Gustav Brock
gustav at cactus.dk
Wed Sep 1 11:48:39 CDT 2004
Hi Francisco Well, to Charlotte this seems to be no problem, but it was considerations as those you describe I was thinking of. So far my thoughts would be to store date/time as UTC time, then do the conversion (shift) later to whatever time zone you may need. But what is your question really? Daylight savings is just another twist to time zone issues. And remember, some time zones are offset by 30 or even 15 minutes only. Did you see my code previously posted (2004-07-14)? /gustav > Here is an issue that I never quite took into consideration: In a > (now in production) database all times are stored as UTC, however to > display I AM using an Offset of the local time, however if i were to > query an "OLDER" timeframe that was outside the scope of the daylights > savigns, then my math would be incorrect right? > How would you manage that? would it be wise to wrap the date inside a > function that could distinguish the daylights savings w/ non for the > local area? This way the dates not shift between daylights and > non-daylights savings schedules. > On Wed, 1 Sep 2004 09:11:44 -0700, Charlotte Foust > <cfoust at infostatsystems.com> wrote: >> World wide corporations exist too. We generally look at the clock to >> see whether we can reach one of our clients in the North Sea or Yemen or >> Canada or South America. What are you envisioning? In our business, >> the relative times aren't relevant, only the local time is important, so >> we don't have to worry about it. >> >> Charlotte Foust >> >> >> -----Original Message----- >> From: Gustav Brock [mailto:gustav at cactus.dk] >> Sent: Wednesday, September 01, 2004 8:54 AM >> To: Access Developers discussion and problem solving >> Subject: Re: [AccessD] Time zones (was: internationalization) >> >> Hi Charlotte >> >> Really? But "nation wide" corporations exist - and the dates of Honolulu >> and Miami can be a date apart for many hours? >> >> /gustav >> >> > We don't generally deal with it except in our heads. I've never had >> > to address it in an application, although probably I would use an >> > offset to UTC as well.