How does this compare to what he was saying in 2004? Has he changed his mind about the brain or about AI? Maybe these things about the brain are foundational and we shouldn’t expect him to change his mind, but surely his beliefs about AI should have changed.
Thanks. I guess I got confused from mainly using greaterwrong, but I did test it before posting.
In the past you (Raemon) have referred to this as the “shortform feed,” but in this post you don’t. Is this intentional? (But then in the comments, you say “feed” again.)
To me “feed” suggests a chronological order. Have you considered making the shortform posts sort by age of comment, just as lists of posts usually do?
Similarly, when LW migrated from OB, adding nesting and voting, it made the OB posts sort their comments in chronological order, to preserve the conversation structure. With the advent of 2.0, these posts have lost their special status and are sorted by votes, making the comment conversation unreadable (example). If you do implement a default sort nudge, you should apply it to these posts, too.
[I guess this comment should have gone here.]
Did your English class distinguish “How do you do?” from “How’s it going?” etc?
Katy seems to be making that distinction, but in my experience people eavesdropping on the masses, most people don’t treat any of the variants as a question, but are substantially more likely to respond with another greeting, rather than something that can be interpreted as an answer.
You already have lots of fitbit sleep data, right?
You should look at that data first and use it to guide the experiment. Should you study other metrics of sleep quality?
In particular, you should determine the standard deviation in onset times and do a power analysis to see how long the experiment has to run. I guess there’s a problem that you might not have labels indicating which days are straight home from work days. You should try to remember that. Even without that, a simple filter like Mondays might be a good choice.
Who were the other people who tried to short housing in 2004-2005? Does Michael Lewis talk about them?
No, really, it was perfectly clear. The problem is that it was wrong.
Sure, you use a delta function when you want to make a simplifying assumption. But this post is about questioning the assumption. That’s exactly when you wouldn’t use a delta function. Your third answer flatly contradicts Shminux. No, he does not believe that there are any perfect dice. Sometimes it’s right to contradict people, but if you don’t notice you’re doing it, it’s a sign that you’re the one who is confused.
The White House spends the vast majority of its resources putting out false press releases. My impression is that that’s what Kalil did, too. Probably he shifted things in a positive direction, but the shape of the marginal effort doesn’t have much to do with the shape of the total effort. That is, how much time he spent shaping the CDC actions vs NIH funding vs conferences of outsiders doesn’t tells us much about how much of his useful actions fell in those categories. He had practically no direct power, so in a sense the CDC and NIH were outsiders to be coordinated, too.
Cummings burnt a lot of bridges by saying important negative things. I’m suspicious of Kalil sounding so positive. The first hour of the podcast gave me an extremely negative view of him, but then he mentioned a lot of trade-offs and strategies that seemed valuable regardless of the average level of government function. Still, I worry that he sold his soul to function in this environment and lost the ability to tell good projects from bad.
Before asking how something happened, we should ask if it happened.
I highly recommend this piece by Dominic Cummings on how government works in practice. He is certainly optimistic that it could accomplish a lot, but my interpretation of the Kalil quotes is pretty much the opposite of yours. I’ll listen to the podcast to get more context.
When I think of “drill,” I think of arithmetic problems. The feedback does not come from the world, but from the teacher. A large part of that is that the child does not want to learn arithmetic because of not having an application for it. An application might provide better feedback.
The second thing I think of with “drill” is a sports coach making the kids break down the game, rather than just playing. The coach makes them do free throws, which they wouldn’t do on their own, but the world provides the feedback and the value of free throws is pretty clear. Erikson says that the coach is right to drill, that it is suboptimal to just play. Even professional players are famous for not practicing free throws. A child at the basketball court alone, unable to just play is unlikely to drill free throws, but is likely to drill layups, which are closer to the standard game and may be a good choice.
But there is a tradeoff between different readers. The “bad” version looks better to me for web readers, who are probably the vast majority. Maybe the writer should think about whether it actually is better, on a link-by-link basis, but I think that they will leave it alone. (Writers should also think about phone vs computer, especially if writing is on computer and reading is on phone.)
Here is a compromise: leave the text the same, but also collate a bibliography of works cited. This would provide a new benefit to web readers, while also providing a hint to print readers.
Try getting the article from greater wrong instead.
Mendeleev received a доктор degree in 1865. Although cognate to doctor, this is usually translated as habilitation. The PhD is usually considered equivalent to the кандидат (candidate) degree, which he received in 1856.
During the Industrial Revolution in Britain, the effect was big enough that wages doubled from 1850-1900.
The Industrial Revolution is not usually dated as beginning in 1850. Often it is dated as ending in 1850. Wages were stagnant 1750-1850. I think it’s considered pretty mysterious why wages behaved differently in the two periods.
The average person spends about 10 years (87k hours) in a house before moving, which already goes against people’s tacit models.
It seems to me that this is the key point, which is why you made it the title and the first sentence of the original post, but I feel like it has been lost in the discussion and maybe swamped by the rest of your post. The longer the tenure, the better buying is. 10 years seems pretty short to me, maybe not a mistake for the average buyer, but definitely a mistake for the marginal buyer. I’d expect most buyers to have an explicit model of more than 10 years, not just a tacit model. But translating it into hours seems to lengthen or obfuscate it and distract from the point.
In 1956, AT&T was banned from selling anything other than telecom. I assume that’s why Shockley left to found his semiconductor company that year. I’m unclear on whether it could license patents, but its existing patents were all seized and put in the public domain. AT&T still had a lot of internal needs for computers, so it kept funding research.
But it does use MCTS in training. You might say that it uses MCTS to generate a better player to learn from.
academic and commercial
I think the main issue is retail vs wholesale. (BGI’s $600 probably means volume pricing.) Also, I think NHGRI publishes averages over deployed machines, including old ones, which overestimates the cost of buying a new machine to create new capacity.
Dante is a retail commercial product of $700 for a whole genome. Dante and Veritas had previously had sales of $200 and $300 (which was probably measuring the demand curve and doesn’t say much about cost). Two days after writing this comment, Veritas cut its price in half to $600, below Dante.