[Bug 4273] [FS] data on element()

http://www.w3.org/Bugs/Public/show_bug.cgi?id=4273





------- Comment #18 from mike@saxonica.com  2007-10-23 09:15 -------
On reflection, I agree with comment #16. While the lack of substitutability in
the dynamic semantics is something that people need to watch out for, it's
probably liveable with. On the static side, I have a feeling that one thing
that we could really do with is an "all data is untyped" switch (in the static
context) so that users working in a non-schema-aware world don't suffer all
these usability problems. Perversely, if you know that all data is untyped then
you can often do much better type inference than when you don't.

Received on Tuesday, 23 October 2007 09:15:25 UTC