Re: Comments on Last-Call Working Draft of RDF 1.1 Semantics

Hi Guus,

sorry for the late reply, I was just drafting on a reply. I will send it 

Concerning indication, I'm sorry to say that I cannot live with the 
response, but the main reason for this only became clear to me during my 
implementation attempt. I now believe that the current solution is 
technically broken and needs to be fixed, otherwise it will become a 
serious problem not only to RDF but also to other core Semantic Web 
Standards in the future. On the other hand, a fix would be pretty easy 
and would not have much impact on other parts of the Semantics document.

For the details, please wait for my reply tomorrow.


Am 05.12.2013 12:29, schrieb Guus Schreiber:
> Hi Michael,
> Could you indicate whether you can live with the WG response on
> ISSUE-165 (see below)?
> Thanks in advance for your reply,
> Guus
> On 23-10-13 18:10, Pat Hayes wrote:
>> Michael, greetings.
>> This is a response to your comments on RDF 1.1 Semantics, recorded by
>> the RDF WG as ISSUE-165 and ISSUE-166. I will deal with ISSUE-165
>> first and the various parts of ISSUE-166 in-line, below.
>> Regarding ISSUE-165, this matter was debated extensively within the
>> WG, and most of your points were made during this discussion. (see
>> and subsequent threads.) The primary reason for the change was to
>> simplify the presentation of the RDF semantics, which was an
>> overarching goal of the WG. The actual mathematics has not altered,
>> as the 2004 semantics required D-interpretation mappings to conform
>> to the datatype map, so the datatype map is simply a part of (a
>> restriction of) the interpretation mapping itself. Once this is
>> recognized, it is clearly simpler to treat it in this way rather than
>> as a separate mapping. In addition, it had been noted by several
>> commentors that the 2004 definitions allowed for 'pathological' D
>> mappings, such as one which permutes the meanings of the XSD datatype
>> IRIs. It was felt that disallowing such maps was a laudable
>> by-product of the change. We also note that this change does not
>> alter any entailments.
>> In response to your comment, I have extended the Change Note in
>> section 7 (and moved it to section 7.1) so that it reads as follows:
>> "In the 2004 RDF 1.0 specification, the semantics of datatypes
>> referred to datatype maps. The current treatment subsumes datatype
>> maps into the interpretation mapping on recognized IRIs. The
>> <dfn>datatype map</dfn> corresponding to D is exactly the restriction
>> of a <a>D-interpretation</a> mapping to the set D of
>> <a>recognize</a>d datatypes. The 2004 definitions permitted
>> "non-standard" datatype maps (such as one that maps the IRI
>> '<code></code>' to the
>> datatype identified by
>> <code></code>). Semantic
>> extensions based on such non-standard mappings are not sanctioned by
>> this specification."
>> As you will see, this provides a linkable definition of the term
>> "datatype map" in terms of the new specification, as well as giving
>> more explanation and motivation for the change.
>> The WG realizes that you may still have concerns regarding this
>> issue. If you feel it would be useful, we invite you to attend a WG
>> teleconference to discuss this issue in more depth.
> [..]
>> Pat Hayes (for the RDF WG)
>>> Best regards, Michael Schneider
>> ------------------------------------------------------------ IHMC
>> (850)434 8903 home 40 South Alcaniz St.            (850)202 4416
>> office Pensacola                            (850)202 4440   fax FL
>> 32502                              (850)291 0667   mobile
>> (preferred)

Received on Thursday, 5 December 2013 21:10:56 UTC