FW: Updated Australian time zone names/strings

[David J N Begley]

On Thu, 22 Mar 2001, Paul Eggert wrote:

> Thanks for your well-researched letter.

You're welcome - sorry again that it's so long, but I didn't want to just say,
"Hey, I reckon they should be changed" without giving some evidence that I
wasn't dreaming up those abbreviations by myself.

> That summary contains a December 2000 letter from Richard Brittain of
> the Australian National Standards Commission endorsing a suggestion
> identical to yours.  I had been meaning to bring this up on the tz
> list at some point and your letter provides a good opportunity.

I have read some of the NSC's papers on time standards, which lead me to
browsing the 1991(-ish) archives of the Commonwealth Hansard - the last
instance the Federal Government tried to nationally legislate something about
time.

Basically what happened was that politicians representing States that do not
adhere to daylight savings at all (such as Queensland) did not wish to see
daylight savings imposed on them and so fought against the bill (one of the
things the bill proposed to do was reduce the number of time zones across
Australia).  The end result is that the Commonwealth left the issue of time
with the States and so the confusion about time zone names and abbreviations
continues.

[...use of term "daylight savings" vs "summer time"...]
> I suspect that Mr Mackin was referring to the suffix used in time zone
> time zone names (e.g. phrases like "Eastern Daylight Time" versus
> "Eastern Summer Time"), not the general notion of daylight saving
> time.  This may explain part of the discrepancy between his experience
> and yours.

Ahh, fair enough.  I honestly didn't mean any disrespect to Mr Mackin, just
disagreed with what I thought he was saying...

> The Australian Tranport Safety Bureau uses EST/ESuT; see e.g.
> <http://www.basi.gov.au/occurs/ob200000765.htm>.  I assume the "Su"
> stands for "Summer".  I've never seen that abbreviation before...

Me neither!  It could be a correction for the "EST/EST" ambiguity in the
existing tz database...

> ...but it is apparently common enough in Australia that Novell supports it;
> see <http://www.ithowto.com/Novell/clienttime.htm>.

Probably as with PostgreSQL, someone's come up with their own "corrected" time
zone abbreviations.  Interesting to see both "AEST" and "AEDT" also included.

> One more thought.  An important goal of changing the abbreviations is
> to reduce ambiguity.  "AEST" is ambiguous in practice, since some
> people use AEST/AEDT and others use AET/AEST.  (ABC News uses both. :-)

Quite right!  The best one could hope for is whatever seems "most common"
(and yes, I accept that this is hard to measure).

I hypothesise (and readily accept others may disagree) that in addition to the
lack of standards, I suspect the time zone database's use of "EST" for both
standard and summer times has contributed (even if just a little) to people
creating their own abbreviations.

> One solution would be to use unambiguous abbreviation pairs like
> AET/AEDT or AET/AESuT, but as far as I can tell these combinations are
> hardly ever used in practice.

I'm sure there's *someone* out there using them, just as with all the other
variations.  Apart from "EST" or "EDT" in various corners, I believe "AEST"
and "AEDT" have the next most legitimate claim to being "in common use".

On Fri, 23 Mar 2001, Eric Ulevik wrote:

> But the most common case is Eastern Standard Time / Eastern Summer Time.

Both terms are used in a local context, most assuredly;  however, their
obvious (EST) abbreviations are not necessarily "most common".  For example,
if Channel Nine is announcing the broadcast times for the cricket they will
say something like "2:30pm eastern standard time" - the "Australian" is
redundant/assumed in this context because it's not of relevance to New Zealand
or anyone else.  In cases whereby things get too complicated, they will simply
say, "Check your local guides for details".

Yet, if you look specifically at abbreviations, "AEST" and "AEDT" are quite
common (even in Nine's own electronic news service);  not universal, sure, but
that's due to the lack of standards so everyone's making up their own.

[...pause...]

Okay, I just conducted a totally unscientific experiment with two colleagues
who happened to be walking by (yeah, hard to prove, but anyway)... they both
agreed on "EST" in the local context because "A" (Australian) was redundant,
but would both use "AEST" in the global context.  For summer/daylight savings
time, one would use "EDT" whilst the other would use "EDST" - again, local
context, prefixing with an "A" for the global context.

When asked if they would recognise or accept "AEST" for standard time and
"AEDT" for summer/daylight savings time, they both were happy to accept those
abbreviations (since they had already encountered them elsewhere, such as the
sources I've previously given as usage examples).

> It's my view that the tz database should attempt to record current practice,
> rather than invent new systems.

I hope I have demonstrated in some small way (short of pointing to every URL,
news clipping or other document that ever has, or ever will, use "AEST" or
"AEDT") that by switching from "EST/EST" to "AEST/AEDT" is not really
inventing a "new system" but just updating the "common usage" (acknowledged,
not universal - but you'll never get a universal usage in the absence of a
standard anyway).

I don't know what sort of poll would be sufficient for the keepers of the tz
database to decide one way or another;  you could post to a newsgroup or two,
but the results could always be argued to be skewed towards a certain type of
user anyway.

Whatever decision you ultimately make, thank you for at least being open to
comment and discussion on the matter (unlike certain others).

Cheers..


dave