steven0461 comments on Do we want more publicity, and if so how? - Less Wrong

17 Post author: Wei_Dai 22 August 2011 05:43PM

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

Comments (54)

You are viewing a single comment's thread. Show more comments above.

Comment author: steven0461 22 August 2011 11:47:55PM 2 points [-]

What I mean is in general the flood of people not familiar with the existing material, not reading it, asking old questions, engaging in half-arsed arguments, misinforming others, etc.

Those sound like low-quality contributions to me. If they're net harmful, we should agree to vote them down.

Comment author: gwern 23 August 2011 12:05:15AM *  7 points [-]

Any one comment is not harmful. There are tons of those in LW already on the -2 to +2 range. What's harmful is all of them put together. The response is to either not have huge surges, or to collectively fight the surge by collectively imposing higher standards and voting much more actively.

And as with collective solutions, there are problems with defection... No individual benefits much from going through a new page and up or downvoting each comment. To quote Buffet, I think:

"Overall, however, we've done better by avoiding dragons than by slaying them."

Comment author: steven0461 23 August 2011 12:30:55AM 3 points [-]

It isn't obvious to me that additional clueless comments are superlinearly harmful, that their harm outweighs the benefits of greater publicity, or that the problems with defection that you mention are serious enough to prevent a collective solution from working.

Comment author: Davorak 23 August 2011 05:28:15AM 0 points [-]

Can you think of other online communities that suffer or at least go through great and unpredictable change due to a high influx of new people?

Comment author: khafra 23 August 2011 12:38:37PM 2 points [-]

The only one I can think of that's stayed relatively high-quality for a long time is Hacker News, and they actively discourage large influxes--for example, by flooding the front page with posts on Erlang internals when mentioned in the mass media.

Paul Graham also does very active experimentation with HN's reputation system, which I like: There are karma threshholds for voting down comments, higher ones for voting down posts; you cannot vote down a direct reply to your own comment; you cannot vote down a comment more than a few days old (this one wouldn't work as well here). The most radical change he's made is that only you can see the exact karma for your own comments (although comments below zero are progressively lighter shades of grey).