- From: Danny Ayers <danny.ayers@gmail.com>
- Date: Thu, 29 Mar 2012 15:40:07 +0200
- To: Jeni Tennison <jeni@jenitennison.com>
- Cc: "www-tag@w3.org List" <www-tag@w3.org>
Thanks Jeni, slipping 'nominal' in there makes a lot of difference, given the constraints you mention. On 29 March 2012 15:29, Jeni Tennison <jeni@jenitennison.com> wrote: > Danny, > > I have reworded to: > > If there is a nominal representation Z from the probe URI (a 2XX response), > then this is equivalent to there being a nominal URI documentation carrier > for the probe URI that says that Z is a current representation of the > resource identified by the probe URI. > > If the application is aware of a ‘describedby’ relationship of which the > probe URI is the object, which may be the case because either > > • the probe URI is itself a URI linked to through one of the > mechanisms listed in Section 4.1 > or > • Z itself contains a statement in which the probe URI is the > object of a ‘describedby’ relationship > > the application can also infer that the identified resource is an > "information resource" (see below). In other cases, no such inference > about the nature of the resource can be made (the application cannot > tell whether the probe URI identifies an information resource or not). > > Note that the first paragraph is essentially redundant; this is just reiterating what HTTPbis says. > > The 'probe URI' phrase is lifted from Jonathan's document; in accordance with the call for change proposals, the change proposal attempts to stick as closely to that as possible. > > I agree that specifying things are 'information resource's isn't helpful; were this approach adopted I'd hope that the resulting document would be merged with other changes that seek to switch in the content vs description notion or something else more useful. > > Jeni > > On 29 Mar 2012, at 14:05, Danny Ayers wrote: > >> I'm fairly neutral on the meat of the proposal : I don't think it's >> necessary, but then augmenting existing material with 'describedby' >> does seem a good idea. >> >> But I do believe it needs some editorial tweaks around this: >> >> "...a 200 response to a probe URI no longer by itself implies...that >> the response is a representation of the resource identified by the >> probe URI" >> >> - especially alongside - >> >> "Rather than simply telling these people that they are Doing It Wrong..." >> >> Instead of telling the people that aren't aware of the distinction >> made by httpRange-14 they are doing it wrong, in its current wording >> the proposal amounts to a redefinition of resource and representation, >> which tells *everyone* that's followed the specs since around RFC2616 >> (1999) they are doing it wrong (in principle, even if they got lucky >> in practice). >> >> Going down the IR path doesn't seem to work, content vs. description >> opens another can of worms. So maybe an alternative might be to say >> that although what you get with a 200 is a representation, it isn't an >> "authoritative representation" until other information (describedby >> etc) is taken into consideration. >> >> Also talking about a "probe URI" seems misleading, taken literally it >> implies every request for data should be preceded by a request for >> metadata. Not sure exactly how this might be reworded, but perhaps >> something along the lines of "probe activity" by the client on the >> response. >> >> Cheers, >> Danny. >> >> -- >> http://dannyayers.com >> >> http://webbeep.it - text to tones and back again >> >> > > -- > Jeni Tennison > http://www.jenitennison.com > -- http://dannyayers.com http://webbeep.it - text to tones and back again
Received on Thursday, 29 March 2012 15:31:19 UTC