• From: Vasileios Papadimos <vpapad@c...>
• To: xml-dev@l...
• Date: Thu, 01 Mar 2001 21:55:04 -0800

On Thu, Mar 01, 2001 at 12:08:13PM -0500, Jonathan Robie wrote:

> Would an operator that indicates that order is irrelevant satisfy your wish 
> here? 

Absolutely! I think such an "unorder" operator was proposed in "A Formal Data
Model and Algebra for XML" by David Beech, Ashok Malhotra and Michael Rys, in
1999.

The query processor could propagate the "not-ordered" property to the inputs
of the unorder operator (so we would only need to mention unorder only once
in a typical database-oriented query). 

An unordered join would also be associative (right?), and that would turn 
XML query optimization back into the exponential problem we know and love :-)

Regards,

Vassilis Papadimos.

• Follow-Ups:
  • Re: Why not reinvent the wheel?
    • From: Jonathan Robie <Jonathan.Robie@S...>
  • Re: Why not reinvent the wheel?
    • From: Jonathan Robie <Jonathan.Robie@S...>

• References:
  • Re: Why not reinvent the wheel?
    • From: Jonathan Robie <Jonathan.Robie@S...>

• Prev by Date: Re: PSVI
• Next by Date: Re: [xsl] ANNOUNCE: Petition to withdraw xsl:script from XSLT 1.1
• Previous by thread: Re: What Ordering Does XQuery Need? (was:Re: Why not reinvent thewheel?)
• Next by thread: Re: Why not reinvent the wheel?
• Index(es):
  • Date
  • Thread