[tz] [PROPOSED] Remove many invented abbreviations in 'asia'
eggert at cs.ucla.edu
Tue Jan 17 00:06:11 UTC 2017
Michael H Deckers via tz wrote:
> Why not just avoid every proleptic use of WITA
Thanks for your thoughtful comments.
We don't know when acronyms like WITA started to be used, so we don't know which
usages are proleptic and which are not. In the absence of information I was
reluctant to insert arbitrary transitions that change only abbreviations, as
that would indicate a degree of knowledge greater than what we have.
More generally, the tzdb does not strictly attempt to avoid proleptic
abbreviations. Common abbreviations like "CET" are proleptic when used for
100-year-old timestamps, and tzdb does this routinely.
> My other concern is the removal of the acronym JST for
> UT + 09 h in favor of +09 when used in occupied territory.
JST wasn't removed everywhere, only in areas that typically use numeric
abbreviations. The idea was to use JST in occupied regions like Asia/Hong_Kong
that have alphabetic traditions like HKT, and to use +09 in occupied regions
like Asia/Yangon which lack such traditions. That way, historical tables in
these areas will tend to use a consistent style, which is a plus.
Some of the previous tables' JST transitions were dubious anyway. Was it really
called "Japan Standard Time" in Jakarta on September 22, 1945? And why wasn't
"JST" used for Asia/Jayapura during 1942-1944 when Jayapura was at +09 and was
under Japanese Imperial control? It was too much trouble to decide when to use
JST vs +09, and it was a relief to remove this questionable political trivia
In an attempt to better document the above, I installed the attached proposed patch.
-------------- next part --------------
From 7c9253c7b3026f2519f01f117bca35c8e9205342 Mon Sep 17 00:00:00 2001
From: Paul Eggert <eggert at cs.ucla.edu>
Date: Mon, 16 Jan 2017 15:53:45 -0800
Subject: [PROPOSED] Document preferences for historical abbreviations
(Problems reported by Michael Deckers.)
* NEWS, Theory: Document this.
NEWS | 3 +++
Theory | 13 +++++++++++++
2 files changed, 16 insertions(+)
diff --git a/NEWS b/NEWS
index 3975295..900ccbb 100644
@@ -80,6 +80,9 @@ Unreleased, experimental changes
Changes to documentation and commentary
+ The 'Theory' file now better documents choice of historical time
+ zone abbreviations. (Problems reported by Michael Deckers.)
tz-link.htm now covers leap smearing, which is popular in clouds.
diff --git a/Theory b/Theory
index 533ee93..33e4606 100644
@@ -209,6 +209,19 @@ in decreasing order of importance:
If there is no common English abbreviation, use numeric offsets like
-05 and +0830 that are generated by zic's %z notation.
+ Use current abbreviations for older timestamps to avoid confusion.
+ For example, in 1910 a common English abbreviation for UT +01
+ in central Europe was 'MEZ' (short for both "Middle European
+ Zone" and for "Mitteleuropäische Zeit" in German). Nowadays
+ 'CET' ("Central European Time") is more common in English, and
+ the database uses 'CET' even for circa-1910 timestamps as this
+ is less confusing for modern users and avoids the need for
+ determining when 'CET' supplanted 'MEZ' in common usage.
+ Use a consistent style in a zone's history. For example, if a zone's
+ history tends to use numeric abbreviations and a particular
+ entry could go either way, use a numeric abbreviation.
[The remaining guidelines predate the introduction of %z.
They are problematic as they mean tz data entries invent
notation rather than record it. These guidelines are now
More information about the tz