March 2012 - tz - lists.iana.org

Australian Time Zone Abbreviations Discussion Review
by Shaun Bouckaert March 5, 2012

March 5, 2012

I started a discussion on this mailing list in November 2011 in regards to the incorrect time zone abbreviations used for Australian time zones. Initially there were several responses, the majority of which were positive to change, and yet the conversation seems to have died off. Recently published on this list was the IETF RFC 6557 "Procedures for Maintaining the Time Zone Database". The relevant section of that document to this discussion, Section 3 "Making Updates to the TZ Database" specifies at the bottom "To be clear, the TZ Coordinator SHALL NOT set time zone policy for a region but use judgement and whatever available sources exist to assess what the average person on street would think the time actually is, or in case of historical corrections, was." Also, point 3.3 states "Changes to existing entries SHALL reflect the consensus on the ground in the region covered by that entry." There were several people involved in the discussion that took place: Myself (Shaun Bouckaert <shaun.bouckaert(a)gmail.com>) Edwin Groothuis <edwin(a)mavetju.org> Eric Ulevik <eulevik(a)gmail.com> Elliot Lear <lear(a)cisco.com> Greg Black <gjb(a)yaxom.com> I reside in Australia, however I am not aware of the nationality or residency of any of the other contributors. I will summarise my points then attempt to do so with the points made by others. It is not my intention to mislead and if my summaries of their contributions are inaccurate then I hope they will correct me. Their contributions to the original discussion are available to view. I originally started the discussion after finding the Australian Federal Government had a web page of information regarding time zones in Australia. This can be accessed at http://australia.gov.au/about-australia/our-country/time This lists the timezones as Australian Eastern Standard Time: UTC +10 hours abbreviated as AEST Australian Central Standard Time: UTC +9 ½ hours abbreviated as ACST Australian Western Standard Time: UTC +8 hours abbreviated as AWST Where daylight savings is concerned, NSW, ACT, Victoria and Tasmania move from AEST to Australian Eastern Daylight Time UTC +11 hours abbreviated as AEDT SA and Broken Hill in NSW move from ACST to Australian Central Daylight Time UTC +10 ½ hours abbreviated as ACDT After this initial posting Edwin replied with his support. Eric then replied stating that Time Zones are legislated at the state level, not the federal level, seeming to imply that only state law should matter. Considering the RFC specifically states that the time zone policy should be based on available sources to assess what the average person on the street would think, and the fact that state legislation doesn't specify an abbreviation, there's no reason that this federal government information, which accurately reflects what everyone I have discussed this matter with understands to be true, should be discounted. This was reenforced by Edwin who pointed out two ways forward, stay where we are based on the absence of any specification of abbreviations in state legislation and continue bickering on this issue or move forward with the clearly defined federal information that fixes ambiguity that exists on an international level with the existing abbreviations and hopefully stop the bickering. Elliot then put forward two points, the first in line with the now existing guidelines in regards to what people on the street think, and the second point being to taking caution to possible side effects that could be caused by changing this. He points out that we could be going from ambiguous labels to something that is also ambiguous, but is there a way of finding out if the AEST, ACST, etc. are ambiguous in any way? Considering the significance of EST and CST to North America, it certainly seems more significant to reduce the ambiguity there. Edwin then found several Australian organisations and reviewed what they use. Broadcasters: The ABC (National Broadcaster): AEST SBS: AEST Channel 10: Mostly AEST, some instances of EST News Organisations: Sydney Morning Herald: EST The Australian: AEST The Daily Telegraph: Both EST and AEST Government Organisations: The Australian Bureau of Meteorology: Uses both, but specifies using AEST in their submission guidelines. Appended to this list I can add the Queensland Department of Transport and Main Roads, which I found is using AEST in their website Terms of Service. I also pointed out at the time that many instances of the use of EST could easily be due to the current state of the TZ database and would likely be corrected to AEST (or other where appropriate) with the update to the TZ database. It was also pointed out that Australians would on large understand the use of the A prefixed abbreviations due to their widespread use, especially by major broadcasters and news media. Greg Black added that he received three emails from different Australian organisations that used ADST for a timezone that he would personally label AEDT. I do not see this being an argument against the change, but indeed an argument for. There is existing confusion, some of it due to the incorrect labels in the TZ database. The fact that the federal government has presented a source for this change means that there's significant grounds for using it as a correct version or standard. Of the 5 people mailing the list during this discussion, the most significant contributions were made by myself and Edwin, both of us in support of the change to the A prefix, which also seemed to be supported by Greg Black who said he would personally label the timezone that the ADST label had referred to as AEDT. I could find no arguments put forward that demonstrated any advantage of the current system over the proposed alternative besides maintaining the status quo. I respect the caution put towards unintended side effects, but where these are due to corrections on what is arguably wrong information, isn't it better to fix it sooner rather than later? Shaun Bouckaert

4 4

apparent-solar-time zone possible cleanup
by Zefram March 4, 2012

March 4, 2012

The apparent-solar-time zones seem to be misusing the "zzz" indicator, and they have poor boundary behaviour. The attached patch makes these changes that seem like a good idea to me: * abbreviation "LAT", for "Local Apparent Time", during the period where the zone provides LAT behaviour * additional apparent-time offset entry for Dec 31 of previous year, so that the ruleset provides LAT correctly for the first half of Jan 1 * range of validity of the LAT data defined strictly as starting and ending at Jan 1 00:00 UT * zone marked as disused outside the range of validity of the LAT data Also, thoughts on apparent solar time more generally: the existing solar?? files inconveniently can't be merged because the result would have more than 256 distinct offsets. However, if the resolution of the offsets were dropped from 5 s to 10 s, then there would be only 186 distinct offsets within the range of the equation of time, so we could have an apparent-time ruleset covering as many years as we're interested in. That could allow incorporating those rules into the regular Asia/Riyadh zone, rather than requiring these hacky separate zones. The apparent-time ruleset could also be formulaically generated at build time, rather than explicitly included in the source tarball. -zefram

1 1

IANA website error: 2012b -> 2012a
by Andrew Paprocki March 2, 2012

March 2, 2012

The IANA website currently says: "Time Zone Data v. 2012b (Released 2 March 2012) tzdata2012b.tar.gz (202kb)" Yet the href points to: http://www.iana.org/time-zones/repository/releases/tzdata2012a.tar.gz -Andrew

2 1

tabs in zone.tab
by Zefram March 2, 2012

March 2, 2012

My automatic repackager for CPAN has just barfed on the zone.tab in tzdata2012a. I should have checked the diff more carefully, I guess. The fault it finds is that the new America/Creston line has *two* tab characters between "America/Creston" and the region description. The comments at the top of the file include "Columns are separated by a single tab.". Previously the file adhered to the single tab rule, and I made my parser correspondingly strict. Obviously I'll manually intervene to get 2012a repackaged, but I think the file should be fixed for 2012b. -zefram

3 4

Reports of Leap Year problem...
by Robert Elz March 1, 2012

March 1, 2012

I cannot confirm any of this, except that the URL's included (the relevant ones for this list anyway) do point at web pages with content as claimed, but this seems like the kind of information that readers of this list would be amused by... kre

1 0

tzcode2012b & tzdata2012b available
by Robert Elz March 1, 2012

March 1, 2012

The files ftp://munnari.oz.au/pub/tzdata2012b.tar.gz ftp://munnari.oz.au/pub/tzcode2012b.tar.gz are now available. These reflect the changes circulated earlier today on the list. There is just one change to tzcode2012b (compared witth 2012a): the Makefile that was accidentally included with 2012a has been replaced with the version that should have been there, which is identical with the previous version (from tzcode2011i). For people upgrading directly from 2011i to 2012b, the changes in tzcode are just to the README and tz-art.htm and tz-link.htm and reflect changes to the mailing list address & subscription URLs, and the code distribution URLs and methods. tz-link.htm also has a link to DateTime::TimeZone::Tzfile added. There are just two changes in tzdata2012b compared with 2012a. Most significantly, summer time in Cuba has been delayed 3 weeks (now starts April 1 rather than March 11). Since Mar 11 (the old start date, as listed in 2012a) is just a little over a week away, this change is urgent. Less importantly, an excess tab in one of the changes in zone.tab in 2012a has been removed. Compared to the last long lived distribution (given tzdata2012a existed less than a day) which was tzdata2011n, the other major changes are: Chile 2011/2012 and 2012/2013 summer time date adjustments. Falkland Islands onto permanent summer time (we're assuming for the foreseeable future, though 2012 is all we're fairly certain of.) Armenia has abolished Summer Time. Tokelau jumped the International Date Line back last December (just the same as their near neighbour, Samoa). America/Creston is a new zone for a small area of British Columbia There will be a leapsecod 2012-06-30 23:59:60 UTC. Other minor changes are: Corrections to 1918 Canadian summer time end dates. Updated URL for UK timezone history (in comments) A few typos in Le Corre's list of free French place names (comments) Once again, crypto hashes of the distribution files, generated using NetBSD's pkgsrc digest program, are appended so you can verify the distribution files you fetch have not been corrupted. kre MD5 (tzcode2012b.tar.gz) = 6137322ffd36e1fd5128885be1c57008 RMD160 (tzcode2012b.tar.gz) = 25acce807c9e19d8294fe4dd8eac08fba9d796af SHA1 (tzcode2012b.tar.gz) = 7172f71a23fb6fe8f99099f810d05366ed5d7130 SHA512 (tzcode2012b.tar.gz) = 12b60b6b00db51651a85e4448c8795c8f0d0fbeeec66bec4e8613dff41fec3324a0b20f7d66183541a389eff0d2a93b3346a045a58a0d977739924f06d017fd3 TIGER (tzcode2012b.tar.gz) = a4378982416395d5d632270db4fb3c5c568ebf57b7130636 WHIRLPOOL (tzcode2012b.tar.gz) = 2a343f318b737a2b4e85f29416d25b07a04f3395a49c6e863a3bf4492b63de17545d30b0377ab4d1587fd0fc85ecf234fe2c45bff4bd2f0f97b33b82b3f2e3f0 MD5 (tzdata2012b.tar.gz) = 0615fd29def380a917e528433c820368 RMD160 (tzdata2012b.tar.gz) = b58bb8a633b664f5babed8a6681ecc25a039733d SHA1 (tzdata2012b.tar.gz) = cd7c43bf8ccf56d174a2b85d2da35e83fd8f1107 SHA512 (tzdata2012b.tar.gz) = 08f3f07e0abf75d278210fc9237344b6613b6b055bedf974b75437bfa153bdaf238135487acde38f352e55805b5bc9b3db93d9c08fdb3cfba68b3609c06445f9 TIGER (tzdata2012b.tar.gz) = 0b67a1f5408d8b6addd7c07df6e87fa9e54bb9102ae758fd WHIRLPOOL (tzdata2012b.tar.gz) = 530a6bcdf4fe4c92c478e709093c19fc2424c52a55756ea2e6cf66854f659bd67c6503e6235aa23348f5caeb5e92348a34d731334ca168a410feaa33f85533b0

1 0

Proposed update to the tz package files
by Robert Elz March 1, 2012

March 1, 2012

The combination of an accidental change to the Makefile in tzcode2012a that was never intended to be released, and the change of the summer time start date for Cuba, that we discovered just too late for 2012a (and the stray extra tab in zone.tab) has led me to the conclusion that we need 2012b very soon - like today... There will be both code & data files - the sole change in the code is to return the Makefile to the exact same version it had in 2011i, which was what it was supposed to be in 2012a. A diff of the data change from 2012a to (what will probably become) 2012b is appended. The only real change is to the start date for summer time in Cuba this year. The end date is not yet certain, so I'll just leave it as we have it (last Sun in Oct) and we'll see what happens in September or October (probably). Since there is no discernable pattern to the start dates, I'm also proposing leaving that for 2013 as we had assumed it would be for 2012 (second Sunday in March). It has been that more years recently than anything else (just not last year, or this year, but they've not been the same either.) The excess tab from the zone.tab file is also gone. For anyone considering upgrading their system to 2012a, who hasn't already done it, I'd advise waiting on 2012b, it is not going to be a long wait. kre --- northamerica 2012/03/01 11:22:55 1.5 +++ northamerica 2012/03/01 20:46:07 @@ -1,5 +1,5 @@ # <pre> -# @(#)northamerica 8.52 +# @(#)northamerica 8.53 # This file is in the public domain, so clarified as of # 2009-05-17 by Arthur David Olson. @@ -2820,8 +2820,9 @@ Rule Cuba 2009 2010 - Mar Sun>=8 0:00s 1:00 D Rule Cuba 2011 only - Mar Sun>=15 0:00s 1:00 D Rule Cuba 2011 only - Nov 13 0:00s 0 S -Rule Cuba 2012 max - Mar Sun>=8 0:00s 1:00 D +Rule Cuba 2012 only - Apr 1 0:00s 1:00 D Rule Cuba 2012 max - Oct lastSun 0:00s 0 S +Rule Cuba 2013 max - Mar Sun>=8 0:00s 1:00 D # Zone NAME GMTOFF RULES FORMAT [UNTIL] Zone America/Havana -5:29:28 - LMT 1890 --- zone.tab 2012/03/01 11:26:34 1.6 +++ zone.tab 2012/03/01 15:30:24 @@ -131,7 +131,7 @@ CA +690650-1050310 America/Cambridge_Bay Mountain Time - west Nunavut CA +6227-11421 America/Yellowknife Mountain Time - central Northwest Territories CA +682059-1334300 America/Inuvik Mountain Time - west Northwest Territories -CA +4906-11631 America/Creston Mountain Standard Time - Creston, British Columbia +CA +4906-11631 America/Creston Mountain Standard Time - Creston, British Columbia CA +5946-12014 America/Dawson_Creek Mountain Standard Time - Dawson Creek & Fort Saint John, British Columbia CA +4916-12307 America/Vancouver Pacific Time - west British Columbia CA +6043-13503 America/Whitehorse Pacific Time - south Yukon

1 0

Cuba delays DST to April
by Steffen Thorsen March 1, 2012

March 1, 2012

According to Radio Reloj, Cuba will start DST on Midnight between March 31 and April 1. Radio Reloj has the following info (Spanish): http://www.radioreloj.cu/index.php/noticias-radio-reloj/71-miscelaneas/7529… Our info on it: http://www.timeanddate.com/news/time/cuba-starts-dst-2012.html Best regards, Steffen Thorsen - timeanddate.com

2 1

tzcode2012a & tzdata2102a released.
by Robert Elz March 1, 2012

March 1, 2012

The files ftp://munnari.oz.au/pub/tzdata2012a.tar.gz ftp://munnari.oz.au/pub/tzcode2012a.tar.gz are now available. These reflect the changes circulated earlier this week, as corrected based upon reports received, and summarised below. The changes in tzcode2012a (compared to the previous version, 2011i) are entirely to the README and tz-art.htm and tz-link.htm files, if none of those concern you, you can ignore the code update. The changes reflect the changed addresses for the mailing list and the code and data distribution points & methods (and a link to DateTime::TimeZone::Tzfile has been added to tz-link.htm). In tzdata2012a (compared to the previous release, which was 2011n) the major changes are: Chile 2011/2012 and 2012/2013 summer time date adjustments. Falkland Islands onto permanent summer time (we're assuming for the foreseeable future, though 2012 is all we're fairly certain of.) Armenia has abolished Summer Time. Tokelau jumped the International Date Line back last December (just the same as their near neighbour, Samoa). America/Creston is a new zone for a small area of British Columbia There will be a leapsecod 2012-06-30 23:59:60 UTC. A few more minor changes are also included: Changes to Canadian 1918 summer time end dates (so it occurred on a Sunday, rather than Thursday, which is more likely correct.) A new URL (in the comments in the europe zone file) to the article describing the history of timezones in the UK (this should probably be promoted to a link in tzlink.htm or something, but that hasn't happened yet.) A couple of typo corrections to the list of French place names reported by Henri Le Corre that define the upper limit of the free zone where (as I understand it) German rules were not applies during (a part of) WWII. These are based on the rendition supplied by John Halloran, and unless I botched it, should make what we have be a correct copy of the source, as nearly as is possible in strict ASCII (the accent marks are missing...) I know I said we would hold off on this until the next version, but with the data being available, and the corrections needed so simple, I thought I could include it in this version. Now we can discuss what how we should represent the data with perhaps more modern (or better ascii versions of) place names in the next version, but at least now we're working from a common (external) base source. I still have not implementewd (even really seriously investigated) the procedures necessary to generate signed files, so once again I am appending to this message the results of several (crypto strength) hash functions applied to the two files (and this time, all this has been prepared on my laptop at a time when it isn't connected to any network at all, so there's an even higher level of assurance than normal that these hashes really do represent the data I intend to distribute.) The "digest" program from NetBSD's pkgsrc (which can be installed on many different OSs) was used to generate these values (but anything running any of the same algorithms should produce the same results.) Finally, it is my hope (though without intending to apply any pressure) that the next release will be back under the coordination of Arthur David Olson (I'll continue if he's unwilling, and Paul Eggert doesn't want to take over instead, but I'd much prefer to see ado back in control, for as long as he wants to continue.) Thanks for all the assistance from everyone that I've received over the few months I've been doing this, this is truly one of the nicest and most satisfying of the network communities in which to participate. kre MD5 (tzcode2012a.tar.gz) = a72bb8c29893ee3bfd1101aca29d4064 MD5 (tzdata2012a.tar.gz) = cc43867f4768e5c2e8315d9b3cac62ed RMD160 (tzcode2012a.tar.gz) = b498355392e7e141e12fc4ca3592cfec6e96ef00 RMD160 (tzdata2012a.tar.gz) = b95722a2d42744bb217e11ace7b5bc7195f86d77 SHA1 (tzcode2012a.tar.gz) = 80e0c9efb067950f34826d0708ff61e72e019da5 SHA1 (tzdata2012a.tar.gz) = 5bd828d0845937da92c0518060a4aa8fbad81f8d SHA512 (tzcode2012a.tar.gz) = c37d95081a1097ba46d1887ef13bf112216bc45f21542bf559744f329602e109c3a9a7a7705ed41432049ad3905f87a747ddbc6f7c11470fcf44fc5bf946c9e3 SHA512 (tzdata2012a.tar.gz) = dff197b7ae7a204fa2c716d27ad83c0f29fba3bdede0059d0507ad3f4117849c90e578df5bba62c397e1709bce1c26382a61b0ea9c8face671e7228dbf5b7bab TIGER (tzcode2012a.tar.gz) = 3f38fc0884f3eef800289a9e6cd9ee69b0405f75447319fc TIGER (tzdata2012a.tar.gz) = d0609c3f36e735ab5708f473336f66edb33f68d58333b9dc WHIRLPOOL (tzcode2012a.tar.gz) = 5a98d5ecb0956eba9f52403ea58125933998f3d1f402fba5b04815b5c2d861e9c010df38258a8f479c8a26a7d301c5d45e63a5a0586e24dcdd97007051e0ca5b WHIRLPOOL (tzdata2012a.tar.gz) = 4c3e7e372d708baf5cf15ece01c31f38880e50fbf402ef346d0cd27f4203282862660322a76efb1144b84ff08c86407e714fc1d858e1c7ce22f7cbb10923fb52

3 3

Re: [tz] Chilean Government annuounced new date for leaving and entering DST in 2012
by Glenn Eychaner March 1, 2012

March 1, 2012

On Feb 29, 2012, at 9:00 AM, tz-request(a)iana.org wrote: > Thank you so much for the diff. - as you said, we can only hope this change gets formalized before March 11 > There is another thread about this ([tz] Updated Proposed time zone package changes (Armenia, Chile, Falklands, Tokelau, Antarctica, Creston(Canada))), according to the contents in that thread an updated timezone package will be released on March 1st 2012 Since Microsoft also seems to be taking this seriously and preparing a hotfix http://blogs.technet.com/b/latam/archive/2012/02/27/chile-dst-2012-fallback… I am currently assuming that this change is "official", and recommending including it in the March 1 timezone update. Finally, I realized that once again I forgot to do the Antarctica file, so here is an update for that file which patches in all the recent time zone changes over the last few years. Officially, ChileAQ should have the same offset and rules as mainland Chile, but I don't know what is done in actual practice. --- antarctica.orig 2011-09-20 14:55:03.000000000 -0300 +++ antarctica 2012-02-29 10:15:23.000000000 -0300 @@ -43,6 +43,15 @@ Rule ChileAQ 1998 only - Sep 27 4:00u 1:00 S Rule ChileAQ 1999 only - Apr 4 3:00u 0 - -Rule ChileAQ 1999 max - Oct Sun>=9 4:00u 1:00 S -Rule ChileAQ 2000 max - Mar Sun>=9 3:00u 0 - +Rule ChileAQ 1999 2010 - Oct Sun>=9 4:00u 1:00 S +Rule ChileAQ 2011 only - Aug Sun>=16 4:00u 1:00 S +Rule ChileAQ 2012 only - Sep 2 4:00u 1:00 S +Rule ChileAQ 2013 max - Oct Sun>=9 4:00u 1:00 S +Rule ChileAQ 2000 2007 - Mar Sun>=9 3:00u 0 - +Rule ChileAQ 2008 only - Mar 30 3:00u 0 - +Rule ChileAQ 2009 only - Mar Sun>=9 3:00u 0 - +Rule ChileAQ 2010 only - Apr Sun>=1 3:00u 0 - +Rule ChileAQ 2011 only - May Sun>=2 3:00u 0 - +Rule ChileAQ 2012 only - Apr 29 3:00u 0 - +Rule ChileAQ 2013 max - Mar Sun>=9 3:00u 0 - # These rules are stolen from the `australasia' file. -- Glenn Eychaner (geychaner(a)lco.cl) Telescope Systems Programmer, Las Campanas Observatory

2 5