Another take: This site is dead with practically no traffic. LW2.0 has various issues and missing features: from a development team perspective it’s still in a lengthy beta phase but practically speaking and from a general user viewpoint it can be considered to have fully replaced this site.
Sorry, I used the ambiguous term “traffic”, meant “amount of new discussion/comments” rather than web traffic.
were it not for curi’s recent flurry there would almost be nothing here.
The biggest “issue” with LesserWrong right now is not whatever “features” are missing; it’s that performance on that website sucks, to the point of making it quite simply unusable. It feels like LW 0.2, not LW 2.0 - it’s even a lot worse than Arbital, which is hardly a high-performing website itself! The way I see it, everything else is secondary—unless this situation is improved well before the vote, I can only assume that lots of people will be voting against the merge, since LW-as-we-know-it would be dead either way, and the “against” option at least keeps archives easily accessible!
For the record, this is not what I, or even most of us (or so I would imagine) actually want! We want a usable LesserWrong, of course. But will we get it?
Well, that’s because the developers decided that it’s going to be a fancy-pants website where the browser is not much more than a Javascript VM. Let’s do everything in JS! That’s what all the cool kids are doing nowadays!
I think you’re right in a way, but it’s definitely a problem. LW 1.0 is easily usable on a mobile internet device (a tablet or even a phone!). LW 0.2 (sorry, I meant “2.0”!) is horribly sluggish even on a fairly reasonable desktop. How can we honestly expect such a site to ever become popular among the “cool kids” of today? (And come to think of it, plenty of “cool kids” read, say, Scott’s blog, and that’s a lot closer to LW 1.0 than to the newer version of the site—it certainly performs reasonably!)
The developers know that performance is the most important thing and are working on it. From my own experience, it’s already much better today than it was at the start.
I think I know what you mean—the site has recently become just barely usable on the simplest of its pages. But as soon as you do something that happens to poke the “JavaScript VM” the wrong way (crazy things like, idk, looking at recent postings by date, viewing a user’s recent contributions to the site, or even just opening a popular post w/ lots of comments!), it just grinds to a halt. It’s maddening.
The Lean Startup way suggest releasing a project early even when the first release has problems. I don’t think there’s a problem with LW2.0 being developed according to those principles.
The “first release” of LessWrong is, well, this site. What’s happening with LW 2.0 is actually called “introducing regressions”, and I don’t think the startup folks would endorse that. The combination of a full rewrite-from-scratch and a stringent deadline—the switchover was originally supposed to happen around this time, as far as I understand, albeit it has likely been postponed by now—is considered especially unwise.
Hopefully the LesserWrong folks can come up with something that’s genuinely usable—there are quite a few things I do like about the new site. But the challenges are just as real.
Another take: This site is dead with practically no traffic. LW2.0 has various issues and missing features: from a development team perspective it’s still in a lengthy beta phase but practically speaking and from a general user viewpoint it can be considered to have fully replaced this site.
I can’t speak for 2.0 but 1.0 still has traffic.
Sorry, I used the ambiguous term “traffic”, meant “amount of new discussion/comments” rather than web traffic. were it not for curi’s recent flurry there would almost be nothing here.
I suspect that if curi were not talking, other people would be.
The biggest “issue” with LesserWrong right now is not whatever “features” are missing; it’s that performance on that website sucks, to the point of making it quite simply unusable. It feels like LW 0.2, not LW 2.0 - it’s even a lot worse than Arbital, which is hardly a high-performing website itself! The way I see it, everything else is secondary—unless this situation is improved well before the vote, I can only assume that lots of people will be voting against the merge, since LW-as-we-know-it would be dead either way, and the “against” option at least keeps archives easily accessible!
For the record, this is not what I, or even most of us (or so I would imagine) actually want! We want a usable LesserWrong, of course. But will we get it?
Well, that’s because the developers decided that it’s going to be a fancy-pants website where the browser is not much more than a Javascript VM. Let’s do everything in JS! That’s what all the cool kids are doing nowadays!
I think you’re right in a way, but it’s definitely a problem. LW 1.0 is easily usable on a mobile internet device (a tablet or even a phone!). LW 0.2 (sorry, I meant “2.0”!) is horribly sluggish even on a fairly reasonable desktop. How can we honestly expect such a site to ever become popular among the “cool kids” of today? (And come to think of it, plenty of “cool kids” read, say, Scott’s blog, and that’s a lot closer to LW 1.0 than to the newer version of the site—it certainly performs reasonably!)
The developers know that performance is the most important thing and are working on it. From my own experience, it’s already much better today than it was at the start.
I think I know what you mean—the site has recently become just barely usable on the simplest of its pages. But as soon as you do something that happens to poke the “JavaScript VM” the wrong way (crazy things like, idk, looking at recent postings by date, viewing a user’s recent contributions to the site, or even just opening a popular post w/ lots of comments!), it just grinds to a halt. It’s maddening.
The Lean Startup way suggest releasing a project early even when the first release has problems. I don’t think there’s a problem with LW2.0 being developed according to those principles.
The “first release” of LessWrong is, well, this site. What’s happening with LW 2.0 is actually called “introducing regressions”, and I don’t think the startup folks would endorse that. The combination of a full rewrite-from-scratch and a stringent deadline—the switchover was originally supposed to happen around this time, as far as I understand, albeit it has likely been postponed by now—is considered especially unwise.
Hopefully the LesserWrong folks can come up with something that’s genuinely usable—there are quite a few things I do like about the new site. But the challenges are just as real.