The Writer’s Coin

I’ve always believed that word of mouth was the best way to learn about new books that I might like to read. That’s why I use Amazon so much—I love seeing what lists people have that include books I’ve enjoyed. It’s fun and it exposes me to new books I never would’ve gotten into (like The Raw Shark Texts—which I thought was OK but was very interesting and different).

But finding books that way relies on subjective opinions by people who often aren’t very good at expressing themselves in writing. Which leads to a whole lot of fluff or outright gushing about a book (“I loooooved it!”). Well, yesterday I found a new way to find new books based on the ones you’ve already read and liked.

Billed as the “Pandora for books,” BookLamp recommends books based on things like writing style and pacing. Granted, it’s in beta right now, but you have to check out the long-ish video to get the full effect:

If you’re familiar with Pandora, then the idea of the site will immediately hit home: they analyze books and classify them according to certain parameters. You tell the system which books you’ve read and liked, and it’ll go out there and find books that you may like based on those you already read. Brilliant!How do they do it? Well, you have to watch the video to find out but by scanning books and looking at things like adjective counts and first vs. third person, there is a surprising amount of stuff you can learn about a book. It may sound like hogwash, but during grad school we took several classes on writing style and this stuff works, trust me. When you start paying attention to how many commas and how long sentences are, you start to see patterns that tell you quite a bit about the books.

I can’t wait to see what this recommends to me once it’s up: Infinite Jest and People of Paper = ??

Tags: amazon, Booklamp, Books, Pandora, reading, recommending new books, scanning, word of mouth