On Saturday 02 July 2005 23:17, Dimitre Novatchev wrote:
> On 7/3/05, Frans Englich <frans.englich@xxxxxxxxx> wrote:
> > Hi,
> >
> > I wonder, what is the difference between the xdt:anyAtomicType and
> > xs:anySimpleType? It is a type(duh) and hence can code and definitions
> > depend on it, but other than that, does it have any "effective" impact?
> >
> > Why does it exist? If it didn't exist, anySimpleType would have to derive
> > from the imaginary "itemType"; is that the reason?
>
> No. xdt:anyAtomicType is not identical (see below) to xs:anySimpleType.
>
> > Can the anyAtomicType be considered a "marker interface" for atomic
> > values, but that it in practice is an anySimpleType?
>
> Any instance/subtype of xdt:anyAtomicType is an instance/subtype of
> xs:anySimpleType but the reverse is not true.

Yes, such kind of relationships emerges as soon one have different types.

>
> > In the XML.com article titled "The XPath 2.0 Data Model"[1] there's a
> > small hint:
> >
> > "The Data Model document adds five new types to the 19 primitive types
> > defined in the Part 2 Recommendation: [...] the xdt:anyAtomicType, an
> > abstract type that plugs a newly-discovered architectural hole [...]"
> >
> > What was the architectural hole(or where can I read about it), and has it
> > any relation to my question?
>
> To represent the set of all types, whose instances are atomic (but not
> list or union) types.

Yes, good point, missed that. Except for that one(separating it from 
union/list types), does there exist any other reason to why anySimpleType 
could not be the direct base class of the atomic types?

>
>
> xdt:anyAtomicType is a true subtype of xs:anySimpleType.

What derivation/inheritage/type is not a "true subtype"?


Cheers,

		Frans