- From: Dave Beckett <dave@dajobe.org>
- Date: Thu, 03 Jul 2008 12:59:27 -0700
- To: "Booth, David (HP Software - Boston)" <dbooth@hp.com>
- CC: Story Henry <henry.story@bblfish.net>, "semantic-web@w3.org" <semantic-web@w3.org>, mnot@pobox.com
Booth, David (HP Software - Boston) wrote: >> From: Story Henry <henry.story@bblfish.net> >> [ . . . ] >> Clearly that atom in the content has to be interpreted as a literal, >> otherwise a feed with a number of entries saying contradictory things >> could produce on GRDDL extraction a nonsensical graph. > > Would named graphs help here, i.e., having one named graph per entry? > http://www.w3.org/2004/03/trix/ Since each atom:entry contains an rdf/xml block, and that corresponds to a graph, I was thinking of named graphs; although how the graphs are named is not something I've gone into, it could be based on the unique atom:id. We'd also should say something about the scope of blank node IDs in the rdf/xml block; they should be distinct for each graph. So another way to think of this is to ship a set of named graphs about resources. The set may be ordered by e.g. atom:updated date and it's possible the same resource may be seen multiple times, we should probably say something about that too, although it's related to the atom feed paging spec too, RFC 5005 http://www.ietf.org/rfc/rfc5005.txt Dave
Received on Thursday, 3 July 2008 20:00:59 UTC