- From: Mark Baker <distobj@acm.org>
- Date: Thu, 5 Dec 2002 12:20:09 -0500
- To: "Champion, Mike" <Mike.Champion@SoftwareAG-USA.com>
- Cc: www-ws-arch@w3.org
On Thu, Dec 05, 2002 at 11:54:01AM -0500, Champion, Mike wrote: > coordination, etc. at the messaging layer. The Linda / JavaSpaces / > TupleSpaces / XML Spaces stuff seems particularly relevant here. I believe > that Linda-esque coordination is considered RESTful by the cognoscenti, but > I'm not sure. (yes, I know, there's a page on this in the REST Wiki ... > I'll read it!) Well, kind of. There are a whole lot of coordination languages out there, but only one is "RESTful"; the one whose coordination semantics are applicable to all things with identity. But each of them shares the same desirable property of visibility (via late binding), since the agreement to communicate is also an agreement to coordinate. This is in contrast to the typical use of SOAP where the communication agreement is just an agreement to send bits. The Wiki page you're probably referring to is; http://internet.conveyor.com/RESTwiki/moin.cgi/CoordinationLanguage which is pretty raw. I just created it as placeholder to capture what I felt was an important point; that application protocols are coordination languages. BTW, it's also interesting to note that David Gelernter basically predicted the Web in his book MirrorWorlds (the first page, even), and most of his research at that time was around coordination languages, in particular Linda, which shares similar goals (and coordination semantics) to the Web. It just wasn't as well designed. MB -- Mark Baker. Ottawa, Ontario, CANADA. http://www.markbaker.ca Web architecture consulting, technical reports, evaluation & analysis
Received on Thursday, 5 December 2002 12:15:44 UTC