All of jdfaben's Comments + Replies

jdfaben3215

I mean, I'm pretty sure the real Eliezer would in fact just say you shouldn't send any messages to a superintelligent AI, so this seems like the correct responses. Does seem to render this somewhat useless as an approach though.

I havent, but I likely will now. I'm tempted to try and come up with some design based on the patterns and put it on a t-shirt.

My first thought was "why is there an API and not a 'download data' button?" But on second thoughts I have to admit that the set of people who want their baby's sleep data in csv format and are unwilling to use a simple API is probably quite small...

Also, there doesn't seem to be a convenient way to download the data, which is kind of frustrating

2jefftk
Have you tried https://pypi.org/project/snoo/ ? (I haven't, but I liked http://www.relevantmisc.com/r/python/2020/05/26/visualizing-baby-sleep/)
Answer by jdfaben120

I can't speak for less wrong as a whole, but I looked into this a little bit around that time, and concluded that actually it looked like things were heading in the sensible direction. In particular, towards the end of 2014, the US government stopped funding gain of function research: https://www.nature.com/articles/514411a, and there seemed to be a growing consensus/understanding that it was dangerous. think anyone doing (at least surface level) research in 2014/early 2015 could have reasonably concluded that this wasn't a neglected area. That does leave open the question of what I did wrong in not noticing that the moratorium was lifted 3 years later...

4GeneSmith
Eliezer seemed to think that the ban on funding for gain of function research in the US simply led to research grants going to labs outside the US (Wuhan Institute of Virology in particular). he doesn't really cite any sources here so I can't do much to fact check his hypothesis. Upon further googling, this gets murkier. Here's a very good article that goes into depth about what the NIH did and didn't fund at WIV and whether such research counts as "gain of function research". Some quotes from the article: ... ... ... There are differing opinions on whether or not what the researchers at WIV did counts as gain of function research: So to summarize: from what we know, researchers at WIV inserted a spike protein from a naturally occuring coronavirus into another coronavirus that was capable of replicating in a lab and infecting human cells. But the genome of this resulting virus seems too different from that of coronavirus for it to have been a direct ancestor of the pandemic causing coronavirus. Overall I don't feel like enough people are linking their sources when they make statements like "I'd give the lab leak hypothesis a probability of X%".

It seems that when there's a discussion of a dangerous practice being stopped pending safety review it makes sense to shedule into the future a moment to review how the safety review turned out. 

Maybe a way forward would be:

Whenever there's something done by a lot of scientists is categorically stopped pending safety review, make a metaculus question about how the safety review is likely to turn out. 

That way when the safety review turns out negatively, it triggers an event that's seen by a bunch of people who can then write a LessWrong post about it?

That leaves the question whether there are any comparable moratoriums out there that we should look at more.  

jdfaben190

I didn't really see much public discussion early outside of epidemiology Twitter. I'm married to an epidemiologist who stocked our flat with masks in December when there were 59 confirmed cases in Wuhan, and we bought enough tins of food to eat for a few weeks in January, as well as upgrading our work-from-home set up before things sold out. Although I completely failed to make the connection and move my pension out of equities, that doesn't actually seem to have cost me very much in the long run (for those keeping count, S&P 500 is up 17% year-on-year... (read more)

Good on your spouse! Very impressed. 

(Also, I don't get the S&P being up so much, am generally pretty confused by that, and updated further that I don't know how to get information out of the stock market.)

I think epistemics is indeed the first metric I care about for LessWrongers. If we had ignored covid or been confident it was not a big deal, I would now feel pretty doomy about us, but I do think we did indeed do quite well on it. I could talk about how we discussed masks, precautions, microcovids, long-lasting respiratory issues, and so on, bu... (read more)

jdfaben240

I'm actually confused by that response, and I don't think it's really part of your best attempt to explain what you meant by 'rationalists pwned covid'. I'll try to explain why I'm unimpressed with that response below, but I think we're in danger of getting into a sort of 'point-scoring' talking past each other. Obviously there were a few rhetorical flourishes in my original response, but I think the biggest part of what I'm trying to say is that the actual personal benefits to most people of being ahead of the curve on thinking about the pandemic were pre... (read more)

Third, as I said above, it’s a pretty low bar. If you’re rich enough (and don’t work at a hospital), avoiding personally getting infected is relatively straightforward, and while obviously it has some benefits, I don’t think it would be enough of an incentive to convince me to take on a whole new worldview.

My personal experience is consistent with this take, for what it’s worth. I think that “rationalists didn’t get COVID” is indeed mostly due to substantially higher average income (perhaps not even among ‘rationalists’ but specifically among Jacob’s friends/acquaintances).

jdfaben780

Seriously, in what sense did rationalists "pwn covid"? Did they build businesses that could reliably survive a year of person-to-person contact being restricted across the planet? Did they successfully lobby governments to invest properly in pandemic response before anything happened? Did they invest in coronavirus vaccine research so that we had a vaccine ready before the pandemic started? Did they start a massive public information campaign that changed people's behaviour and stopped the disease from spreading exponentially? Did they all move to an islan... (read more)

0leonidasmith
Thank you for this comment.  I have very mixed feelings about this website. On the one hand, it's got interesting articles and reading HPMOR was very entertaining. But, on the other, there's so many people who are just writing posts that are embarrassingly the opposite of what they claim they are all about: self-consciousness, in particular.  Me and my friends are rational, that is, all that is right and correct, by definition, and we call ourselves Rationalists. Can't you see the paradox in this claim? And yet I thought people here didn't find Spock rational, and correctly assessed that his character is "defined" as such by the story, but it fails to fulfill itself. As a human being you will always fail. The website is called "less wrong", for a random divinity's sake. It should be about striving to be less wrong while admitting we cannot avoid failure, not about jerking each other off about how rational and better we are than others. And yet...  In general I am highly suspicious of any lover of truth who will willingly call themselves a sophist: for a Rationalist, there is little praise as high as being a Rationalist, and therefore calling yourself and the people who agree with you with this name is very much patting your own back. Especially when using to criticize and compare yourself to people you disagree with. Having said this, there are interesting things in this post too. I'm not saying it's completely bad, but a lot of the language and framing here leads me to think there is also a lot of unnecessary arrogance. I just wanna say, more actual thinking, less implicit self-praise.
3voxelate
I saw the supply chain disruptions coming and made final preparations for it, I saw layoffs coming in my aviation-related job so I updated my resume, took a good severance package, and found a new, remote-based job with significantly higher pay.  And yes, I also significantly re-balanced my portfolio and took advantage of the crash early this year.  In all, I expect about 40% additional income/unrealized gains this year than last.  To me that's more than minimal. Rationalists that were paying attention get the 1st chance of understanding the implications and making moves (big or small) before a mass of people finally took it seriously in the US.  I'll admit part of it is certainly luck since I can't really time the market or precisely know how gov't policies and actions will affect my stocks.   It's also hard to know how much of that on my part was explicit reason, I was certainly reading up on the literature about it, but there was not a ton of data.  I did use some social cognition based on the Chinese response under the presumption that they knew more about it since it originated there. I don't think the COVID response is even the best measure to judge the benefits of being rational, it's just one part of it.  If you want to solve problems, you have to be rational... being irrational is a bad way to solve problems.  

FWIW I left a decent job that required regular air travel to deep red "COVID is a liberal hoax" areas of the US based heavily on content here.  I had alternatives lined up but I probably would've stuck it out otherwise and I think that would've been a mistake. 

We didn't get COVID, for starters. I live in NYC, where approximately 25% of the population got sick but no rationalists that I'm aware of did.

Ben Pace*210

The best startup people were similarly early, and I respect them a lot for that. If you know of another community or person that publicly said the straightforward and true things in public back in February, I am interested to know who they are and what other surprising claims they make.

I do know a lot of rationalists who put together solid projects and have done some fairly useful things in response to the pandemic – like epidemicforecasting.org and microcovid.org, and Zvi's and Sarah C's writing, and the LW covid links database, and I heard that Median gr... (read more)

I'd stress the idea here that finding a "solution" to the pandemic is easy and preventing it early on based on evidence also is.

Most people could implement a solution better than those currently affecting the US and Europe, if they were a global tsar with infinite power.

But solving the coordination problems involved in implementing that solution is hard, that's the part that nerds solving and nobody is closer to a solution there.

In case you haven't already found out, the free version has been updated to be a smaller version of GPT-3. Confirmed on twitter https://twitter.com/nickwalton00/status/1284842368105975810?s=19

3gwern
Although smaller is not very interesting, especially if you want to probe the model's understanding and intelligence. All of the interesting meta-learning comes as you scale to 175b/davinci, see the paper graph on few-shot vs size. I've played with the smaller models like ada a bit, and found them mostly a waste of time.

It is now literally not true that car accidents kill more people, in either the UK or Italy, and won't be true in the US in about a week.

I've found the time-delayed log graphs like this one pretty convincing: https://images.app.goo.gl/iKrfzw9Wt7hAqkVB6