[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: XHTML m12n XSD
- From: Ann Navarro <ann@webgeek.com>
- To: Mike.Champion@SoftwareAG-USA.com, xml-dev@xml.org
- Date: Thu, 04 Jan 2001 17:14:03 -0500
At 03:49 PM 1/4/01 -0500, Mike.Champion@SoftwareAG-USA.com wrote:
>DTDs are too limited to be used effectively for "data", but never heard that
>XSDs are not fully suitable for "documents".
>
>Clearly the XHTML entity declarations would have to be in a DTD (since
>there is no way to declare entities in XSD), but in other ways would an
>XSD and a DTD version of a document "schema" (broadly defined) be
>non-equivalent?
The issue is post-declaration arbitrary changes to content models.
I have module A, B, C, D, E, and F, which have the basic content models of
Block, Inline, Flow, etc.
New Module Z is introduced and combined with A-F to create a new Host
Language Conformant Document Type, where the foo element created in Z must
be inserted in Block, or some other content set that's been pre-defined in
another module.
With DTDs, this is done with heavy parameterization and the use of PEs such
as %Block.extra which are defined as empty when the initial Block entity is
defined, and can then be altered in the new modules to hold these additions.
Schemas, to the best of our understanding, don't provide this facility --
though there's been suggestion that there's a way, but the one person that
understands it hasn't had the time to explain it, making it an unproven
assertion.
We could instantiate a real rathole if we discussed why Schemas are
incapable of entity declarations in the first place, which has lead to
questions of "why bother" if so much DTD segments must be used to replicate
the facilities of XHTML to begin with.
Ann
---
Ann Navarro, Author and Chief Geek
WebGeek, Inc. http://www.webgeek.com
Now Available! - XHTML By Example -
http://www.webgeek.com/books/xhtmlbyexample/