xml-dev - Re: Should we care about elements order ?

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

To summarize what has been told so far:

- I should have been more precise in my terminology and specify that my
question was about sibling order (or, in other words, the document order
of elements having the same parent) within "data centric" elements ([1],
[2] and [3]).

- Except for single pass streaming based applications --unable to
compute forward references--, constraining the order of elements is
usually not necessary ([1], [4] and [5]) and there don't seem to have
any reason to add such a constraint if the application doesn't require
it.

- The data modeling should be done first and a validation tool chosen
(or created) to fit the model ([6] and [7]).

- Schematron is flexible enough to accommodate unordered elements ([8]).

- Some RDBMS/XML interfaces don't guarantee siblings order, other do
([1], [9] and [10]).

I am quite comfortable with these conclusions which are kind of
confirming what I thought, but still disappointed that it invalidates
the usage of XML Schema, a specification which given the weight of the
W3C is likely to play a major role in the future.

The usage of Schematron is still an interesting option, of course.

Another one could be a specific validation package composed of a XSLT
transformation "normalizing" the document followed by a XML Schema (or
DTD or Relax) validation.

Such a mechanism could even be used for DTD validation of multi
namespaces documents.

It's likely to be far more complex to deploy than a single pass standard
validation process, though.

Thanks for your answers !

Eric

[1] http://lists.xml.org/archives/xml-dev/200010/msg00777.html
[2] http://lists.xml.org/archives/xml-dev/200010/msg00782.html
[3] http://lists.xml.org/archives/xml-dev/200010/msg00785.html
[4] http://lists.xml.org/archives/xml-dev/200010/msg00781.html
[5] http://lists.xml.org/archives/xml-dev/200010/msg00786.html
[6] http://lists.xml.org/archives/xml-dev/200010/msg00780.html
[7] http://lists.xml.org/archives/xml-dev/200010/msg00792.html
[8] http://lists.xml.org/archives/xml-dev/200010/msg00771.html
[9] http://lists.xml.org/archives/xml-dev/200010/msg00783.html
[10] http://lists.xml.org/archives/xml-dev/200010/msg00795.html
-- 
------------------------------------------------------------------------
Eric van der Vlist       Dyomedea                    http://dyomedea.com
http://xmlfr.org         http://4xt.org              http://ducotede.com
------------------------------------------------------------------------