RE: DASL and W3C XML-QUERY Requirements

Thank you for your feedback on the XML Query Requirements document.

>DASL at least could get by with a far less ambitious XML Query than the
one you have outlined.

This point became quite obvious during Kevin Wiggen's presentation to the
recent XML Query F2F.  One of his slides states:

   There?s an obvious place to add an XML query operator to basicsearch
   - analogous to like
   - XML query must be expressible in XML
   - response should give Boolean predicate

Our meeting pointed out to Kevin that if an implementer was looking for a
solution to this problem today then the syntax in XPath at least partially
meets this requirement.  But we will certainly take into account your
requirements.

/paulc

PS: Thanks for helping arrange Kevin's talk at the Query F2F - it was very
useful.

/paulc

Paul Cotton, DB2 Language Architecture & Standards
IBM Canada Ltd, 17 Eleanor Drive, Nepean, Ontario K2E 6A3
Phone: (613) 225-5445   Fax:  (613) 226-6913
email: cotton@ca.ibm.com


---------------------- Forwarded by Paul Cotton/Toronto/IBM on 02/11/2000
01:41 PM ---------------------------

Jim Davis <jrd3@alum.mit.edu> on 02/11/2000 09:35:11 AM

Please respond to Jim Davis <jrd3@alum.mit.edu>

To:   Massimo Marchiori <massimo@w3.org>, www-xml-query-comments@w3.org
cc:   www-webdav-dasl@w3.org
Subject:  DASL and W3C XML-QUERY Requirements

At 01:31 PM 2/8/00 -0500, Massimo Marchiori wrote:
>The W3C XML Query Working Group has just published its first
>public Working Draft "XML Query Requirements" at
>http://www.w3.org/TR/xmlquery-req

Thank you for informing  the DASL community about this document.

I am pleased to see that DASL compatibility listed as a desirable feature
for XML query.

The requirements in your draft are a super-set of DASL's requirements.  If
you meet all of them, then XML Query will also be sufficient for DASL.   I
see no need to elaborate on those requirements.

I will  however  say that they seem quite extensive.  DASL at least could
get by with a far less ambitious XML Query than the one you have outlined.
In fact the only problem I can imagine would be if the final XML Query
design is both  very rich and not partitionable.  In DASL we've seen that
the general sentiment of the Internet community has been to keep things
simple.  If the XML query language were both large and entirely mandatory,
then undoubtedly some implementors would refuse to use it.

In particular, I hope it will not  be necessary that all XML query
processors support structural combinations (3.4.5)  and aggregation (3.4.6)
and  transformations (3.4.10).  So perhaps you can interpret this as
suggesting a meta-requirement: that the design of XML query be such that it
allows for clean subsets of functionality, so that one need only implement
as much as one requires, and not more.

If there is anything that I or the DASL community can do to assist you with
XML Query please do not hesitate to ask.

Regards

Jim Davis