XML to RDF transformation

Sjoerd Visscher: Sam Ruby made an XSLT tranformation from Atom to RDF. He said it was hard to do. On the #echo IRC channel I said that XR would probably make it a lot easier. Then Sam asked me to make an XR transformation that would do the same thing. I did, and here it is.

Sweet!

 That's really cool.  Does that mean that it would be possible to basically leave the current Atom syntax alone and just specify a canonical (is that the right terminology) XR-based transformation as the "official" RDF representation of a given Atom feed?  That would certainly make the view-sourcers like me happy -- does it go far enough for the RDF afficionados?

Posted by Dave Walker at
The result of the transformation is good enough for the RDF afficionados, but what I hear is that they don't want to do the transformations.

Posted by Sjoerd Visscher at

Sjoerd: you're absolutely right.  Now that they have their transformation, they're complaining that they don't want to use it:

http://www.intertwingly.net/blog/1557.html#c1060877012

Sure, it would be simpler for a certain (very small) class of people if everything in the world were pure, perfect RDF that re-used existing popular ontologies and so forth.  But for the other 99.999% of the people who will touch this, that would impose an unacceptable burden.

This transformation seems like an excellent compromise.  People who want to use Atom as RDF can use Atom as RDF -- really useful RDF, not this halfway "hey I put an rdf:RDF element around everything, look at the pretty graphs" stuff.  We didn't even say "well screw you, come up with a transformation then" -- we actually worked with them to come up with one.  And now we're hearing that they (well, at least some of them) don't want to use it.

Just no pleasing some people, I guess.

Posted by Mark at

Mark:  I never doubted for a second that such a tranform could be done, nor have I ever underestimated it's value.  Saying that I "don't want to use it" is a simple misrepresentation, or misreading, of everything I've said on this issue.  As a hacker, I don't have the slightest problem using it.  I simply happen to believe that in the near future, non-hackers are going to start using RDF tools to do interesting things, and even hackers are going to use those tools for quick "I wonder if..." experinents which lead to interesting apps.  And I suspect that being able to seamlessly plug it into those tools will be the deciding factor in whether Atom will be used in those experiments or tossed aside as "too much trouble".

I'm not a master prognosticator.  I could be all wet on this.  Having the data in XML is a damn good start to future-proofing it.  I just believe that in the near future, direct RDF compatibility will be just as important to lowering the bar for higher-level toolsmiths and tinkerers as keeping the format eyeball-readable is to the current crop of potential aggregator and blogware authors.  Time will tell.

Posted by Avdi at

RDF, about recent documents found

A number of comments reflect on the activities described in Sam Ruby's Quantifying the "RDF tax" and XML to RDF transformation.... [more]

Trackback from the iCIte net development blog at

Add your comment












Nav Bar