Can we make the following unoffical principle for DSDL schemas then?
"Schema languages are responsible for their own construction and
composition"
I was not sure whether Part 10 would look after this, so I kept minimal
in Part 3,
but it is good if there is some unifying principle.
I think we should track the principles. Off the top of my head:
* little languages: targetted and combinable
* XML-in/XML-out: no non-XML PSVI outputs
* validation not type annotation: annotation/tranformation/reporting can
be a side-effect of course
* schema languages responsible for their own construction: modular
* re-use concepts and names where possible: easier to learn
* new languages minimally implementable in XSLT 1 where possible
* avoid W3C XML Schemas mistakes, but provide superset of their features
for validation
* include formal characterization/approximation in some logic system,
where possible
Cheers
Rick
Martin Bryan wrote:
> Thanks Rick. Looks a good starting point for discussion. My personal
> preference would be alignment with RELAX NG wherever possible as that
> means less of a learning curve for people working with the whole set
> of our standards. (But then I'm biased!)
>
> Martin
-- DSDL members discussion list To unsubscribe, please send a message with the command "unsubscribe" to dsdl-discuss-request@dsdl.org (mailto:dsdl-discuss-request@dsdl.org?Subject=unsubscribe)Received on Thu Apr 12 04:23:52 2007
This archive was generated by hypermail 2.1.8 : Thu Apr 12 2007 - 07:03:02 UTC