had any issues with this on my Huawei Nova, and in fact I only charge it every 2 or 3 days with normal use. Would you still recommend dual wielding if battery concerns are not an issue? Comment by themajor on Can we really prevent all warming for less than 10B$ with the mostly side-effect free geoengineering technique of Marine Cloud Brightening? · 2019-08-08T09:51:28.080Z · score: 6 (3 votes) · LW · GW

Is this based on https://www.wired.com/story/airline-emissions-carbon-offsets-travel/ ? I think https://slatestarcodex.com/2019/08/05/links-8-19/#comment-783425 might be relevant - it seems only the increase compared to 2019 is being offset, not the whole flight.

Comment by themajor on Can we really prevent all warming for less than 10B\$ with the mostly side-effect free geoengineering technique of Marine Cloud Brightening? · 2019-08-08T09:48:58.514Z · score: 3 (2 votes) · LW · GW

I think a very significant (probably even dominant) fraction of this geoengineering project would not be the industrial aspect but the organisational and political aspects. Building some ships sounds very doable (although I don't know to what extend "spraying water" and "autonomous" are assembly-line projects, do we already have industries that make ships like this?) , coordinating around letting them sail around and alter the atmosphere less so.

Comment by themajor on Insights from Linear Algebra Done Right · 2019-07-17T11:40:38.020Z · score: 1 (1 votes) · LW · GW

Just to share my two cents on the matter, the distinction between abstract vectors and maps on the one hand, and columns with numbers in them (confusingly also called vectors) and matrices on the other hand, is a central headache for Linear Algebra students across the globe (and by extension also for the lecturers). If the approach this book takes works for you then that's great to hear, but I'm wary of `hacks' like this that only supply a partial view of the distinction. In particular matrix-vector mulitplication is something that's used almost everywhere, if you need several translation steps to make use of this that could be a serious obstacle. Also the base map that limerott mentions is of central importance from a category-theoretic point of view and is essential in certain more advanced fields, for example in differential geometry. I'm therefore not too keen on leaving it out of a Linear Algebra introduction.

Unfortunately I don't really know what to do about this, like I said this topic has always caused major confusion and the trade-off between completeness and conciseness is extremely complicated. But do beware that, based on only my understanding of your post, you might still be missing important insights about the distinction between numerical linear algebra and abstract linear algebra.

Comment by themajor on The Competence Myth · 2019-07-01T21:21:35.826Z · score: 1 (1 votes) · LW · GW

I think 'competent' should in this context mean something like 'has the ability to, after being pointed to a gap in the market, build and/or keep functional a company that fills this gap'. This agrees fully with what you said, there is an extreme lot of wiggle room between 'total retard' and this sense of competent (in fact, I think almost everybody lives in this wiggle room). Furthermore, I think it makes sense to naively think that the abundance of successful companies suggests a lot of people are competent in this sense, whereas I claim this is not the case.

Comment by themajor on The Competence Myth · 2019-07-01T12:16:26.665Z · score: 6 (4 votes) · LW · GW

Very interesting observations! Personally I'd perhaps phrase it the other way around, not 'incompetence is killing corporations' but more something like 'what changed in the past 70 years that allowed people to build long-living corporations back then and not now, assuming today's regular company deaths are caused by incompetence?'. My personal guess is that either back when these long-living companies were founded (~1890's) there was much more low-hanging fruit on the market, allowing less efficient companies to still survive, or alternatively that today's economic environment is much more risk-tolerant so the selection for competence happens much more *after* founding a company.

I agree fully with the government bureaucracy remark, although I suspect there are a ton of other very important effects at work there too (for example, out of all organisations I expect governments in particular to have high accountability and regular run-ins with Chesterton's fence, both of which increase bureaucratic load).

Comment by themajor on The Competence Myth · 2019-07-01T09:22:24.722Z · score: 5 (3 votes) · LW · GW

I personally think we don't need to posit a mechanism that explains why people's wrong beliefs don't cause immediate disaster for companies. In my worldview this is fully explained by selection effects in the market, both at the level of organisations and at the level of individual employees. Since long-term views are very hard to link to individual outcomes, the selection pressure is weaker here.

I'd like to point out that this does suggest that organisations and companies fail and go bankrupt regularly, we just don't hear that much about the quick failures (which I think fits reasonably well with observations, but I haven't looked into this all that much).

This is in fact also an/my answer to the non-rhetorical question why anything works at all. I disagree with Kirkpatrick in attributing this to individuals, which seems to suggest there is some class of millions of managers who have attained some mystical level of competence that somehow doesn't scale to groups.

Comment by themajor on Coherent decisions imply consistent utilities · 2019-05-13T14:14:24.615Z · score: 7 (5 votes) · LW · GW

This is part of the meaning of 'utility'. In real life we often have risk-averse strategies where, for example, 100% chance at 100 dollars is preferred to 50% chance of losing 100 dollars and 50% chance of gaining 350 dollars. But, under the assumption that our risk-averse tendencies satisfy the coherence properties from the post, this simply means that our utility is not linear in dollars. As far as I know this captures most of the situations where risk-aversion comes into play: often you simply cannot tolerate extremely negative outliers, meaning that your expected utility is mostly dominated by some large negative terms, and the best possible action is to minimize the probability that these outcomes occur.

Also there is the following: consider the case where you are repeatedly offered bets of the example you give (B versus C). You know this in advance, and are allowed to redesign your decision theory from scratch (but you cannot change the definition of 'utility' or the bets being offered). What criteria would you use to determine if B is preferable to C? The law of large numbers(/central limit theorem) states that in the long run with probability 1 the option with higher expected value will give you more utilons, and in fact that this number is the only number you need to figure out which option is the better pick in the long run.

The tricky bit is the question whether this also applies to one-shot problems or not. Maybe there are rational strategies that use, say, the aggregate median instead of the expected value, which has the same limit behaviour. My intuition is that this clashes with what we mean with 'probability' - even if this particular problem is a one-off, at least our strategy should generalise to all situations where we talk about probability 1/2, and then the law of large numbers applies again. I also suspect that any agent that uses more information to make this decision than the expected value to decide (in particular, occasionally deliberately chooses the option with lower expected utility) can be cheated out of utilons with clever adversarial selections of offers, but this is just a guess.

Comment by themajor on Tales From the American Medical System · 2019-05-10T10:26:08.916Z · score: 1 (1 votes) · LW · GW

I think your first remark is exactly the point. If the visits are useless then this is a crappy doctor scamming money and time out of patients and insurance companies, if the visits are important then asking OP's friend to come in (for being over 4 months late on a 3-month checkup) sounds very reasonable to me. I think Zyryab's suggestion of asking a doctor to Turing Test this makes a lot of sense - maybe the checkups are more valuable in certain life stages/demographics/early after diagnosis? Maybe the checkup is something more complicated than recording the HbA1c levels? I'm surprised to hear that without outside medical information the doctor is guilty until proven innocent.

Comment by themajor on Tales From the American Medical System · 2019-05-10T10:15:38.823Z · score: 1 (6 votes) · LW · GW

I'm really surprised this is being downvoted so much.

As far as I can tell (and frankly I don't care enough to put serious effort towards finding more information, but I do note nobody in the comments started with "I am a doctor" or "After talking about this with my own doctor, ...") OP's friend was in a life-threatening situation, the solution to which is a renewed insulin prescription. On top of that, the doctor/medical establishment enforces the rule that people (only young people? only people who recently developed diabetes? There could be a good medical reason here, I don't know) with Type I Diabetes have regular checkups.

Now I imagine there are all sorts of reasons for wanting to skip this checkup. Maybe the checkup isn't needed, and is just a money scam (small aside: if my doctor tells me I need a regular checkup, this is not my first thought. But individual situations can vary). Maybe the doctor's schedule is so unreasonable that it's impossible to make an appointment. There could be thousands of valid reasons. The problem as I see it is that, from the point of view of both the doctor and the nurse, they are only negotiating over the checkup. You mention right at the start that the nurse offered a solution ("drop everything and come see your doctor tomorrow") - from that point on the situation was no longer life-threatening! There was no realistic scenario in which this would cost your friend more than the plans they made for the next day! You were just haggling over what is more important, your friend's schedule or the rules set by the medical establishment that you need an active prescription to get insulin and you need a checkup to renew your prescription. Guess which one the nurse is going to find more important.

I understand if it feels like your friend is being blackmailed by the doctor (and in fact it seems like they are), but by refusing to visit the next day you are the ones who escalated the situation. And then escalated even further by threatening with media exposure. I think from the point of view of the nurse your friend is showing rather hostile behaviour. I'll take the liberty of going through the phone call as you posted it, filling in how I expect nurses to act:

The nurse tells my friend he needs to go see his doctor, because it has been seven months, and the doctor feels he should see his doctor every three.

Probably standard procedure. At any rate this decision it out of the nurse's hands, so they are just providing information here.

My friend replies that he agrees he should see his doctor, and he has made an appointment in a few weeks when he has the time to do that.
The nurse says that he can’t get his prescription refilled until he sees the doctor.

Still standard. Nurses don't get to overrule conditions doctors set for medication, if the doctor says a checkup is needed then the nurse has no way of handing over insulin.

My friend explains that he does not have the time to drop what he is doing and see the doctor the next day. That he is happy to see the doctor in a few weeks. But that until then, he requires insulin to live.
The nurse says that he can’t get his prescription refilled until he sees the doctor. That if he wants it earlier he can find another doctor.

Still the same issue. The nurse doesn't have the authority to overrule the conditions set by the doctor. Also I'm missing a sentence here, who introduced talking to the doctor the very next day?

My friend explains again that he does not have the time to see any doctor the next day, nor can one find a doctor on one day’s notice in reasonable fashion. And that he has already made an appointment, and needs insulin to live. And would like to speak with the doctor.
The nurse refuses to get the prescription filled. The nurse does not offer to let him speak to the doctor, and says that he can either wait, make an appointment for the next day, or find a new doctor.

So apparently making an appointment one a one-day notice is very doable on the doctor's side. By this point you are solidly haggling about time, not medicine. I also think the nurse could have let you speak with the doctor here. But I think it's also plausible that they get/did in the past get phone calls from all kinds of entitled weirdos who refuse to show up to appointments, and at this moment it's really not clear your friend is not one of them. Why would their day plans be more important?

My friend points out that without insulin, he will die. He asks if the nurse wants him to die. Or what the nurse suggests he do instead, rather than die.
This seems not to get through to the nurse, because my friend asks these questions several times. The nurse does not offer to refill the prescription, or let my friend talk to the doctor.
My friend says that if the doctor does not give him access to life saving medicine and instead leaves him to die, he will post about it on social media.
The nurse now decides, for the first time in the conversation, that my friend should perhaps talk to his doctor.

Really? Your friend escalates from "I don't want to visit you tomorrow" to "that means you must want me to die", which of course the sensible nurse ignores, and your strategy was to repeat it a few more times? Yeah, you really showed them there. I bet the nurse immediately realised they were wrong the first time, and connected you through with the doctor before you got to the third repetition. From their point of view you've refused a good solution to the problem and are now just bugging them to make your life easier (who likes going to checkups? Nobody. So who haggles about not wanting to show up? Well, not everybody, but more than just your friend I bet). And at that point your strategy is to escalate even more by threatening media exposure, and put even more pressure on that poor nurse? I'm not surprised the doctor claimed you are blackmailing them after this.

What was your goal of the conversation with the nurse in the first place? You need a doctor's prescription for the insulin, so shouldn't you have aimed for talking with the doctor? And if that was your goal, what purpose did it serve to tighten the screws on the nurse? You should have acted like a model patient and calmly requested you speak with the doctor, who can (and did) overrule the normal medical process just to give you life-saving medicine.

I guess that became a far longer monologue than I planned, I'm not going to go through the phone call with the doctor because it's just more of the same. I think OP is in the wrong here, at the very least in their interaction with the nurse. And I do agree that this is a bad medical system, but you really can't throw the co-pay costs, the lack of automatic prescription extensions/sufficiently large prescriptions to last you a long time and your interaction with the nurse and doctor on one heap and pretend this is the fault of "the American medical system". The overall structure sucks, but some of these people are just local actors who cannot make a change and your friend threatened them to not have to change their schedule.

Comment by themajor on Best reasons for pessimism about impact of impact measures? · 2019-05-03T16:54:19.675Z · score: 13 (5 votes) · LW · GW

I have a bit of time on my hands, so I thought I might try to answer some of your questions. Of course I can't speak for TurnTrout, and there's a decent chance that I'm confused about some of the things here. But here is how I think about AUP and the points raised in this chain:

• "AUP is not about the state" - I'm going to take a step back, and pretend we have an agent working with AUP reasoning. We've specified an arcane set of utility functions (based on air molecule positions, well-defined human happiness, continued existence, whatever fits in the mathematical framework). Next we have an action A available, and would like to compute the impact of that action. To do this our agent would compare how well it would be able to optimize each of those arcane utility functions in the world where A was taken, versus how well it would be able to optimize these utility functions in the world where the rest action was taken instead. This is "not about state" in the sense that the impact is determined by the change in the ability for the agent to optimize these arcane utilities, not by the change in the world state. In the particular case where the utility function is specified all the way down to sensory inputs (as opposed to elements of the world around us, which have to be interpreted by the agent first) this doesn't explicitly refer to the world around us at all (although of course implicitly the actions and sensory inputs of the agent are part of the world)! The thing being measured is the change in ability to optimize future observations, where what is a 'good' observation is defined by our arcane set of utility functions.
• "overfitting the environment" - I'm not too sure about this one, but I'll have a crack at it. I think this should be interpreted as follows: if we give a powerful agent a utility function that doesn't agree perfectly with human happiness, then the wrong thing is being optimized. The agent will shape the world around us to what is best according to the utility function, and this is bad. It would be a lot better (but still less than perfect) if we had some way of forcing this agent to obey general rules of simplicity. The idea here is that our bad proxy utility function is at least somewhat correlated with actual human happiness under everyday circumstances, so as long as we don't suddenly introduce a massively powerful agent optimizing something weird (oops) to massively change our lives we should be fine. So if we can give our agent a limited 'budget' - in the case of fitting a curve to a dataset this would be akin to the number of free parameters - then at least things won't go horribly wrong, plus we expect these simpler actions to have less unintended side-effects outside the domain we're interested in. I think this is what is meant, although I don't really like the terminology "overfitting the environment".
• "The long arms of opportunity cost and instrumental convergence" - this point is actually very interesting. In the first bullet point I tried to explain a little bit about how AUP doesn't directly depend on the world state (it depends on the agent's observations, but without an ontology that doesn't really tell you much about the world), instead all its gears are part of the agent itself. This is really weird. But it also lets us sidestep the issue of human value learning - if you don't directly involve the world in your impact measure, you don't need to understand the world for it to work. The real question is this one: "how could this impact measure possibly resemble anything like 'impact' as it is intuitively understood, when it doesn't involve the world around us?" The answer: "The long arms of opportunity cost and instrumental convergence". Keep in mind we're defining impact as change in the ability to optimize future observations. So the point is as follows: you can pick any absurd utility function you want, and any absurd possible action, and odds are this is going to result in some amount of attainable utility change compared to taking the null action. In particular, precisely those actions that massively change your ability to make big changes to the real world will have a big impact even on arbitrary utility functions! This sentence is so key I'm just going to repeat it with more emphasis: the actions that massively change your ability to make big changes in the world - i.e. massive decreases of power (like shutting down) but also massive increases in power - have big opportunity costs/benefits compared to the null action for a very wide range of utility functions. So these get assigned very high impact, even if the utility function set we use is utter hokuspokus! Now this is precisely instrumental convergence, i.e. the claim that for many different utility functions the first steps of optimizing them involves "make sure you have sufficient power to enforce your actions to optimize your utility function". So this gives us some hope that TurnTrout's impact measure will correspond to intuitive measures of impact even if the utility functions involved in the definition are not at all like human values (or even like a sensible category in the real world at all)!
• "Wirehead a utility function" - this is the same as optimizing a utility function, although there is an important point to be made here. Since our agent doesn't have a world-model (or at least, shouldn't need one for a minimal working example), it is plausible the agent can optimize a utility function by hijacking its own input stream, or something of the sorts. This means that its attainable utility is at least partially determined by the agent's ability to 'wirehead' to a situation where taking the rest action for all future timesteps will produce a sequence of observations that maximizes this specific utility function, which if I'm not mistaken is pretty much spot on the classical definition of wireheading.
• "Cut out the middleman" - this is similar to the first bullet point. By defining the impact of an action as our change in the ability to optimize future observations, we don't need to make reference to world-states at all. This means that questions like "how different are two given world-states?" or "how much do we care about the difference between two two world-states?" or even "can we (almost) undo our previous action, or did we lose something valuable along the way?" are orthogonal to the construction of this impact measure. It is only when we add in an ontology and start interpreting the agent's observations as world-states that these questions come back. In this sense this impact measure is completely different from RR: I started to write exactly how this was the case, but I think TurnTrout's explanation is better than anything I can cook up. So just ctrl+F "I tried to nip this confusion in the bud." and read down a bit.
Comment by themajor on 1960: The Year The Singularity Was Cancelled · 2019-04-23T12:50:19.211Z · score: 3 (2 votes) · LW · GW

Comment by themajor on 1960: The Year The Singularity Was Cancelled · 2019-04-23T09:14:10.958Z · score: 2 (2 votes) · LW · GW

I think the evidence presented is way too weak to support the type of conclusions drawn in this piece. I mean really, we're computing doubling times by taking logarithms of estimated GDP, inserting an arbitrary offset in our definition of the horizontal axis and then plotting THAT on a log-log scale? What were you expecting to find?

More specifically: the horizontal labels of the most recent data points are heavily influenced by the particular choice of 2020 offset. I've taken the liberty of repeating (I hope) Scott's analysis with the data from the paper, and swapping the offset to 2050 or even 2100 bunches the last data points a lot closer together, allowing a linear fit to pretty much pass through them. I think some argument can be made that we need a higher time resolution in an era with a doubling time of ~20 years compared to an era with a doubling time of ~500 years, but I'm still not happy with how sensitive this analysis is and would love to hear why 2020 is a better choice than 2100.

Also I notice that Scott left a bunch of data points from the paper out of the graph. I can live with excluding the really early ones (before 10000 B.C.), but why do you skip over the ones near 0 A.D.? The 1100-1200's? And where are the data points with negative doubling times (i.e. declining GDP)? Maybe I missed it but I don't see mention of these at all.

Comment by themajor on Rule Thinkers In, Not Out · 2019-02-28T23:24:11.376Z · score: 8 (5 votes) · LW · GW

Yes, I think you're right. Personally I think this is where the charitable reading comes in. I'm not aware of Einstein specifically stating that there have to be hidden variables in QM, only that he explicitly disagreed with the nonlocality (in the sense of general relativity) of Copenhagen. In the absence of experimental proof that hidden variables is wrong (through the EPR experiments) I think hidden variables was the main contender for a "local QM", but all the arguments I can find Einstein supporting are more general/philosophical than this. In my opinion most of these criticisms still apply to the Copenhagen Interpretation as we understand it today, but instead of supporting hidden variables they now support [all modern local QM interpretations] instead.

Or more abstractly: Einstein backed a category of theories, and the main contender of that category has been solidly busted (ongoing debate about hidden variables blah blah blah I disagree). But even today I think other theories in that pool still come ahead of Copenhagen in likelihood, so his support of the category as a whole is justified.

Comment by themajor on Rule Thinkers In, Not Out · 2019-02-28T15:27:55.422Z · score: 11 (4 votes) · LW · GW

I feel like I'm walking into a trap, but here we go anyway.

Einstein disagreed with some very specific parts of QM (or "QM as it was understood at the time"), but also embraced large parts of it. Furthermore, on the parts Einstein disagreed with there is still to this day ongoing confusion/disagreement/lack of consensus (or, if you ask me, plain mistakes being made) among physicists. Discussing interpretations of QM in general and Einstein's role in them in particular would take way too long but let me just offer that, despite popular media exaggerations, with minimal charitable reading it is not clear that he was wrong about QM.

I know far less about Einstein's work on a unified field theory, but if we're willing to treat absence of evidence as evidence of absence here then that is a fair mark against his record.

Comment by themajor on So You Want to Colonize The Universe · 2019-02-28T15:14:53.910Z · score: 0 (3 votes) · LW · GW

I think this is an interesting idea, but doesn't really intersect with the main post. The marginal benefits of reaching a galaxy earlier are very very huge. This means that if we are ever in the situation where we have some probes flying away, and we have the option right now to build faster ones that can catch up, then this makes the old probes completely obsolete even if we give the new ones identical instructions. The (sunk) cost of the old probes/extra new probes is insignificant compared to the gain from earlier arrival. So I think your strategy is dominated by not sending probes that you feel you can catch up with later.

Comment by themajor on One Website To Rule Them All? · 2019-01-24T08:48:00.234Z · score: 1 (1 votes) · LW · GW

Well, I still don't have any experience with this. But maybe possible avenues include:

• Looking into moderation rules.
• Including some kind of reputation/point/reward system, and other methods to keep your users engaged.
• Tracking metrics on the growth of the Site, and ideally having some advance expectations/plans on how to respond to different rates of growth/decline.
• A more radical approach might be to give up the phase 2 and beyond in their entirety, and settle for a target audience of people close enough to you that you can reasonably trust them.

The survivorship bias is a very valid point, but [not doing research on how to make websites grow] is also a poor strategy. Personally I'd still look into the advice, but I'm afraid what you're trying to do is simply very difficult.