- From: Kendall Clark <kendall@monkeyfist.com>
- Date: Mon, 16 May 2005 14:21:33 -0400
- To: Dan Connolly <connolly@w3.org>
- Cc: DAWG Mailing List <public-rdf-dawg@w3.org>
On Mon, May 16, 2005 at 01:03:39PM -0500, Dan Connolly wrote: > I agree that <service> can be left aside, but what other > HTTP binding details are missing? I can't think of any. There are a few; but rather than run them all down here, I'm just trying to get them sorted out and into the WSDL file instead. That's more profitable, I think. > I'd like to have most of the details of the HTTP binding in hand when > I put the question. Perhaps I chose the name of the issue poorly, > but I mean for a decision on this issue to specify > the HTTP query parameter names and describe them using WSDL. Ah, I still believe the actual parameter names can be overriden in anyone's <service> section of their WSDL, but maybe this is a "just don't do that" situation. Anyway, time will tell. And, yes, the name is slightly misleading, but no matter now. > I'm considering arguing for SparqlGraph and SparqlBindings > interfaces, and perhaps SparqlQueryAndTransform ... > but I'll do that separately, with suggested text and such, > if at all. As a WG member, I will oppose any XSLT additions. I'm perfectly satisfied by the orthongonality argument; there are existing XSLT services which will happily consume our XML output. That's the cleanest and easiest thing to do IMO. Just FYI. Kendall Clark
Received on Monday, 16 May 2005 18:30:58 UTC