- tz - lists.iana.org

Could anyone help
by Dave's DHL Mail Dec. 29, 1999

Dec. 29, 1999

Hello Everyone, First, thanks to all on the list for the useful (and often amusing) information that's been passed round for the last couple of years. Next, I wonder if anyone can help with a little light hearted excercise - I'm afraid my coding isn't up to it and I don't have access to a compiler at home anyway... I'd like to produce a list of all the distict beginnings of the new millenium - listing the zone and UTC time when they will first see the new millenium, e.g.: Pacific/Chatham Island Dec 31 10:15 UTC .... Europe/Paris Dec 31 23:00 UTC Europe/London Jan 1 00:00 UTC ... etc.. I'd be really grateful if anyone can help with this, or alternatively point me to a list that already exists on the web somewhere. Thanks again, and best wishes for *your* new Millennium Dave Kelly

2 1

Re: Could anyone help
by Bradley White Dec. 29, 1999

Dec. 29, 1999

> I'd like to produce a list of all the distict beginnings of the new > millenium - listing the zone and UTC time when they will first see the new > millenium, e.g.: > > Pacific/Chatham Island Dec 31 10:15 UTC > .... > Europe/Paris Dec 31 23:00 UTC > Europe/London Jan 1 00:00 UTC > ... > etc.. > > I'd be really grateful if anyone can help with this, or alternatively point > me to a list that already exists on the web somewhere. #!/usr/bin/env perl use integer; use strict; use POSIX qw(mktime strftime tzset); # set $tzdir and $bogon as you see fit my $tzdir = '/usr/share/zoneinfo'; my $bogon = '^(right|posix|Etc|SystemV)/|\.tab$' .'|^(Factory|UTC|localtime)$|^GMT.' .'|^(..T|.ST\d.DT)$'; my %mill = (); chdir $tzdir or die "$tzdir: $!\n"; foreach my $zone (`find * -type f -print`) { chomp $zone; next if $zone =~ m:$bogon:o; $ENV{'TZ'} = $zone; tzset; my $when = mktime(0, 0, 0, 1, 0, 2000-1900); $mill{$when} = [] unless exists $mill{$when}; push @{$mill{$when}}, $zone; } foreach my $when (sort { $a <=> $b } keys %mill) { my %zones = (); foreach my $zone (@{$mill{$when}}) { my ($dev, $ino) = stat $zone; next unless defined $ino; $zones{"$dev.$ino"} = [] unless exists $zones{"$dev.$ino"}; push @{$zones{"$dev.$ino"}}, $zone; } print strftime "%b %d %H:%M UTC\n", gmtime $when; foreach my $devino (keys %zones) { print ' 'x4, join('|', sort @{$zones{$devino}}), "\n"; } } exit 0;

2 1

Re: Could anyone help
by Bradley White Dec. 29, 1999

Dec. 29, 1999

Oops ... I forgot to account for daylight savings in the call to mktime. Sorry about that. #!/usr/bin/env perl use integer; use strict; use POSIX qw(mktime strftime tzset); # set $tzdir and $bogon as you see fit my $tzdir = '/usr/share/zoneinfo'; my $bogon = '^(right|posix|Etc|SystemV)/|\.tab$' .'|^(Factory|UTC|localtime|posixrules)$' .'|^GMT.|^(..T|.ST\d.DT)$'; my %mill = (); chdir $tzdir or die "$tzdir: $!\n"; foreach my $zone (`find * -type f -print`) { chomp $zone; next if $zone =~ m:$bogon:o; $ENV{'TZ'} = $zone; tzset; my $when = mktime(0, 0, 0, 1, 0, 2000-1900, 0, 0, -1); $mill{$when} = [] unless exists $mill{$when}; push @{$mill{$when}}, $zone; } foreach my $when (sort { $a <=> $b } keys %mill) { my %zones = (); foreach my $zone (@{$mill{$when}}) { my ($dev, $ino) = stat $zone; next unless defined $ino; $zones{"$dev.$ino"} = [] unless exists $zones{"$dev.$ino"}; push @{$zones{"$dev.$ino"}}, $zone; } print strftime "%b %d %H:%M UTC\n", gmtime $when; foreach my $devino (keys %zones) { print ' 'x4, join('|', sort @{$zones{$devino}}), "\n"; } } exit 0;

1 0

other Nunavut hamlets rebel and stay on Eastern time
by Paul Eggert Dec. 28, 1999

Dec. 28, 1999

We already knew about Pangnirtung, but Michaela Rodrigue, writing in the Nunatsiaq News (1999-11-19), reports: The Baffin communities of Clyde River, Pangnirtung and Sanikiluaq now operate with two time zones, central - or Nunavut time - for government offices, and eastern time for municipal offices and schools.... Igloolik [was similar but then] made the switch to central time on Saturday, Nov. 6 ... after Nunavut Premier Paul Okalik reassured them the unified time zone would be reviewed in one year. Reference: http://www.nunatsiaq.com/archives/nunavut991130/nvt91119_17.html

1 0

data on DST changes all over the world
by Geelen, Paul Dec. 22, 1999

Dec. 22, 1999

Hi Elsie, How surprised I was to find this site.(http://sandbox.xerox.com/stewart/tzconvert.cgi). Since I am corresponding a lot with colleagues in different places all over the world, I would like to gain information on daylight saving changes that will occur in the next year. Could you please give me any data on these DST changes, for example in a spreadsheet or ascii file with the information as given below? country Netherlands timezone GMT + 1 from 1999, march 28, 02:00 AM until 1999, october 3, 02:00 AM shift + 1 hour country Brasil (east) timezone GMT - 3 from 1999, october 3, 02:00 AM until 2000, february 20, 02:00 AM shift + 1 hour I would be very pleased to hear from you. Thanks in advance. > Paul > __________________________________ > > room: VN 5.28 040 - 27 87 903 > E-mail: Paul.Geelen(a)nl.origin-it.com > __________________________________ > > >

1 0

RE: Multiple time zones in Mongolia
by Law, Gwil Jr. Dec. 13, 1999

Dec. 13, 1999

> I think you should not use Dariv as the central place in the UTC+7 > time zone. Take the more obvious Hovd. There are two Hovd in > Mongolia, so you should add the alternative name Jargalant. > > Jargalant is problematic. Rand McNally spells it Jirgalanta. Shanks > spells it Dzargalant (with a hacek over the z), or as Chovd. The CIA > spells it something like Duno-Us [...] I have Dund-Us as the capital of Hovd province, with Hovd and Jirgalanta as older names which may still be in use for this city. Still older names include Khobdo and Kobdo. > We generally prefer the most populous location to represent a region, > so we should use the most populous location in Hovd, Uvs, and > Bayan-Olgiy. I suspect that the candidates are Jargalant (the capital > of Hovd), Ulaangom (the capital of Uvs), and Olgiy (the capital of > Bayan-Olgiy). Do you happen to know which is the most populous? Looking in a 1980 Hammond atlas that I have at hand, I find the following populations for cities in the three provinces in question. They're out-of-date, but they may indicate relative sizes. Bulgan, Hovd 3,100 Dund-Us, Hovd 12,400 * Hyargas, Uvs 1,600 Olgiy, Bayan-Olgiy 11,700 * Tsagaannuur, Bayan-Olgiy 2,000 Tsetseg, Hovd 700 Ulaangom, Uvs 14,000 * It looks as if the three provincial capitals (*) are the only good candidates. Since ambiguity is a criterion for rejecting a location name, I checked the NIMA place names website. It has 6,161 records of populated place names in Mongolia, including variants. The NIMA database agrees with me about the preferred names of the three provincial capitals, as listed above. In the NIMA database, the names Dund-Us and Ulaangom are both unique. Among the other proposed names, Jargalant is the least desirable: there are 13 cities whose preferred names are Jargalant. There are three places whose preferred names are Hovd. There are two places named Olgiy, but one of them is a variant name for Har-Us. Darvi, not Dariv, is shown as a variant for Bulgan. There is a city named Dariv, but it's not in the new time zone. Doing some more research, I found the home page of the Mongolian National Statistical Office, at <http://statis.pmis.gov.mn/home_engl.htm>. There, I found the populations of the three capitals according to the 1989 census and an estimate for year-end 1998. 1989 1998 Dund-Us 24,100 27,400 Olgiy 27,200 21,000 Ulaangom 22,900 26,500 Following the published criteria, Dund-Us is the top choice, with Ulaangom a close second. Dund-Us is more centrally located in the zone, which might give it more of an edge. Yours, Gwillim Law

2 1

FW: Time zone(s) in Mongolia
by Jesper Nørgaard Dec. 12, 1999

Dec. 12, 1999

fyi, this was sent to me by an "editor of someting" in Mongolia (sorry for not being more precise). Seems to me like pretty convincing refutation of several time zones in Mongolia. It should be added though that Mr. Lutaa is situated in Ulaanbaatar, not near Dund-Us. A confirmation from that region would seem appropriate. ---------- From: lut[SMTP:lut@magicnet.mn] Reply To: lut Sent: Domingo 12 de Diciembre de 1999 0.41 To: Jesper Nørgaard Subject: Re: Time zone(s) in Mongolia Dear Mr. Jesper Nørgaard, Thank you for your interest in Mongolia and our magazine. As for the time zoning, indeed Mongolia has time difference. When sun rises in Eastern provinces its still one hour late in Ulaanbaatar and everybody sleeps. While Kazakhs prepare for bed in Westernmost province of Byan Ulgii, the national television program still goes on. My colleagues in radio stations across the country have this problem. However, seemingly this does not pose a big problem as in China. And we adhere to only one time zone GMT +8.0. Yours, B.Lutaa ----- Original Message ----- From: Jesper Nørgaard <jnorgard(a)Prodigy.Net.mx> To: <lut(a)magicnet.mn> Sent: Sunday, December 12, 1999 1:17 PM Subject: Time zone(s) in Mongolia Hi. I just wondered if you know if Mongolia has several timezones, or just one (like China). In Ulaanbaatar you surely have GMT+8.0 e.g. the same as China, but I don't know if there are other time zones. What about Dund-Us, they have the same time as you? Regards, Jesper Nørgaard Email: jnorgard(a)df1.telmex.net.mx Programador/Analista CIMMYT - Centro Internacional de Mejoramiento de Maíz y Trigo Dirección: CIMMYT, Lisboa 27, 06600 Mexico, D.F. Tel.: +52 (5) 7267541 ext. 1374 Fax: +52 (5) 7267558 Tel. Casa: 53-10-05-95 ó 53-10-97-78 CIMMYT home page: http://www.cimmyt.mx

1 0

Multiple time zones in Mongolia
by Paul Eggert Dec. 11, 1999

Dec. 11, 1999

Date: Sat, 11 Dec 1999 06:43:46 +0100 From: Oscar van Vlijmen <o.van.vlijmen(a)tip.nl> I think it is safe to assume that Mongolia has two time zones. Thanks for your information and references; I'll incorporate them into my next set of proposed changes. I'll CC: this message to the tz mailing list in case someone else can contribute info about Mongolia. Shanks writes that the Mongolian UTC+9 time zone didn't exist until 1983; perhaps it vanished some time between then and now. No one mentions DST. The IATA formerly agreed with Shanks that Mongolia observed DST, but the most recent issue of the SSIM (dated 1999-09) no longer says this. For lack of better information, I'll assume that Mongolia stopped observing DST starting this year. I think you should not use Dariv as the central place in the UTC+7 time zone. Take the more obvious Hovd. There are two Hovd in Mongolia, so you should add the alternative name Jargalant. Jargalant is problematic. Rand McNally spells it Jirgalanta. Shanks spells it Dzargalant (with a hacek over the z), or as Chovd. The CIA spells it something like Duno-Us; I can't quite read the scanned image in <http://www.lib.utexas.edu/Libs/PCL/Map_collection/middle_east_and_asia/Mong…>. We generally prefer the most populous location to represent a region, so we should use the most populous location in Hovd, Uvs, and Bayan-Olgiy. I suspect that the candidates are Jargalant (the capital of Hovd), Ulaangom (the capital of Uvs), and Olgiy (the capital of Bayan-Olgiy). Do you happen to know which is the most populous? Olgiy doesn't make it into Goode's, so I assume that it's smaller. If Ulaangom is larger than Jargalant, or if we can't figure out which is the larger, then I'd prefer to use Ulaangom, as its spelling is less problematic.

1 0

UTC Questionnaire response
by Joseph S. Myers Dec. 11, 1999

Dec. 11, 1999

This is a personal response to the UTC Questionnaire, as an individual contributor to the public domain UNIX timezone database at elsie.nci.nih.gov. I had originally not intended to respond to this questionnaire, since my experience is in the study of historical timezone changes and summer-time rules, and my views on the technical issues considered on their own are generally in accordance with those expressed in the response by Markus Kuhn, but on further consideration I decided that historical timezone research shows up potential practical, civil and legal issues that could arise with the decoupling of UTC from UT1, and of which it might be desirable to present a summary. In general, the time used for everyday purposes in most parts of the developed world (where time signals are readily available) is based on UTC, as distributed by time signals, with an offset that is normally an integral number of hours depending on the local timezone and summer-time arrangements. However, the legal definition of the time to be used may vary from country to country; for example, in the United Kingdom the time has been based on Greenwich Mean Time since 1880 (except that the whole of Ireland used Dublin Mean Time until 1916, then a single period of summer time at DMT plus one hour, then time based on Greenwich); while it has been proposed to change to a UTC-based definition, this sort of matter has not been a legislative priority. Although GMT has been replaced as a designation of a timescale by more precise definitions of variants of UT, it seems most likely that GMT would be taken, if UTC were decoupled from UT1, to be UT1 or very close to it, rather than UTC. In France, the time is now based on UTC, but from 1911 until 1978 the definition was based on Paris Mean Time, retarded by 9 minutes 21 seconds. The European Union directives defining the start and end points of summer time in the EU still state the relevant points in terms of Greenwich Mean Time. The time used in Liberia did not change to a Greenwich-based zone until 1972. >From this, it seems likely that many countries have definitions of legal time based on GMT rather than UTC; and, while the difference is insignificant for the purposes of everyday life, changing it has been of no great priority; a second here or there is not generally of great significance in the timing of everyday events. However, if UTC and UT1 are decoupled, it would seem likely that the legal and practical times would for the most part be allowed to drift, at least until this caused notable problems and quite likely for some time thereafter. Examples of the problems arising, taken from the 1847-1880 period when legal time in Great Britain was local time but for most purposes Greenwich time, as used by the railways, was used, may be found in Howse's book (Greenwich Time and the Longitude). In 1858 the case of Curtis v. March (where a court wrongly disposed of a case because the defendant had not appeared on time by the court clock, on Greenwich Time, but appeared on time by local time) lead to the case law that legal time was local time (though almost all clocks showed Greenwich time). Despite the confusion so caused, and the lack of ready availability of local time, the matter was not legislated upon until 1880. It seems plausible that similar disputes, about whether or not something was done within specified times, could arise if UTC and UT1 were substantially decoupled. While it would be clearly desirable for legislation in all countries to establish the legal time as being based on the readily available time scale, UTC, rather than GMT, I do not expect this state to be approached for several decades if it becomes practically important, or much longer if it does not. > A. If the appropriate international bodies decide to eliminate the > insertion of new leap seconds, would you foresee any practical > problems for your institution/instrument/observations? I would expect the problems described above to arise somewhere for some people, to cause confusion in areas where exact timing is legally important, and so to provide a source of undeserved extra income to lawyers. I would not expect to be involved in any such problems myself. It would be appropriate to study what legal definitions of time scales are used around the world, and whether eliminating leap seconds would cause any change in such definitions. > B. Would you be in favor of such a proposal? My preference would be for Markus Kuhn's proposal of leap seconds announced several decades in advance. -- Joseph S. Myers jsm28(a)cam.ac.uk

1 0

FW: partial Macintosh port
by Olson, Arthur David (NCI) Dec. 10, 1999

Dec. 10, 1999

Oscar is not on the time zone mailing list; be sure to direct any replies appropriately. --ado -----Original Message----- From: Oscar van Vlijmen [SMTP:o.van.vlijmen@tip.nl] Sent: Thursday, December 09, 1999 9:53 PM To: tz(a)elsie.nci.nih.gov Subject: partial Macintosh port Dear reader, Partial Macintosh port ---------------- Your excellent timezones information is sadly not available to Macintosh users, apart from the difficult to read text files. Therefore I have made a HyperCard application, based upon the current, so not the historical, timezone information in the tzdata directory. I intend to keep my HyperCard application up to date. It can be found on my site: http://www.tip.nl/users/o.van.vlijmen/AboutTimeZonesHC.html and it has been submitted to Info-Mac. This application is freeware and can be freely distributed, but not to be used for commercial applications and not on CD-ROM's. >From the Read Me ------------- This HyperCard stack by Oscar van Vlijmen can show the time in ALL known time zones, taking the daylight savings time (summertime) into account. Enter a time, set a few time zones, and this stack will compute the correct time in the set time zones. From base time to zone time and vice versa. If you enter a date as well, the correct daylight savings time corrections (summertime) are taken into account. This version can show up to 12 timezones at once. Additions, corrections ---------------- Compared to your database I have made a separate card (record) for each and every timezone, including the uninhabited islands. I have looked up all geographical coordinates, more or less missing from your database. Furthermore I have found a few small errors. For instance, the Brazilian island (Southamerica file) is called Trindade, not Trinidade. I think it should be Manahiki island (Australasia file), part of Cook Islands, not Manihiki. However, the Millennium Edition of the Times atlas says Manihiki Atoll. And also part of the Cook Islands: Rakahanga Atoll is more often used than your Rakehanga. The Times atlas says Rakahanga. For the moment I follow your ONE Mongolian time zone. I did some investigations, however not in Mongolia itself, which give the impression that Mongolia has two timezones. It looks as if the Mongolians themselves don't know, because some Mongolian Internet sites tell you there is one zone, others say two. Lonely Planet says two. If you like, I could send you more details of my investigations. If someone is interested, I could compile and send you a list of all geographical coordinates in my application, including the ones I have found. Regards, Oscar van Vlijmen 10-Dec-1999

2 1