[
Lists Home |
Date Index |
Thread Index
]
- From: Stefan Haustein <haustein@kimo.cs.uni-dortmund.de>
- To: "Simon St.Laurent" <simonstl@simonstl.com>
- Date: Fri, 11 Feb 2000 13:23:55 +0100
"Simon St.Laurent" wrote:
>
> I hope that we can find out during the Candidate Recommendation (or
> possibly sooner) whether the size and scope of the XML Schema
> specifications lead to a large number of less-than-completely-interoperable
> functionality subsets.
The problem is that you cannot just take a subset: concepts are
overlapping,
but you cannot take out one completely without needing some additions
elsewhere.
For example, you have the content-type that can be empty, which seems
equivalent to an empty type definiton. But you cannot remove the
content-type completely because you cannot express mixed content
in the <type> definition. There was a nice suggestion on the
schema-comments list of introducing <pcdata> inside <type>,
removing the need of the content-type, but I did not find
any senseful response on the list...
> I'm hoping that XML Schemas prove simple and efficient to implement, but
> (based on my readings of the latest drafts), I can't say I'm optimistic.
Implementing XML Schema is not really fum: You do more consitency
checks like "what, if content-type is empty and the type definition not?
Throw an error? What is the default for content type?" than
senseful implementation.
Best regards
Stefan
--
Stefan Haustein
University of Dortmund
Computer Science VIII
www-ai.cs.uni-dortmund.de
|