> This is related to the Australian timezone duplication where we have a
> GUC parameter. Could we generalize the GUC stuff to allow arbitrary
> timezone mappings? Just an idea. Not sure it is worth it. The
> Australians complained but I have not hear complaints from other
> countries.
I don't recall that "the Australians complained". Did their consulate
call you? ;)
What was the issue? Was the previous compile-time behavior preferable to
some?
In any case, we do have to understand how to support areas which have
conflicting definitions for character time zones, so if there are
problems we need to be specific or be satisfied with the result.
- Thomas