Jeni's Musings

Hey Jeni,

This is a happening idea. Just today (2008-03-14) Andrew Welch posted XSL-List wanting to know how he could write a conditional testing whether a given construct was schema-valid (no simple trick in current tools). Of course this is closely related to the feature you’re discussing (allowing for differences in schema languages).

If one could wire an RNG processor to an XSLT engine to do this, say by allowing an XSLT function to validate a node or sequence of nodes against a given RNG pattern, wouldn’t you have what you needed to do this fancy bit of structural inferencing, plus just check on the fly what was valid to what?

You can say better than I how close current technologies are to such a thing — but I bet they’re close.

Cheers — Wendell