On Sunday 09 October 2005 23:04, Jack Carroll wrote: > > Jumping to the previous posting, an XML DTD might be a very good way > to encode an engineering parts list. It's simple enough to hack by hand, > and has the tags for automated processing into other formats. (It might > even be formatted as an OpenDoc file, using a template that provides a > paragraph style name for each data element. OpenDoc is an official > standard now, so long-term stability should be there.) I don't know how > popular XML is in data base design these days -- not my field -- but it's > certainly one of the things a systems analyst would think about.
I agree that XML would be a good idea. XML support in Open Source RDBMSes seems a bit limited still though: PostgreSQL only has limited support (http://www.throwingbeans.org/tech/postgresql_and_xml.html) as far as I can tell, and MySQL has nothing. Firebird doesn't seem to have anything either. There is eXist (http://exist.sf.net) which is a native XML database (as opposed to a relational database with XML extensions), but it's still a bit too unstable to trust with any vital data IMHO. Apache Xindice (http://xml.apache.org/xindice/) sounds like the most viable candidate at the moment, although I'm not sure about its stability. On the proprietary side of the fence, all of the big three (Oracle, IBM DB2 and MS SQL Server) have some XML extensions, but how good they are in practice remains to be seen it seems. All in all, XML databases is still a pretty young area of research and development (efficiently updating documents stored in a database is still being worked upon for instance), but it is definitely coming. Lourens (who is doing his MSc thesis on a datastructure for native XML databases)
pgpCWjS8zUxwf.pgp
Description: PGP signature
_______________________________________________ Open-graphics mailing list [email protected] http://lists.duskglow.com/mailman/listinfo/open-graphics List service provided by Duskglow Consulting, LLC (www.duskglow.com)
