Re: Can XML Schemas Support Document Systems (WAS RE: ZDNetSchemaarticle,and hiding complexity within user-friendly products)

[Eric van der Vlist <vdv@dyomedea.com>]

"Henry S. Thompson" wrote:
> 
> And in an XML DTD you express this exactly how?  Come on Eric, this is
> the revolution of rising expectations problem -- you can't say on the
> one hand "I want lots of stuff that XML DTDs don't give me" and on the
> other hand "So you gave me a lot of stuff that XML DTDs didn't give me,
> but since you didn't give me everything I will assert what you've done
> is useless".

When have I said that what you've done is useless ?

On the contrary, I have said that I find it a good fit for data oriented
applications but not such a good fit for other applications.

My point was *** NOT *** to compare W3C XML Schema to SGML or XML DTDs
either (I like W3C XML Schema better than XML DTDs)!

I humbly feel like a newbie compared to most of the people on this list.

I have started working with XML and XSLT only 2 years ago, mostly on
light web publication applications, I have hardly written a XML DTD that
I haven't found adapted to my needs and I must confess that I don't know
much about SGML or SGML DTDs.

I feel that there is a need for a schema language even for lightweight
and flexible applications, though and my hopes when I have started
studying W3C XML Schema was that it could meet this need.

All what I am saying is that it's not the case and that other
alternatives that are doing a better job in that specific case (such as
TREX, Relax, Schematron or even Examplotron to name a few) are useful in
addition to W3C XML Schema.

I don't mean either to give any judgment on the relative "market" share
of the vocabularies that can be described by W3C XML Schema and those
that cannot but just acknowledge that it doesn't cover 100% of the need
(I don't feel qualified to say if it's covering 50%, 90% or 99% of the
range).

> Which takes away the flexibility which document DTD designers have
> always had -- not!  You seem to be forgetting/ignoring the fact that
> neither XML nor SGML allow ND content models.  XML Schema _increases_
> flexibility compared to DTDs, e.g. by allowing content models for
> mixed content, by allowing datatypes for element content, by allowing
> element content to be IDs, by giving you scoped keys (Now _there's_ a
> good example -- the 'document' people on the WG worked hard to make
> sure that part of the design was there), by providing wildcards.

Wildcards are great. If I can afford one additional criticism, though, I
regret that you can't allow any elements (or attributes) from any
namespace that hasn't been defined. This limitation is strongly reducing
the usability of wildcards in multi namespaces application.

Thanks for your answer and please forgive me if I have given the
impression to say that W3C XML Schema is useless, I do feel that it's a
usefull and important specification and that's why I feel concerned
about it!

Eric

> ht
> --
>   Henry S. Thompson, HCRC Language Technology Group, University of Edinburgh
>           W3C Fellow 1999--2001, part-time member of W3C Team
>      2 Buccleuch Place, Edinburgh EH8 9LW, SCOTLAND -- (44) 131 650-4440
>             Fax: (44) 131 650-4587, e-mail: ht@cogsci.ed.ac.uk
>                      URL: http://www.ltg.ed.ac.uk/~ht/
> 
> ------------------------------------------------------------------
-- 
See you in Hong Kong for www10:
                     http://www.www10.org/program/w10-half-tut.shtml#ta5
------------------------------------------------------------------------
Eric van der Vlist       Dyomedea                    http://dyomedea.com
http://xmlfr.org         http://4xt.org           http://examplotron.org
------------------------------------------------------------------------