>I don't like this as it allows other time zones and daylight savings
>times.
I don't think it allows DST, does it? You have to specify an offset, not
a time zone symbol. Your local offset might change across the boundary,
but that's handled by the time library that generates the offset I
assume, and the stamp on an earlier assertion is computed to GMT for
comparison based on the offset at that time.
>I don't think requiring Z is a problem if that is easier technically.
All
>things being equal I would allow either Z or nothing.
I don't know how much easier it is technically, but it puts the burden
on SAML (as schema author) and the parser (as validator) to prevent a
time zone offset other than Z from showing up, instead of the program
examining the XML post-validation.
My reading of schema part 2 is that they intended people to use facets
to constrain the lexical forms, I just don't know if other people are
doing that or not. I'll see if I can find any good examples.
-- Scott
