- From: Roy T. Fielding <fielding@gbiv.com>
- Date: Tue, 16 Jan 2018 13:25:44 -0800
- To: Chris Pacejo <chris@pacejo.net>
- Cc: RFC Errata System <rfc-editor@rfc-editor.org>, Julian Reschke <julian.reschke@greenbytes.de>, ben@nostrum.com, aamelnikov@fastmail.fm, adam@nostrum.com, Mark Nottingham <mnot@mnot.net>, pmcmanus@mozilla.com, ietf-http-wg@w3.org
> On Jan 16, 2018, at 12:34 PM, Chris Pacejo <chris@pacejo.net> wrote: > > Hi Roy and Julian, thanks for the replies. > >> On Tue, Jan 16, 2018, at 2:33 PM, Roy T. Fielding wrote: >> The Original Text is about weak validators, which don't even require that >> the content be the same. The two do not conflict. The suggested change >> would be incorrect. > > The specific text which confuses me, from the section on weak validators, is (emphasis mine): > > "However, two simultaneous representations might share the same *strong* validator if they differ only in the representation metadata, such as when two different media types are available for the same representation data." > > Am I misunderstanding that this is in conflict with the example you gave? (text/plain and application/json representations with same octets must have different strong validators.) > > Similarly: > > "Likewise, a validator is weak if it is shared by two or more representations of a given resource at the same time, unless those representations have identical representation data." > > The "unless" clause would appear to apply to the example you gave, implying that both representations can have the same validator but need not be weak. Yes, it does in both cases. The only distinction between weak and strong validation is that strong can be used for partial updates of the body. That’s why the metadata doesn’t matter. If the server wants the metadata to matter, it has to construct a validator (etag) accordingly to be better than just strong. ....Roy
Received on Tuesday, 16 January 2018 21:26:26 UTC