So I’ve started work on my own small RSS aggregator for some or other web project I may or may not actually complete. And no, it’s nothing like Gregarius, it’s more of an ‘internal function’ of a larger project.
So anyway, after checking out the various RSS version specs and things, I hunted down as many feeds as I could to get an idea of the kinds of data I’m going to end up sifting through.
Wow. Despite the fact that there are standards out there doesn’t seem to mean much. Nearly every feed is a world apart from the next one, either throwing in millions of useless custom tags, renaming standard tags to some other random thing that made sense to the author and nobody else, leaving out loads of actual useful information, mixing and matching the specs as they feel the urge, and a million other randomnesses.
Anyway, on the way to making sense of it all, I fed some of them through MagpieRSS, which actually does a fairly reasonable job of making them a bit more sane. Still, I have to guess a lot of fields and things, and pretty much hope for the best.
At the moment, people’s RSS feeds generally seems more psychotic than some of their use of HTML.