Re: complete graphs

On Sat, 2011-10-01 at 17:09 +0100, Andy Seaborne wrote:
> 
> On 30/09/11 19:46, Sandro Hawke wrote:
> > On Fri, 2011-09-30 at 15:04 +0100, Andy Seaborne wrote:
> >>
> >> On 30/09/11 13:59, Lee Feigenbaum wrote:
> >>> On 9/30/2011 8:44 AM, Peter Frederick Patel-Schneider wrote:
> >>>> From: Sandro Hawke<sandro@w3.org>
> >>>> Subject: Re: complete graphs
> >>>> Date: Fri, 30 Sep 2011 07:31:26 -0500
> >>>>
> >>>> [...]
> >>>>
> >>>>> The restriction on the fourth column is that the fourth column is the
> >>>>> web address of a place (a g-box) currently serving that triple.
> >>>>> (That's the architecture I'm arguing for in this morning's post to
> >>>>> public-rdf-prov [1].)
> >>>>
> >>>> You are going to build this into the formal meaning of RDF?
> >>>> That's a non-starter for me.
> >>>
> >>> If I understand it correctly, I think it's a non-starter for me as well.
> >>> This would prohibit non-HTTP URIs from being used to as the 4th element
> >>> in a quad (i.e. as the identifier of a named graph)?
> >>>
> >>> Lee
> >>
> >> I understood Sandro's remark coming out of the discussion about
> >> provenance on the web and so I took generalising to any URI scheme for
> >> other situations as read.
> >
> > Right.   More formally, I'd say the fourth column is the identifier (IRI
> > or BNode) of an Information Resource which MAY provide representations.
> > If it does provide representations, it SHOULD provide an RDF
> > representation (a g-text).   If you want to use a non-dereferenceable
> > IRI scheme like uuid or tag, that's not good Linked Data but it's fine
> > RDF.  Informally, the fourth column entry denotes a g-box, but I'm not
> > convinced g-boxes should be formalized.
> 
> BNode?  Such a bNode is outside any graph so how does its semantics 
> work? (I'm not saying it does not work - but I don't see how it would.)
> 
> RDF-MT : sec 1.5
> http://www.w3.org/TR/rdf-mt/#unlabel
> 
> If we allow bNodes, at least it forces the decision on bNode label scope 
> in a multigraph format to be "document", but that, IMHO, is the better 
> choice anyway.

Yes, me too, but since we settled the .well-known/genid workaround, I
don't care very much.

     -- Sandro

> >      - Sandro
> >
> >>  Andy
> >>
> >>>
> >>>>
> >>>>> The issue about completeness is that if I want to say, as in [1], that I
> >>>>> agree or disagree with a statement (or otherwise build on it), it's
> >>>>> important the readers see the whole statement (or know that they are
> >>>>> seeing only a partial statement). It's even more important for me to
> >>>>> know if I'm seeing the whole statement before I say if I agree.
> >>>>
> >>>> Please, let's try to be more precise. In particular, there is
> >>>> rdf:Statement, so "statement" is something that has to be carefully
> >>>> used.
> >>>>
> >>>>> -- Sandro
> >>>>>
> >>>>> [1] http://lists.w3.org/Archives/Public/public-rdf-prov/2011Sep/0023
> >>>>
> >>>>
> >>>> peter
> >>>>
> >>>>
> >>>
> >>
> >>
> >
> >
> >
> 
>

Received on Saturday, 1 October 2011 19:48:36 UTC