- From: Mark Baker <distobj@acm.org>
- Date: Fri, 17 Jun 2005 10:37:09 -0400
- To: Paul Downey <paul.downey@whatfettle.com>
- Cc: public-web-http-desc@w3.org
On Fri, Jun 17, 2005 at 03:16:46PM +0100, Paul Downey wrote: > > On 17 Jun 2005, at 14:06, Mark Baker wrote: > > > >>Actually, I thought you meant that people would generally create such > >>libraries for each *MIME type* (or schema), > > > >Yes, that's what I meant, a library per media type, not one for each > >eBay or Google service. This would mean that for a very general media > >type like Atom, that most developers should be content to have a URI, > >HTTP, and Atom library in hand. Having more specific, possibly > >per-service libraries on top (even code-generated ones) may have some > >value in some cases, but as I mentioned before, I wouldn't want them to > >step on the toes of the URI/HTTP/Atom libs by defining things which > >were already handled in those libs... like operations, for instance. > > So would you anticipate one library for the generic XML, XHTML and RDF > media > types, even thought there may be one of a number of micro-data formats > embedded inside the single returned document? Sure, for XHTML+microformats & RDF, at least. Vanilla XML is a different story, and probably not one I need to get into here. Hmm, I suspect I'm missing your point, Paul. Mark. -- Mark Baker. Ottawa, Ontario, CANADA. http://www.markbaker.ca Coactus; Web-inspired integration strategies http://www.coactus.com
Received on Friday, 17 June 2005 14:36:28 UTC