Bubba has some interesting ideas around subscribing to an entire site instead of just "news." He is doing his Master's thesis on blogging so he has been spending a lot of time thinking about htis stuff.
RSS is a part of the puzzle but only captures a part of what is on a site. What would you need to move in to the post browser era?
- A "site.xml" file to describe the type of content available at the site.
- This could point to a set of blogs available on that site. Bubba (and a lot of others) expose more than one blog, or view of their blog. Perhaps this is just a pointer to an RSS file.
- Included in each RSS file a pointer to the archive for that is important too. Winer has some stuff on that a while ago that could easily be integrated here.
- A reference from the "site.xml" file to other resources available on that site. This could include a list of "stories" and photo galleries. Perhaps specific pages like a disclaimer/license page, a contacts page, etc.
- We could use XHTML for all unstructured content that isn't covered otherwise.
This is an incremental start on the semantic web. If we start with XHTML and start pulling stuff out on a piece by piece basis we can start to reduce the amount of unstructured content.
Any other ideas? Send me mail or post them to your blog.