> I agree with date stamps in namespace uris. Minor versions almost
> invariably don't guarantee backwards or forwards compatibility (like xml
> 1.1 to xml 1.0).
I think the fact that everyone pretty much accepts that the XML change
should have been numbered 2.0 and not 1.1 is really an argument in favor
of well-known major.minor semantics.
As a pragmatic matter, it's a royal pain in the neck to do interop debugging
using date-based URL's. Sitting there on the phone with someone reading
dates "041113" vs "040113" is a real pain. Anyone who's worked on following
various drafts of WS-Security and its errata will know -- been there,
done that, don't wnat to do it again.
/r$
--
Rich Salz Chief Security Architect
DataPower Technology http://www.datapower.com
XS40 XML Security Gateway http://www.datapower.com/products/xs40.html
