TZ database content

Robert Elz kre at munnari.OZ.AU
Sun Feb 11 09:12:23 UTC 2001


    Date:        Sat, 10 Feb 2001 12:31:04 -0800
    From:        "John A. Halloran" <seagoat at primenet.com>
    Message-ID:  <3.0.6.32.20010210123104.00dfca50 at pop.primenet.com>

  | The numbers in the previous list are not UTC offsets.  The standard amounts
  | are true time zones.  They can all be defined as "Add to get UTC."

Yes, obviously it is possible to do it either way.  The point was just
that the general convention is that the offset is the value added to UTC
to get the local time, rather than the other way.

  | The following list of 55 world time zones from an Australian source agrees
  | with you about the name, having "Australian Eastern Summer Time",

Of course - it is "Australian Eastern {Summer | Standard} Time" - just as
the one in the US, Canada, probably Mexico and Brasil, maybe other places
is "American Eastern Standard Time" (etc).   "Eastern" means exactly nothing
without  frame of reference (east of what?)

But the timezone here is officially "Eastern Standard Time" or "Eastern
Summer time" - and "here" happens to be Australia...

  | although in order for it to be unique, the abbreviation is still AEDT.

There's no hope (other than by forcing unrelated named) to make timezone
name abbreviations unique.   Most of us simply recognised that as an
impossibility a long time ago.   You can impose your own set of abbreviations
on whatever you control, but you're not going to convince many others to
go along with your set.

A large part of the world don't have names for timezones, let alone
abbreviations for them - there is simply "the time", which you can
pretend is called "The time in XXX" or "XXX time" for some country
name XXX (or region), but that's just you inventing that label.

Numeric zone offsets are easier to manipulate, make just as much
sense, and are much less ambiguous.   We should just stick to using
those.

kre




More information about the tz mailing list