gwern comments on Assessing Kurzweil: the gory details - Less Wrong

14 Post author: Stuart_Armstrong 15 January 2013 02:29PM

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

Comments (6)

You are viewing a single comment's thread.

Comment author: gwern 16 January 2013 05:27:54AM 5 points [-]

18 people initially volunteered to do varying amounts of assessment of Kurzweil's predictions; 9 ultimately did so.

Not uncommon.

Comment author: EricHerboso 16 January 2013 01:53:44PM 4 points [-]

My initial impression was that the volunteer completion rate would be higher among a group like LW members. But now I realize that was a naive assumption to make.

Comment author: falenas108 16 January 2013 11:36:25PM 2 points [-]

Is 50% lower than usual? My intuition says the norm is between 15% and 40%, with ~60% confidence.

Comment author: gwern 17 January 2013 01:02:39AM 8 points [-]

Givewell's volunteer failure rate was apparently ~80%: http://blog.givewell.org/2011/07/13/a-good-volunteer-is-hard-to-find/ (LW discussion).

Comment author: Qiaochu_Yuan 18 January 2013 05:33:04AM 3 points [-]

Email is a horrible method of communication, and one of the things it's horrible at is getting people to do things. This is the price to pay for how convenient it is to send an email. No doubt requiring that volunteers make a phone call to offer to volunteer instead would decrease the attrition rate but it's unclear if you'd actually get more useful work that way.