This is the mail archive of the
xsl-list@mulberrytech.com
mailing list .
RE: XML design of Database
- To: "'xsl-list at mulberrytech dot com'" <xsl-list at mulberrytech dot com>
- Subject: RE: XML design of Database
- From: "Liam R. E. Quin" <liamquin at interlog dot com>
- Date: Tue, 30 May 2000 10:16:01 -0400 (EDT)
- Reply-To: xsl-list at mulberrytech dot com
Vun Kannon, David <dvunkannon@kpmg.com> wrote:
> Andreas' presentation is clear on some things, not on others.
>
> XML as normalised relational tables means table per element, not table per
> nesting level.
Er, there is no one true definition, folks :-)
One table per document is also fine, and for some applications wil
give a thousand-fold or more performance improvement.
When you've spent an hour waiting for a document to be served by
a million-dollar SPARC running Oracle, you know that the "obvious"
relational schema isn't always the most appropriate.
In my book (forthcoming) on the subject, I actually seggest that
if you have XML documents, you store them outside a database
altogether in many situations, using a relational database for
metadata such as author's name, and sometimes for extracted data
such as a list of parts.
Lee
--
Liam Quin, Barefoot Computing, Toronto; The barefoot programmer
Ankh on irc.sorcery.net, http://www.holoweb.net/~liam/
co-author, The XML Specification Guide, Wiley Inc.
forthcoming: The Open Source XML Database Toolkit, Wiley, 2000
XSL-List info and archive: http://www.mulberrytech.com/xsl/xsl-list