On Fri, Sep 2, 2011 at 2:50 PM, Tim Berners-Lee <timbl@w3.org> wrote: > > On 2011-08 -31, at 17:39, Alan Ruttenberg wrote: > > > Spot on. Couldn't agree more. > > -Alan > > > > On Wed, Aug 31, 2011 at 4:52 PM, Ian Davis <me@iandavis.com> wrote: > >> I think there are a number of contributing factors: > >> > >> 1) architecturally the meaning of the fragment is determined by the > media > >> type of the representation. Thus meaning of a hash URI depends on how > you > >> access it. > > Let me first declare this a red herring. It is convenient to > be able to server a file up say in RDF/XML and N3 and content negotiate, > but a condition of doing this is that the meaning is NOT changed between > the two. > Any more than if you served a document saying > "It is sunny" in text/html and one saying "it is rainy" in > application/xhtml+xml. > Typically with for example rdf/xml and turtle, the same RDF graph is > sent. > You are correct. I should have read his words more carefully. I believe, however, that he meant not that "the meaning depends on how you access it" but rather "the protocol for determining what a fragment refers is determined by the media type". In other words, that we are not free to give an (authoritative) interpretation of the meaning of a fragment in every case. Doing so, for a media type that gives a conflicting interpretation for a fragment identifier, would be something like defining the meaning of URIs in domains of which you are not owner. -Alan > > We could quibble around the edges of this, but basically > it i not a valuable way to spend time. > > And the meaning any document you get back even if you *don't* use a > hashURI depends > on the media type in exactly the same way. So the meaning > of the original URI referred to in that document similarly would depend on > the media > type to the same extent. > > Now to the more interesting questions in another message. > > Tim > >
Received on Friday, 2 September 2011 19:12:12 UTC