I hope to blog something that's nonobvious — or at least not entirely obvious — based on the results of this poll (of course, if the results are as I expect them to be); so if you can participate, I'd be much obliged.

A thought about Cryptonomicon I had from reading the comments on the Yamamoto thread, and that was confirmed by the poll results: It's hard to argue with those who love it, or those who hate it. But lots of people — at last count, 158, including me, out of the 459 that voted — say that it's one of their favorite novels (and that doesn't count the ones who said that it's merely excellent but not one of their favorites). My guess is that there are few novels that arouse such enthusiasm, even if many novels arouse less hostility.

So if you spend $9 plus sales tax in a bookstore, or $5-$6 or so (including shipping) from an amazon used book seller, plus 30 minutes or however long it takes for you to get a sense of whether you like the book, you have a decent chance of getting a novel that will become one of your favorites. Since I get tremendous pleasure from novels I really love, don't much care about the $5-$6, and am often willing to risk the 30 minutes, that sounds like a good gamble to me.

More broadly, many choices are not just about how likely you are to like something, but how much of a benefit you're likely to get out of it if you like it, and how much of a cost it will be if you dislike it. With some forms of entertainment, the downside is unlikely, but the upside is pretty low. With others, including certain kinds of books (or for that matter with trying a new restaurant), the downside may be more probable but not very expensive, and the upside can be great. If so, it's worth a try.