Syndication Design Patterns
For a side project at work I’ve been poking around with a plug-in layer for our software that would make it really easy to expose various subsystems as a syndicated feed item. The idea is to provide some out-of-the-box RSS feeds that monitor system status, and show our customers how we could easily add similar functionality for their domain-specific needs.
This sort of feed, the kind where each item is periodically refreshed rather than replaced, is a different pattern than the standard feed most blogs and news services provide. The “typical” feed pattern is a fixed number of items, with the oldest being dropped off when a new one is added.
I have tried to document the plug-in layer in such a way that my coworkers and customers understand these different approaches, and hopefully begin to dream up new ones. Unfortunately, I keep bumping into this need for a new taxonomy to describe the approaches. What I really need is a vocabulary of syndication design patterns. If I need it, my guess is that other people are going to need it too.
As of August 9, 2005, the phrase “Syndication Design Patterns” yields only one hit in Google. It’s a Dutch keyword spam page. So it would seem that this topic is ripe for development.
For the purposes of starting a discussion, I have stood up a page on inkBlots called Syndication Design Patterns. This is absolutely not the best place for it… consider it a placeholder until I find a good community home (a wiki most likely) where some real mindshare can take place. In the meantime, please feel free to drop me an email (see the above page for info) and I’ll evolve the page until we find a better home for it.
