shripaldalal,

Saxon and DataDirect XQuery have very different approaches to data
access / XQuery processing. Saxon does its processing in memory, where
as DataDirect XQuery decomposes and optomizes the XQuery into SQL and
relies on the database (SQL Server 2000 or whatever) to do much of the
heavy lifting. I think you'll find this to be a far much more scalable
approach, especially for XQueries involving joining different database
tables, etc.  Check out this page, i believe it has the answers to the
questions you ask:

http://www.datadirect.com/products/xquery/ddxq_faqs/index.ssp

Hope that helps!

Sincerely,
The Stylus Studio Team
www.---.com

ps - if you're new to XQuery, check out this XQuery Primer:
www.---.com/xquery_primer.html

At the end he talks a bit about
shripaldalal wrote:
> Hi,
>
> If I have to then I dont mind using Java. But will datadirect be really
> really fast ? even faster then saxon ? I really dont mind paying for
> it. Also, then I have to use SQL Server 2000. I guess I dont mind doing
> that. My reporting should be quick.
>
> I had one more question... do XQueries become slow if the XML document
> does not have a proper schema defined ? Then I do not mind defining a
> schema. I just want to make my reporting applications using XQuery
> faster. This is frustrating right now.
>
> Regards,
> Shripal Dalal.
>
> stylusstudio@g... wrote:
> > hi -
> >
> > Is there any chance your data access stuff be written in Java? if so,
> > try DataDirect XQuery:
> > http://www.datadirect.com/products/xquery/index.ssp - you can execute
> > XQuery against SQL Server 2000 and many other databases. You'll find it
> > to be very fast.
> >
> > Sincerely,
> > The Stylus Studio Team
> > www.---.com
> >
> > ps - we agree with your assessment of  AltovaXML
> > www.---.com/altovaxml.html