I've been thinking about rebuilding the architecture and some of the design of this site to adopt to tags and XML. I'm starting to see the site as a large feed reader for my own content. The intruiging part is that if I rebuild this site to work directly off of my RSS feed then it will work on anyone's feed. The site becomes simply a database app for a standard type of data. I've always thought as websites as the result of database programs, but the more I grok RSS as a delivery and storage mechanism the more opportunities I see for working with it as the core architectural structure rather than an export or import protocol. Hopefully these ideas will become more clear as I build the next iteration of this site.
How do we know the origin of a blog posting after it leaves its author's website? If combining a collection of posts, remixing them based on an algorithm or community assesment, and distributing them as a new RSS feed is the model we are ready to embrace, how do we detect and prevent the inevitable fraud? For example, how do I know a map mashup that claims to be delivering an unbiased search engine's results combined with a map isn't actually eliminating selected matches for profit, censorship, or malice? How do I know if a news item with the URL of a famous news source is real or a press release if I don't get it from the new source's site. Of course we've had feeds for years, but after several layers of remixing, the purity of the stream is going to be questionable.
Sam Ruby is doing a thorough review of the Googlebase data formats and he isn't happy about their feeds:
None of the complex types are valid RDF/XML, and therefore can't be used in RSS 1.0 --also personals and news are incomplete. None of the guids in the RSS 2.0 feeds are valid permalinks. ... People who propose extensions should try to validate them first.
RSS Feed