Re: DOM Level 3 XPath: editorial, use case analysis, and counterproposal

>Being able to relax the ordering requirements seems like a key enabler for
>lazy evaluation (and unlocking the value of a return set limit)

Take that up with the XPath group. I think XPath 2.0 has considerered
allowing that relaxation at user request, for just this reason.

But If we're doing an XPath 1.0 API, we need to implement XPath 1.0 as
written. Inventing yet another varient on querying and trying to call it
XPath is not an option. If there's something the DOM _can't_ support, or if
there's something that has to be expressed a bit differently in the return
values due to impedence mismatches between the XPath and DOM data models,
workarounds are OK.

But we can't arbitrarily choose not to support defined behaviors. It's
specifically an XPath API... or it's something else and you have to go back
to ground zero to justify the DOM standardizing it.

______________________________________
Joe Kesselman  / IBM Research