Less Wrong doesn't seem "overgrown" to me. It actually seems dried out and dying because the culture is so negative people don't want to post here. I believe Eliezer has talked about how whenever he posted something on LW, the comments would be full of people trying to find anything wrong with it.

"Overgrown" was probably a bad analogy, I tried too hard to reference the idea of well-kept gardens. What I was trying to say is that there are too many hostile elements who are making this website an unwelcoming place, by unnecessary criticism, ad hominem attacks and downvotes; and that those elements should have been removed from the community earlier. I actually think we agree on this.

I think trying to impose strict new censorship rules and social control over communication is more likely to deal the death blow to this website than to help it. LessWrong really needs an injection of positive energy and purpose. In the absence of this, I expect LW to continue to decline.

OK, from reading this and other comments I accept that this was the weakest part of my post. Also, after re-reading the Wiki entry on Crocker's rule, I don't think I intended to suggest anything quite that extreme. Crocker's rules say that rudeness is acceptable simply in order to provide a precise and accurate signal of annoyance. This is certainly not what I had in mind.

I apologize for my incorrect usage of the term "Crocker's rules", and I recognize that this was probably not a good idea. I hope someone can come up with a policy that achieves the objectives I had in mind when I wrote that sentence.

I agree with this (I probably contribute a bit to the problem, I will try to do better).

edit: I think a lot of Clarity downvotes have to do w/ people not liking how that person comes across.

doesn't seem "overgrown" to me. It actually seems dried out and dying

I think both. There's more post and comment volume than I remember from "the good old days", but it's much lower quality in terms of density of interesting contributions.

because the culture is so negative people don't want to post here.

I don't agree with that as the main cause. I don't think it's a negative culture, so much as a lack of positive culture. The people doing interesting work on rationality, decisionmaking, and AI are mostly not doing it here anymore.

Would there be a way for people who already maintain blogs elsewhere to cross-post to their LW subdomain? (Would this even be desirable?)

We would have to discuss this with people who run blogs elsewhere to find out what solutions would work for them. My preferred solution would be for people to import their old blog posts, and then redirect their domain to the LW subdomain. I do not know whether outside bloggers would find this acceptable. In some cases we may also have to consider the question of advertisement revenues.

Do you envision LW2 continuing to include applied rationality type posts? Does that work with "everything should work towards Aumann agreement"?

My apologies, I did not intend to declare that posts about applied rationality should be avoided. I guess my phrasing reveals my bias towards the "epistemic" part of this community rather than the "instrumental" side. My personal preference is to shift the community focus back towards epistemic rationality, but that is a separate discussion which I did not intend to raise here. The community should discuss this separately.

users may not repeatedly bring up the same controversial discussion outside of their original context

How could we track this, other than relying on mods to be like "ugh, this poster again"?

There would have to be some moderator discretion on this issue. My personal view is that we should err on the side of allowing most content. This language was intended for extreme cases where the community consensus is clear that a line has been crossed, such as Eugine, AdvancedAtheist or Jim Donald.

professionally edited rationality journal

Woah. Is this really a thing that MIRI could (resources permitting) just like ... do?

Yes, they can certainly do this if they have the resources. Initially, academics may not take the journal seriously and it definitely will not be indexed in academic databases. If the quality is sufficiently high, it is conceivable that this may change.

I sometimes semi-jokingly say that the easiest way for me to contribute to LW code would be to reimplement the whole thing in Java.

If someone would be okay with paying me to do this, I would be happy to try. Not sure if that's the best use of the money, though. Making it open-source would have the advantage that if I fail to complete the project, other people may join or replace me later.

(Why Java? First, that is the only programming language I am sufficiently fluent at. Second, Java has the advantage that the source code is already compiled and loaded in the memory when HTTP requests come, so in theory it should be fast. But I don't have an experience with measuring performance in large projects.)