Roger B. wrote:
Requiring URIs in the content to be absolute to begin with is too expensive, and probably completely unenforceable.
Graham: I suppose that, if nothing else, it's a self-correcting problem. When all those hybrid RSS/Atom parsers out there display broken images or links, most publishers will get a clue and make sure they're putting absolute URIs in their feeds.
And if they don't mind broken images and links, well, I guess it wasn't much of a problem in the first place.
You are right. RSS has this problem already, and it isn't really that much of a problem, just a slight annoyance. As aggregators I've only ever used Radio Userland and Bloglines, and they both don't resolve relative URIs correctly.
But there still are plenty of RSS feeds with relative URIs in them. The workaround as user is to simply follow the link to the item first, and then click the link there on the original page.
The only thing that changes with xml:base support in Atom is that aggregators who *do* resolve relative URIs, now have more chance to get the originally intended base URI.
And a side effect is apparently that aggregator writers are alerted to the relative URI problem, instead of just staying ignorant about it.
-- Sjoerd Visscher http://w3future.com/weblog/
