eggert at cs.ucla.edu
Fri Jun 3 05:49:06 UTC 2016
> Defend them from whom?
From discussions like these. :-) For Tomsk, for example, should the
abbreviation be TOMT or KRAT or NOVT or what?
> No-one's saying they should continue *exactly* as they have been
Actually there is some sentiment for leaving things alone, a sentiment I would
ordinarily share. Unfortunately the invented alphabetic abbreviations are too
often broken, and the introduction of Asia/Tomsk is rubbing our noses in the
> these fields are consistently and exclusively referred to
> as *names* in the relevant standards
Not really. ISO C11 calls strftime %Z "the locale's time zone name or
abbreviation". POSIX's formal description of the TZ environment variable uses
the term "designation" to describe these fields. In ordinary English one would
ordinarly use "abbreviation" to describe "PST" etc.; "name" would be used for
phrases like "Pacific Standard Time". The ISO C and POSIX standards allow these
fields to be used for either names (as in MS-Windows) or abbreviations (as in
tzdata); in tzdata they are always abbreviations.
> I certainly don't think "The US and Europe get
> abbreviations and Russia does not" is reasonable.
Quite right, the guideline should not pick on Russia, and that's not the intent.
Russia's new zones (along with some non-Russian arctic and antarctic entries)
merely happen to be the first places the revised guideline is being used. The
idea is to take our time in converting, in order to shake out implementation
bugs. (You can see the guideline in the "Theory" file; look for "Time zone
Victor Sudakov wrote:
> Windows uses abbreviations like RTZ6 for Russian timezones. Is
> Microsoft a reliable independent English-language source in your
Not really, as Microsoft just invented their abbreviations too, and my
impression is that abbreviations like RTZ6 are extremely rare in
English-language text, even rarer than abbreviations like KRAT.
More information about the tz