- From: Pat Hayes <phayes@ihmc.us>
- Date: Fri, 18 Nov 2011 11:07:25 -0600
- To: Richard Cyganiak <richard@cyganiak.de>
- Cc: Ivan Herman <ivan@w3.org>, Antoine Zimmermann <antoine.zimmermann@emse.fr>, public-rdf-wg@w3.org
On Nov 16, 2011, at 3:14 PM, Richard Cyganiak wrote: > Hi Ivan, > > On 15 Nov 2011, at 16:59, Ivan Herman wrote: >> The way I, naïvely, interpret this is that the datatypes maps, their definition, the way the various mappings are associated with a URI, etc, are outside the OWL or RDFS spec. > > They are implementation-defined, in other words, an implementation hardcodes its own datatype map. > >> But that also means is that the OWL entailment that you rely on, based on owl:sameAs, may _not_ apply here. No, it definitely applies. If you have datatypes at all, then you can say that they are equal. > > That is possible, but I think that's beside the point. > > The question was whether I can say the following in RDF Concepts: “If the IRI <bar> is not in the datatype map, then the value of any literal "xxx"^^<bar> is unknown.” > > The RDF Semantics spec answers: “No you can't say that, because there might be another IRI <baz> in the datatype map that denotes the same datatype as <bar>.” Right, exactly. In fact, we specified that the datatype URI denote *the datatype* precisely to make this possible (and have it be that owl:sameAs works between datatypes.) We dont say exactly what a datatype is, note. But whatever they are, they are denoted by the datatype IRI, so equality reasoning applies to them. BTW, its very odd to say that something "is unknown". If this means "we don't know what its value is (yet)", then of course any missing information makes something unknown. But it is tempting to treat "unknown" as a classification, like being human, so that once something is in the unknown category then it is **known** to be 'unknown'. And if we do that, then the logic becomes nonmonotonic and many other semantic assumptions break. I think you guys might be using the word in different ways (?). I'm assuming Richard is using in the second way. > > Now the built-in entailment regimes may not actually be expressive enough to say that two IRIs <bar> and <baz> denote the same datatype. If they are both datatype IRIs, then owl:sameAs is expressive enough. Pat > And maybe not even OWL2 is expressive enough. But having that ability in *some* semantic extension of RDF might certainly be useful in a few situations, no? Are we sure that OWL3 isn't going to add an owl:equivalentDatatype property? It surely would come in handy sometimes, e.g., oracle:NVARCHAR2 owl:equivalentDatatype xsd:string. > > So, the RDF Semantics answer to my question above seems to be a justified one, unless we are willing to close the door on such future semantic extensions. > > Best, > Richard > > > >> After all, owl:sameAs is not some sort of a universal magic that blurs the differentiation of URI-s, it has, instead, some entailment rules associated to it. (Eg, Table 4 of [1]). Ie, I am not sure that it follows, in OWL, that the URI <bar>, in your original example, is properly associated to datatype map LV's, just because it is owl:sameAs to <baz> and the <baz> is indeed associated with a datatype mapping. >> >> Again, I am not an expert here, and I know I do some hand-waiving. If this is really important, we should ask somebody who really know these things, like Boris Motik. >> >> Ivan >> >> [1] http://www.w3.org/TR/owl2-profiles/#Reasoning_in_OWL_2_RL_and_RDF_Graphs_using_Rules >> >> >> On Nov 15, 2011, at 17:39 , Richard Cyganiak wrote: >> >>> On 15 Nov 2011, at 16:30, Ivan Herman wrote: >>>>>>> Is it somehow possible under RDFS-Entailment + D-Entailment to get a value for "foo"^^bar if bar is not in the datatype map? >>>>>> >>>>>> It is not possible. >>>>> >>>>> I think you're mistaken. >>>>> >>>>> if <bar> owl:sameAs <baz>, and <baz> is an IRI in the datatype map, then "foo"^^<bar> may have a well-defined value even if the IRI <bar> is not in the datatype map. >>>> >>>> Just to play the disagreeable guy: owl:sameAs is not an RDFS term. If we are talking about RDFS-Entailment, this will not work... >>> >>> Ok, you're right Ivan, under RDFS-Entailment "foo"^^<bar> won't have a well-defined value. >>> >>> But to quote again the phrase from Section 5.1 that I quoted earlier: >>> >>> [[ >>> The condition does not require that the URI reference in the typed literal be the same as the associated URI reference of the datatype; this allows semantic extensions which can express identity conditions on URI references to draw appropriate conclusions. >>> ]] >>> >>> My original question was: Is it true that "foo"^^<bar> has an L2V-assigned value if and only if <bar> is in the domain of the datatype map? The answer to that is: “There might be entailment regimes where it's not true, OWL's RDF-based semantics being an example.” >>> >>> Best, >>> Richard >> >> >> ---- >> Ivan Herman, W3C Semantic Web Activity Lead >> Home: http://www.w3.org/People/Ivan/ >> mobile: +31-641044153 >> FOAF: http://www.ivan-herman.net/foaf.rdf >> >> >> >> >> >> > > > ------------------------------------------------------------ IHMC (850)434 8903 or (650)494 3973 40 South Alcaniz St. (850)202 4416 office Pensacola (850)202 4440 fax FL 32502 (850)291 0667 mobile phayesAT-SIGNihmc.us http://www.ihmc.us/users/phayes
Received on Friday, 18 November 2011 17:08:13 UTC