Alternate titles: What Comes Next?, LessWrong is Dead, Long Live LessWrong!
You've seen the articles and comments about the decline of LessWrong. Why pay attention to this one? Because this time, I've talked to Nate at MIRI and Matt at Trike Apps about development for LW, and they're willing to make changes and fund them. (I've even found a developer willing to work on the LW codebase.) I've also talked to many of the prominent posters who've left about the decline of LW, and pointed out that the coordination problem could be deliberately solved if everyone decided to come back at once. Everyone that responded expressed displeasure that LW had faded and interest in a coordinated return, and often had some material that they thought they could prepare and have ready.
But before we leap into action, let's review the problem.
The people still on the LW site are not a representative sample of anything. With the exception of a few people like Stuart Armstrong, they’re some kind of pack of unquiet spirits who have moved in to haunt it after it got abandoned by the founding community members. At this point it’s pretty much diaspora all the way down.
--Yvain on his Tumblr.
One of the problems is that people who control the LW website are running it in pure maintenance mode. LW was put out to pasture -- there have been no changes to functionality in ages.
--Lumifer
LW's strongest, most dedicated writers all seem to have moved on to other projects or venues, as has the better part of its commentariat.
In some ways, this is a good thing. There is now, for example, a wider rationalist blogosphere, including interesting people who were previously put off by idiosyncrasies of Less Wrong. In other ways, it's less good; LW is no longer a focal point for this sort of material. I'm not sure if such a focal point exists any more.
This dwindling content can be seen most clearly in the "Top Contributors, 30 Days" display. At the time I write this there are only seven posters with > 100 karma in the past 30 days, and it only takes 58 to appear on the list of 15. Perhaps the question should not be whether the content of LW should be reorganised, but whether LW is fulfilling its desired purpose any longer.
As nearly all the core people who worked the hardest to use this site to promote rationality are no longer contributing here, I wonder if this goal is still being achieved by LW itself. Is it still worth reading? Still worth commenting here?
--qsz
LW does seem dying and mainly useful for its old content. Any suggestions for a LW 2.0?
--signal
So let's talk suggestions for a LW 2.0. But just because we can restart LW doesn't mean we should restart LW. It's worth doing some goal factoring first (see Sacha Chua's explanation and links here). Before getting into my summary, I'll note that The Craft and the Community Sequence remains prescient and well worth reading for thinking about these issues. And before we can get into what our goals and plans are, let's talk some about:
What went wrong (or horribly right):
So why did LessWrong fade? One short version is that LW was a booster rocket, designed to get its payload to a higher altitude then discarded. This is what I mean by what went horribly right--MIRI now has a strong funding base and as much publicity as it wants. Instead of writing material to build support and get more funding, Eliezer (and a research team!) can do actual work. Similarly, at some point in one's personal growth it is necessary to not just read about growing. We should expect people who aren't habitual forum-posters to 'grow out' of heavy reading and posting on LW.
Another short version is that there was only so much to say about rationality (in 2012, at least), and once it was said, it wasn't clear what to say next. Whether something is on topic for LW and whether it belongs in Main, Discussion, or an Open Thread is unclear and so less and less content is created, and so less and less people visit, leading to even less content. The easiest example of friction is whether or not 'effective altruism' is a core LW topic; this comment by iceman expresses the problem better than I could.
Relatedly, while rationality is the Common Interest of Many Causes, in that many different causes all potentially benefit from someone coming to LessWrong and adopting its worldview and thought patterns, LessWrong seems flavored enough by MIRI and Eliezer in particular that we mostly see the Many Causes free riding instead of contributing to the upkeep of LW (in terms of content, not hosting funds). Even CFAR, the most closely related of the Many Causes to LessWrong's stated mission, mostly overlaps with LW instead of supporting it. (To be clear, this is a decision I endorse; CFAR has benefited from not being tied to the idiosyncrasies of LessWrong. CFAR staff are also some of the most frequent contributors left of the founding community members.)
I should elaborate that by Many Causes I explicitly mean a broader tent than Effective Altruism. Anyone who is sympathetic to the Neo-Enlightenment Eliezer talks about in Common Interest of Many Causes strikes me as enough of a fellow traveler, regardless of whether or not they have found something to protect or whether or not that something to protect is the kind of thing Givewell would consider altruistic or a top priority.
What roles LW served, and what could do it better:
First, some roles that LW (the website) doesn't or can't serve:
- Getting direct work done. Open-sourcing things is powerful, but it remains true that money is the unit of caring. When people really want something done, they have an institution with an office and employees that get the thing done. Direct work on any of the Many Causes is going to be done by people working directly on that task, not by posts on an internet forum.
- Physical interaction with like-minded people. You can organize a meetup on LW, but you can't attend one.
- Practical rationality training. The Sequences are great at giving people a philosophical foundation, but they can only do so much. There's a reason why CFAR has workshops instead of writing articles and books.
- Focal Point / News Organization
- Welcoming Committee / Rationality Materials
- Meetup Organizer / Social Club
Focal Point / News Organization
If your values and interests are similar to a community's, the main benefit you get out of the community and the community gets out of you have to do with correlating your attention. If something of interest to me happens, be it a blog post or a book or an event or a fundraiser, I won't know unless it enters one of my news streams. Given the high degree of shared interests between supporters of the Many Causes or by virtue of social ties to the community, treating the community's attention as a shared resource makes great sense. (Every promoted post since Julia Galef's in April seems like an example of this sort of thing to me.) For example, MIRI's Winter Fundraiser is going on now. But there are Many Causes, not one cause, and as much as possible the ability to direct shared attention should respect that.
Many people in the community also have interesting thoughts, which they typically post to their blog (or twitter or tumblr or ...). Aggregating those into one location reduces the total attention cost of keeping up with the community. (This is especially important if one wants to maintain people who are time-limited because they are working hard on their Important Project!) The experience of SSC seems to suggest that it's way better for authors to have control over their branding. I suspect much of the mainstream attention that Scott's received is because he's posting to a one-man blog, and thus can be linked to much more safely than linking to LW.
So compared to when most things were either posted or crossposted to LW, it seems like we currently spend too little attention on aggregating and unifying content spread across many different places. If most of the action is happening offsite, and all that needs to be done is link to it, Reddit seems like the clear low-cost winner. Or perhaps it makes sense to try to do something like an online magazine, with an actual editor. (See Viliam's discussion of the censor role in an online community.) I note that FLI is hiring a news website editor (but they're likely more x-risk focused than I'm imagining).
If we were going to modify LW to serve this role better, multiple reddits seems like the obvious suggestion here (and a potentially interesting innovation may be tag reddits, where categories are not exclusive). "Main" and "Discussion" do not at all capture the splits in what the audience wants to pay attention to. Integrated commenting across multiple sites, if possible, seems like it might be a huge win but may be technically very difficult (or require everyone to agree on a platform like Disqus).
Welcoming Committee / Rationality Materials
Someone is interested in learning more about thinking better; probably they have tons of confusion about philosophy, how the world works, and their own goals and psychology. Someone mentions this LessWrong place, or links them particular articles, or they read HPMoR and follow the links in the Author's Notes.
But then they realize just how long Rationality: From AI to Zombies is, or they don't understand a particular part. Without social reinforcement that it's interesting and without other people to ask questions of, they likely won't get all that far or as much out of it as they could have.
And then there's all the other things that someone picks up by being part of a community--who the various people are, what they're working on, what options are out there.
It seems to me that the the optimal software for something like this is perhaps more like Wikipedia or Stack Overflow than it is like Reddit. If we're building a giant tree of rationality-related concepts and skills, it doesn't quite make sense to have individual blog posts written by individual authors, instead of community-maintained wiki pages with explanations and links.
Meetup Organizer / Social Club
You can't do the physical meeting up online, but you can alert people to meetups near them. At time of writing, according to the map on the front page, one of the five closest meetups to me (in Austin, Texas) is in Brussels.
Part of that is groups moving to other communication channels to organize meetups. In Austin, for example, the email list is a much more reliable way to contact people--especially since many of them don't regularly check LW! But the lost advertising potential seems significant, and something like the EA Hub seems like a better solution.
There's also a role to be played in colocating rationalists, either through helping form group houses and shared apartments or moving subsidies / loans. It's not clear it's efficient for more people to move to the Bay Area relative to secondary or tertiary hubs, but it does seem likely that we should put resources towards growing the physical community.
There's also a much longer conversation that could be had about effectively employing more social technology to develop and strengthen the community, but I get the sense that most of those organizations, be they formal institutions or churches or families or mastermind groups or taskforces, are categorically unlike online forums, and the community they will be developing and strengthening will not be "LessWrong readers" so much as "meatspace rationalists." So I'll ignore this for now as off topic, except to note that I am very interested in this subject and you should contact me if also interested.
Why not have that and LW?
So far, I've talked about things that would serve various roles better than LW, though perhaps not at the same time. One could easily imagine them existing side by side: it's not like Scott Alexander needed to shut down his Yvain account to start posting at SSC, he just made the alternative and started posting to it. Similarly, a reddit for the rationalist diaspora already exists (though it doesn't see much use yet), as do two (well, one and a half) for SSC.
The trouble is the people who have noticed that people have left, but not where they're going, and the links to LessWrong over all the old material. If LessWrong is a ghost town that's being haunted by a pack of unquiet spirits, well, better to be upfront about that than give people the wrong impression about what rationalists are like.
TL;DR
I think we should either develop a plan that makes LW fully functional at the three roles mentioned above (and any others that are raised), or we should close down posting and commenting on LW (while maintaining it as an archive). The shutdown could either happen at the end of December, or March 5th to correspond to the opening of LW, but the most important factor is that there be replacements to point people to. It seems likely we should leave open the LW wiki (and probably make the LW landing page point to a wiki page, so it can be maintained and updated to point to prominent parts of the diaspora). The Meetups functionality should probably be augmented or replaced (either static links to dynamic objects, like Facebook groups, or with functionality that makes it easier on the organizers, like recurring meetups).
(I wrote that as an 'or', but at present I lean heavily towards the 'archive LW and embrace the diaspora' position.)
A tangential note on third-party technical contributions to LW (if that's a thing you care about): the uncertainty about whether changes will be accepted, uncertainty about and lack of visibility into how that decision is made or even who makes it, and lack of a known process for making pull requests or getting feedback on ideas are incredibly anti-motivating.
This is probably the single most important obstacle to making a better LW on the technical side.