- From: Graham Klyne <GK@ninebynine.org>
- Date: Mon, 05 Sep 2011 22:04:21 +0100
- To: Luc Moreau <L.Moreau@ecs.soton.ac.uk>
- CC: public-prov-wg@w3.org
My previous comment was made without fully understanding the role of the ASN. I think I now better understand that. But I *still* think the interpretation of the ASN is inadequately explained (contributing to my earlier misunderstanding). And also, it *is* a kind o formalism, though whether it is a formalism of the concepts depends on what you consider to be "concepts". In another message, I gave an example in which the expression of a constriuct was presented and explained via the ASN - I think that would be one way to provide a description of how to interpret the ASM, i.e. on a case-by-case basis. Another way would be to expand the section in which the formal syntax is given to describe how each of the syntax productions is intended to be interpreted. So, when you say: > -to explain how the example(s) can be encoded according to PIDM, and For me, it explains nothing if I have no guidance on how to interpret the ASN constructs. > -to list the constituents of each of the constructs. Again, this doesn't help if I don't know what the constituents are intended to denote. ... I now recognize in the provenance abstract syntax notation (PASN) a pattern similar to that which has been used in describing both RDF and OWL, but in this case I am less convinced the pattern is useful. In the case of RDF, the abstract syntax provides a fundamental underlying structure, independent of specific serialization, over which the formal model-theoretic semantics are defined. For OWL, there are necessary structures that are not easy to express within the simple RDF data model - and again I understand there is a separate model theoretic semantics defined over the OWL abstract syntax. As far as I can tell, neither of these considerations applies to the Provenance model - as far as I can tell, the basic structures could just as easily be expressed directly in RDF, with a "schema" (*) expressed in OWL, and thereby provide a much more direct relationship between the intended model and the exchange serialization, *and* provide a well-established platform on which a formal semantics for the provenance could be based. Thus, it's not clear to me that the Provenance Abstract Syntax Notation is actually providing sufficient expressive benefit to justify the additional cognitive load of understanding what it all means, given that there is an existing semantic web framework that can apparently express most of what is expressed by the ASN. To make the PASN worthwhile, I think it needs to be *much* easier to understand, to the point that it becomes easier to write a provenance assertion in PASN rather than in RDF. (Rather like the way it's easier to write OWN class expressions using the "Manchester" or similar syntax rather than expressing them in RDF.) Currently, the exposition of PASN isn't satisfying this criterion, IMO. #g -- (*) the term "schema" is not strictly correct, but the expectation of satisfiability for defined OWL classes provides what amount to a set of structural constraints on what can be recognized as reasonable assertions about some class(es) of things. On 05/09/2011 21:23, Luc Moreau wrote: > Hi Graham, > > We indicate that the abstract syntax notation is used to express instances of > the data model. > > In essence, it's a form of serialization. It's useful > -to explain how the example(s) can be encoded according to PIDM, and > -to list the constituents of each of the constructs. > > It was not intended to formalize concepts. > > Cheers, > Luc > > On 01/09/11 17:42, Provenance Working Group Issue Tracker wrote: >> PROV-ISSUE-87 (Model-concepts-formalism): Formalism used is not explained, not >> applied to concepts [Conceptual Model] >> >> http://www.w3.org/2011/prov/track/issues/87 >> >> Raised by: Graham Klyne >> On product: Conceptual Model >> >> The "Abstract Syntax Notation" formalism used in the concepts document is not >> fully explained - only a syntax is given. Yet, all the uses of the formalism >> are in examples: it is not (cannot be?) used to formally describe the actual >> concepts. As such, it doesn't really seem to serve any useful purpose. Or if >> it does, that purpose is not clear. >> >> See also: http://lists.w3.org/Archives/Public/public-prov-wg/2011Aug/0383.html >> >> >> >> >
Received on Monday, 5 September 2011 21:05:12 UTC