ranking

It’s old news (Boing Boing and Slashdot covered it a month ago), but Flickr’s patent application is a bit troublesome. It’s not that they’re trying to patent tagging (they’re not), it’s that they’re trying to patent the things library folks have been wanting to do (and in some cases actually doing) for some time.

Media objects, such as images or soundtracks, may be ranked according to a new class of metrics known as ”interestingness.“ These rankings may be based at least in part on the quantity of user-entered metadata concerning the media object, the number of users who have assigned metadata to the media object, access patterns related to the media object, and/or a lapse of time related to the media object.

See, interestingness is what you get when you link two or more metrics — think $interestingness = ($circulation * $comments * $rating); — together to get a number you can rank items by. I’d been playing with that sort of thing with bsuite, does that mean I might be subject to a lawsuit?

Danah Boyd posted about the biases of links over at Many2Many the other day. She looked for patterns in a random set of 500 blogs tracked by Technorati as well as the 100 top blogs tracked by Technorati. She found patterns in who keeps blogrolls and who is in them, as well as patterns about how bloggers link in context and who they link to.

The patterns Boyd points to would certainly effect the Google Economy, our way of creating and identifying value based on linking structures. And though she’s emphasizing gender differences, the patterns show broad differences in linking patterns between content types as well.

Discussion?