- From: Graham Klyne <Graham.Klyne@MIMEsweeper.com>
- Date: Sat, 14 Apr 2001 17:30:05 +0100
- To: Brian McBride <bwm@hplb.hpl.hp.com>
- Cc: w3c-rdfcore-wg@w3.org
Brian, I fully appreciate the desire, and need, to keep the work items within well-managed bounds. My comments about N3 were based, in part, on my attempts to articulate an application (CC/PP) that uses RDF -- I feel that the XML form of presentation was very difficult to read for someone not already familiar with RDF principles. But, as you imply, that is not directly a problem for the RDFcore WG, and I'll concur that it should not be an assumed activity... unless we find it needed or helpful to achieve the group's chartered goals. ... As for "executable specification", I'm currently unconvinced. Not opposed, just not yet convinced. I have a feeling that computer-interpretable specifications don't often make good human-interpretable specifications. RDF-in-XML being a case in point. But I'm not religious about this and I promise I will listen to any ideas with an open mind. ... And finally, I think there has been something happening on the RDF-logic list recently that hasn't yet made it to the issues list (that I'm aware of -- I must re-check the issues list). This is the relationship between RDF and formal semantics, if indeed RDF is to have a formal semantics. I'm not quite sure how to phrase this as an "RDF issue", but I guess I must now take a shot at it: """ RDF semantics. Does RDF, of itself, have a formal semantics? Or is it so general that a formal semantics serves little useful purpose? GK has suggested [<link to my original message>] that it may be sufficient and more appropriate to regard RDF as an "abstract syntax", and RDF schema as the first steps of a "theory", without trying to ascribe a specific formal semantics to the RDF core. This would then leave the field clear for further developments and RDF applications to define required "semantics" within a an abstract framework for processing of RDF statements. """ Another way to lodge this on the issues list would be to fish out the relevant postings from Pat Hayes, Peter F. Patel-Schneider, Drew McDermott (and others?) #g -- At 06:39 PM 4/13/01 +0100, Brian McBride wrote: >Hi Graham, > >Graham Klyne wrote: > > > > > 5. N3 -- for human discourse about graphs: the RDF/XML syntax is not an > > easy syntax for human discourse. Nodes and arcs are all very well, but > > difficult to capture in emails and documents. N3 seems to neatly capture > > some middle ground here, and it _might_ be worth this group's while to > > clean up the N3 spec (possibly as a W3 NOTE) to provide an easier means to > > talk about RDF ideas in email and other documents. > >Found myself in strong agreement with the sentiments you expressed >though less sure about N3. > >We are specifically not chartered to develop a new RDF syntax, but we >also clearly need something to use to communicate amongst ourselves as >we discuss the issues and for when we get to the spec writing phase. >Whatever we use, especially in the final specs, must support the goals >you articulated of clarity and precision. > >I'd been envisaging building on the simplest imaginable triple syntax >for representing RDF graphs as that leaves the least ground for >ambiguity and gives us the least work in defining it. The current m&s >uses (p,s,o). Which is not say that we can't use richer syntax's where >appropriate, but ultimately, everything needs to be grounded in a triple >syntax. > >I had the idea also, that it would be good if we were able to provide an >'executable specification' of the translation from the RDF XML syntax >into triples and I suspect this will be easier and clearer if this >translation is to a very simple syntax. I've done an informal >experiment using an attribute grammar with an XSLT implementation and >that seemed to work ok. > >Our main task with model and syntax is 'rearticulation' and I'd hate to >see us get distracted from that. How about starting off focussing on >the issues from the issues list and seeing what notations we find we >need to support our discussions. And in the meantime it would be a good >idea if everyone were to make sure they were familiar with N3. > >Brian ------------------------------------------------------------ Graham Klyne Baltimore Technologies Strategic Research Content Security Group <Graham.Klyne@Baltimore.com> <http://www.mimesweeper.com> <http://www.baltimore.com> ------------------------------------------------------------
Received on Saturday, 14 April 2001 14:09:06 UTC