Google Data

One of the great things about Reader, and feed readers in general, is that they let you follow sites much more efficiently than you could just by visiting them directly. This means that once you get into the flow of it, you can subscribe and keep track of many more sites than before. The problem then becomes: what should you subscribe to? For example, there are millions of blogs, and while a few are well known, there are many more interesting (micro) niche blogs that would be great to subscribe to, if only you could find them.

To help with the discovery of interesting sites to subscribe to, we just released personalized recommendations in Reader. When you visit our discovery page, you'll see quite a few feeds that we think you may find interesting. "Interesting" here is determined by what other feeds you subscribe to, as well as your Web History data, all taken into account in an automated, anonymized fashion. (To learn more about how our recommendations work, see our help article about them). Hopefully you'll find your recommendations interesting and helpful in getting your information-triggered dopamine squirt.

Recommendations make it even easier to subscribe to lots of feeds, so then the question becomes: how do you organize those feeds better? As luck would have it, one of the other features we're announcing today is drag-and-drop support for your subscriptions and folders. You can now easily move feeds between folders, as well as reorder things up and down within the list.

On the right, you can see Mihai's subscription list as he uses drag-and-drop to reorganize his froworkers folder. (Though you may not want to let your friends know how you're ranking then, lest it become a competition.) Of course, we can't take too much credit for drag-and-drop subscription management -- desktop aggregators have had it for a while, as have online feed readers like Bloglines and NewsGator.

What these two features have in common is that (ex-)interns played a significant role in delivering both of them. Nitin Shantharam (UC Irvine) helped create the user interface for recommendations, while Olga Stroilova (MIT) had a hand in the algorithms that generate them. Meanwhile, drag-and-drop was the brainchild of Brad Hawkes (UMass Amherst), who was our intern the summer of 2006, and returned to join the Reader team full-time earlier this year. As RIE (Reader Intern Emeritus), Brad paved the way for the great student help that we've had this year. Also to be mentioned is Jason Hall (Kent State), who was behind many interface improvements that have been live for a few months already. All told, our interns (and ex-interns) have done great work this year!