Re: consequences of hashes controlling cache refresh
Bill Newman <[email protected]>
| Newsgroups | gmane.ietf.medfree |
|---|---|
| Message-ID | <[email protected]> |
Larry Masinter wrote, quoting me > > Since > > I use CONNEG indirection to propagate changes in corporate policy, > > This is an unreasonable application of CONNEG indirection. > The purpose of the syntax is to represent the capabilities > of a recipient. It sounds like you're trying to use indirection to > solve a very different problem than the one initially proposed, > which is to compactly represent what otherwise would be a large > capability expression. > > It sounds like your goal is to reflect a centralized policy > maintained by a policy service, in a regime where the content server > obtains the policy from the centralized policy service but > the client does not! > > If this is really what you want to do, then the hash isn't > needed at all, you just want the URL. However, it's a little > odd to posit that the client doesn't actually find out the > capabilities that it's requesting! > > [etc.] This isn't actually *my* goal, it's my understanding (and quite possibly my mistaken understanding) of a scenario from Graham Klyne's "Common feature sets (V2)" posting of February 1: + (S3) Providing an expression of corporate policy with respect to + media capabilities. For example, an organization might declare a common + capability for incoming Internet faxes, so that detailed capabilities of + each internal system do not have to be published. My story was only intended to show why I think this scenario is not very consistent with the use of the hash both as an abbreviation for the feature set and as notification when to flush the cache. I wrote the story in the first person, but I didn't intend any special significance to that -- certainly not that I have a pressing need to have a corporate fax policy. My primary interest is in saving bandwidth on slow mobile links; if corporate policy can be expressed by indirection, too, that's nice, but I don't consider it of primary importance. Sorry about the confusion.. Bill Newman [email protected]