[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

RE: Why XML Schema is so large

• To: xml-dev@l...
• Subject: RE: Why XML Schema is so large
• From: Eric van der Vlist <vdv@d...>
• Date: Mon, 22 Dec 2003 15:38:14 +0100
• In-reply-to: <BKELLDAGKABIOCHDFDBPMEJAENAA.danny666@v...>
• Organization: Dyomedea (http://dyomedea.com)
• References: <BKELLDAGKABIOCHDFDBPMEJAENAA.danny666@v...>

Hi,

On Mon, 2003-12-22 at 14:53, Danny Ayers wrote:
> > What does 'a technology is complex' mean?
> 
> I guess this relates to the number of discrete moving parts. XML Schema is
> full of little cogs, 

True enough.

> but XSLT layers on other abstractions more, black
> boxes. If you were to decompose each down to their atomic parts (probably a
> horrid tangle of FOL) they'd probably fill similar-sized bags...

No, I don't think so. I haven't done the exercise with XSLT, but having
written a book on WXS and another on RELAX NG, I can compare both on
that respect.

WXS looks like a pile of semi-independent user features, each of them
having its own sets of elements and/or attributes. That could be handy
to explain if these features were really independent but, since that's
not totally the case (and also since the pile isn't built on a firm
model), you fall into the nightmare of explaining a huge number of
mutual restrictions.

RELAX NG is based on two main features (patterns and named classes) and
everything is derived from these two concepts and grounded on a solid
mathematical model. 

The result is that to explain RELAX NG you must more or less translate a
set of mathematical expressions into user features and that's not as
easy as I had expected when I have started writing this book. Of course,
you don't have to spend time on restrictions, but you can't either fully
describe each element or attribute at once because each of them
participate to many different user features. You end up with something
more like math classes where each year you study the same concepts with
an increased level of depth.

The bags that are the atomic parts of RELAX NG and WXS are thus very
different in nature. But, yes, I do think those of RELAX NG are much
simpler (even if you restrict WXS to its part one to compare similar
feature scopes).

Eric

-- 
Don't you think all these XML schema languages should work together?
                                                         http://dsdl.org
Upcoming XML schema languages tutorial:
 - Santa Clara  -half day- (15/03/2004)        http://masl.to/?J24916E96
------------------------------------------------------------------------
Eric van der Vlist       http://xmlfr.org            http://dyomedea.com
(W3C) XML Schema ISBN:0-596-00252-1 http://oreilly.com/catalog/xmlschema
------------------------------------------------------------------------

• References:
  • RE: Why XML Schema is so large
    • From: "Danny Ayers" <danny666@v...>

• Prev by Date: RE: Why XML Schema is so large
• Next by Date: XPath predicate and context
• Previous by thread: RE: Why XML Schema is so large
• Next by thread: RE: How to do a choice on aggregation with w3C XML schema
• Index(es):
  • Date
  • Thread