For "what kind of building do you live in," I'm not sure any option reflects my living situation? I'm a college student, and I spend ~40 weeks/year in various sorts of dorms with roommates and 50+ people in the building, sharing a common kitchen and lounge/living spaces (not really a "large apartment building") and ~12 weeks/year in a single-family house with my parents. I have therefore left the question blank.
The building question seemed to focus on the building style, but my choice is largely based on more practical considerations, especially regarding roofs. I wouldn't choose a building with much in the way of flat roofs, because these inevitably spring leaks before long, and are expensive to fix. Conversely, I don't like houses with steep sloping roofs, "all roof" so to speak, such as Dutch roofs, as this means upstairs rooms all have sloping walls, which is inconvenient and a massive waste of space. (An exception to my "no flat roofs" rule is a railed off balcony.)
He separated “apartment” from “townhouse/condo/etc”, which seems to be more about ownership structure than style. It doesn’t distinguish whether you have upstairs and downstairs neighbors, or only neighbors attached on the sides, but does (largely) distinguish whether you own the unit or are renting it.
“Condo” is an ownership structure. It means ownership of one unit in a larger structure, usually a single building divided into apartments but sometimes a bunch of townhomes with some shared spaces or even single family homes with shared spaces that are co-owned.
A ten story building with four apartments on each floor can either be structured as a condo, where residents own each unit individually (with shared control of things like elevators and the building frame) or as what we usually refer to as an “apartment building”, where the building as a whole is owned by a single company, that rents out individual units to people.
The way Scott wrote things suggests different answers in those two cases, even though the building looks identical in those two cases.
Yes. Technically, I rent a condo in a 10 unit condo building, but I checked the < 10 unit apartment building box because I think that's what makes sense to check. Probably some split about # of shared walls/neighbors/whether you share an entrance/stairs would make more sense?
Agreed, practical considerations are important. I would be reluctant to buy a 2-storey house, regardless of its charm, because that would rule out me doing roof inspection and cleaning eavestroughs.
Agreed on flat roofs - water always finds a way. (It's moot for me; I don't like the look of most "architect" houses anyway.)
Similar problem but I went with my legal address where I can vote. I own a house in the surburbs of a big city. I'm currently working and living for a bit (probably a year) out in the unincorporated country up a dirt road.
Yeah, I live in a multi-family group house and the "group house" option specifically excludes living with family. Really couldn't figure out what to do.
Yeah I went with the one that didn't have the stupid lawn that would take forever to maintain. But like that's probably not what the question was about
I was sad to see the loss of the "Been here since it was a LiveJournal" option, but Catmint makes a good point that that data is probably inferrable from old surveys. Unless there's some weird unexpected pattern in which particular ages of old readers have/haven't stuck around this long.
Me! I found Less Wrong while looking into that whole LK-99 room temperature superconductor thing a while back (July 2023), and found Scott's writing there.
This is what happened to me too. I found specific SSC posts that interested me, eventually browsed through SSC, discovered the whole ACX stuff, and now read every ACX post. As such, I selected for only the best content while on SSC, so the quality did go down, but only because the nature of my reading changed. I ended up just saying I learned about it from a blog (got into rationalism through Gwern) and didn't do anything for quality.
Not to dunk, but I used to love intuiting the hypotheses you were going for when filling out these surveys, and I'm extremely disappointed this one was themed around criminality and the homeless, with a side of architecture (which is also politicized now!). The shift from a blog about science and curiosity to one about politics is basically completed now.
Yes, you're right, but most of the time it was but one of the topics Scott was curious and explorative about (hence his posts about his fictional country), not hot topics (again, most of the time). And yeah, I'm glad he's at least doing the research, but I'm still annoyed overall.
I am quite curious about criminal science, social science (homelessness), the future of architecture and many other things that also happen to be influenced by "politics". Hot or not - and "the poor will be always there with you", said some rabby 2000 years ago. The most popular post is still ivermectin - some may say that was/is a "hot political topic", too. I can do without news from Raikoth.
Yes. I worry that when I indicate that I’m “interested” in politics, many people assume that means I’m interested in party politics and individuals and Go Team! Whereas I am interested in political issues. I generally trust that Scott means the same.
I preferred SSC because Scott had less to lose at that time so he was more freewheeling and open, but the issues of homelessness and societal breakdown are pretty important ones.
If the link I saw (and mentioned in the immediate aftermath of Luigi's capture) was accurate and Luigi actually read ACX I understand completely why Scott didn't include questions grading his readers' sympathy for him (or related questions) but back in the SSC days I think he would have. Here we have an actual terrorist attack and a solid percent of Americans (and non-Americans) support him. That seems like a bigger issue to look at than whether people should be penalized for using the 10 items or less aisle.
The police haven't released much of what they have but they did release included a quote (which 3 minutes of search failed to find - of course - so here's the paraphrase) 'August 15th: Things are coming together, the target is insurance. It checks all the boxes..."
He's a freedom fighter or terrorist or revolutionary of a serious sort. No doubt he thinks the justice system is corrupt, and doesn't think very highly of either the president or the president-elect but he thought his actions through quite seriously. More seriously than he thought about his own escape plan.
He's not some guy who killed a rapist or a corrupt cop or a POS lobbyist or anyone of any fame whatsoever. He thought this shit through and he was right. He's got people deep in the culture wars rethinking who is actually waging war against them.
If Americans were literally any other people (save Canadians or Australians) the reaction would be On The Streets -- that's how much support he has.
Back in the SSC days I think Scott would have jumped into the fray. I don't hold his more "professional" and careful approach against him at all. He is incredibly brave, bold, and honest. One can not be all things to all people. I think this blog is great, but I hear where you're coming from.
Science and curiosity have long been political, and no one can control what things they care deeply about that also make other people in their society mad, which is what it really means for something to be political. I think that no one at all, not Scott Alexander nor anyone else, should have any qualms about blogging about things they care about that have political implications, out of some misguided feeling that there is something wrong with being about politics.
"In percent, what do you think is the profit margin of the average American company?"
This could do with more precision. Gross or net profit margin? Average how? Weighted by revenue or just a simple arithmetic mean? Publicly traded companies, or all businesses?
Many different reasonable ways this could be reasonably answered.
Another confused person here. Many companies with high gross margins have tiny net margins or make losses after depreciation and taxes. I answered as though the question was asking about net margins.
median profit margin in 2015 was xy% - for whatever that means. While "the public" seems to assume yx% - for whatever that means. "The Public Thinks the Average Company Makes a Profit Margin, Which Is About ... "
Speaking from total eco-science ignorance, this was very confusing in the “average company” aspect for me: I have no idea, so I’d like to imagine an “average company”, but… what does that mean? Sum all profit and divide by the number of companies? Sort the companies by their profit margins and choose the median one? Etc. Basically, this seems (to me, a total know-nothing) as unspecific as “average country” (what is the [any random economic measure like GDP growth% or unemployment rate] of an average country?).
I also struggled with this question, because I don't know what company is average. I know tech companies have significantly higher profit margins than many traditional companies, but I don't know what percent of companies fall under tech. I know a lot of standard companies like grocery stores have very low profit margins, but again don't know what percent of companies this encompasses.
I don't feel like my answer to this question is very useful or meaningful. Even to me. I suspect he doesn't care what the actual answer is, so much as perception, but if even I don't trust my own answer at all, I'm not sure how useful that is.
I suspected this question was kind of along the same lines as "when did you last think about the Roman empire?" It may be compared to an actual answer, and then you can see what your answer says about you.
I think net profit is the figure that captures the obvious reasonable interpretation of the question.
If a company is losing money, I expect its profit to be negative.
(I don't have strong views on the other questions, but don't think what you choose to answer to them should have all that big an impact on the answer).
The standard here is EBITDA. I’m quite confident Scott will end up using this.
What I’m less confident about how this would be weighted. Median profit margin of all listed stocks? Market cap weighted profit margin? Also, for the current year? Smoothed over the last 3 years? The business cycle and interest rates make this very noisy.
The answer was higher than I expected, when I looked it up (the search results appeared to be market cap weighted, but I’m not sure), but I’m predicting most people anchor on tech profit margins and will be surprised how low any reasonable “average” is.
Limiting it to publicly trade companies seems like much too harsh of a limit. There are on the order of tens of millions of businesses in the US, and only on the order of tens of thousands of publicly traded companies.
If we are going to go down this route (and I think your questions are reasonable!) then we also want to be clear on whether money LOSING companies are included in the average.
A negative business net profit margin is possible if we add in all the money losing businesses. Certainly in an industry (hello, restaurant industry!).
*) Net vs Gross matters
*) Including money losing businesses matters
*) Privately held (e.g. large law firms, family run restaurants) or not matters
I answered with my best guess for net profit for 'all' businesses. But many other interpretations are possible.
Especially with the US market being so heavily weighted towards the biggest tech companies (which are often unusually high margin) the number is radically different depending on whether it's listed, weighted by size, includes every incorporated entity (in which case the number will be very low). I gave a sort of generic number but the question is, I think, fundamentally broken and can't be used for analysis.
Yeah, I really wasn't sure about the weighting of various companies. Like I just figured many companies are running at a loss right now, and if I think of the set of companies relevant to be "a company that is started on America", then there are so many negative values in that average. But then... does Microsoft get weighted 10 million times as much as some guys food truck? Or is that just two companies whose margin percentages get averaged?
Is it the average profit margin of an American company? Averages of ratios are fairly meaningless, and companies of vastly different scales are a concern. Median profit margin quantified across American LLCs might make sense, but 90% of those are going to employ <10 people, and work across vastly different industries (net profit in legal companies is going to be quite a bit higher than retail).
I'm actually a bit confused why anyone would interpret it to mean anything other than net margin. That's the portion of total revenue that stockholders get to take home (or reinvest in the company). That's what the average "man on the street" thinks when he thinks profit margin. Why would internal accounting metrics like gross margin matter?
I think the HBD question is ambiguous in an actually-confusing way. e.g. if someone believes that such differences are an unfortunate inequality that must be rectified, are they supposed to answer "favorable" (since they believe that the differences exist), or "unfavorable" (since they're against this state of affairs)?
Agree. It is not at all clear whether favorable means 'believes there is some scientific evidence for it' or 'believes it is a good thing' or 'believes it should be used to inform policy' or 'opposes all mention of sex/race in laws or policies.'
I read that question as clarified, "eg the belief that races differ genetically in socially relevant ways?". You're favorable to the idea if you believe it's true, independent of if you believe it should be true, or if you believe that, if true, it should be corrected.
Yeah I agree that's a reasonable interpretation, but it's not clear. If someone asks me whether I'm "favorable to the idea of eugenics", I'm pretty sure they're asking whether I think it's a good idea, not whether I think it's possible.
I am pretty certain that, as expressed in the survey, "the belief that races differ genetically in socially relevant ways", is pretty much an inevitable consequence.
Supposing the concept of "race" more or less points to a cluster of factors that are associated with each other (and thus recognizable as a cluster) and those factors being genetically determined (or at least, so much influenced by genetics as to be determined, in practical terms, let's say 80%/20% nature/nurture), and those factors being hereditary, thus (unless mixed with other races) they persist across generations. Unless you deny some or all of the preceding, you must admit to the existence of "race" as a concept, if a fuzzy one.
Then, supposing there are at least 2 different human groups, such as they can be distinguished by some of those clusters of factors, it would be strange if they, as groups, didn't have different means, distributions, etc. of values across several possible measures. E.g. height. Not all measures must differ, of course. Or be correlated. But some of them will, or else you don't really have different groups (or you're not correctly carving reality at the joints, see e.g. bleggs - https://www.lesswrong.com/posts/4FcxgdvdQP45D6Skg/disguised-queries). So, after accepting there are some measures that differ, it's not a stretch to postulate some of them are "socially relevant", since society measures us all in a dizzying amount of ways. Even if it's on the extremes of the outliers. E.g. Kenyan athletes are over-represented among Olympians in long distance running.
What you choose to do with those facts, of course, can be unfortunate. But I have problems to understand why is the fact itself unfortunate. Or rather, I understand that considering the mere fact "unfortunate" is a strong social signifier, but apart from social signalling, I don't really see why. That's part of the problem of making the definition almost maximally vague.
You might want to equalize outcomes. But even supposing you could (I don't mean it in terms of political capital, but as results of individual persons, all of this being the realm of averages and groups and standard deviations), the optimal amount of effort to equalize requires at least a reasonably close estimation of the types of differences and the amount of the differences. That's the strangest part. Seeing people who would like a certain result, but not able or willing to (even with great distaste) do the needed calculations to reach it. And I, for one, would really like to see it all out in the open. Anything else is just bickering about how to indirectly affect the reality without ever reaching a consensus of what even *is* the reality to be affected, much less when will we reach the desired result.
Race is just the phenotypically recognizable evidence of ancestry. Abilities and disabilities tend to cluster and persist for a while in families. Big deal. Nothing is permanent.
I think we may assume he means genetic relations; your third stepfather's cousin's second husband may be nutty as a fruitcake but that has no bearing on your mental health or propensity to develop mental problems.
Biological parents (if you know 'em), biological siblings (ditto), grandparents, aunts/uncles, cousins - any of which developed problems or had them from the start. That gives the hereditary background from which you can say "yeah, there is X in my family and there were concerns that I would pass it on to my kids" or "developed Y but nobody else has it".
What about adopted children? I am a former foster parent who adopted 3 of my foster children, and needless to say, my answers would be very different if I were to include them.
I always answered genetically. I don't think he cares who my brother-in-law is or what conditions he may have, related to the rest of my answers on the survey. The lack of specifics (two generations with no details) and no details on number of people or severity makes it almost impossible that he would try to use it as a general survey of frequency or anything else other than genetics.
I sort of remember he commented in a previous year that he meant for it to be a genetic connection, but I might be conflating that with how I decided to answer it after reading the comments.
It would be nice to have that clarified on the survey. Though, since there's no distinction between people who have X and also have family with X vs people who have X without family who have X, I think we can safely interpret the family option as a low-but-nonzero probability that I have X without realizing it.
I wish it had clarified "big city" or maybe provided a an option for "Midsize city". I'm in St Pete FL, which is certainly no Chicago or LA, is also not a suburb, but certainly isn't a small town and it's in the midst of a huge metro area with suburbs of its own. Most places I've lived (Ann Arbor MI, Akron OH, Ft Lauderdale FL also) have been in this size range. Only lived in one big city (Baltimore-- I think it still qualifies although it's definitely the Poor Relation of east coast cities) and did suburbia briefly after college and spent an unhappy year and a half in a semi-rural but summer resort area.
I think that if it counts as a big city in its area (the most populous or important) then the literal size doesn't matter as much - no it's not going to be as big as New York city but it's a "big city" for Florida and that's the important distinction there.
Seems like a small city. Above a million is my personal cutoff.
Edit: I think of city size in a global context, and China has completely changed the context of cities over the last few decades. Population >1m used to be a very large city within living memory.
IMO, a small city would be like my home town of Ypsilanti MI (20K people, and another 30K in the township), or Salisbury MD or Jamestown NY. The sort of place you can comfortably bicycle across.
It depends also on the country; a city of a million is a big city if the entire population is around five million, but by comparison with a country population of fifty million, it's a small city or even a large town.
OECD doesn't say cities as such but rather urban areas:
I don't see a detailed explanation but it looks like OECD uses commuting patterns as part of their definition. For economic purposes that makes a lot of sense, but because the question in the survey specifically includes "suburb" as an option it feels skewed against this type of definition -- a suburb is basically just "a lower density area that functions economically as subsidiary housing for a nearby city" and so I would expect most suburbs to be part of the larger OECD metro by definition.
I still don't understand what Americans mean by the term "suburb". Like, is Beverly Hills a suburb or not? Is Sunset a suburb of San Franciso? Is Flushing a suburb of New York?
Generally we mean a region of predominately non-dense residential housing (not apartments) near but not in a city: Beverly Hills is a suburb, but Flushing is not.
If 50000 is a large city, what larger size categories do you use for cities in the millions and tens of millions? Because putting Tokyo (~37 million) in the same category as Buffalo NY seems...questionable.
Man, that is not at all what I would consider a large city. I live in a suburb that has 150,000 people in a metro area with 1.3 million people, and I chose the small city option.
The suburb I live in is touching the main city. I'm a ten minute drive from the Central Business District. That main city in my region though is less than 500,000 people, so I couldn't consider it a large city. In reality, I would call it a mid-sized city, but that wasn't an option.
Similarly if I live in a small city that's a suburb of a bigger city, should I have picked small city or suburb? Is the important distinction whether it's a self-contained community or an adjunct of a bigger city, or are we asking about single-family sprawl vs apartments and businesses and dense public transit?
Given that suburb is an option I would have gone with suburb, but there are definitely edge cases. What is Fort Worth in the DFW area? If the "suburb" is a 40 minute drive through countryside but is legally part of the metro?
As always, building discrete classifications for something on a continuous spectrum is hard.
My best guess is that the intent of the question is more along the lines of "how densely built/populated is it" which is why I chose small city but yeah it's pretty unclear.
And don't forget e.g. Fairfax County Virginia, which is almost fully urbanized, but in which the vast majority of the population lives in unincorporated ares (there are only 3 towns in the whole county, and cities are independent of counties in Virginia).
I wasn't sure what to answer for the kidney donation question. I ended up with "I haven't started the process," but in reality I started researching it, and I even emailed a couple of organizations, but kidney donation to a stranger is illegal where I live. So maybe I was rejected? But I don't think that's what you were getting at with that option, either.
I would have liked a "I'm open to it, but haven't done anything yet to start the process." I'm not against it, but had to answer Not Interested because the other answers were more incorrect.
As my long-term monogamous partner and I live together and have a child together but aren't legally or religiously married and don't plan to (which is not a particularly rare arrangement in Europe nowadays), I'm not really sure how to answer the "Married people:", "Single people:" and "People in a relationship other than marriage:" questions -- I answered the first two by interpreting "future spouse" broader than literally, skipped the following two, answered the following one by interpreting "marriage" broader than literally, and skipped the last two.
If two people are cohabiting, share finances (and benefits), are sleeping together and intend the situation to endure then they are de facto married. Until SSM became legal same sex couples generally thought of it that way and would refer to their partner as "husband" or "wife".
For "what kind of building do you live in," I'm not sure any option reflects my living situation? I'm a college student, and I spend ~40 weeks/year in various sorts of dorms with roommates and 50+ people in the building, sharing a common kitchen and lounge/living spaces (not really a "large apartment building") and ~12 weeks/year in a single-family house with my parents. I have therefore left the question blank.
The building question seemed to focus on the building style, but my choice is largely based on more practical considerations, especially regarding roofs. I wouldn't choose a building with much in the way of flat roofs, because these inevitably spring leaks before long, and are expensive to fix. Conversely, I don't like houses with steep sloping roofs, "all roof" so to speak, such as Dutch roofs, as this means upstairs rooms all have sloping walls, which is inconvenient and a massive waste of space. (An exception to my "no flat roofs" rule is a railed off balcony.)
He separated “apartment” from “townhouse/condo/etc”, which seems to be more about ownership structure than style. It doesn’t distinguish whether you have upstairs and downstairs neighbors, or only neighbors attached on the sides, but does (largely) distinguish whether you own the unit or are renting it.
He never asked about ownership and none of the options reflect ownership at all, plenty of people rent townhouses/condos/etc.
He also distinguished between smaller apartment buildings and larger apartment buildings, so I'm pretty sure he was focused on the types of buildings.
“Condo” is an ownership structure. It means ownership of one unit in a larger structure, usually a single building divided into apartments but sometimes a bunch of townhomes with some shared spaces or even single family homes with shared spaces that are co-owned.
A ten story building with four apartments on each floor can either be structured as a condo, where residents own each unit individually (with shared control of things like elevators and the building frame) or as what we usually refer to as an “apartment building”, where the building as a whole is owned by a single company, that rents out individual units to people.
The way Scott wrote things suggests different answers in those two cases, even though the building looks identical in those two cases.
I stand corrected, thanks for the explanation.
Yes. Technically, I rent a condo in a 10 unit condo building, but I checked the < 10 unit apartment building box because I think that's what makes sense to check. Probably some split about # of shared walls/neighbors/whether you share an entrance/stairs would make more sense?
Agreed, practical considerations are important. I would be reluctant to buy a 2-storey house, regardless of its charm, because that would rule out me doing roof inspection and cleaning eavestroughs.
Agreed on flat roofs - water always finds a way. (It's moot for me; I don't like the look of most "architect" houses anyway.)
Sounds like the nearest thing to your situation would be "group living" as that's the majority of your time (in student accommodation).
I also split my time between a dorm and my parents.
Similar problem but I went with my legal address where I can vote. I own a house in the surburbs of a big city. I'm currently working and living for a bit (probably a year) out in the unincorporated country up a dirt road.
Yeah, I live in a multi-family group house and the "group house" option specifically excludes living with family. Really couldn't figure out what to do.
Being in a semi and being relegated to the condo/townhouse/whatever category hurt
Yeah I went with the one that didn't have the stupid lawn that would take forever to maintain. But like that's probably not what the question was about
It saddens me a bit that LessWrong isn't even an option for 'How did you find ACX?' anymore.
Sonnet 12
When I do count the clock that tells the time
And see the brave day sunk in hideous night,
When I behold the violet past prime
And sable curls all silvered o’er with white;
When lofty trees I see barren of leaves,
Which erst from heat did canopy the herd,
And summer’s green all girded up in sheaves
Borne on the bier with white and bristly beard;
Then of thy beauty do I question make
That thou among the wastes of time must go,
Since sweets and beauties do themselves forsake
And die as fast as they see others grow;
And nothing ’gainst Time’s scythe can make defense
Save breed, to brave him when he takes thee hence.
How many people weren't around for Slate Star Codex but have found ACX through Less Wrong? Can't be many?
Really, I also don't like that all of us from before ACX are lumped together and don't matter in these questions anymore.
Presumably he still has our answers from previous surveys.
I was sad to see the loss of the "Been here since it was a LiveJournal" option, but Catmint makes a good point that that data is probably inferrable from old surveys. Unless there's some weird unexpected pattern in which particular ages of old readers have/haven't stuck around this long.
I came in through his tabletop roleplaying games, I don't think I've ever had a survey question ask about that XD
Me! I found Less Wrong while looking into that whole LK-99 room temperature superconductor thing a while back (July 2023), and found Scott's writing there.
I got into Lesswrong just a few months ago (but learned of it 1.3 years before that). Then I went
Lesswrong -> SSC -> ACX
It's good to hear that while the old pipeline might be rusty it is still functional.
This is what happened to me too. I found specific SSC posts that interested me, eventually browsed through SSC, discovered the whole ACX stuff, and now read every ACX post. As such, I selected for only the best content while on SSC, so the quality did go down, but only because the nature of my reading changed. I ended up just saying I learned about it from a blog (got into rationalism through Gwern) and didn't do anything for quality.
I did. Learned about Less Wrong while going down an AI rabbit hole, stumbled upon Meditations on Moloch, read through most of Scott's old posts.
Not to dunk, but I used to love intuiting the hypotheses you were going for when filling out these surveys, and I'm extremely disappointed this one was themed around criminality and the homeless, with a side of architecture (which is also politicized now!). The shift from a blog about science and curiosity to one about politics is basically completed now.
Shift? It's always been a blog "about" politics. Plenty of anti-SJ stuff back in 2014 and 2015. The anti-reactionary FAQ is from 2013!
Also, while criminality and homelessness are politicised, they're also somewhat scientific, and I think Scott explores that scientific angle.
Yes, you're right, but most of the time it was but one of the topics Scott was curious and explorative about (hence his posts about his fictional country), not hot topics (again, most of the time). And yeah, I'm glad he's at least doing the research, but I'm still annoyed overall.
I am quite curious about criminal science, social science (homelessness), the future of architecture and many other things that also happen to be influenced by "politics". Hot or not - and "the poor will be always there with you", said some rabby 2000 years ago. The most popular post is still ivermectin - some may say that was/is a "hot political topic", too. I can do without news from Raikoth.
Yes. I worry that when I indicate that I’m “interested” in politics, many people assume that means I’m interested in party politics and individuals and Go Team! Whereas I am interested in political issues. I generally trust that Scott means the same.
Human psychology, health, genetics, and behavior isn't science? Haven't the surveys always been about that?
I'm looking forward to finding out how architectural tastes are correlated with pectus excavatum.
I suspect that spending any time at all in the bay area, while simultaneously having children, sort of forces the mind to reorient its priorities
I preferred SSC because Scott had less to lose at that time so he was more freewheeling and open, but the issues of homelessness and societal breakdown are pretty important ones.
If the link I saw (and mentioned in the immediate aftermath of Luigi's capture) was accurate and Luigi actually read ACX I understand completely why Scott didn't include questions grading his readers' sympathy for him (or related questions) but back in the SSC days I think he would have. Here we have an actual terrorist attack and a solid percent of Americans (and non-Americans) support him. That seems like a bigger issue to look at than whether people should be penalized for using the 10 items or less aisle.
The police haven't released much of what they have but they did release included a quote (which 3 minutes of search failed to find - of course - so here's the paraphrase) 'August 15th: Things are coming together, the target is insurance. It checks all the boxes..."
He's a freedom fighter or terrorist or revolutionary of a serious sort. No doubt he thinks the justice system is corrupt, and doesn't think very highly of either the president or the president-elect but he thought his actions through quite seriously. More seriously than he thought about his own escape plan.
He's not some guy who killed a rapist or a corrupt cop or a POS lobbyist or anyone of any fame whatsoever. He thought this shit through and he was right. He's got people deep in the culture wars rethinking who is actually waging war against them.
If Americans were literally any other people (save Canadians or Australians) the reaction would be On The Streets -- that's how much support he has.
Back in the SSC days I think Scott would have jumped into the fray. I don't hold his more "professional" and careful approach against him at all. He is incredibly brave, bold, and honest. One can not be all things to all people. I think this blog is great, but I hear where you're coming from.
Science and curiosity have long been political, and no one can control what things they care deeply about that also make other people in their society mad, which is what it really means for something to be political. I think that no one at all, not Scott Alexander nor anyone else, should have any qualms about blogging about things they care about that have political implications, out of some misguided feeling that there is something wrong with being about politics.
"In percent, what do you think is the profit margin of the average American company?"
This could do with more precision. Gross or net profit margin? Average how? Weighted by revenue or just a simple arithmetic mean? Publicly traded companies, or all businesses?
Many different reasonable ways this could be reasonably answered.
I second this. I really am confused as to the gross/net profit.
Another confused person here. Many companies with high gross margins have tiny net margins or make losses after depreciation and taxes. I answered as though the question was asking about net margins.
median profit margin in 2015 was xy% - for whatever that means. While "the public" seems to assume yx% - for whatever that means. "The Public Thinks the Average Company Makes a Profit Margin, Which Is About ... "
Agreed, this confused me (I went with net, publically traded - I didn't even think of different kinds of weighting, to my shame!)
Speaking from total eco-science ignorance, this was very confusing in the “average company” aspect for me: I have no idea, so I’d like to imagine an “average company”, but… what does that mean? Sum all profit and divide by the number of companies? Sort the companies by their profit margins and choose the median one? Etc. Basically, this seems (to me, a total know-nothing) as unspecific as “average country” (what is the [any random economic measure like GDP growth% or unemployment rate] of an average country?).
Exactly: does each company count as one company, or are we weighting by the size of the company?
I also struggled with this question, because I don't know what company is average. I know tech companies have significantly higher profit margins than many traditional companies, but I don't know what percent of companies fall under tech. I know a lot of standard companies like grocery stores have very low profit margins, but again don't know what percent of companies this encompasses.
I don't feel like my answer to this question is very useful or meaningful. Even to me. I suspect he doesn't care what the actual answer is, so much as perception, but if even I don't trust my own answer at all, I'm not sure how useful that is.
I suspected this question was kind of along the same lines as "when did you last think about the Roman empire?" It may be compared to an actual answer, and then you can see what your answer says about you.
There is no actual answer! The question is too vague for that.
I think net profit is the figure that captures the obvious reasonable interpretation of the question.
If a company is losing money, I expect its profit to be negative.
(I don't have strong views on the other questions, but don't think what you choose to answer to them should have all that big an impact on the answer).
The standard here is EBITDA. I’m quite confident Scott will end up using this.
What I’m less confident about how this would be weighted. Median profit margin of all listed stocks? Market cap weighted profit margin? Also, for the current year? Smoothed over the last 3 years? The business cycle and interest rates make this very noisy.
The answer was higher than I expected, when I looked it up (the search results appeared to be market cap weighted, but I’m not sure), but I’m predicting most people anchor on tech profit margins and will be surprised how low any reasonable “average” is.
Limiting it to publicly trade companies seems like much too harsh of a limit. There are on the order of tens of millions of businesses in the US, and only on the order of tens of thousands of publicly traded companies.
If we are going to go down this route (and I think your questions are reasonable!) then we also want to be clear on whether money LOSING companies are included in the average.
A negative business net profit margin is possible if we add in all the money losing businesses. Certainly in an industry (hello, restaurant industry!).
*) Net vs Gross matters
*) Including money losing businesses matters
*) Privately held (e.g. large law firms, family run restaurants) or not matters
I answered with my best guess for net profit for 'all' businesses. But many other interpretations are possible.
Also, does this include unprofitable companies? What about sole-proprietor/pass-through LLCs?
Especially with the US market being so heavily weighted towards the biggest tech companies (which are often unusually high margin) the number is radically different depending on whether it's listed, weighted by size, includes every incorporated entity (in which case the number will be very low). I gave a sort of generic number but the question is, I think, fundamentally broken and can't be used for analysis.
Yep, gross / net is a huge difference.
Yeah, I really wasn't sure about the weighting of various companies. Like I just figured many companies are running at a loss right now, and if I think of the set of companies relevant to be "a company that is started on America", then there are so many negative values in that average. But then... does Microsoft get weighted 10 million times as much as some guys food truck? Or is that just two companies whose margin percentages get averaged?
I ended up putting 0 as my answer.
Is it the average profit margin of an American company? Averages of ratios are fairly meaningless, and companies of vastly different scales are a concern. Median profit margin quantified across American LLCs might make sense, but 90% of those are going to employ <10 people, and work across vastly different industries (net profit in legal companies is going to be quite a bit higher than retail).
I'm actually a bit confused why anyone would interpret it to mean anything other than net margin. That's the portion of total revenue that stockholders get to take home (or reinvest in the company). That's what the average "man on the street" thinks when he thinks profit margin. Why would internal accounting metrics like gross margin matter?
"Is there a single person in the world who will genuinely be confused/upset with this wording?"
must... resist... the temptation to meta-nitpick *that* question itself
don’t fight it. be a pedant. i’ll be rooting for you.
Whatever, Deiseach already made the exact point I had in mind anyway
There is always going to be at least one person who will be confused/upset about something, no matter what that thing is.
Perhaps, but maybe not one person currently reading this blog who is engaged enough to take the survey!
No, there won't.
😁
It would not be an ACX Survey without pedants, nit-pickers, contrarians, and the awkward squad despite all the Rightful Caliph's exhortations.
https://en.wikipedia.org/wiki/Awkward_squad
❤️
I think the HBD question is ambiguous in an actually-confusing way. e.g. if someone believes that such differences are an unfortunate inequality that must be rectified, are they supposed to answer "favorable" (since they believe that the differences exist), or "unfavorable" (since they're against this state of affairs)?
Agree. It is not at all clear whether favorable means 'believes there is some scientific evidence for it' or 'believes it is a good thing' or 'believes it should be used to inform policy' or 'opposes all mention of sex/race in laws or policies.'
Strong agree here. Ambiguous and I struggled to answer this question.
I read that question as clarified, "eg the belief that races differ genetically in socially relevant ways?". You're favorable to the idea if you believe it's true, independent of if you believe it should be true, or if you believe that, if true, it should be corrected.
Then the wording be a more neutral like 'agree with the proposition'.
What to do after that agreement is a wholly separate thing, really.
Yeah I agree that's a reasonable interpretation, but it's not clear. If someone asks me whether I'm "favorable to the idea of eugenics", I'm pretty sure they're asking whether I think it's a good idea, not whether I think it's possible.
Yeah, I think it's a very unfortunate but likely true hypothesis about the state of the universe.
I am pretty certain that, as expressed in the survey, "the belief that races differ genetically in socially relevant ways", is pretty much an inevitable consequence.
Supposing the concept of "race" more or less points to a cluster of factors that are associated with each other (and thus recognizable as a cluster) and those factors being genetically determined (or at least, so much influenced by genetics as to be determined, in practical terms, let's say 80%/20% nature/nurture), and those factors being hereditary, thus (unless mixed with other races) they persist across generations. Unless you deny some or all of the preceding, you must admit to the existence of "race" as a concept, if a fuzzy one.
Then, supposing there are at least 2 different human groups, such as they can be distinguished by some of those clusters of factors, it would be strange if they, as groups, didn't have different means, distributions, etc. of values across several possible measures. E.g. height. Not all measures must differ, of course. Or be correlated. But some of them will, or else you don't really have different groups (or you're not correctly carving reality at the joints, see e.g. bleggs - https://www.lesswrong.com/posts/4FcxgdvdQP45D6Skg/disguised-queries). So, after accepting there are some measures that differ, it's not a stretch to postulate some of them are "socially relevant", since society measures us all in a dizzying amount of ways. Even if it's on the extremes of the outliers. E.g. Kenyan athletes are over-represented among Olympians in long distance running.
What you choose to do with those facts, of course, can be unfortunate. But I have problems to understand why is the fact itself unfortunate. Or rather, I understand that considering the mere fact "unfortunate" is a strong social signifier, but apart from social signalling, I don't really see why. That's part of the problem of making the definition almost maximally vague.
You might want to equalize outcomes. But even supposing you could (I don't mean it in terms of political capital, but as results of individual persons, all of this being the realm of averages and groups and standard deviations), the optimal amount of effort to equalize requires at least a reasonably close estimation of the types of differences and the amount of the differences. That's the strangest part. Seeing people who would like a certain result, but not able or willing to (even with great distaste) do the needed calculations to reach it. And I, for one, would really like to see it all out in the open. Anything else is just bickering about how to indirectly affect the reality without ever reaching a consensus of what even *is* the reality to be affected, much less when will we reach the desired result.
Race is just the phenotypically recognizable evidence of ancestry. Abilities and disabilities tend to cluster and persist for a while in families. Big deal. Nothing is permanent.
The mental health questions about what conditions my "family" has are unclear as to whether they mean only people who are genetically related to me.
After 3 years of this he still hasn't clarified that?
I think we may assume he means genetic relations; your third stepfather's cousin's second husband may be nutty as a fruitcake but that has no bearing on your mental health or propensity to develop mental problems.
Biological parents (if you know 'em), biological siblings (ditto), grandparents, aunts/uncles, cousins - any of which developed problems or had them from the start. That gives the hereditary background from which you can say "yeah, there is X in my family and there were concerns that I would pass it on to my kids" or "developed Y but nobody else has it".
What about adopted children? I am a former foster parent who adopted 3 of my foster children, and needless to say, my answers would be very different if I were to include them.
I always answered genetically. I don't think he cares who my brother-in-law is or what conditions he may have, related to the rest of my answers on the survey. The lack of specifics (two generations with no details) and no details on number of people or severity makes it almost impossible that he would try to use it as a general survey of frequency or anything else other than genetics.
I sort of remember he commented in a previous year that he meant for it to be a genetic connection, but I might be conflating that with how I decided to answer it after reading the comments.
It would be nice to have that clarified on the survey. Though, since there's no distinction between people who have X and also have family with X vs people who have X without family who have X, I think we can safely interpret the family option as a low-but-nonzero probability that I have X without realizing it.
Just curious if I selected the way other people would: I live in a city of ~560,000. Is that a small city or a large city?
I grew up half in the middle of Dallas and half in a little town of ~14,000 so my sense of scale is all over the map.
that's on the lower end of a large city. I would put the boundary at maybe 200k.
I wish it had clarified "big city" or maybe provided a an option for "Midsize city". I'm in St Pete FL, which is certainly no Chicago or LA, is also not a suburb, but certainly isn't a small town and it's in the midst of a huge metro area with suburbs of its own. Most places I've lived (Ann Arbor MI, Akron OH, Ft Lauderdale FL also) have been in this size range. Only lived in one big city (Baltimore-- I think it still qualifies although it's definitely the Poor Relation of east coast cities) and did suburbia briefly after college and spent an unhappy year and a half in a semi-rural but summer resort area.
I think that if it counts as a big city in its area (the most populous or important) then the literal size doesn't matter as much - no it's not going to be as big as New York city but it's a "big city" for Florida and that's the important distinction there.
For that I think the better term is perhaps "regional hub". That defines the town in relation to the area rather than in terms of size.
Our regional hub is definitely Tampa. But St Pete is no mere suburb of Tampa.
Seems like a small city. Above a million is my personal cutoff.
Edit: I think of city size in a global context, and China has completely changed the context of cities over the last few decades. Population >1m used to be a very large city within living memory.
IMO, a small city would be like my home town of Ypsilanti MI (20K people, and another 30K in the township), or Salisbury MD or Jamestown NY. The sort of place you can comfortably bicycle across.
I would call it a town
Two comments, half an order of magnitude of disagreement. I at least feel vindicated in my uncertainty!
It depends also on the country; a city of a million is a big city if the entire population is around five million, but by comparison with a country population of fifty million, it's a small city or even a large town.
OECD doesn't say cities as such but rather urban areas:
https://www.oecd.org/en/data/indicators/urban-population-by-city-size.html
"Urban areas in OECD countries are classified as:
large metropolitan areas if they have a population of 1.5 million or more;
metropolitan areas if their population is between 500 000 and 1.5 million;
medium-size urban areas if their population is between 200 000 and 500 000; and,
small urban areas if their population is between 50 000 and 200 000."
So a city of 560,000 would be a big(gish) city; not the largest, but not a small city either.
I thought about using the population of my metropolitan statistical area rather than the city proper, but the MSA here is ridiculous (https://en.wikipedia.org/wiki/Albuquerque_metropolitan_area); it's 80-90% uninhabited land.
I don't see a detailed explanation but it looks like OECD uses commuting patterns as part of their definition. For economic purposes that makes a lot of sense, but because the question in the survey specifically includes "suburb" as an option it feels skewed against this type of definition -- a suburb is basically just "a lower density area that functions economically as subsidiary housing for a nearby city" and so I would expect most suburbs to be part of the larger OECD metro by definition.
I still don't understand what Americans mean by the term "suburb". Like, is Beverly Hills a suburb or not? Is Sunset a suburb of San Franciso? Is Flushing a suburb of New York?
Generally we mean a region of predominately non-dense residential housing (not apartments) near but not in a city: Beverly Hills is a suburb, but Flushing is not.
Anything bigger than 50,000 is a large city. So Buffalo NY is a large city, while Jamestown NY is a small city.
If 50000 is a large city, what larger size categories do you use for cities in the millions and tens of millions? Because putting Tokyo (~37 million) in the same category as Buffalo NY seems...questionable.
You could say that about Tokyo and most cities. Berlin for instance.
If we decide a large city is a reasonable fraction of the worlds biggest city then most of what is thought of as large cities wouldn’t make the grade.
I'd say that 50K to 100K is a large town, 100K-500K is a small city.
Man, that is not at all what I would consider a large city. I live in a suburb that has 150,000 people in a metro area with 1.3 million people, and I chose the small city option.
Perhaps distance to the nearest larger city is a more important factor than absolute number of people?
The suburb I live in is touching the main city. I'm a ten minute drive from the Central Business District. That main city in my region though is less than 500,000 people, so I couldn't consider it a large city. In reality, I would call it a mid-sized city, but that wasn't an option.
Similarly if I live in a small city that's a suburb of a bigger city, should I have picked small city or suburb? Is the important distinction whether it's a self-contained community or an adjunct of a bigger city, or are we asking about single-family sprawl vs apartments and businesses and dense public transit?
Given that suburb is an option I would have gone with suburb, but there are definitely edge cases. What is Fort Worth in the DFW area? If the "suburb" is a 40 minute drive through countryside but is legally part of the metro?
As always, building discrete classifications for something on a continuous spectrum is hard.
My best guess is that the intent of the question is more along the lines of "how densely built/populated is it" which is why I chose small city but yeah it's pretty unclear.
That will depend on the country as well.
For China, 1 million isn't a large city. For Czechia, 250 thousand is a large city.
I live in a city that size and I said small after hesitation. It feels big to me, but I guessed that Scott would consider it small.
And don't forget e.g. Fairfax County Virginia, which is almost fully urbanized, but in which the vast majority of the population lives in unincorporated ares (there are only 3 towns in the whole county, and cities are independent of counties in Virginia).
I wasn't sure what to answer for the kidney donation question. I ended up with "I haven't started the process," but in reality I started researching it, and I even emailed a couple of organizations, but kidney donation to a stranger is illegal where I live. So maybe I was rejected? But I don't think that's what you were getting at with that option, either.
I would have liked a "I'm open to it, but haven't done anything yet to start the process." I'm not against it, but had to answer Not Interested because the other answers were more incorrect.
Good point. I wasn't sure how to answer as it's something I would seriously consider, but am medically not eligible.
As my long-term monogamous partner and I live together and have a child together but aren't legally or religiously married and don't plan to (which is not a particularly rare arrangement in Europe nowadays), I'm not really sure how to answer the "Married people:", "Single people:" and "People in a relationship other than marriage:" questions -- I answered the first two by interpreting "future spouse" broader than literally, skipped the following two, answered the following one by interpreting "marriage" broader than literally, and skipped the last two.
I think it’s clearly people in a relationship other than marriage, common law marriage being not a thing anymore.
All ambiguity is out the window if you pop down to the registry office and sign a document.
Re: common law marriage
IMO, it should be.
If two people are cohabiting, share finances (and benefits), are sleeping together and intend the situation to endure then they are de facto married. Until SSM became legal same sex couples generally thought of it that way and would refer to their partner as "husband" or "wife".