- From: Mark Baker <distobj@acm.org>
- Date: Sun, 19 Jun 2005 21:24:22 -0400
- To: Mark Nottingham <mnot@mnot.net>
- Cc: public-web-http-desc@w3.org
On Sun, Jun 19, 2005 at 08:28:17AM -0700, Mark Nottingham wrote: > > On Jun 16, 2005, at 7:01 AM, Mark Baker wrote: > > > >Ok, Mark. I have some issues with those interfaces, but those are > >secondary to my major concern; is it your intent that the service > >provider be the one authoring instances of the description language > >which describe those interfaces? If so, I don't see it, for the > >reasons > >I've already given. If not, then we're in synch, but then this is a > >very different use case than what we've been assuming to date. > > That was one use case I had in mind, yes. Could you give a little > more context on "reasons I've already given"? Sure. It's just that the service provider has already defined the operations to their service by virtue of adopting HTTP. The operations a client might want to use from their programming language (e.g. "searchByISBN") are none of its business, and moreover, they'd be lying if they said they offered them... or else, if they actually did offer them, then they wouldn't be RESTful. Either way, it's not anything I think a Web friendly description language should involve itself with. Mark. -- Mark Baker. Ottawa, Ontario, CANADA. http://www.markbaker.ca Coactus; Web-inspired integration strategies http://www.coactus.com
Received on Monday, 20 June 2005 01:23:44 UTC