RE: Sync'ing triplestores

Yeah, there are some other detailed proposals along this line.  As I have said, any proposal which requires endpoints to support ontology, RDFS, or reification is DOA in terms of deployment. 

> -----Original Message-----
> From: Elias Torres [mailto:eliast@gmail.com] 
> Sent: Monday, April 04, 2005 12:28 PM
> To: Joshua Allen
> Cc: Geoff Chappell; Bill de hÓra; semantic-web@w3.org; Danny Ayers
> Subject: Re: Sync'ing triplestores
> 
> I saw this thread very early on, but I have not followed it 
> on an on-going basis. Have you guys read this?
> 
> http://www.w3.org/DesignIssues/Diff
> 
> Elias Torres
> 
> On Apr 4, 2005 3:13 PM, Joshua Allen <joshuaa@microsoft.com> wrote:
> > 
> > Yeah, I was thinking just don't support bnodes -- replace them with 
> > some urn:bnode:random-guid syntax
> > 
> > 
> > > -----Original Message-----
> > > From: Geoff Chappell [mailto:geoff@sover.net]
> > > Sent: Monday, April 04, 2005 11:58 AM
> > > To: Joshua Allen; 'Bill de hÓra'; semantic-web@w3.org
> > > Cc: 'Danny Ayers'
> > > Subject: RE: Sync'ing triplestores
> > >
> > >
> > >
> > > > -----Original Message-----
> > > > From: semantic-web-request@w3.org
> > > [mailto:semantic-web-request@w3.org]
> > > > On Behalf Of Joshua Allen
> > > > Sent: Monday, April 04, 2005 2:03 PM
> > > > To: Bill de hÓra; semantic-web@w3.org
> > > > Cc: Danny Ayers
> > > > Subject: RE: Sync'ing triplestores
> > > >
> > >
> > > > As for identifying the triples, I would just let each quad
> > > > (context+triple) identify the triple.  The context would
> > > be the URI
> > > > for the source feed, perhaps.
> > >
> > > That might not be enough if the statement contains a blank node.
> > >
> > > For example if your repository contains:
> > >
> > > _:a1 a rss:item; dc:title "Item 1".
> > > _:a2 a rss:item; dc:title "Item 2".
> > >
> > > And your update statement says:
> > >
> > > Delete:
> > >
> > > _:b1 a rss:item
> > >
> > > What happens? Both get deleted? Neither? It's tricky to identify 
> > > blank nodes. You have to rely on identification by 
> description and 
> > > if you don't have an inverse function property to work with, you 
> > > just have to hope that your description is unambiguous.
> > >
> > > -Geoff
> > >
> > >
> > 
> >
>

Received on Monday, 4 April 2005 19:41:32 UTC