On 2026-03-04 02:19, Philip Paeps wrote:
What's wrong with "-07"? That's what's used in other parts of the
world. Why is North America exceptional?
Inertia. For example, the earliest email standard, Internet RFC 561
(1973)[1], coauthored by the late Ray Tomlinson (widely credited for the
1971 invention of email![2]), was designed mostly for North America
because when that RFC was written almost all the ARPANET's nodes were in
North America (there was only one exception, in London). RFC 561
specified alphabetic abbreviations, but unsurprisingly only for North
American and the UK. A subset of those abbreviations survive not only in
the current email standard[3], but also in the drafts for its planned
successor[4].
Another example: Version 7 Unix (1979) supported only [ACEMP][DS]T + GMT
(good enough for AT&T's North American operations), and similar sets of
North-American-only or -mostly abbreviations are still hardwired into a
lot of software, including the Thunderbird email program that I'm
composing this missive on (see, e.g., [5]).
Unfortunately alphabetic abbreviations do not scale world-wide. If we
had to do it over again we'd surely stick to numeric abbreviations as
that's simpler and cleaner. But that ship sailed long ago.
It'd be a stretch to change America/Edmonton, America/Whitehorse,
America/Los_Angeles, etc. to use numeric abbreviations by default -
rightly or wrongly, too many programs and people would object. And it'd
be pretty strange if America/Vancouver used numeric abbreviations even
though all its neighbors used alphabetic. Although we could add an
option to zic to always generate numeric abbreviations, for people who
prefer that, that's not likely to become the default any time soon.
Chris Walton's recent email[6] lists reasons to prefer the alphabetic
abbreviation "MST" over the alternatives, and that's the direction I'm
leaning as well.
[1]: https://datatracker.ietf.org/doc/html/rfc561
[2]:
https://www.nytimes.com/2016/03/08/technology/raymond-tomlinson-email-obituary.html
[3]: https://datatracker.ietf.org/doc/html/rfc5322
[4]: https://datatracker.ietf.org/doc/draft-ietf-emailcore-rfc5322bis/
[5]:
https://github.com/mozilla/releases-comm-central/blob/7913b03ecb5d21754eaa821b61ed0909ba9dbbbe/mailnews/mime/jsmime/jsmime.mjs#L1374
[6]:
https://lists.iana.org/hyperkitty/list/[email protected]/message/7Q52N6O2OJSIESKKWV7UKZ6TXYDJJXUQ/