Re: Datatyping Summary V4

Brian,

For some appropriate definition of d:double-de and d:double, I think the 
entailment you describe does follow; e.g.

IEXT(I(d:double)) contains <10.5,"10.5">, and no other pair of the form 
<x,"10.5">

and

IEXT(I(d:double-de)) contains <10.5,"10,5">, and no other pair of the form 
<x,"10.5">

then the *only* value (denotation) for the bnodes that can satisfy:

    _:s1       <xsd:double-de>  "10,5" .

OR

    _:s2       <xsd:double>  "10.5" .

is 10.5.   So the bnodes here must both denote the same value, i.e. 10.5.

BUT, this depends on a very specific understanding of d:double and 
d:double-de, but I think we have to accept this kind of specific (i.e. 
commonly understood out-of-band specification) is in the nature of datatypes.

#g
--

At 03:24 PM 2/4/02 +0000, Brian McBride wrote:
>At 16:10 04/02/2002 +0200, Patrick Stickler wrote:
>
>[...]
>
>> > does not entail:
>> >
>> >  <foo> <eg:size> _:s .
>> >  <bar> <eg:size> _:s .
>>
>>Firstly, I presume you meant
>>
>>   <foo> <eg:size> _:s2 .
>>   <bar> <eg:size> _:s1 .
>
>Nope.
>
>[...]
>
>>But even with the S-A idiom, if we have
>>
>>    <foo>      <eg:size>   _:s1 .
>>    _:s1       <xsd:double-de>  "10,5" .
>>
>>    <bar>      <eg:size>   _:s2 .
>>    _:s2       <xsd:double>  "10.5" .
>>
>>this also does not entail
>>
>>    <foo> <eg:size> _:s2 .
>>    <bar> <eg:size> _:s1 .
>
>
>It is my understanding that it does, and also:
>
>>    <foo> <eg:size> _:s .
>>    <bar> <eg:size> _:s .
>
>Opinions please.
>
>Brian

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

Received on Monday, 4 February 2002 16:39:29 UTC