[
Lists Home |
Date Index |
Thread Index
]
5/13/2002 11:15:20 AM, Jonathan Robie <jonathan.robie@datadirect-technologies.com> wrote:
>I can understand the desire for fewer types. But if you support any data
>types in documents, its hard to justify not supporting those data types
>found in the documents that are being queried, which may belong to any
>built-in type.
OK, but we're talking about a minimum conformance level here, and the whole
point is to find the smallest subset that works for the greatest number of users.
Most people, in my experience, will call an integer an "integer."
Those who choose to distinguish between "unsignedLong" and
"positiveInteger" will just not be able to use the lower conformance level.
Michael Kay noted the other day that namespaces seemed like just
as much of a nightmare in XSLT 1.0 as W3C Schema
does now; people got used to namespaces, and are starting to get used to XSD,
so maybe there is hope for the world after all. Sure ... but I'd suggest that
people like the idea of declaring their elements to be strings, integers,
floating point numbers, dates,etc.; not very many people want to worry about the
distinctions between int,long, and short -- to say nothing of unsignedLong and
positiveInteger -- unless absolutely forced to.
>
>One niggle: the primitive types do not include short and long integers,
>etc., these are found in built-in derived types. It would be possible to
>support only primitive types and not derived types. I'm not sure that would
>be much of a simplification.
All I can say is that a quick review of http://www.w3.org/TR/xmlschema-2/#integer
makes me [personally, blah blah blah] appreciate the truth of what Matthew
Gertner said earlier: "As far as I'm concerned, the W3C set itself up for
this by releasing the XML Schema spec."
|
