Corrections to timezone database

Mark Davis mark.davis at jtcsv.com
Fri Feb 4 15:25:35 UTC 2005


I do want to call people's attention to the CLDR project work on timezone
identifiers (http://unicode.org/cldr/).

We are in the process of gathering localization data for this coming
release. We have gathered some data for the locales {locale identifiers: ar
cs da de en es fi fr hu it ja nb nl pl pt pt_PT ru sv tr zh zh_Hant}, and
are working on fleshing those out and adding more (we're not done yet!). See
http://unicode.org/cldr/data/dropbox/timezones/.

For example, the following shows the current translations using the data we
have (for German):

http://unicode.org/cldr/data/dropbox/timezones/de_current.html

And the following shows a checklist for translators (note: in many cases
those don't *need* translation, since they are the same as the default
names.)

http://unicode.org/cldr/data/dropbox/timezones/de_to_localize.xml

=====

The mechanism we use for the generic (wall time) tzid is:

If a country has a single zone, we use the country name. Example (from
Chinese simplified):

Europe/Brussels =>  比利时

Otherwise we use the last field of the TZID, followed by the country name in
parens. Example:

Australia/Melbourne =>  墨尔本 (澳大利亚)

We are going to be adding additional information where one zones in
*certain* multi-zone countries will just use the country name. Example, we
have currently:

 Asia/Shanghai => 上海 (中国)
 Asia/Kashgar => 喀什葛尔 (中国)

Once we have that data, we'll generate:

 Asia/Shanghai => 中国
 Asia/Kashgar => 喀什葛尔 (中国)

(All of this can be overridden by specific localizations for generic,
standard, or daylight names. So if the equivalent of "Beijing Time" is what
is customary in Chinese, we could use that.)

‎Mark

----- Original Message ----- 
From: "Paul Eggert" <eggert at CS.UCLA.EDU>
To: "Funda Wang" <fundawang at gmail.com>
Cc: "Tz (tz at elsie.nci.nih.gov)" <tz at lecserver.nci.nih.gov>
Sent: Friday, February 04, 2005 00:08
Subject: Re: Corrections to timezone database


> Funda Wang <fundawang at gmail.com> writes:
>
> > In fact, we do not have an official abbreviation for "The Time of
> > Beijing", as we do not use English :)
>
> OK, in that case we should probably stick with "CST", as it
> abbreviates the most commonly used English phrase for the time in
> China.
>
> The tz database is an English database, by and large.  For example, it
> uses the abbreviation "CET" for Central European Time even though most
> residents of Central Europe use other abbrevations like "MEZ".  We
> don't currently have the resources to localize all the abbreviations,
> so we stick with English ones.  If and when the database is localized,
> we will transliterate "CST" in China to the appropriate Chinese
> characters (e.g., the characters for "The Time of Beijing").
>
> > the main reason that there should be a "Beijing" is that we call
> > UTC+8 as "The Time of Beijing".
>
> There is longstanding precedent in the tz database for decoupling tz
> zone labels from informal time zone names.  For example, the tz label
> "Europe/London" denotes local time in England, even though the people
> in England call their local time "Greenwich Mean Time".  The case of
> Beijing is somewhat similar.
>
>



More information about the tz mailing list