- From: Williams, Stuart <skw@hplb.hpl.hp.com>
- Date: Fri, 3 Aug 2001 12:09:00 +0100
- To: "'Mark Nottingham'" <mnot@akamai.com>
- Cc: "'xml-dist-app@w3.org'" <xml-dist-app@w3.org>
[Resending to xml-dist-app] > Wow - now that's a message. *grin* I'll take that as a compliment, Ta :-). > The whole hasn't sunk in (and I'm not sure it will by Thursday!), but > one thing caught my eye - > > > One of the ways to maintain context is through cookies (of course > > values encoded in request-uri's is another). If an HTTP server were > > to anticipate a response to a message it sends in an HTTP POST > > response, it could include a cookie (a locally significant > > reference) that would get returned by HTTP in a subsequent POST > > request bearing a response from the client to the server. Again, I > > haven't work this through fully, but it gives a glimmer of a > > mechanism that enables the '...with corellation' to be symmetric > > over HTTP. > > I'd hope to discourage this, as cookie-stripping HTTP intermediaries > are becoming increasingly common, and there has been discussion of > P3P being implemented by intermediairies as well (to the same > effect). > > OTOH, most any block information can be serialized into an extension > HTTP header (e.g., "X-SOAP-Block-Correlation: foo; > ns=http://bar.com/" > ), which is really where this is leading to. No problem... it was literally a half-baked idea that had been in my head around ways (with HTTP) that one-way with causality/correlation might be made 'functionally' symmetric even if the mechanism involved were assymetric. I hadn't considered cookie stripping HTTP intermediaries which would create 'a bit of a problem' (UK understatement!). A different HTTP extension header in the HTTP wrappings could certainly do the trick, as you suggest, although I'm not at all sure how the pragmatics would work with installed base HTTP servers and implementation of the semantics that would go with such a header - I defer to those who know better than I how these things work. Another alternate is also Glen's public facet approach which would do a similar job, but place the 'header' inside the XML envelope... but of course that means cracking the XML to get at it... > -- > Mark Nottingham, Research Scientist > Akamai Technologies (San Mateo, CA USA) Regards Stuart
Received on Friday, 3 August 2001 07:09:05 UTC