Yep, I think these three perspectives roughly cover why I think it might have been a good idea. I also think that a good number of people we now think of as having had a large impact on x-risk and who were kind of similar to rationalists (e.g. some of the Manhattan Project scientists) had that impact because they participated in that effort (and the followup cold-war period) for roughly the three reasons you cite.
Yeah, I agree and I do apologize deeply for things not being as fast as they should be.
Yeah, the whole platform is definitely still pretty buggy.
My experience was great after 30 initial minutes of chaos, and then I had an hour of pretty good conversation.
Respond to this comment if you want to ask a question that will be asked in the Q&A section.
This is the fixed link: https://www.youtube.com/watch?v=4qYNjZZGLDQ&feature=youtu.be
Promoted to curated: I’ve been thinking about this post a lot since it has come out, and it is just a really good presentation of all the research on AI Alignment via debate. It is quite long, which has made me hesitant about curating it for a while, but I now think curating it is the right choice. I also think while it is reasonably technical, it’s approachable enough that the basic idea of it should be understandable by anyone giving it a serious try.
Hey, I’ve been thinking of curating this post. Would you be up for crossposting it’s full content now that it’s been out for a while, which would allow us to curate it?
Oops, fixed. It’s noon PDT.
I think the model uses a much shorter time for active infections than 2.5 weeks. Not sure what it is, but I think it’s closer to 5 days or something like that, which seems to actually fit the behavior of the disease best, on a broad scale.
Agree that it looks weird. I’ve asked the authors of the project to add a cumulative graph, which makes these assumptions a lot clearer.
That’s active infections. That number corresponds to something like 70% of the population having been infected at some point.
This problem cannot be cleanly solved analytically (because f is discontinuous and obviously lacking a clean closed form), but is expressed by a beautiful and simple differential equation.
Huh, this is a great example. My historical relationship to differential equations has been mostly from the perspective of “it’s a method by which you eventually arrive at good analytic descriptions of systems”. I think this example really concretely illustrated why that’s a wrong perspective.
I expect to learn some lessons from it, in particular in how to design high-density UIs in a LessWrong context, that I expect will show up in other parts of LessWrong. I can also imagine that at some point we would more generally support complicated tables and databases on LessWrong, but that seems quite a while away.
Promoted to curated: This post is a bit more technical than the usual posts we curate, but I think it is still quite valuable to read for a lot of people, since it’s about a topic that has already received some less formal treatment on LessWrong.
I also am very broadly excited about trying move beyond a naive bayesianism paradigm, and felt like this post helped me significantly in understanding what that would look like.
Replaced the rot13 in your comment with spoiler tags. (Accessible via >! in the WYSIWYG editor)