Which next step for examplotron?

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

Hi,

I need some feedback concerning future directions for examplotron
(http://examplotron.org/)...

The idea behind examplotron is to define a schema language in which the
schema would as close as possible to sample instance documents.

I think that there is a niche for such languages for all the people
(including me) who like to start writing sample documents for new
vocabulary, check if they meet their needs and eventually derive a
schema from these samples documents.

The current approach has been to define a specific semantic for
examplotron, borrowing concepts from Relax NG and Schematron and write a
specific compiler.

Another approach would be to consider it as an alternative syntax for
Relax NG with embedded Schematron or xvif/xpath rules.

Basically, this would come to the point of allowing people to write in a
schema:

<foo @bar="xxx"/>

instead of

<element name="foo">
  <attribute name="bar"/>
</element>

or

element name {attribute foo {text}}

The question is: do we need a third syntax for Relax NG?

This syntax might be useful for some, but adding new syntaxes is always
confusing.

OTH, do we need another schema language with yet another semantic and is
there a good reason to take the pain to finalize it (defining it on top
of Relax NG would be much more trivial) ?

Its (tiny) users community seems pretty happy with it... are they wrong
and is examplotron just useless? 

Thanks

Eric
-- 
Rendez-vous a Paris.
                                         http://www.edifrance.org/ebd03/
------------------------------------------------------------------------
Eric van der Vlist       http://xmlfr.org            http://dyomedea.com
(W3C) XML Schema ISBN:0-596-00252-1 http://oreilly.com/catalog/xmlschema
------------------------------------------------------------------------