- From: Damian Steer <d.steer@bristol.ac.uk>
- Date: Thu, 4 Aug 2011 16:52:18 +0100
- To: Alexander Dutton <alexander.dutton@oucs.ox.ac.uk>
- Cc: Damian Steer <d.steer@bristol.ac.uk>, Linked Open Data <public-lod@w3.org>
On 4 Aug 2011, at 15:58, Alexander Dutton wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Hi Damian, > > On 04/08/11 15:41, Damian Steer wrote: >> >> On 4 Aug 2011, at 14:22, Alexander Dutton wrote: >> >>> <#fragment> a fragment:Fragment ; >>> fragment:within <http://example.org/something.xml> ; >>> fragment:locator "/some/path[1]"^^fragment:path . >> >> I think you're mostly describing XPointer? [1][2] > > Ooh, yes — that's maybe a better fit for what I meant when I said XPath. > I think I'd like a way of encoding "use this XPointer expression to pull > something out of that document" as RDF. As far as I know that > expression-in-RDF bit is missing. XPointer can be used as fragment-id-from-hell, so use in RDF is straightforward: <http://www.example.com/something#xpointer(//para[1])> ex:mentions <http://www.example.com/something-else> >> By 'more generally' do you mean non-xml documents? > > Yes. Ah, sorry I wasn't sure. There's quite a bit of work on fragments and multimedia out there, the most familiar being youtube's time offsets, and there is [1] from the W3C. Damian [1] <http://www.w3.org/TR/media-frags/>
Received on Thursday, 4 August 2011 15:54:05 UTC