New Data Model draft

Volume 9, Issue 65; 11 Jul 2006; last modified 08 Oct 2010

XPath 2.0 and XQuery 1.0 Data Model fans, a new draft was published today. It'll be of interest to those especially concerned about the finer details of element and attribute node types and type value determination.

Truth lies within a little uncertain compass, but error is immense.

Mea culpa. What happened was, a month or two ago, the Working Group's agreed to a proposal for changes to sections 3.3.1.1 Element and Attribute Node Type Names and 3.3.1.2 Typed Value Determination of the XQuery 1.0 and XPath 2.0 Data Model (XDM).

Wearing my editor hat, I dutifully accepted, reviewed, and implemented those changes. I know this because I remember doing it, I remember calling one of the authors of the proposal to discuss part of it, and I closed the relevant bugs in the bug database. And then…I don't know what happened.

There's no record of my commit in the CVS log and the document published on 8 June 2006 with all of the other specifications did not include the changes.

It's an area of the specification that's a bit subtle, and the changes are ones that we wanted to give the public an opportunity to review. So I argued, persuasively, at the last face-to-face meeting, that the most responsible course of action was to republish the document immediately with the necessary revisions. That's what happened today.

Unfortunately, even the revisions, which I do think are an improvement, have to address the specific details in the schema specification with a little less precision than would be ideal. This was deemed unavoidable given the organization of the relevant details in the schema specification.

Anyway, it's out now and you should take a look if you have interest in this level of detail.