To the discussion about possible alternative zone names in the tz-data files: I wrote:
If one doesn't like those names as zone names, just add them as a comment. And Paul Eggert replied:
That has already been done in the latest tz sources. Sorry, I missed this one! It's good that this happened, because the alternative names are a valuable feature of the tz data, worth mentioning.
About the Greenland names:
From the postal codes Greenland page (http://www.dpc.dk/Guide/PostalCodes.html): "3970 Pituffik (Dundas, Thule Air Base)" "3971 Qaanaaq (Thule)" See also: http://www.greenland-guide.dk/gt/maps/default.htm
Paul Eggert wrote:
... But this is another hint that perhaps we should leave the tz Zone names alone, at least for now.
In conclusion for the Greenland case I would like to suggest: Keep the current zone names because Thule and the other Danish names are globally more common, keep the commented Inuit names, but change Pituffik to Qaanaaq. However small Qaanaaq is, it refers to Thule; Pituffik refers to the air base. There are several other small inconsistencies in the zone names and comments of the tz data files. I took a look at the Asia file, with the latest Times atlas, the huge millennium edition, in front of me: * Bhutan: the name of the capital is Thimphu, not Thimbu. It is possible that Thimphu (not an f but an aspirated p) is a closer transliteration of the Dzongkha prononciation than the old British transliteration Thimbu. * Cambodia: Phnom Penh says the tz data, Phnum Penh is a more modern transliteration. * China: Chungking: seems to me an old and wrong French transliteration of Chongqing; the q is not a k-sound, but a ts-sound. * Cyprus: Nicosia. Isn't this called Lefkosia at the moment? * Kazakhstan: Aqtau: could be a pre-1940 transliteration; since the Cyrillic script was reused, the correct transliteration, according to the Times atlas, is Aktau. Where on earth is Aqtobe? TZ data says it's in the Aktyubinsk time zone. Since Aktyubinsk is on the map (both as an oblast' name and as the capital of that county), why not change Aqtobe to Aktyubinsk. * Kirgizstan: should be Kyrgyzstan. * Laos: Vientiane. Put in a comment the better transliteration of Lao name: Viangchan. * Qatar: Al Dawhah: it's Ad Dawhah or the more Latin alternative Doha. * Bahrain: no need for '-' in Al-Manamah, so: Al Manamah. * Syria: Damascus; comment: Dimashq. * Thailand: no, I won't suggest a comment for the (abbreviated) formal name Krung Thep; just Bangkok. If you really like the long one, here it comes: Krungthep Mahanakhon Amorn Rattanakosin Mahintara Yudthaya Mahadilok Pohp Noparat Rajathanee Bureerom Udomrajniwes Mahasatarn Amorn Pimarn Avaltarnsatit Sakatattiya Visanukram Prasit (ref.: http://www.geocities.com/Tokyo/Flats/1233/bkk.htm). * Turkmenistan has since 1991 the Latin alphabet instead of the Cyrillic. I'm not sure if it's Ashkhabad or Ashgabat at the moment. I think the latter. * Vietnam: tzdata says: Saigon's official name is Thanh-Pho Ho Chi Minh. How nice, but isn't the capital Ha Noi? The fact that probably more Americans actually have been in Saigon should not play any role ;) So, there is still a bit of work to do if one would like to dot some i's. Oscar van Vlijmen 2000-09-26
Oscar van Vlijmen wrote:
* Cambodia: Phnom Penh says the tz data, Phnum Penh is a more modern transliteration.
Translitteration has nothing to do here (furthermore, it depends of the standard you are referring to, for my side I did not see of a widely accepted standard for the Khmer script). Names are the English ones; this calls for transcription rather than translitteration. And furthermore, the common usage (the layman point of view) rather than the scientific one.
* China: Chungking: seems to me an old and wrong French transliteration of Chongqing; the q is not a k-sound, but a ts-sound.
FWIW: none of them in my (not recent) dictionnary. Furthermore, as far as French is concerned, if the initial is an affricate, then the real "old and wrong French" name looks more like Tch'... something. BTW, there is a 3M inhabitants (in Sichuan) city whose traditional French name is Tchong-K'ing.. (Chongqing in pinyin will correspond to Tch'ong-k'ing in EFEO, so I wonder if it the same place... BTW, EFEO have both k' and ts' to transcribe pinyin q, so I do not know what the real prononciation is, and if it varies according to the place). Antoine
Oscar van Vlijmen wrote:
* Cyprus: Nicosia. Isn't this called Lefkosia at the moment?
The Greek name of this city is <lambda> <epsilon> <upsilon> <kappa> <omega> <sigma> <iota-acute> <alpha>, usually transliterated Levkosia or Lefkosia. Its Turkish name is Lefkosa, with a cedilla on the s. In English it's Nicosia; French, Nicosie; and German, Nicosia or Nikosia. Even Cyprus government Web pages in English (e.g. http://www.pio.gov.cy/cygov/localgov.htm) usually call it Nicosia. Jesper Nørgaard wrote:
I believe that tz data has an (unstated?) policy of defining the default city of a country not necessarily to be the capital, but the city with the highest number of inhabitants [...]
It's explicit. In the tzcode tar file, under the heading "Names of time zone rule files", there are eleven guidelines. The seventh says, "Use the most populous among locations in a country's time zone, e.g. prefer `Shanghai' to `Beijing'. [...]" Paul Eggert wrote:
[...] I'm not convinced that we should switch entirely to Pinyin. [...]
We shouldn't, for a reason not mentioned. I think it's fine to put alternative place names into the comments, but I hope we'll resist making frequent changes to the zone names. Please remember that every change has a cost. In this case, there's the labor of updating databases, resource files, source code, etc. It may be negligible, but the tz archive is widely replicated, and hundreds or thousands of people might be affected (I have no way of estimating how many). In fact, I would like to add a new guideline for choosing zone names: Do not change established names unless they become ambiguous or shockingly incorrect. Yours, Gwillim Law
participants (3)
-
Antoine Leca -
Gwillim Law -
Oscar van Vlijmen