Re: what is discovery - One concrete proposal

On Thu, Oct 10, 2002 at 02:51:35PM -0400, Heather Kreger wrote:
>       - Services are described with WDSL (i.e. a formal document)

Well, our definition currently says;

  "whose interfaces and bindings are *capable* of being defined,
   described, and discovered as XML artifacts" (emphasis mine)

I think this is important, since for those late-bound services whose
interface is described by an application protocol, WSDL is nowhere
near capable enough to describe them.  RFC 2616 isn't 175 pages
long for the heck of it 8-); it really does take that much prose to
describe the interface.

That's why I think we can't require WSDL.  But I still think our
definition is ok, since RFC 2616 could conceivable be "mapped into XML"
(whatever that means).  And I'm also happy to continue to document a
SOAP+WSDL centric architecture.

MB
-- 
Mark Baker, CTO, Idokorro Mobile.  Ottawa, Ontario, CANADA.
http://www.markbaker.ca             http://www.idokorro.com