RE: storing xml files into database

["Champion, Mike" <Mike.Champion@SoftwareAG-USA.com>]

> -----Original Message-----
> From: Bill Lindsey [mailto:bill@blnz.com]
> Sent: Monday, September 10, 2001 1:08 PM
> To: xml-dev@lists.xml.org
> Subject: Re: storing xml files into database
> 
> 
> A typical, naive definition of a "nodes" table does lead to 
> unacceptable 
> performance due to the necessity of many self-joins.  It is possible, 
> however, to devise a scheme for encoding nodes' context in a compact 
> form, optimized for an RDBMS' indexing facility, and build a
> generic table structure, capable of storing any well-formed
> XML, yet does not exhibit the self-join problem.

Interesting ... Is this a technique that end-users can use to normalize a
specific XML schema into a form suitable for practical RDBMS storage, or is
it a  technique that vendors (such as B-Bop?) use to develop middleware?
Also, are you just asserting that this technique exists (i.e., the details
are your trade secret), or is this technique public  knowledge?

If public knowledge, it would be interesting to compare your technique to
the Monet system
http://monetdb.cwi.nl/xml/Reports/Monet_report/monet_report.html

(Uhh, is this getting to DB-nerdy?  I would sympathize with a motion to move
the discussion to xmldb@xmldb.org ... on the other hand, since this list
tolerates discussions of haiku, ecology, legal arcana, and post-medieval
ontology, I sortof hope it's not considered off-topic!).