How Buildings Learn, for social software

LibraryThing, a site for booklovers who catalog and review the books on their bookshelves is the opposite of those FaceBook junkfood applications designed to get you to use them once or twice, annoy all your friends, and move onto the next big thing. LibraryThing is deep. The social features of LibraryThing aren’t about popularity and making lots of friends, but the opposite — they are designed to help people find a few “like minds” with the same obscure shared interests. From the LibraryThing blog: the fun of LibraryThing isn’t just in the widely held books, it’s in those that are shared by only 10 or 20 other members. It’s easy to find someone who has read The Hobbit. Finding someone to discuss your more obscure books isn’t quite so simple.
Recently I listened to Jon Udell’s interview of Tim Spalding, founder and programmer at LibraryThing. Spalding designs LibraryThing for engagement and depth. It’s best customers are booklovers who put in the time, not only to catalog, rate and review, but to disambiguate titles, variants, translation, and authors helping to build a coherent database out of a gnarly ontological problem, and making the tool more useful for all.
In the interview, Spalding has an interesting insight about why Amazon’s customers don’t tag. When you’re browsing Amazon, your goal is to find books to buy, and to leave. You don’t have an incentive to stick around, to make the site better for yourself and others. LibraryThing is for connoissieurs, not shoppers. LibraryThing’s customers appreciate their collective bookshelp and want to keep organizing it and making it better.
Spalding approaches LibraryThing as a tinkerer, experimenting, remodeling and building wings and extensions. His recommendation tools are works in progress. He’s been gradually adding social features: groups, discussions, recommendations of others with similar tastes. It’s not a site designed to get big and get bought. It’s designed to continually add engagement for members.
Several takeaways from the interview about the design of social software
* Social Software doesn’t get “finished”. It’s more like a building, in Stewart Brand’s How Buildings Learn. Brand writes about how buildings are continually adapted, remodeled and refitted over time for new uses as its occupants’ needs change.
* Social software rewards depth over time. The joy of Facebook, Friendfeed, and Twitter is about letting people know what their friends are doing moment by moment. LibraryThing enables you to make a friend because you have the same 15-year old book (tip: you can run LibraryThing through FriendFeed to get the immediacy, too)
* Social software rewards deep engagement. The reason to add features isn’t because there’s a checklist, it’s because people can continue to do more valuable and enjoyable things in the environment over time

Leave a Reply

Your email address will not be published. Required fields are marked *