- From: Jonathan A Rees <rees@mumble.net>
- Date: Sun, 1 Apr 2012 16:40:54 +0200
- To: Larry Masinter <masinter@adobe.com>
- Cc: Ashok Malhotra <ashok.malhotra@oracle.com>, "public-awwsw@w3.org" <public-awwsw@w3.org>
oops, sent from wrong account. On Sun, Apr 1, 2012 at 12:21 AM, Jonathan A Rees <jonathan.rees@gmail.com> wrote: > > > -- apologies for brevity / using handheld gizmo -- > > On Mar 31, 2012, at 12:40, Larry Masinter <masinter@adobe.com> wrote: > >>> Reviews are welcome >> >> A quick review: >> >> The document assumes some things that I don't believe: >> >> - there's a determinable distinction between information resources and non-information resources, and thus between metadata and data. > > Good, I dont believe this either. > The doc does not assume it > >> - that URIs have (or can have) "definitions" ( except in a few obscure cases, e.g., with a URN, the naming authority could supply a definition). > > In fact, in some communities and contexts, some http uris do have definition. This is not an assumption, it's empirically true. > > Certainly most don't though, and this is not assumed in the document. > >> - that the 'meaning' of a URI depends (can depend, should depend) on the behavior of the HTTP protocol > > In both 1998 rdf and in linked data what's retrieved does tell some people something about what they will think is identified. In the HTTP spec 200 means you got a representation of the resource, and that's telling you something, even if it's not clear what. > > I dont get what 's not clear. Let's cover this on Monday, it's not rocket science. > > Jonathan > >> >> >> >>
Received on Sunday, 1 April 2012 14:41:22 UTC