It took longer than we hoped, but LessWrong 2.0 is finally ready to come out of beta. As discussed in the original announcement, we’re going to have a vote on whether or not to migrate the new site to the lesswrong.com URL. The vote will be open to people who had 1,000 or more LW karma at the time we announced the vote back in September, and they’ll receive a link by email or private message on the current LessWrong.com. If you had above 1000 karma in September and did not receive an email or PM, send an email to habryka@lesserwrong.com and we will send you the form link.

If we move, we will not break old incoming links to LessWrong.com.

We take rationalist virtues seriously, and I think it’s important that the community actually be able to look at the new implementation and vision and be able to say “no thanks.” If over half of the votes are to not migrate, the migration will not happen and we’ll figure out how we want to move forward with the website we’ve built.

Unfortunately, the alternative option for what will happen with the lesswrong.com URL is not great. Before I got involved, the original dominant plan was to replace it with a static HTML site, which would require minimal maintenance while preserving the value of old Sequences articles. So in the absence of another team putting forward heroic efforts and coordinating with Trike, MIRI, etc. that would be the world we would be moving towards.

Why not just keep things as they are? At the time, it was the consensus among old regulars that LW felt like an abandoned ghost town. A major concern about keeping it alive for the people still using it was that newcomers would read Sequences articles linked from elsewhere, check out the recent discussion and find it disappointing, and then bounce off of LW. This reduced its value for bringing people into the community.

More recently, various security concerns have made it a worse option to just keep old websites running – Trike has run into some issues where updating the server and antiquated codebase to handle security patches proved difficult, and they would prefer to no longer be responsible for maintaining the old website.

In case you’re just tuning in now, some basic details: I’ve been posting on LW for a long time, and about two years ago thought I was the person who cared most about making sure LW stayed alive, so decided to put effort into making sure that happened. But while I have some skills as a writer and a programmer, I’m not a webdev and not great at project management, and so things have been rather slow. My current role is mostly in being something like the ‘senior rationalist’ on the team, and supporting the team with my models of what should happen and why. The actual work is being done by a combination of Oliver Habryka, Raymond Arnold, and Ben Pace, and their contributions are why we finally have a site that’s ready to come out of beta.

You can read more about our vision for the new LessWrong here.

New Comment
65 comments, sorted by Click to highlight new comments since:

This has been some heroic work. This place is back to one of my favorite places to read for inspiration and learning. Huge congrats and thanks to the whole team.

Here's the main consideration from my point of view as an organizer:

Lesswrong.com is still the #1 source of newcomers to the San Francisco meetup.

We haven't had much chance to try out the new meetup functionality on lesserwrong.com yet, but we really need it to work. This is our community's primary source of new blood, and it is super important for that to be all set before the migration happens.

Agree. Something the old LessWrong did that the LW2 community page doesn't currently do, was that it displayed upcoming meetups on the side of every page, and in my experience as an organizer sometimes people would just stumble across the site for the first time, immediately see there was a meetup in my city, and show up. From a design perspective I know there's no way Oliver will go for having meetups display on the side of every page, but maybe we can do something to make nearby meetups comparably visible, because that does seem really important.

Yes, I've had many of our newcomers tell me that they showed up because of that feature (not because they went looking for in-person meetups). Discoverability matters.

Wanted to share our thoughts on this so far (we're currently pretty uncertain about how to weigh a large number of concerns together, but agree that meetups should be discoverable and are a really important part of the LessWrong community, and I'm interested in thoughts on how to go about this).

Throughout most of the web, you're barraged with distractions. You can't read a blogpost without seeing links to related content, notifications, etc, hyperlinks that force at least a mild decision of whether to keep reading the current thing or switch contexts or at the very least pause to open it in a new tab for later.

And this adds up to a major cost to attention, which is one of our most rare and precious resources. And a major design philosophy of LesserWrong is, as much as possible, to not be doing that. Each page on the site tries to focus your attention on one thing (with a plausible exception for the frontpage, but at least there's a one-thing of "figure out what you want to read in detail")

This does make a lot of design choices harder – if you're not willing to bombard your user with lots of things-to-maybe-pay-attention-to, it's obviously harder to offer them the ability to pay attention to things they haven't yet considered.

Sidebars lean heavily into the "offer lots of other things to think about" mentality, which is why we'd like to avoid them.

We do plan to feature meetups in some more obvious-fashion on the frontpage, although it seems like this may involve some careful rework of what the frontpage is currently doing. This wouldn't address the use-case of "gets linked to a particular blogpost, reads it, sees on the side that there's a meetup in their city, goes to meetup."

We are leaning towards tweaking the header-bar to contain more information (something like "transform what is currently the hidden hamburger menu into a menu-bar with more discoverability), which'll help a bit, but it'd feel weird to me to use that particular space for "meetup in your area!"

It seems plausible that once you finish a given article that maybe it should show some things you might want to do next. (I think if this is put all the way at the bottom of the comments it'd be so obscured as to be mostly-useless, and if it's put above the comments, it's a bit differently weird)

There's a whole separate issue, where it's making some kind of statement about what LessWrong is when social hangouts ["board game meetup near you!"] are presented on the front page (and a different statement if the only meetups on the frontpage are more intellectual-leaning), and figuring out how to set the right tone for the culture is a pretty big question.

This is all to say "yes, we're taking this seriously, but also it's hard." Solving the issue requires a fair amount of background on how everything fits together so most simple solutions will probably not be sufficient, but I'm interested in additional thoughts people have on it.

I vote for checking to see if there is a meetup coming up soon near the user's IP address, and if so, putting a notification below every post, above the comments.

[-]Elo20

It seems plausible that once you finish a given article that maybe it should show some things you might want to do next.

No. You can encourage authors to back link. But this is way too much like distraction.

So on one hand, this particular solution wasn't something I was seriously considering regardless. But, I'm curious:

a) are you therefore saying it's basically not worth it to include any kind of prompt for meetups on individual blogposts? (this seems fine and coherent, but my previous comment was basically oriented around "how to give meetups as much exposure as we can without being distracting or producing weird cultural effects" and am curious what your own thoughts on that are.)

b) I'm not actually sure what the concern here is (I do have a vague ugh field around sites that do this – i.e. when tumblr shows me three random posts I might want to read after a post, something about that feels annoying, but I'm not sure why. Distraction doesn't feel like the right word – I just finished a post, so it's not like it's ripping me out of my train of thought)

[-]Elo60

For B: I have an analogy - it's more like a Wikipedia article with relevant links interspersed than a clickbait article with a list of "you won't believe what happens when you use this formula on yourself", at the bottom.

[-]Elo50

For A: I prefer a permanent sidebar. I don't like floating, disappearing or overly complicated "features".

I'm going to second a permanent sidebar. Frankly, I wouldn't play coy with the concerns of in-person organizers. If they're not among your top three priorities then your priorities happen to be wrong.

We haven't had much chance to try out the new meetup functionality on lesserwrong.com yet, but we really need it to work. This is our community's primary source of new blood, and it is super important for that to be all set before the migration happens.

I agree that meetup functionality is one of the core features of the site; it just so happened that it was the last core feature that we built, and so hasn't had as much time to be polished as the others.

(For some context on our decision-making process, we thought that the site shouldn't be brought out of beta until we had all of the core features that current LW has, but also that we should bring it out of beta as soon as possible, and continue making improvements at the lesswrong.com domain.)

I have submitted my vote (my vote is "migrate later", so I did not check either box -- why was this not an explicit option?), but I want to repeat my voting comment here, in the hopes that other voters might see it and be inspired to say the same. Original comment follows:

My vote is "migrate later". What do you need to do first? FIX COMMENTING. The current comment composer is seriously bad. I have commented on this repeatedly on LW2.0 to no reply. (Examples: https://www.lesserwrong.com/posts/vtZsEerABCjhtgizX/beta-first-impressions#aamPf38rosSaiRw8s https://www.lesserwrong.com/posts/SQugQ9j7GeWzAuZec/1-5-2018-update-comment-styling#cTtt7yqyELnEHd9cs and more detail over on SSC at http://slatestarcodex.com/2018/02/25/ot96-snopen-thread/#comment-605324 )

[Added note not in original comment: I couldn't even copy and paste the above paragraph without it screwing things up, because not only did the stars get converted to rich text italics, an extra space was added after each, screwing up the spacing (there was a space between "bad" and the immediately following period.]

Seriously. Fix this. Its current state is not acceptable. (You should also make Markdown work in profiles again, but that's a lesser matter.)

I'm not at all disputing your points about the brokenness of the comment editor, so this comment is kinda unrelated, but since we are giving the devs criticism about some of the editing features, I also want to give them some praise:

I wrote my meditation post in a Google Doc originally, and then when I copy-pasted it to the post editor from there, it just worked perfectly right away, no editing needed. That is not the case with most online WYSIWYG editors that I've used. That felt amazing, and I imagine some special effort must have gone into that. Now I can just write any extended posts in a GDoc, collect comments and feedback on the draft, and then just copy it to the site without needing to do any editing to e.g. fix unnecessary newlines the way I would need to if I copy-pasted it into a WordPress editor.

The same happens when I copy paste my articles from the Wordpress.com platform, whether I do so from the editor or the published page. Everything except for images just magically moves and looks good with only minimal editing needed.

Huh. That's not been my experience with Wordpress.

Yep, I think for people who preferred the old commenting experience, greaterWrong is great. I personally do much prefer the WYSIWIG editor experience, even if it sometimes runs into trouble with markdown, but it seems totally reasonable to differ on that (and I do also really want to fix the bugs in the current editor, which also frustrate me quite a bit).

Actually, on that topic: We could just turn off automatic markdown parsing in our editor for now, and just have it do rich-text via keyboard shortcuts and the toolbar that shows up when you select text. And then we already have the option for a super-plain markdown editor that you can activate in your profile to replace the normal comment editor. Would people prefer that?

That's basically what I suggested with the defaults reversed, so yes.

Again, part of the problem, if you read my linked comments, is that the editor isn't WYSIWYG, it's worst-of-both-worlds. WYSIWYG rich text (with buttons always visible) would be an improvement, as would ordinary Markdown with no WYSIWYG component. I think the latter is preferable to the former, but either is an improvement, and allowing both is even better. I personally would have set the default differently, but that disagreement has no real relevance.

Cool, I created a new Meta post for this. If enough people agree then I think removing the markdown parsing is the way to go for now.

I am not a huge fan of permanently displaying formatting options at the top, because I do think it takes up a lot of unnecessary real estate and makes formatting comments feel a lot more intimidating than it has to. (Though it's definitely a much harder UI challenge to then still allow people to format things the way they want). I would be happy to discuss that more though, but if we do, let's continue that in the relevant meta-thread.

I would. WYSIWYG is a terrible editing paradigm, but some people like it, so I won't argue against providing it. Trying to mix WYSIWYG with markup-based editing, though, is far worse.

I further suggest that the plain option not be hidden out of the way. Make it selectable when commenting, and remember the selection. I wasn't even aware it existed until just now.

(edit: To be fair, I'm probably going to keep using greaterwrong regardless. Discount my opinion to whatever extent applies)

I would STRONGLY prefer that. I'm sorry for using strong language and I'm trying to be emphatic here, not upset or judgemental, but: I would describe the current state of the editor as "unacceptably broken" and "totally unusable (for anything at all complex)".

If you don't agree with me about "unacceptably broken", please reread the bug I filed about this: https://github.com/Discordius/Lesswrong2/issues/314 -- it's not just that automatically applying formatting is bad, it's not even just that it COMPLETELY breaks copy-paste and links that have underscores in them -- which would alone be reason enough to turn it off and leave it off -- the implementation of autoformatting also DOES NOT WORK RIGHT AT ALL.

[-]Elo20

if that were fixed?

Otherwise I think lesserwrong is awesome and is totally ready to replace LW. I have <1000 LW karma though, so I'm not vote-eligible. :-)

[-]Elo10

You tried GreaterWrong?

I don't think the existence of an alternate interface has any bearing on whether the main interface is good, or should be promoted out of beta. Practically speaking almost nobody will use the alternate interface. Me using it does not solve any problems except my personal ones, which are of no import.

[-]Elo20

As far as I can work out - most users I talk to are using gw.

While I agree there are quite a few people using GW, which I think is good and makes GW worthwhile, LesserWrong.com currently has 30 times as much traffic as GW. So I do think that the main site interface should be the key determining factor.

I do however think that the meta point, which is the existence of an API that allows users to create third-party clients and the proof-of-concept of GW makes a large difference of whether we should move to the new codebase, since it does make a much deeper level of customization possible.

I'm not. And I have given up on many comments because commenting on Android has been broken. It appears to be almost working right now, except the text is flowing outside the comment box. :S

Pushing a fix for that tonight! Sorry about that!

Which users you talk to is going to be such a biased sample though. I would expect lurkers both to be more likely to use lesserwrong and to be a majority of overall users.

I don't have strong intuitions about whether this applies to commenting though - it's plausible it doesn't.

[-]Elo40

We could just turn off automatic markdown parsing in our editor for now, and just have it do rich-text via keyboard shortcuts and the toolbar that shows up when you select text. And then we already have the option for a super-plain markdown editor that you can activate in your profile to replace the normal commentEditor. Would people prefer that?

I don't know about other people but I'm not sure what half of that means.

Ah, sorry. Let me clarify:

We have our current editor, which is a what-you-see-is-what-you-get editor that automatically tries to translate markdown you enter into bold. e.g. if you type <asterisk> text <asterisk> then as soon as you press space after the second asterisk, the text gets bolded and the asterisks disappear. This both preserves the WYSIWYG abstraction, while also allowing people to style their content via markdown.

But this also leads to the majority of bugs people have been experiencing with the editor. Sometimes when people post links with underscores, the underscores get translated as italics, and then the link breaks because the underscores disappear, which is what happened in the above comment. And so one option that would remove a lot of bugs from the editor would be to deactivate that automatic markdown parsing.

[-]gjm40

My understanding is that the single thing that causes most annoyance about the comment composer is that pasting in a URL with underscores in it tends to have horrible results. I agree that that's really bad, but note that there is a fairly decent workaround: enter some link text, select it, and then paste the URL into the URL-specific box.

(For the avoidance of doubt: I know Sniffnoy knows this, not least because the first URL in the parent of this comment is to a comment from Sniffnoy containing a link made in that way.)

Just voted Migrate, because Archive is not a free action - it makes a ton of content frozen at its original URLs with no way for the original authors to edit it.

For UI I'm mostly using greaterwrong.com anyway, so editor bugs etc don't bother me much. I only come to LW2 for the inbox, GW hasn't implemented it yet.

Oh, and thanks so much to Matthew/Oliver/Ben/Ray/others for reviving LW! It is again a great place to hang out.

Yep, thanks for your great work. I voted to migrate, especially if the old links stay; I'm currently posting most of my work here, rather than old LW or agent foundations.

When does voting close? EDIT: "This vote will close on Sunday March 18th at midnight PST."

I don't have 1000 karma, but I'd like to ask something:

Are you going to keep backups of the original LW1.0 database (preferably multiple copies/locations)? I'm worried that some of the content might not have made it through the conversion to LW2.0. So far, I've noticed that old polls (both the questions and the results) are missing, and there might be other little things like that even if the main content is preserved.

I'm tempted to use wget or something like that to download all pages off LW1.0, but I don't have much memory and it could take a long time. If I knew that you were going to retain all the original data for preservation purposes, I wouldn't be as worried.

Are you going to keep backups of the original LW1.0 database (preferably multiple copies/locations)?

We plan to. It's not obvious that those backups will be public by default--we receive database dumps from Trike, which contain both public content like comments and private content like private messages and email addresses--but if there's interest it doesn't seem too difficult to create a sanitized version of the database.

Yep, I think we should definitely keep a few copies of the db dumps we have, and have them saved on a few different machines.

It would be amazing if you published a sanitized version of the LW 1.0 DB. In addition to ensuring the content is preserved, that would also make it easier for people to do interesting statistical analyses.

This is a concern for me too. A suggestion I made in feedback: Don't break inbound links. Keep the old site, static, under archive.lesswrong.com or something, and redirect classic-format url paths to the archive.

There is a lot of valuable material on the classic site. It might not be useful for current discussion, but let's not lose it, or let it get buried on archive.org.

(come to think of it, if maintaining an archive is itself unworkable, a redirect to archive.org might be an acceptable next-best alternative)

Don't break inbound links.

I think I've said this before, but the most important feature of a redesign has always been "don't break inbound links."

We tried pretty hard to not break any incoming links, and have been watching the google analytics for the old site to make sure we covered all the inbound links.

Okay, cool. As long as it's on your radar.

I wouldn't trust a redirect to archive.org, because some of the content might have randomly been missed by the Wayback Machine crawler or the last crawled version might be missing comments that were added later. It also might have systematically missed certain things, such as deeply nested comment chains where you have to click "continue this thread" and posts with a lot of replies where you have to click "load more comments" (which is even less likely to be preserved, as it relies on AJAX).

Are you asking people about migrate now vs. not at all or are you also including migrate later as an option? Some people may want to hold off migrating because of a single bug, ie. not being able to comment on Android or lack of meetup support?

We do now actually have (very basic) meetup support! (And commenting on Android is also fixed!)

On the more object-level: I would quite strongly prefer to move now, but there is a commenting field in the form, so if a lot of people are saying yes conditionally on us fixing one specific bug, then I think it would make sense for us to fix that bug/missing feature before we move. In generla

Just tried commenting on Android, and it indeed now works!

[-]Elo30

Can you announce that. Can you also not ask for locations and instead ask people to input location? Like the old site. I don't like the privacy violation of letting my browser know where I am, but I want to be able to put in my city and find my local meetup. I also travel to other cities and would like to be able to see meetups before I get there.

Also my meetup has several organisers who will want to post events. Is there support for that?

Can you announce that.

We're basically giving it a day or so of "make sure we've caught obvious bugs" and "get a bunch of meetups listed" and then will have a big announcement.

Also my meetup has several organisers who will want to post events. Is there support for that?

Right now it's actually both the case that:

  1. I think you can add multiple organizers to a group (not 100% sure we shipped that but it was in the works), but, meanwhile:
  2. There are not currently restrictions on who can post an event for a given group. (This is just an interim thing because implementing a robust permission system is a fair amount of work, but meanwhile we wanted to remove trivial inconveniences to getting events up and populated)
Can you also not ask for locations and instead ask people to input location?

We do plan to implement both options.

[-]gjm50

If the vote comes out in favour of making LW2 the official New Less Wrong rather than giving up and just archiving, is old-LW karma somehow going to be ported over to new-LW?

Yep, that will happen before the move.

If you had over 1000 karma on LW 1.0, and we couldn't contact you at your email, I've now PM'd you (on LW 1.0) with your vote link.

The new site still really sucks on mobile. I have trouble logging in and also reading notifications and such. (More specifically, closing the announcement tab so I can actually read the reply I'm being notified about. On the other hand, the way it should work is indeed better than the old site. I'd say go ahead and migrate but please do something about my user experience on my iPhone 6S.

Yeah - fixing browser compatibility issues is among the top priorities. As Habryka noted last week we enabled a pure markdown editor which is automatically used by iOS and Android which avoids some of the problems interacting with mobile keyboards and the Draftjs editor. (Incidentally we've also added an option to enable the pure-markdown editor more generally in your profile settings)

Next week I'm hoping to get the site working properly on a couple versions of Safari that seem to be having trouble with it.

By the time I got around to casting my vote 7 days and a few hours after the vote was announced, the vote was already closed.

The reason I mention that is that I hope LW2 will in the future try to accommodate people who usually go more than 7 days between visits to LW2.

Alas, I am sorry! We definitely want to accommodate people who go more than 7 days between visits, and sent people an email and a PM for precisely this reason (so they wouldn't have to visit LW2 to notice the vote). But it might have still been better to have the vote for a bit longer.

I haven't gotten the voting link (I've now emailed to ask), but I am sadly already pretty negatively surprised at how lesserwrong.com has turned out (props to the maker of greaterwrong, though) and very much hope that it doesn't completely replace LessWrong.com. Even if LessWrong.com is just killed and made read-only (since after all the efforts to migrate people here, it is even more unlikely that the original lesswrong will get any new use), that's a better outcome for me.

I wouldn't even post this, but I hear a lot more people sharing the same opinion (selection effects apply), but (selection effects again) few of them are here to actually say it.

[-]gjm60

Would you care to explain what harm you find LesserWrong to be doing, that would make replacing LW with a static archive preferable to letting LW2 continue?

(That's a genuine question, not an attempt at snark or whatever.)

I would be interested in chatting to anyone who is disappointed by the new site. Also note that greaterwrong.com will continue to be available when we move, so if you like that site more, then you will be able to continue using that.

I was disappointed by the new site, but still voted to migrate. The conversation is here, and content is king. Despite my bitching, your team deserves a great deal of credit just for breathing life back into the community.

That being said:

Performance was a big complaint, and kept me off lesserwrong until greaterwrong showed up, but you already know about that and for all I know it may have been fixed. My complaints are less with lesserwrong itself than with modern web design in general, and are mainly variants on "use of javascript as a first resort instead of a last resort," "interfaces that want you to notice them," and "overly complex underlying mechanics." In short, lesserwrong may well be a fantastically engineered site, designed under a paradigm I am predisposed to despise. Discount my opinions accordingly.

(if I may nerdrage for a moment, the top navbar that folds down as soon as I scroll up, covering the text I scrolled up to see, is a common web misfeature that should die and its inventor should be forced to play a variant of the transparent newcomb's problem where both boxes contain tigers.)

Note that the fact that greaterwrong can even exist (that is, that there's an API with enough power and stability to make an alternate interface) is a huge win, and you and whoever else made the decision to allow third party clients deserve +gazillion karma for it. ...but I still have to complain that said API is not documented.

Checking lesserwrong itself for the first time since GW became available, it looks as if it has improved. The comment box in particular is less odious, and the site as a whole no longer seems to grind my browser to a halt. These are good improvements. I'm sure there are other things not obvious at first glance.

If you do want to chat with a quasi-naysayer, I'm on the LW Slack as Error, on Freenode #lesswrong as ehs, and on xmpp as error@xmpp.feymarch.net. I'm best reached in the afternoons, eastern time.

the top navbar that folds down as soon as I scroll up, covering the text I scrolled up to see, is a common web misfeature

That definitely gets my vote for my biggest irritant on this site. The first thing I did after creating an account is to look in preferences for a way to prevent the bar from ever appearing. I suppose I am in the habit of being able to use the up-arrow key to scroll the window up a little bit, reliably, so that once I've hit the up arrow key and the page changes in any way (and my browser is the app that has focus) I can assume the scroll happened after the briefest glance at the page, and I can immediately turn my attention to something else.

When I need to interact with the elements that now live on the bar, I would be prepared to scroll to the top of the page or (if I don't want to lose my place on the page) to visit LW2 in a new tab.

I'm also prepared to install a Chrome extension if that is the easiest way for someone to implement what I'm asking for.

Happy to answer questions sent to hruvulum@gmail.com.

If we migrate, will some lesswrong.com urls become broken?