On 08/25/2010 09:19 AM, Richard Esplin wrote: > XML databases like Mark Logic and eXist are aware of the structure of > the XML content stored within the database. This makes it easier and > faster to do queries based on the underlying structure of the data > then would be possible in a purely relational system.
Good explanation, Richard. I still think "XML database" is a poor name. Reminds me of people that say they "program in XML." That said, there are turing-complete languages that are composed of XML tags, like XSLT... blech. > This is especially useful when the input to the system and the output > to the system are both XML (such as web content management). In those > cases, storing the binary XML content in an XML database removes the > need for transforming the content into rows in a database and then > back into XML while preserving the ability to query based on specific > XML properties. Very true. And if a transformation can get the XML to HTML for presentation, that's pretty slick (but still makes me feel somewhat nauseous). /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
