[tz] Reason for removal of several TZ

Tom Lane tgl at sss.pgh.pa.us
Mon Dec 4 02:26:41 UTC 2017


Thomas M Steenholdt <tmus at tmus.dk> writes:
> Paul Eggert wrote:
>> As far as I can see, it's an invented abbreviation propagated from
>> tzdata, and there isn't much independent support for it. I've been
>> trying to omit these inventions, as tzdata should record timekeeping
>> not invent it.

> I have gathered a few sources that mention WGT/WGST, but I honestly have
> no clue if these are based of off tzdata (which could very well be the
> case) or something else.:
> ...
> In any case I can't help feeling, that with all this "knowledge" out
> there, removing the WGT/WGST timezone abbreviations from tzdata seems
> wrong. These are the TZ abbreviations that we work with on a daily basis
> up here (I'm in Nuuk, Greenland) and suddenly they are unknown in our
> favorite timezone data.

FWIW, I tend to share the suspicion that these zone abbreviations have
become normalized simply because they've been in widely-recognized
reference data (to wit, tzdb) for years.  Very few non-experts will
know that they had no standing otherwise.

The Postgres project has so far refrained from removing them from what
we recognize in timestamp input, because we fear the backlash if we do.
Most of our timestamp output formats represent timezone offsets
numerically, so that the issue doesn't arise so much on that side, and it
happens that our list of abbreviations recognized for input is separate
from the tzdb data proper.

			regards, tom lane



More information about the tz mailing list