I started writing this today and lost it. So I am starting over - 8 hours later and after I've read Danny's two responses to Udell's piece.
First an intro to what's going on. It's kind of interesting. Jon Udell posts a piece called - An RSS/RDF epiphany - which mentioned emails and conversations with specific players in the 'RDF camp' - true believers in RDF and the future of the semantic web - done THEIR way.
I mean here's Jon Udell, trying to make hide or hare of all this stuff, and Bill Kearney gets all pissy - 'cause he doesn't go ALL THE WAY. Danny Ayers (of course) is upset at Udell over the same things that Kearney is, but I'm afraid these guys are missing the big picture - Udell's trying to extend an olive branch and say: "can't we all play together?"
Udell's recent focus on thinking about having a namespace work in both RDF/RSS 1.0 and RSS 2.0 is heroic. I applaud him for it. Just because he doesn't quite get it ALL doesn't mean we have to get all upset at him.
OK - so here's the complaints from Bill and Danny - and then I'll put MY answer below..... [I hope that makes it clear what's going on here - you have to read Udell's piece.....]
Enlightenment is soooo elusive. Jon Udell: An RSS/RDF epiphany I've met Jon, read his book and have followed his comments over the years. He's... [Bill Kearney]
Jon Udell quotes me (and danbri and others) but the order in which he presents the quotes is a little strange. Anyhow, below is how I responded to a mail from him which he quotes somewhere in the middle of his piece, without the response....... [Danny Ayers]
I was just out watering the garden (in the dark) and the reason Jon Udell's post irritated me dawned. The recent problems under discussion (and Winer's funkiness thing) are all about RSS 2.0 lack of real support for extensions.
Namespaces only started in the Userland branch in 2.0, following demand from XML developers. With namespaces came the problem of interpreting the extra material. But this problem had already been solved in RSS 1.0! It was only bloody-mindedness that prevented RSS 2.0 being a unified syntax which inherited the solution.
The namespace interpretation issue does extend a lot wider that RSS 2.0. But that's no reason to go on the offensive against RDF, something that's actually solved the problem! [Danny Ayers]
Marc's response.....
There's good stuff in all this. Udell's article proudly points out (from a quote by Stefano Mazzocchi) :
The mental model that XML promotes is basically a tree of couples.
The mental model that RDF promotes is basically a collection of triples.
Sounds familiar doesn't it? The Hierarchical vs. Relational war over again 30 years later?
He's right - it is about graphs versus trees. And don't get me wrong, we make an on-line outliner, so I'm all into hierarchical trees. But needless to say the world is rich place, and arguing that everything can be done simply - is just naive.
One of Danny's comments brings up something Tim Berners-Lee said: Every aspect of the internet should function as a web, rather than a tree structure.
That just about sums it up. You can have simplicity, but you're never gone build anything complex or interesting. Simple is as simple does (is how Forest Gump said it.)
The kind of research that Udell calls for in this article is actually happening right now - and once Matt and Paolo put out their k-collector plug-in for Moveable Type - everyone can do it. At that point RDF heads can use ENT and build topics into their RDF/RSS 1.0 feeds - which can then be read by ENT aware aggreagtors and new school tools - regardless of whether or not the feed is a RSS 1.0 or 2.0 format!
That will be the first 'deployed' cross-namespace - to my knowledge. It's time to connect these two worlds together - and get on with it.
