Re: A summary of the proposal for resolving the issues with rdf:text --> Could you please check it one more time?

On May 21, 2009, at 12:16 PM, Boris Motik wrote:

> Hello,
>
> [snip]
>
>> OWL 2 and RIF accept, in their RDF transcriptions, RDF plain literals
>> using the current RDF syntax without change; but treat them, for
>> internal purposes, as 'invisibly' typed by a datatype which I will
>> here call foo:text. This foo:text is unique among RDF datatypes in
>> that it can apply to literals with language tags. The lexical space  
>> of
>> foo:text is the set of all strings and pairs <string, tag>, and its
>> L2V mapping is the identity map. (Or, if y'all prefer, it could be
>> restricted to the case with tags, and just use xsd:string for the
>> untagged case.)
>>
>
> Before I voice my objections to this solution, I would like to point  
> out
> something that probably should have been mentioned in the discussion  
> thus far,
> and which, I believe, renders the compatibility argument moot: OWL 2  
> Full works
> with plain literals AS THEY ARE. That is, in OWL 2 Full -- which is  
> built
> directly on top of RDF syntax and semantics -- plain RDF literals  
> exist in
> exactly the same form as they exist in RDF: no changes, no special  
> cases, no
> nothing. Thus, if you are purely in the RDF world, nothing changes.

But everyone will want to be compatible with the -DL version, just as  
they are with OWL 1.

I find your attitude extraordinary, I have to say. You speak of the  
RDF world as though it were some foreign country you have diplomatic  
relations with. You really must bear in mind that the semantic web IS  
the RDF world, right now, and is likely to continue to be for some  
time in the future. All pieces of SWeb content that are published or  
transmitted on the Web are written in RDF. Most of the deployed RDF  
uses only fragments of OWL, many engines support only parts of the  
RDFS or OWL specs, and this situation is likely to continue. Under  
these circumstances, the 'structural' specifications of OWL and RIF  
are likely to be of little more than theoretical interest to most  
users and developers. What will actually matter are how they cash out  
in terms of the RDF transcription, which is what all the software is  
going to have to deal with.

> It is only in the structural specification of OWL 2 and RIF that the
> compatibility argument might potentially apply. Both of these  
> specifications
> have their independent conceptual models and, to enable  
> interoperability with
> RDF, they provide RDF import/export functionality. But precisely  
> because the
> conceptual models of these specifications are not layered on top of  
> RDF, the
> compatibility argument does not apply to them. At best, the  
> compatibility
> argument can be applied to the RDF import/export features of these
> specifications.

I am only talking about RDF "interactions". As I believe I said, my  
proposal should make no difference at all to the conceptual models of  
RIF or OWL.

>
>
> Now as to the above solution, I think it is not viable simply  
> because, in XML
> Schema, the lexical space of any datatype (and rdf:text is going to  
> be a
> datatype just like any other XML Schema datatype) is a set of  
> strings; thus, it
> cannot be a set of pairs.

This is just trivial definition-tweaking. Have the strings be as they  
are for rdf:text, and treat the RDF literal "foo"@eng  as a shorthand  
for the string  "foo@eng" In this case, you can even use rdf:text as  
defined 'internally' to the OWL and RIF specs, but still allow the RDF  
syntax to be unchanged.

> Introducing yet another exception is unacceptable, as
> this would cause major disruptions of various definitions in the  
> structural
> specifications of OWL 2 and RIF.

I have to say,. that is not of the first importance. The definitions  
in the spec document should be designed in the service of user  
interoperability, not to be an end in themselves. But in any case, I  
would not expect that many changes would be needed. Probably only an  
addition of a section explaining the somewhat idiosyncratic  
relationship of the new datatype to the existing RDF surface syntax  
would be required, is all.

Pat


>
>
> The best compromise we can come up with is to keep rdf:text as is,  
> but place
> restrictions on the cases when OWL 2 ontologies and RIF rule sets  
> are written
> into RDF. These already exist in the structural specification of OWL  
> 2 (see
> Section 5.7 of the Syntax document). Therefore, I strongly believe  
> we are simply
> in good shape to close this discussion.
>
> [snip]
>
> Boris
>
>
>

------------------------------------------------------------
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 Thursday, 21 May 2009 18:09:30 UTC