Re: let's specify meaning rather than processing

I've just been looking at this in a little more detail.  It seems to me 
this could be a very important development.  I'm inspired to try to find 
a few spare moments and hack it into some stuff I'm doing around Jena. 
Maybe we could apply this to the SWCG calendar?

One suggestion:

As I understand things, XHTML is currently a special case, i.e. 
namespaces are mapped to scripts that extract the RDF, except XHMTL has 
the name of the script embedded in it.  It would be good to generalize this.

So I suggest that a script can return either - some RDF or the URI of 
another script to run.  These are easy to distinguish. This iterates.

That way:

   - a single generic process can be written to determine the script to run
   - any arbitrary XML language can have the script name embedded in it 
and still have the generic process find the right script

   - XHTML is no longer a special case

Brian

... another level of indirection.



Dan Connolly wrote:

> On Mon, 2004-01-19 at 08:41, Dominique Hazaël-Massieux wrote:
> 
>>Le sam 17/01/2004 à 00:00, Dan Connolly a écrit :
>>
>>>That was a great start. After working thru various technical
>>>details (grokRDDL and such) I found some inspiration to do
>>>some writing today. What do you think of this new draft?
>>
>>Looks great! (FWIW, I've fixed a few typos spotted by ,spell)
> 
> 
> I moved the history/rationale to separate page:
> 
> GRDDL Background: Design History and Rationale
> http://www.w3.org/2004/01/rdxh/specbg.html
> $Revision: 1.2 $ of $Date: 2004/01/20 23:23:51 $ by $Author: connolly $
> 
> <-
> http://www.w3.org/2004/01/rdxh/spec
> $Revision: 1.19 $ $Date: 2004/01/20 23:24:05 $
> 
> Tim, I'm thinking about specbg as a TAG finding on
> issue 35.
> 

Received on Friday, 23 January 2004 12:00:17 UTC