Less Wrong: Progress Report
Less Wrong is emerging from beta as bugs continue to get fixed. This is an open-source project, and if any Python-fluent programmers are willing to contribute a day or two of work, more would get done faster.
The character of the new site is becoming clear. The pace of commenting is higher; the threaded comments encourage short replies and continuing conversations. The pace of posting exceeds my fondest hopes—apparently not being able to post automatically on OB was a much greater barrier to potential contributors than I realized.
We’ve had 12,428 comments so far on 113 articles, 100 of them posted since contributing was enabled for all users over 20 karma on March 5th.
Browsing to the Top Scoring articles on Less Wrong will give you an idea of how things are developing. A quick view of all posts can be found here, with the current top scorer being “Cached Selves” by Salamon and Rayhawk, followed by “Rational Me or We?” by Hanson. If this looks like a blog you like, go ahead and add it to your blog roll now, please!
Yvain has emerged as a prolific and highly upvoted contributor with too many excellent posts to mention, but The Apologist and the Revolutionary (on brain damage and rationalization) and The Least Convenient Possible World (an exercise in not avoiding painful questions) are two places to start.
Our most highly commented thread was And’s Closet survey #1: What do you believe that most people on this site don’t? with 314 comments.
Johnicholas brings us Information Cascades showing how taking other people’s ratings into account in your own vote vastly decreases the information content...
...which inspired Marcello to build the anti-kibitzer Firefox extension that hides comment authors and vote counts. (We’d like to integrate these sorts of features into the site, but we need more Python programmers!)
Alicorn observes that systems of rationality must have two tiers: an ideal tier, and a tier that can actually be implemented (analyzing consequentialism as an example).
Kaj_Sotala asks whether blind review slows down science by preventing old scientists from championing new ideas.
MBlume asks what resources are available for raising young rationalists.
Jimrandomh notes that support can sound like dissent, creating a false picture of the overall reaction, and suggests prefixing “I agree with your conclusion, but...”
Thomblake on “Is Santa Real?”
Carl Shulman warns us, “Don’t Revere The Bearer of Good Info”
Patri Friedman on “Individual Rationality is a Matter of Life and Death”
...and the list goes on. And any number of highly intelligent comments are being upvoted to the prominence they deserve.
It might be just my imagination or my prior hopes, but it looks to me like the threaded, rated, and sorted comments create a completely different experience of reading a post—the first comment you encounter is going to be something highly intelligent, and then right away, you’re going to see the most intelligent reply and a well-sorted discussion all in one place. Much more of the action is in the comments.
The karma system is giving me valuable (if not always pleasant) feedback about which of my posts and comments my readers actually like. I shall try not to be too influenced by this.
An on-site wiki is on the way, and meanwhile there’s a temporary Wiki hosted at Wikia, currently with 163 articles.
I definitely agree that LW’s structure encourages participation (I have rarely contributed to OB) and the to and fro of comments gives valuable information on what/who people are in this “rationalist community” and where you stand.
However “the first comment you encounter is going to be something highly intelligent” is sales-talk and highly ridiculous. The first coment you encounter is just as likely to be runaway conformity. I would suggest that the pressure to conform is high and much of the intelligence is being used to signal logical dexterity on things with very little practical benefit.
It is my impression that LW is a tight community with little tolerance for what falls outside Eliezers definition of rationality and how rational people express themselves. I do not think this description will be accepted by Eliezer or the other contributors (and it would never be one of the first comments you met on a thread) but maybe they are Just Wrong.
I fled Him, down the nights and down the days; I fled Him, down the arches of the years; I fled Him, down the labyrinthine ways Of my own mind; and in the mist of tears I hid from Him
I think there is some truth to Marshall’s critique and that the situation could be easily improved by making it clear (either on the “about” page or in some other high-visibility note) what the guidelines for voting are. That means guidelines would have to be agreed upon. Until that happens, I suspect people will continue to just vote up comments they agree with, stifling debate.
I’ve previously suggested a change to the voting system, but this might require more man-power to implement than is available.
I don’t pay much attention to karma—but it is weird what gets voted up and down.
For a rationist community, people seem to go for conformity and “applause signs” much more than I would have expcted—while criticisms and disagreements seem to be punished more than I would have thought.
Anyway, interesting raw material for groupthink studies—some day.
My first comment is a month old (as is Eliezer’s original and now edited post). A month is a long time and in that time I was “hounded” out of LW. I think I was the first who experienced Eliezer’s idea of policing the walled garden from vandalism and entropy. However to an outsider this looks more like a recipe for political correctness. In other words the uniformity of thought on LW is rather high and the place does seem rather boring. Motivated rationalists searching for a raison d’être—be it nusery rhymes, losing weight, procrastination or missionary behaviour with daring programmes for eradicating other’s irrationality. In a sense it all rather seems like one big Multi-Player-Game in Language. Scoring points and distributing hits with rather transparent strategies for how to proceed. In my worst moments I sometimes think, that Eliezer has opened “the box” and the “thing” needs an army of obedient servants. LW would be the first place to start recruiting.
Looking at:
http://google.com/search?q=Marshall+site:lesswrong.com
...there were about 500 comments involving “Marshall”—and now they all appear to have been deleted—leaving a trail like this:
http://lesswrong.com/lw/9/the_most_important_thing_you_learned/53
Did you delete your account there?
Presumably he was karma farming—under the old system you got 1 point of karma for each comment you made, so he made a bunch of comments and then deleted them. (This has been corrected in a recent update.)
Yes Tim I deleted my account. Eliezer explained to me that I was not “ready” to comment om LW but I was welcome to continue reading. It is thus a little insulting, when Eliezer now says, I was “karma farming”. I was contributing as best I could and consequently down-voted. I asked several times why I was down-voted and Eliezer himself answered by asking everyone to down vote me—just deserts because of “vagueness”. Several of the articles I posted were never commented on. This does not sound like farming to me and why on earth would an adult man wish to collect pixel points? Eliezer’s answer here reveals a moral weakness. And that is bad karma.
The general rule in groups with reasonably intelligent discussion and community moderation, once a community consensus is reached on a topic, is that:
Agreement with consensus, well articulated, will be voted up strongly
Disagreement with consensus, well articulated, will be voted up and start a lengthy discussion
Agreement with consensus, expressed poorly, will be voted up weakly or ignored
Disagreement with consensus, expressed poorly, will be voted down viciously
People who complain about groupthink are typically in the habit of doing #4 and then getting upset because they don’t get easy validation of their opinions the way people who agree inarticulately do.
As an example on LW, consider Annoyance, who does both #2 and #4 with some regularity and gets wildly varying comment scores because of it.
Once we have more programming resources, we really need to test out separate agree/disagree and low/high-quality buttons.
EY,
Don’t you think having a button for low/high quality comments will increase comment length (which is bad)?
Marshall, I would keep in mind that good intentions are not sufficient for getting your comments up-voted. They need to contribute to the discussion. Since your account was deleted, we can’t to judge one way or the other.
I don’t actually know that separate agree/disagree and low/high quality buttons will be all that helpful. I don’t know that I personally can tell the difference very well.
I downvoted a comment once. I happened to disagree with it, but I downvoted it because the reasoning seemed egregiously specious. However, I’m not sure that that can really be separated in people’s minds as much as they think. It could be that, in reality, the main reason was I disagreed with it.
I doubt that there’s any way around this. People are biased against being able to tell the difference, I’d wager. Less Wrong isn’t going to have any better voters than other community content sites. Just Another Slashdot. (Mind you, I do sometimes read Slashdot threads. It’s not a terrible system.)
Do I have a configuration/preferences problem on LW?
When I follow the link to top scoring articles, I don’t see “Cached Selves”, I see “Instrumental vs. Epistemic—A Bardic Perspective” by MBlume. I never downvoted “Cached Selves”, in fact I believe I voted it up (the “+ Vote up” button is highlighted). After “Instrumental...” on the Top page comes “Rational Groups Kick Ass” with a score of only 17. Other posts on that page have scores of 1 and less. Is this a general bug or am only I seeing it?
Also, as for the claim that “the first comment you encounter is going to be something highly intelligent”, that doesn’t happen either. This post by Yvain on Extreme Rationality: It’s Not That Great has comments sorted by “Popular” yet the first comment has a score of 1 point, it’s from a few hours ago and has no responses. It’s not a terrible comment but it’s not one of the highest rated.
So what’s going on here?
Hal, sorting by “popular” doesn’t seem to do what you think it does (what it actually does I can’t figure out). Sort by “Top” to get ordered by score.
Also, anecdotally, I can tell the difference between “agree” and “high quality” to some extent. I’ve up-voted several comments I disagreed with because I thought they presented their idea well.
Thanks, ASA, that helps. How about http://lesswrong.com/top/, do you see Anna and Steve’s post first? Or Mike Blume’s?
Okay, with further study I fixed my problem on that page; there’s a (rather inconspicuous IMO) “Links from” button over in the sidebar which I had somehow/sometime set to “Today”. That apparently limits what is shown on every page. So I’ve had a skewed view of the site for some time!
Apparently the correct URL was http://lesswrong.com/top?t=all. I will try to get this fixed so as to be all-time by default.
If top is by score, then I think popular is by total upvotes, ignoring downvotes.
I prefer OB and don’t read LW as much, but that’s partly because there are other things competing for my time and you’d have to sift through a lot of mediocre stuff at LW to find the same sort of quality. I expressed my disagreement with Eliezer in his post on gardens, which I’d rather not reiterate here as others can read it there. Personally, I don’t do any voting at all.
This summary was very interesting. Even if you pull back your time commitments, It would be nice if you published progress reports from time to time. I’ve been trying to engage with LessWrong out of interest in it’s purpose and trajectory, but I missed a lot of the stuff you linked to. It was very helpful and has improved my opinion of the place.
I’m still mostly reading, but there is something nascent about it that intrigues me. I’m not sure that it’s right to judge things based on what they might become, but that’s all I have at the moment.