Less Wrong is a community blog devoted to refining the art of human rationality. Please visit our About page for more information.

Alexei comments on On the importance of Less Wrong, or another single conversational locus - Less Wrong

84 Post author: AnnaSalamon 27 November 2016 05:13PM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (362)

You are viewing a single comment's thread.

Comment author: Alexei 27 November 2016 06:42:19AM 21 points [-]

I strongly agree with this sentiment, and currently Arbital's course is to address this problem. I realize there have been several discussions on LW about bringing LW back / doing LW 2.0, and Arbital has often come up. Up until two weeks ago we were focusing on "Arbital as the platform for intuitive math explanations", but that proved to be harder to scale than we thought. We now pivoted to a more discussion-oriented truth-seeking north star, which was our long-term goal all along. We are going to need innovation and experimentation both on the software and the community levels, but I'm looking forward to the challenge. :)

Comment author: John_Maxwell_IV 28 November 2016 05:21:05AM *  5 points [-]

We now pivoted to a more discussion-oriented truth-seeking north star, which was our long-term goal all along.

Exciting stuff!

Are you planning to engage with the LW community to figure out what features to implement?

I know that Eliezer was heavily involved with Arbital's product management. But I think it's a mistake to make him the BDFL for LW 2.0, because LW 1.0 failed, and this was plausibly due to actions he took. Beware the halo effect: someone can simultaneously be a great blogger and a lousy product manager/forum moderator. I think we should let someone else like Vaniver have a try.

If you're planning to engage with the community (which I would strongly recommend--ignoring their userbase is the kind of thing failed startups do), I suggest waiting a bit and then creating a new thread about this, to simulate the effect of a sticky.

Comment author: Alexei 30 November 2016 04:33:03PM 2 points [-]

Are you planning to engage with the LW community to figure out what features to implement?

Eric R and I read all the comments in this thread. We've also met with multiple people in person to discuss exactly what the platform should look like. So the broad answer is "yes", but if you have a specific mode of engagement in mind, then it might be "no".

I know that Eliezer was heavily involved with Arbital's product management.

He is an adviser. There are no advocates to make him a BDFL as far as I know.

I suggest waiting a bit and then creating a new thread about this, to simulate the effect of a sticky.

I expect we'll have a public beta ready in two weeks. I plan to write a blog post of my own to explain Arbital in more details.

Comment author: John_Maxwell_IV 01 December 2016 09:22:50AM 0 points [-]

Sounds great!

if you have a specific mode of engagement in mind, then it might be "no".

Well, if you created a new thread called "Eric and I are taking suggestions for Arbital", I imagine you might get a lot more relevant ideas and feedback :)

Comment author: malcolmocean 27 November 2016 01:32:22PM 5 points [-]

I'm very excited to have an Arbital-shaped discussion and writing platform. I've been thinking for awhile that I want some of my online writing to become less blog-like, more wiki-like, but I don't actually want to use a wiki because... yeah. Wikis.

Arbital seems way better. Is it at the point now where I could start posting some writing/models to it?

Comment author: Alexei 27 November 2016 05:55:03PM 1 point [-]

Not yet; hence Anna's comment above.

Comment author: AnnaSalamon 27 November 2016 07:01:11AM *  17 points [-]

I am extremely excited about this. I suspect we should proceed trying to reboot Less Wrong, without waiting, while also attempting to aid Arbital in any ways that can help (test users, etc.).

Comment author: RyanCarey 27 November 2016 09:45:15PM *  9 points [-]

If half-hearted attempts are doomed (plausible), or more generally we're operating in a region where expected returns on invested effort are superlinear (plausible), then it might be best to commit hard to projects (>1 full-time programmer) sequentially.

Comment author: Mqrius 05 December 2016 08:05:15AM 0 points [-]

Does that take into account, for example, Arbital seeming less promising to people / getting less engagement, because all the users have just sunk energy into trying to get by on a revived LW?

There's an intuition pump I could make that I haven't fully fleshed out yet, that goes something like, If both Arbital and Lesswrong get worked on, then whichever seems more promising or better to use will gain more traction and end out on top in a very natural way, without having to go through an explicit failure of the other one.

There's caveats/responses to that as well of course — it just doesn't seem 100% clear cut to me.

Comment author: casebash 27 November 2016 06:00:56PM 1 point [-]

If Arbital provides a solution, then that would be great, but I think it is best to have multiple projects operating at the same time.

Comment author: Alexei 28 November 2016 12:28:49AM 4 points [-]

Why?

Comment author: casebash 28 November 2016 02:13:37PM 2 points [-]

Gives us two changes to succeed.

Comment author: Qiaochu_Yuan 30 November 2016 07:50:02PM 4 points [-]

But also weakens both options' ability to be a Schelling point.

Comment author: Drea 11 December 2016 07:59:36PM 0 points [-]

I can see value in having LW as a prototype or scratch pad, making simple modifications of existing discussion platforms (e.g. improved moderator powers as discussed above). Then Arbital can do the harder work of building a collaborative truth-seeking platform, adding in features to, for example, support Double Crux, fine-typed voting, or evidence (rather than comments).

Perhaps in the end there's a symbiosis, where the LW is for discussion, and when a topic comes up that needs truth-seeking it's moved to Arbital. That free's Arbital from having to include a solved problem in it's code base.