[foaf-protocols] First WebID Teleconference minutes (July 27th 2010)

Sarven Capadisli info at csarven.ca
Wed Aug 4 22:34:37 CEST 2010


On Tue, 2010-08-03 at 15:38 +0100, Kingsley Idehen wrote:
> Sarven Capadisli wrote:
> > On Tue, 2010-08-03 at 10:41 +0100, Nathan wrote:
> >   
> >> Before I go mixing up and conflating what I'm saying - my preferred 
> >> approach would be to simply say 'MUST be machine readable'...'see the 
> >> section on serializations/formats' then use RDFa examples through the 
> >> primer with turtle/n3 for clarity in certain bits.
> >>     
> >
> > +1
> >   
> Ditto.
> > I think emphasising RDFa examples would at least accomplish the
> > following:
> >
> > * Show something that they are perhaps more familiar with i.e., (X)HTML
> > and let them follow up on RDF if they need to.
> >   
> 
> Yes.
> 
> But again, what do you mean by "RDF" ? This is the inherent problem with 
> "RDF" do you mean data representation formats or data model? Everyone 
> repeats the same mistake which is the root of its fundamental problems 
> when *clarity* is required.
> 
> I suspect you really mean: RDF/XML. In the purest sense you mean one of: 
> RDF/XML, RDFa, NTriples, Turtle, N3, TriX, TriG, JSON-* etc.. re. 
> formats associated with the RDF Data Model.
> > * It will show that machine readable or RDF is not all about RDF/XML.
> >   
> Yes, but you are conflating matters already by presenting HTML+RDFa and 
> RDF as being distinct. HTML+RDFa is just one of several data 
> representation formats associated with the RDF data model.

Sorry, I should have emphasised on 'data representation formats'. I just
meant to say that by using RDFa examples, it will put RDF/XML to the
side momentarily so that when developers see or hear RDF, they don't
equate it to RDF/XML being the only way to represent the RDF data model.
It wasn't my intention to suggest that HTML+RDFa and RDF are distinct.

> > Hence, it will tone down that misconception, and let the developers go
> > ahead with their choice of serialization.
> >   
> 
> Eventually, when articulated with clarity.
> > * And obviously because it is both machine and human readable (in a
> > sense that the data is visible on a Web page).
> >   
> 
> Yes, the point I've been making repeatedly without overloading "RDF" :-)
> > Using turtle/n3 in parts would give them other exit points and not
> > misinterpret WebID as MUST be in RDFa.
> >   
> 
> Will show other data representation formats associated with the core 
> data model plux associated semantics of the actual WebID protocol.
> 
> I agree with your goal conceptually, but we have to be clear about "RDF".
> 
> As you can see there is a reason why I preoccupy myself with this "RDF" 
> matter. We can't be loose about it, since this has been the historic 
> issue that ultimately makes "RDF" borderline unexplainable in a 
> consistent manner.

Totally agree about being careful with the RDF term. Will follow with
proposed changes to the spec.

-Sarven



More information about the foaf-protocols mailing list