Where I've found the XML data type useful has been in situations where you can't control schema changes to a relation (table) and want to record all data from it even if columns are added, modified or removed.
Logging triggers, or archiving processes in these cases can take advantage of "select * from table for xml" and insert into an XML column in a log/archive/audit table.
For semi-structured data that has variable schema elements, XML is pretty ideal. Beyond that, relational data structures are better.
- Gus "GSquared", RSVP, OODA, MAP, NMVP, FAQ, SAT, SQL, DNA, RNA, UOI, IOU, AM, PM, AD, BC, BCE, USA, UN, CF, ROFL, LOL, ETC
Property of The Thread
"Nobody knows the age of the human race, but everyone agrees it's old enough to know better." - Anon