> Now if Windows users are expecting displayed timezone to remain
> unchanged, then I'd certainly agree with you that that trumps
> consistency with Unix. However, if instead it's just something they
> could live with, then I'd argue we should strive for consistency and
> map the timezone to daylight savings if applicable.
> Having said all that, time calculations and I are old enemies, so I
> may well have all this bass-ackwards.

I may be wrong, but last time I checked, Windows had an incomplete time
zone system. I think they say London time is GMT, but it's GMT+BST.  MS
Outlook was unable to handle daylight-saving properly, but that is
probably unrelated. I didn't look at the APIs though, it's only from a
user point of view.

In POSIX, the automatic change CST->CDT, EST->EDT, GMT->BST, etc., is
normal, except when it's not; it depends on the full zone name -- e.g.
"America/Montreal" is EST in the winter, EDT in the summer, with switching
dates following the (1987?) north-american standard. However, if the full
name is "GMT", then it's only GMT; for "Asia/Tokyo", it's only JST. Try it
using "(export TZ=Asia/Tokyo; ruby)"

matju