Re: Take 2: how does existing RDF software handle this datatypes test?

>
>In the interest of exploring this issue with the expectation
>of datatyping rather than its total absence, could those of
>you who have provided results based on Dan's original example
>please indicate results from the following extended examples,
>which includes explicit datatyping:

This still boils down to whether or not a system treats literals as tidy or
untidy (i.e. what determines identity). In our case (RDFQL) we currently
treat literals as tidy.  That could be altered with relative ease if
necessary. I imagine we'll have to do that if the WG "sanctions" _any_
datatyping idioms that assume untidy literals (because you could easily
support datatype idioms that assume tidy literals in a query system that did
not with the addition of a rule or two to define identity based on labels,
but it would be hard to go the other way)

[...]

>And, per Dan's original example, for each case:
>
>Suppose I asked you:
>    does that document say that
>    there's something with a dc:title?
>I hope you'd say: yes, of course.
>Now... suppose I ask:
>    whatever that title is, let's call it X.
>    Does that document also say that there's
>    something with ex:age of that same X?

I imagine few if any query systems have a particular datatype idiom
"hard-wired" (yet?). If you anticipate a particular datatyping scheme, it
should be straight-forward to craft a query to get the results you desire
(mod the difficulties of dealing with untidy literals in a system that
assumes they are tidy).

>
>Do you consider the results to be logically correct?

Really depends on the rules of the game, doesn't it? If literals are defined
(by WG) as always tidy, then any query that failed to bind literals of the
same label would appear non-logical; or if they're untidy, any query that
bound literals simply on the basis of the their labels, would appear
non-logical.

>
>Thanks,
>
>Patrick


Geoff

Received on Thursday, 31 January 2002 07:28:00 UTC