From: Dan Connolly ([email protected])
Date: 02/01/01
"Peter F. Patel-Schneider" wrote: > > I would turn things around and ask why there would be any reason to not use > XML Schema datatypes directly. I presume you mean "... any reason not to use the *concrete syntax* from the XML schema datatypes spec directly"? I do expect to use the types; just not the surface syntax. In answer to your question... Well... I know how to treat RDF syntax as logical formulas; e.g. I've automated translation to KIF in the general case. http://www.w3.org/2000/07/hs78/KIF Combine that with some axioms, and you've got a complete theory for whatever you want to talkd about. (yes, there are details about whether classes and bags can contain themselves or not that haven't been worked out; but daml+oil+concrete doesn't address those either.) I sort of know how to treat XML Schema concrete syntax as logical formulas; we could rigorously specify it in our spec. I think the easiest way to do that is to specify a translation to RDF syntax. In any case, we can't just put a one-liner in our spec that says "and you can use XML schema concrete syntax in there; the meaning is obvious." At least: I would find that very unsatisfactory. And RDF parsing software would burp. We can show what they should do in stead of burping, but that's a non-trivial amount of work. > Using XML Schema datatypes directly instead of translating them is much > better in terms of reuse and extension. Er... really? I don't think so; but I would have to see a specification of how to use them "directly" first. I don't see one in http://www.cs.man.ac.uk/~horrocks/daml+oil/Datatypes/ e.g. what semantics is assigned to declarations like this one? <xsd:simpleType name="over12"> <!-- over12 is an XMLS datatype based on positiveIntege --> <!-- with the added restriction that values must be >= 13 --> <xsd:restriction base="xsd:positiveInteger"> <xsd:minInclusive value="13"/> </xsd:restriction> </xsd:simpleType> I don't see anything relevant in http://www.cs.man.ac.uk/~horrocks/daml+oil/Datatypes/semantics-concrete.html > If reuse and extension is not one > of our goals, then I have lots of suggestions on how to change the syntax > of DAML+OIL for the better. > > Translating XML Schema datatype constructors requires work. Using them "directly" requires work too. > Adding XML Schema datatype-derived constructors to DAML+OIL (abstract) > classes requires a rethink of the semantics. How so? > peter > > From: Dan Connolly <[email protected]> > Subject: daml+oil+concrete includes XML Schema syntax? [was: ... next steps?] > Date: Thu, 01 Feb 2001 12:03:39 -0600 > > > Are you proposing to extend DAML+OIL syntax to include > > XML Schema syntax? I don't see any reason to do that. > > We can express things like minInclusive in the > > same style that we express domain, range, cardinality, > > etc., no? -- Dan Connolly, W3C http://www.w3.org/People/Connolly/ office: tel:+1-913-491-0501 pager: mailto:[email protected] (put return phone number in from/subject)
This archive was generated by hypermail 2.1.4 : 04/02/02 EST