Monday, 16 June 2003
Weblog data modeling
Couple of things out of the way first;
‘log’ is confusing; I thought he was talking about logfile formats when I first saw it. Call it a Weblog, don’t try to invent new terminology! (and see below for an even better reason not to take the ‘Web’ out of ‘Weblog’)
‘well-formed’ doesn’t make much sense to folks coming from an XML angle, where the phrase means syntactic correctness; ‘valid’ might be more appropriate. What he really wants to do, I think is create an ontology, but let’s not go there for now…
OK, now that the inner pedant is sated, I’m really glad that Sam’s bought into doing public work on Wikis. It makes sense, and I’m fairly optimistic about the outcome.
When I think about this problem space, I get confused really quickly. ‘What is a Weblog’ is a question that has been doing the rounds for a little while, and I honestly don’t think it’s much more than a group of people’s ideas about how to use the Web. Therefore, any discussion on modeling Weblogs is really about modeling the Web and its content, and providing tools for extending it.
One of those extensions of the Web is RSS, which has a lot to do with Sam’s Wiki (obviously). I’m not sure that going back to first principles on RSS is going to get us any further, but what the heck. We’ve tried everything else.
I do see clearly that people’s interests in RSS are continuing to separate into two very different groups; weblogs (content syndication) and news (metadata syndication). If Sam and crew make that split more pronounced, maybe we can make some progress, but I have a suspicion it’ll all need to come back together in the end.
I’ll make my technical comments in the Wiki, where Sam intended.