Then again, webtests seem to get more comments, according to my recent entries page.

[chess]

Maybe it's not just that number of comments are inversely proportional to the length of the post - they're inversely proportional to the *amount of content* in the post, except for extreme cases (starting a meme obviously gets you lots of comments; being depressed tends to get you more comments).

If I had any invite codes left or any time, I might set up a new journal to test these theories. Maybe a statistical survey of my LJ-friends would do.

Comments

[originalauthor.livejournal.com]

This one is spot-on. I think that it has something to do with the ability to pick out *which* nugget of content to reply to out of a post with many. Does that make sense?

[rho]

Hmmm. The number of comments is inversely proportional to the number of seperate ideas made in the post. It is also proportional to how interesting the ideas expressed in the entry are, on average, and how important the ideas are (saying "I'm depressed" or "I'm in love" seems to get comments).

Which would mean, by my reckoning:

C = k Sigma(i) Sigma(I)
    -------------------
           n^3

This would mean that the effect of posting more than one thing in an entry is only canceled out if it what you are saying is both interesting and important.

This entry, for example, contains only one core idea, which is fairly interesting, but not very important, so it appears to be getting comments.

Oh, and it might be worth adding a constant to the end allowing for an extra comment or two for any entry that specificly mentions comments.

Or maybe I'm babbling.

[chess]

Meta-posts (posts about LJ, rather than about the poster's life) also seem to benefit from the extra-comment effect like posts mentioning comments.