Conjecture: The set of popular textual advertisement themes for porn (e.g. subject lines of porn spam) undergoes systematic collective fluctuations. (New popular themes arise and disappear) These fluctuations contain a nontrivial signal about the current zeitgeist among the primary target audience (suitably subsection the audience if there is enough breadth in the market to justify so doing) and could be usefully mined for a lot of interesting information about what's going on in the country. Its fast response time makes it particularly interesting.
This could be a better signal than e.g. tabloid covers, since while they have similar fluctuation patterns, they seem to have settled down into a fairly constant set of themes that doesn't fluctuate nearly as broadly as porn.
Opinions?

March 8 2004, 10:40:43 UTC 11 years ago
March 8 2004, 15:11:59 UTC 11 years ago
opinion 2 = dude, you SO need a better excuse for staring at piles of
porn.
opinion 3 = i am disappointed. i think a Ph D should be able to craft a
better excuse.
opinion 4 = pathhhhhhhhhhhhhhhhhhbt! ;P
March 8 2004, 17:37:09 UTC 11 years ago
But a side effect of the Ph.D. is that, seeing porn, one is struck by the patterns, and becomes curious about whether they mean anything, or could be analyzed to tell us something about the world we live in.
(And yes, btw, this conjecture is perfectly serious)