Strawman XML ID
The W3C’s architecture folks have a strawman up on the problem of IDs.
Simply put, the problem is this: It’s nice to be able to uniquely identify a given element in an XML document. It’s pretty much essential if you ever want to link to it, for example. Markup traditionally does this by blessing particular attributes with IDness, which requires them to be unique inside their documents and allows references to them to be checked for the existence of the target.
Problem is, IDness is conferred by a DTD or schema, and if that DTD or schema is stripped away at some point (as with merely well-formed XML), the IDness is lost.
Still reading… I have been following this issue, but I haven’t a stake in any particular solution, as long as they bloody well solve it.
Via XMLHack.