Originally Posted by kovidgoyal
@Hadrien: Are you going to support RSS feeds that don't have content embedded in them? That way you can undercut Amazon's newspaper service as well.
You mean extracting the content like you do with web2lrf ? My first priority with RSS feeds is to create a real cache system and improve the overall speed and support of the feeds.
The only problem that I have with extraction pattern is that they're really geeky, it's hard to make any of this easy to use for a normal user. Some partial feeds are transformed and redistributed as full content feeds by random people on the net (BBC). It's also possible to create your own extraction pattern using tools like Dapper.