Michael Rys (and who would know better/sooner about this) just announced on his blog that XQuery 1.0 and associated specs (including XPath 2.0 and XSLT 2.0) are now official W3C recommendations. Congratulations to the working group on this. The specification process is also underway for a standard XQuery Update Facility and XQuery/XPath Full-Text query facility. The specs are available on the W3C website.

I'll have to revisit/reread the specs, now that they are finalized. The one that always seem to cause the most controversy in classes that I've taught has been the XQuery 1.0/XPath 2.0 Data Model, which is based on sequences, rather than being a document object model, Infoset, or Post Schema Validation Infoset, or a description of the serialized form of XML (Extensible Markup Language (XML) spec). The inclusion of both nodes and atomic values is usually a bit disconcerting, and I get comments like "that's not XML" when people see it. Although SQL Server 2005's XML data type doesn't exactly follow the XQuery 1.0/XPath 2.0 Data Model, rumor has it that the next version of the ISO/ANSI SQL spec (SQL2007?) may have some something to say about this, as well as something to say about XQuery in general. Right now, the SQL2003 spec doesn't specify a query language.

It will also be interesting to see what the SQL Server folks do with regards to updates to support the new specs in the next release, and support of a larger portion of the language constructs. Also, it would be interesting to see support for sequences in other XML specs and/or client and middle-tier APIs. So that "other folks" can read the results…