- From: Marko Smiljanic <markosm@cs.utwente.nl>
- Date: Fri, 1 Nov 2002 12:28:38 +0100
- To: <www-ql@w3.org>
First many thanks to Michael, Michael, Jonathan and Maurice (who contacted me offline) for their comments on the problem... Your first reaction and common conclusion was that using "positional grouping" for modeling important concepts is a BAD DESIGN! I completely agree. Unfortunately, XML schema design is not the problem I'm facing. I'm doing a research in XML schema integration, and in that area I might be facing the problem of integrating XML schemas, one of which can have a bad design.... but let's drop those problems for the moment. My next BIG question is: Having in mind the problems with querying "positional groups" (and other related ambiguities ... ) should we try to state a STRONG argument for removing minOccurs and maxOccurs attributes xs:all, xs:sequence and xs:choice elements in XML schema? To rephrase, do you know for any applications where you can only solve things by using those attributes with values different then 1? ( Similar story was the one between structural programming and "goto" statement . It exists (in C++) but you should never use it...But look what happened with "goto" in Java ... well, it was "replaced" but still banned from Java) Thanks, Marko. ps. this might be a little off the WWW-QL line but I'm still approaching the problem from XQuery's side of the story. ************************* Research assistant, Database Technology Department of Computer Science, University of Twente P.O. Box 217, 7500 AE Enschede, The Netherlands Phone +31 53 4894520, Fax +31 53 4892927 E-mail: markosm@cs.utwente.nl WWW: www.cs.utwente.nl/~markosm
Received on Friday, 1 November 2002 06:28:41 UTC