- From: Jonathan Borden <jborden@mediaone.net>
- Date: Tue, 28 Nov 2000 15:02:32 -0500
- To: "McBride, Brian" <bwm@hplb.hpl.hp.com>, <www-rdf-interest@w3.org>
McBride, Brian wrote:
> I've been out and have been trying to catch up on this discussion.
>
> I was kinda wondering how it got here. I have this idea stuck
> in my head that the syntax is there to represent the model, not
> the other way round.
1) I'm thinking more and more that the model is just an assembly language
and perhaps that is its beauty. As such the only serialization syntax it
needs is:
<rdf:RDF>
<rdf:Statement><rdf:predicate ...
<rdf:Statement><rdf:predicate ...
<rdf:Statement><rdf:predicate ...
...
</rdf:RDF>
2) Many people are more interested in how RDF can be used to process
documents which are on the Web, rather than merely as a way to build a
triple database that has no connection to the Web.
> So if I turn a model into an XML serialization,
> I'd like to be able to get exactly that model back again from a
> parser, without it feeling a desparate need for it tell me about
> the structure of the XML serialization.
See above serialization...
> Now if a parser offers
> me a choice, thats fine. But I don't see anything in M&S that
> says a parser MUST generate all that structure.
>
> Shouldn't simple things be simple.
Most things ought be exactly no more complex than they need be.
Jonathan Borden
The Open Healthcare Group
http://www.openhealth.org
Received on Tuesday, 28 November 2000 15:15:02 UTC