Self-experimentation with comments

Since we believe in self-experimentation (by the way, Seth Roberts is now guest-blogging over at Freakonomics), I recently ran an experiment with comments on this blog.  The results:

1. Visitor stats rise considerably.  But this happens so quickly, I believe it is people hitting "reload" to read additional comments, rather than more readers.

2. The more that comments are regularly available, the more rapidly the quality of comments falls.  The quality of comments stays high when it is periodic, not automatic, and when we request comments specifically.

3. The quality of comments is highest when the matter under consideration involves particular facts and decentralized knowledge.  Posts which mention evolution, free will, or Paul Krugman do not generate the highest quality of comments.

So my current sense (Alex chooses his own course, though I believe he agrees) is to ask for comments periodically rather than always having comments open.  The goal is to maximize the real value of comments, rather than the number of comments (or measured visits) per se.

Comments on this post are not open!