question about GMT and UTC

[Marcus]

Hello,

I need a timezone to store timestamps in that is independent of daylight
saving. From everything I've read, it seems I want either GMT or UTC.
The only problem is I can't figure out from any documentation I've read
if there is any difference (practically speaking, not scientifically and
taking into account the earth's orbit etc).

I can set my server timezone to GMT, but if I want UTC I will have to
use putenv('TZ=UTC' ). Could someone please confirm if they are in fact
the same or if one is better, and if I am correct in stating that
neither is affected by daylight saving? Lastly, if I set my server to
GMT (or UTC for that matter), am I also correct in saying that in that
case gmmktime() and gmdate() are the exact same as mktime() and date()?
Thank you very much in advance for your help.

[Marcus]

Marcus wrote:

Hello,

I need a timezone to store timestamps in that is independent of daylight
saving. From everything I've read, it seems I want either GMT or UTC.
The only problem is I can't figure out from any documentation I've read
if there is any difference (practically speaking, not scientifically and
taking into account the earth's orbit etc).

I can set my server timezone to GMT, but if I want UTC I will have to
use putenv('TZ=UTC' ). Could someone please confirm if they are in fact
the same or if one is better, and if I am correct in stating that
neither is affected by daylight saving? Lastly, if I set my server to
GMT (or UTC for that matter), am I also correct in saying that in that
case gmmktime() and gmdate() are the exact same as mktime() and date()?
Thank you very much in advance for your help.

I should elaborate on my first point and ask additionally if a call to:

mktime(12, 0, 0, 5, 10, 2007);

for example, will always return the exact same timestamp in either GMT
or UTC timezone and vice versa, i.e. 1 timestamp resulting in the exact
same 12:00:00 readout via date() whether GMT or UTC. Thanks.

[Snef]

date() returns server time. gmdate returns GMT (UTC) time.

All the dates in my database are stored as GMT/UTC and I display the users local
time onscreen.

Marcus wrote:

Marcus wrote:

>Hello,

I need a timezone to store timestamps in that is independent of
daylight saving. From everything I've read, it seems I want either
GMT or UTC. The only problem is I can't figure out from any
documentatio n I've read if there is any difference (practically
speaking, not scientifically and taking into account the earth's orbit
etc).

I can set my server timezone to GMT, but if I want UTC I will have to
use putenv('TZ=UTC' ). Could someone please confirm if they are in
fact the same or if one is better, and if I am correct in stating that
neither is affected by daylight saving? Lastly, if I set my server to
GMT (or UTC for that matter), am I also correct in saying that in that
case gmmktime() and gmdate() are the exact same as mktime() and
date()? Thank you very much in advance for your help.

I should elaborate on my first point and ask additionally if a call to:

mktime(12, 0, 0, 5, 10, 2007);

for example, will always return the exact same timestamp in either GMT
or UTC timezone and vice versa, i.e. 1 timestamp resulting in the exact
same 12:00:00 readout via date() whether GMT or UTC. Thanks.

[Marcus]

Snef wrote:

date() returns server time. gmdate returns GMT (UTC) time.

All the dates in my database are stored as GMT/UTC and I display the
users local time onscreen.

Thanks for the reply. In my initial post I said I was planning on
setting the server time to GMT, so in that case date() and gmdate()
would be the same, yes?

Also, from what you are saying, does that mean there is no difference
between GMT and UTC as far as timestamps are concerned? Thanks.

[Snef]

As far as i have read, UTC is the 'new' GMT. When you see the output of gmdate,
you will see that the result is in UTC.

Snef

Marcus wrote:

Snef wrote:

>date() returns server time. gmdate returns GMT (UTC) time.

All the dates in my database are stored as GMT/UTC and I display the
users local time onscreen.

Thanks for the reply. In my initial post I said I was planning on
setting the server time to GMT, so in that case date() and gmdate()
would be the same, yes?

Also, from what you are saying, does that mean there is no difference
between GMT and UTC as far as timestamps are concerned? Thanks.

[Marcus]

Snef wrote:

As far as i have read, UTC is the 'new' GMT. When you see the output of
gmdate, you will see that the result is in UTC.

Snef

Sorry to beat a dead horse here, but from what you are saying and the
documentation I've read, does that mean that the only real difference is
UTC is more accurate in its precision for a timestamp, such as if I want
to know the time right now it might be slightly more accurate to the
"official" time?

In other words, if I just want a representation for May 5, 2007 at 12:00
pm (or any other future time), are GMT and UTC identical and just as
good as each other? (i.e. I don't care if it reports this exact moment
in time as 1 microsecond off of whatever the atomic clock says, I just
want a reliable way of representing various future points in time via
timestamps that aren't affected by daylight saving).

Thanks.

[Snef]

as far as I understand (and use): yes

Marcus wrote:

Snef wrote:

>As far as i have read, UTC is the 'new' GMT. When you see the output
of gmdate, you will see that the result is in UTC.

Snef

Sorry to beat a dead horse here, but from what you are saying and the
documentation I've read, does that mean that the only real difference is
UTC is more accurate in its precision for a timestamp, such as if I want
to know the time right now it might be slightly more accurate to the
"official" time?

In other words, if I just want a representation for May 5, 2007 at 12:00
pm (or any other future time), are GMT and UTC identical and just as
good as each other? (i.e. I don't care if it reports this exact moment
in time as 1 microsecond off of whatever the atomic clock says, I just
want a reliable way of representing various future points in time via
timestamps that aren't affected by daylight saving).

Thanks.

[Jerry Stuckle]

Marcus wrote:

Snef wrote:

>As far as i have read, UTC is the 'new' GMT. When you see the output
of gmdate, you will see that the result is in UTC.

Snef

Sorry to beat a dead horse here, but from what you are saying and the
documentation I've read, does that mean that the only real difference is
UTC is more accurate in its precision for a timestamp, such as if I want
to know the time right now it might be slightly more accurate to the
"official" time?

In other words, if I just want a representation for May 5, 2007 at 12:00
pm (or any other future time), are GMT and UTC identical and just as
good as each other? (i.e. I don't care if it reports this exact moment
in time as 1 microsecond off of whatever the atomic clock says, I just
want a reliable way of representing various future points in time via
timestamps that aren't affected by daylight saving).

Thanks.

Try a little search:

http://en.wikipedia.org/wiki/Coordinated_Universal_Time

--
=============== ===
Remove the "x" from my email address
Jerry Stuckle
JDS Computer Training Corp.
js*******@attgl obal.net
=============== ===

[Bob Devine]

In article <f9************ *************** ***@insightbb.c om>, JumpMan222
@aol.com says...

it seems I want either GMT or UTC.
The only problem is I can't figure out from any documentation I've read
if there is any difference

GMT is the obsolete name. The Greenwich observatory has been out
of the role of an official clock source for a long time.

UTC (Universal Coordinated Time) is the shorthand name for the
modern reference time at the prime meridian.

So, use UTC as the designation for your timezone reference.

The US military (and probably NATO) likes to call the
zero-offset timezone "zulu time", just to confuse everyone...

And another good practice is to use numerical offsets for
timezones instead of the 3-letter names. For example, saying
"EST" is confusing because it means something different
in Canada compared to the US. When control of timezone rules
remain in the hands of politicians, don't hard-code any assumptions.

Bob Devine

[Marcus]

Jerry Stuckle wrote:

>
Try a little search:

http://en.wikipedia.org/wiki/Coordinated_Universal_Time

I actually read that entire page and spent the last 2 days googling but
never answered my question definitively. I only post to the newsgroup
after I exhaust all my search options and still am not sure.