Allow me to start off with full disclosure: I work for a corporation that
has committed itself to XML Schema. All statements in this message are
strictly my own opinions and may or may not reflect the opinions of anybody
else, including my employer.
In addition, I should note that I represent my company on both the XML
Query Working Group and on the XML Schema Working Group (but I am a very,
very recent addition to the Schema WG, so you can't blame me for very much
there!). My comments may or may not reflect the opinions of any other
participants in those Working Groups.
I, like many of you, find XML Schema extraordinarily complex. Like you, I
am often very frustrated by it. And, like you, I tend to use a relatively
small subset of its capabilities. I am also unconvinced that it is
error-free ;^}
However, let me suggest some points worth considering:
* Certain specifications, notably XML Query, must have a well-defined (XML)
data model in order to be properly specified.
* The W3C arguably has several XML data models under its control (e.g., the
DOM, the Infoset, the PSVI, and now the XQuery/Xpath data model).
* While XML Schema is undoubtedly complex, and many agree that it is too
complex, it has the necessary characteristics to be a foundation of the
data model used for XQuery's purposes. The Infoset alone does not, and the
DOM has such a completely different orientation that it also does not.
* It is unreasonable to expect a W3C specification (such as XQuery) to
adopt as its basis a data model not under control of the W3C, when there is
a W3C data model that is acceptable.
* Since XPath 2.0 and XQuery 1.0 are (properly, in my opinion!) closely
linked, and since XSLT 2.0 (again, properly) depends heavily on XPath 2.0,
it is difficult to justify using a different data model for XPath 2.0 and
XSLT 2.0 than for XQuery 1.0.
My conclusion is this: As bad as many observers and participants think XML
Schema to be, it is appropriate for it to be the basis for XQuery 1.0,
XPath 2.0, and XSLT 2.0. To hope that the various Working Groups will "see
the light" and choose to use a schema-like facility defined outside the W3C
is highly unlikely. An tremendous investment has already been made in XML
Schema by many companies (including mine, of course) and I am highly
skeptical that they will toss that investment away lightly.
Like Mike Kay said:
> I wish XML Schema would go away (I still wish namespaces would go
> away...) but it won't.
This has little or nothing to do with corporate representatives refusing to
admit mistakes. It has much more to do with appropriateness and
investment. For good or ill, XML Schema isn't going away any time
soon. Is RELAX NG better, as James Clark says? Maybe, even probably, when
evaluated by the criteria James uses in his discussion at
http://www.imc.org/ietf-xml-use/mail-archive/msg00217.html. But those are
not the only criteria that might be used, and the needs of the IETF are not
necessarily the needs of all of corporate users.
It might be more fruitful for commenters to identify the appropriate subset
of XML Schema to be used by XQuery 1.0, XPath 2.0, and XSLT 2.0. That is
something that might cause the Working Groups to pay attention and alter
their directions.
Sigh...
Jim
========================================================================
Jim Melton --- Editor of ISO/IEC 9075-* (SQL) Phone: +1.801.942.0144
Oracle Corporation Oracle Email: mailto:jim.melton@xxxxxxxxxx
1930 Viscounti Drive Standards email: mailto:jim.melton@xxxxxxx
Sandy, UT 84093-1063 Personal email: mailto:jim@xxxxxxxxxxx
USA Fax : +1.801.942.3345
========================================================================
= Facts are facts. However, any opinions expressed are the opinions =
= only of myself and may or may not reflect the opinions of anybody =
= else with whom I may or may not have discussed the issues at hand. =
========================================================================
XSL-List info and archive: http://www.mulberrytech.com/xsl/xsl-list