Re: Denotation of datatype values

On 2002-04-15 23:54, "ext Pat Hayes" <phayes@ai.uwf.edu> wrote:

>> 
>> This is a *very* important point to grasp. Even when you have the datatyped
>> literal pairings you still can't be sure if the values they represent
>> are not equal. You can only be sure if they are equal, insofar as the
>> pairings themselves are concerned..
> 
> Why? If I have the mappings and the literals, and I can compute the
> values (in some canonical form) and then I may well be able to
> determine inequality. Eg I know that <decimal, "25"> and <octal,
> "7" > are not equal.

Note the qualifying phrase at the end of my paragraph "insofar as
the pairings themselves are concerned".

Without full knowledge of what the datatypes 'decimal' and 'octal'
are, you cannot be sure that the two pairings do *NOT* represent
the same value. The only sureties you can have with datatyped
literal pairings are (a) a pairing represents only one value,
(b) if two pairings are fully identitical, then they represent
the same value.

Thus, to take a simpler case, without knowing the full semantics
of xsd:integer, you cannot know that <xsd:integer,"5"> and
<xsd:integer,"005"> denote the same value. Now, if you *do* know
about xsd:integer, then you can determine that they both map
to the same value -- but that's not what I was saying.

At the level of datatyping interpretation, such equality cannot
be determined, because at that level, datatypes are fully
opaque to RDF.

This is just like not knowing, without full knowledge of http: URIs,
whether "http://abc.com/" and "HTTP://ABC.COM/" resolve to the same
representation, because similarly URI schemes are opaque to RDF.

Patrick

--
               
Patrick Stickler              Phone: +358 50 483 9453
Senior Research Scientist     Fax:   +358 7180 35409
Nokia Research Center         Email: patrick.stickler@nokia.com

Received on Tuesday, 16 April 2002 02:33:13 UTC