xsl-list
[Top] [All Lists]

RE: Xquery for 'hard data probs' was RE: XSLT vs Perl

2004-02-21 11:50:06
[mailto:owner-xsl-list(_at_)lists(_dot_)mulberrytech(_dot_)com] On Behalf Of 
Michael 
Kay
Sent: 21 February 2004 12:41
Subject: RE: Xquery for 'hard data probs' was RE: [xsl] XSLT vs Perl

Hmmm, perhaps it will ultimately be performance characteristics that
become the main reason to use Xquery over XSLT.
 
If you think the choice will be made rationally on objective criteria,

you are more of an optimist than I am. Since when did anyone choose 
their programming languages rationally?

admittedly I am rationalising XQuery usage in terms of performance. 

I guess the question I really am struggling with is 'Is XQuery relevent
in a world with XSLT and XPATH', and what I am hearing from students,
peers, and from a few folks on the list is that XQuery is nominally
relevent.

There is a need for fast efficient querying of xml data, which is also
concise and compact...will XQuery be that language? As a technical
architect, with projects that span across years, I am finding technology
selection difficult at the moment; Which schema technology XML Schema,
DTD, RelaxNG, Schematron? Which long term storage...XMLDB or RDBMS?
Should XSLT be constrained to lightweight processes such as
presentation? Which update technology (e.g. Xupdate)? Should I consider
using XForms, is infopath relevent ? And so on....now I have to choose a
query technology..XQuery, XPATH, XSLT?

Is the W3C willing to spend X years specifying languages that in the end
are not used or are not relevant...perhaps it would be more useful to
specify technologies with real use cases, for example why would I not
want to use XQuery via a GET request...seems reasonable enough to me
since I am dealing with HTTP everyday of my programming
existance....instead antiseptic scenarios are agreed upon because they
conform with current language theory and well known optimisations.
Perhaps there is a need to incubate certain specifications, akin to
Apache's incubation of new technologies...there is benefit to pure
research that doesn't have the stress of generating some output e.g. a
recc. or standard.

I see vendor support gearing up for XQuery implementations, but a
distinct lack of vendor support for XSLT 2.0 support..the crystal ball
is quite hazy on this one. Oh well, will have to hedge bet on all the
horses. 

cheers, Jim Fuller


 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list