Re: simplified datatyping proposal

At 11:41 AM 2/20/02 -0600, Pat Hayes wrote:
>>Oh dear, it's looking as if I seriously dropped the ball on this...
>>
>>With my CC/PP hat on I don't see the following "long-range" usage is 
>>supported:
>>
>>    _:SomeClientComponent client-property:dpi "100" .
>>
>>     :
>>
>>    client-property:dpi rdfs:range datatype:number .
>>
>>i.e. does not define support for idiom B in the datatyping desiderata 
>>document:
>>
>> 
>>http://lists.w3.org/Archives/Public/www-archive/2002Jan/att-0133/00-gk.htm
>>
>>What also now seriously bothers me is that I can't see how the full 
>>proposal [1] supports this either.  I had earlier convinced myself that 
>>this was all OK, but now I can't see it.   Aaargh!
>
>No, it definitely does not support it, in fact in its current form it 
>effectively deprecates it, since literals always denote themselves, and 
>there is no way to change the meaning of an in-line literal. But I thought 
>that was what we had all decided to do?? The proposals have been saying 
>this loud and clear in paragraph 1 from the beginning, and people were 
>sending me only positive comments, so....

Yeah... I don't know how I missed it...  I guess I've been looking at so 
many proposals over the past couple of weeks I'm just not seeing the wood 
for the trees.

>To fix this, at the very least, you would have to use rdfs:drange instead 
>of rdfs:range. Right now that would not work either, but I can tweak the 
>semantics to make it work.  But at a cost: we have to allow  nonmonotonic 
>datatyping, in the sense that adding an rdfs:drange assertion *alters* the 
>interpretation of the in-line idiom. There's no way around that, seems to 
>me. But maybe we can live with this much nonmonotonicity  when considering 
>datatypes.
>
>Can I ask y'all for some clarification. Do people want to support BOTH 
>in-line and bnode forms at the same time? That is, should the following 
>mean the same thing and be affected in the same way by a drange assertion 
>on ex:age ??:
>(1)
>person:Jenny ex:age "10" .
>(2)
>person:Jenny ex:age _:x .
>_:x rdfs:dlex "10" .

I think both have their place:  (1) is how CC/PP currently works, but (2) 
provides a more flexible way of modelling going forwards.

>As things are at present, (1) means that Jenny's age is a character 
>string, no matter what else you say, whereas (2) says her age is something 
>that can be described by a character string, so can be modified by other 
>datatyping. We can change this, as I say, but only at a cost.

I have no problem with that bit ... I just want to be able to say that not 
all strings are valid here, only those which can represent (in this case) 
Jenny's age.  (It doesn't really matter to CC/PP what the "10" formally 
denotes, but I don't want to avoid the possible conclusion that 
person:Jenny ex:age "Humpty Dumpty" is equally meaningful.)

Also, I don't think they have to be available for the same property, though 
they may be needed for different properties in the same graph;  e.g.

    person:Jenny ex:age "10" .
    person:Jenny ex:weight _:x .
    _:x xsd:number "10" .

(Did you mean to use rdfs:dlex in your example above?)

#g


------------------------------------------------------------
Graham Klyne                    MIMEsweeper Group
Strategic Research              <http://www.mimesweeper.com>
<Graham.Klyne@MIMEsweeper.com>

Received on Wednesday, 20 February 2002 18:16:53 UTC