« Discussion

Do You Trust a Computer to Tell You What to Read?

Some of the books I’ve enjoyed most are precisely those I’d have never picked up myself. Can computers replicate serendipity?

By Edward Nawotka

library computers

Let’s face it, the idea of a computer telling you what to read kind of creepy. But as algorithms become more accurate — largely by tracking the way you use a site like Amazon.com or B&N.com — they are closing in on the ability to predict what you might enjoy reading. With BookLamp’s book scanning project, it’s getting even more sophisticated. And it’s not happening just with books. AOL’s Editions iPad magazine actually promises to “read you,” which seems even creepier.

Still, as a reader, I know that some of the books I’ve enjoyed most are precisely those I’d have never picked up myself and either came to me through a friend, word-of-mouth, a helpful bookseller or simple chance. (I used to wander the stacks at Columbia University looking for books no one had seemed to check out.) Can computers replicate serendipity?

Let us know what you think in the comments.

This entry was posted in Discussion and tagged , , , , , . Bookmark the permalink. Both comments and trackbacks are currently closed.

3 Comments

  1. Posted August 24, 2011 at 8:20 am | Permalink

    My policy is very simple. Since I am using a Kindle, I am strictly ignoring any recommendation coming my way from Amazon. I make my own choices.
    I also had a Borders Rewards card, and I used to get quite some recommendations – and I ignored them right away.
    There are two reasons for my attitude towards algorithms based on my past preferences or orders:
    first of all, my tastes and preferences change all the time. And secondly: I am not a bandwagon person, and blatant advertising has always turned me off.

  2. Posted August 24, 2011 at 10:21 am | Permalink

    I have not doubt that a computer could make a recommendation that I would enjoy. But, I really don’t think it’s possible for a computer to often predict what I would MOST enjoy.

    Why? The algorithms they use will, at it’s most basic level, always be a function of my past behavior. The best finds, however, always have a randomness that is, literally, impossible to predict.

    The art of randomly browsing will, on average, yield more enjoyable results for me.

  3. Posted August 25, 2011 at 11:19 am | Permalink
  • Get Publishing Perspectives in your inbox each day and stay up-to-date on international publishing.