OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help



   Re: [xml-dev] Streaming XML (WAS: More on taming SAX (was Re: [xml-dev]

[ Lists Home | Date Index | Thread Index ]

On Mon, 27 Dec 2004 22:14:54 -0500, Liam Quin <liam@w3.org> wrote:

> Don't forget that a projection of relational tables into the
> XPath 2 Data Model is also possible, so that it becomes possible to
> use XQuery over "normal" (non-XML) relational tables.  There are
> implementations (both open source and proprietary) doing that today.

Don't forget the points that Dare Obasanjo
http://lists.xml.org/archives/xml-dev/200412/msg00443.html and
Soumitra Sengupta
http://lists.xml.org/archives/xml-dev/200412/msg00445.html  have made
about the difficulties of making this work in practice. :-)   The
*story* about XQuery in middleware sounds great.  The reality gets
pretty ugly.

All I'm really arguing is that people should have realistic expections
about XQuery.  It has a very useful role to play in the XML ecosystem,
and we need to complete the Recommendation in a timely manner.  IMHO
it shouldn't get wrapped around everything that might be possible to
do with it, because that will cause a lot of frustration and  give the
language as a whole a bad name.  Maybe I'm just disillusioned because
my last two employers invested very heavily in XQuery and have gotten
mostly frustration so far in return, largely because of the delays
caused by trying to do too much. I think the way forward is to focus
on what it does that nothing else does well -- querying XML stores --
and not promoting it as an alternative to existing, more mature

Arguably XQuery is good enough to be an XML query language, a SQL
integration language, an XML transformation language, and a native XML
development language.  The problem is that the network effect from
interoperability that made XML succeed isn't at work in the XQuery
domain, so it will have to win mindshare battles separately against
SQL for relational integration, XSLT for transformation, and all sorts
of alternatives for building XML applications.  I don't think it will
decisively win any battles except the one for which there is no
opponent, i,e, querying XML data sources.

Of course, I'm partly throwing this out to see if anyone  chews it to
pieces :-)  If there is a critical mass of people out there happily
using XQuery to do all this other stuff, I want to hear about it!


News | XML in Industry | Calendar | XML Registry
Marketplace | Resources | MyXML.org | Sponsors | Privacy Statement

Copyright 2001 XML.org. This site is hosted by OASIS