W3C home > Mailing lists > Public > public-rdf-wg@w3.org > November 2012

Re: [All] Agenda for 21 Nov

From: Pat Hayes <phayes@ihmc.us>
Date: Sat, 17 Nov 2012 21:22:48 -0800
Cc: David Wood <david@3roundstones.com>, RDF Working Group WG <public-rdf-wg@w3.org>
Message-Id: <CA8E681A-EC1A-4DB3-BABF-26A517F184BF@ihmc.us>
To: Richard Cyganiak <richard@cyganiak.de>

On Nov 17, 2012, at 6:50 AM, Richard Cyganiak wrote:

> Hi Pat,
> 
> On 17 Nov 2012, at 00:46, Pat Hayes wrote:
>>> For ISSUE-109 (Consequence of ill-typed literals) I think we can state a concrete proposal to drive discussion:
>>> 
>>> PROPOSAL: Graphs containing ill-typed literals are inconsistent in
>>> all entailment regimes that reference a datatype map.
>> 
>> Can you expand on exactly what "that reference a datatype map" means? 
>> 
>> I would prefer to say, a graph containing a literal which is ill-typed is inconsistent in any entailment regime that recognizes the datatype in the literal. 
>> 
>> And then "recognizes a datatype" means that it is D-entailment for some D containing the datatype in question. 
> 
> Elsewhere there has been talk about separating the semantic conditions of D-entailment into:
> 
> 1. those that recognizes the equivalence of equal-valued literals, and
> 2. those that treat datatypes as classes and hence make rdfs:range work with datatypes.
> 
> This separation would produce an entailment regime that is simpler than RDFS-entailment, and still recognizes that "1"^^xsd:integer and "+1"^^xsd:integer are the same.
> 
> The intention of my proposal above is that a triple containing an ill-typed literal would be an inconsistency in all entailment regimes that make use of the semantic condition in group 1, whatever these entailment regimes may be in the end.

Fine, I am cool with all that. But I still want to say that ill-formedness is only recognized as such when the datatype in question is being considered to be part of the entailment regime. Surely we have to do something like this, for what is one to do when faced with some RDF containing literals typed iwth an unknown datatype? You don't know if they are ill-formed or not. So I think we have to say that such literals are treated simply as denoting unknown objects in the universe. 

Pat

> 
> Best,
> Richard
> 
> 
> 
>> 
>> Make sense?
>> 
>> Pat
>> 
>> 
>> 
>>> 
>>> I'm not sure we will get to consensus on 107 or 109, but we should at least get all concerns on the table.
>>> 
>>> Best,
>>> Richard
>>> 
>>> 
>>> On 16 Nov 2012, at 17:08, David Wood wrote:
>>> 
>>>> Hi all,
>>>> 
>>>> The agenda for our 21 Nov telecon is available at:
>>>> http://www.w3.org/2011/rdf-wg/wiki/Meetings:Telecon2012.11.21
>>>> 
>>>> Chair: David Wood
>>>> Scribe: Yves Raimond
>>>> Alternate: Zhe Wu
>>>> 
>>>> The focus of the meeting will be on RDF Concepts.
>>>> 
>>>> Regards,
>>>> Dave
>>>> 
>>>> 
>>>> 
>>>> 
>>> 
>>> 
>>> 
>> 
>> ------------------------------------------------------------
>> 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
>> 
>> 
>> 
>> 
>> 
>> 
> 
> 
> 

------------------------------------------------------------
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 Sunday, 18 November 2012 05:23:25 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 16:25:53 GMT