May 2012 - tz - lists.iana.org

northamerica - Nevada - Mountain Time / Pacific Time
by Tobias Conradi May 22, 2012

May 22, 2012

was "Idaho tz correction" On Tue, May 22, 2012 at 12:04 AM, Alan Mintz <Alan_Mintz+TZ_IANA(a)earthlink.net> wrote: > I see someone tried to address this last month but it kind of fell flat. The > solution seems clear to me, as described here: > >> In ftp://ftp.iana.org/tz/data/northamerica v8.54 dated 2009-05-17 (the >> current version), it says" ... > If we wanted to give the same detail treatment to NV and OR: > > # US Pacific time, represented by Los Angeles > # ... > # - Nevada (except for West Wendover) Alan, that would introduce inconsistency, since for Mountain time, represented by Denver it says "Nevada border (Jackpot, Owyhee, and Mountain City)" and then West Wendover wouldn't be covered anywhere. I suggest to keep "most of Nevada", but to amend the comments for the Nevada part that is on Mountain time, so the exceptions are only mentioned at one place. More info the exceptions at: http://en.wikipedia.org/wiki/Jackpot,_Nevada http://en.wikipedia.org/wiki/West_Wendover,_Nevada -- Tobias Conradi Rheinsberger Str. 18 10115 Berlin Germany http://tobiasconradi.com/

1 0

Re: [tz] Bug in tzdata2012c/northamerica 2x Idaho County in America/Los_Angeles
by Steve Jones May 21, 2012

May 21, 2012

At 05:03 5/21/2012, Ian Abbott wrote: >Except that Idaho City is in Boise County, which is covered by a >different rule, so maybe it's not referring to Idaho City after >all. It all depends which rule covers Idaho City. > >So which rule covers Idaho City, America/Los_Angeles or America/Boise? > >Whichever rule it is, I guess the comment(s) could be improved. Here is an image that shows Idaho County, Idaho City, and a magenta line indicating the path of the time zone border in that area: <http://ontimezone.com/images/idahocounty.gif>http://ontimezone.com/images/idahocounty.gif It might be be more accurate to say "northeastern three-quarters of Idaho county", given the portion of the America/Boise rule which juts northwards to the west of the Salmon River. Cheers, Steve Jones

2 1

Theory - proposal to delete the reference to population
by OvV_HN May 21, 2012

May 21, 2012

In reply to: [tz] Theory - proposal to delete the reference to population Tobias Conradi tobias.conradi at gmail.com Fri May 18 18:20:39 UTC 2012 http://home.tiscali.nl/~t876506/Multizones.html says "As soon as the territorial waters of a nation is reached, the clock is set to the local time of that (part of the) nation." Reply: Probably interesting to know: - What you referred to is my page (not interesting, but preliminary to the next). - The chapter about international waters is the result of a discussion on the TZ list a couple of years ago. So, that information is in fact peer reviewed. - The discussion was based on the renowned Bowditch publication The American Practical Navigator. In the chapter mentioned above a table and an algorithm can be found, that can be used to find a kinda logical time for an uninhabited zone. In my opinion: if an authority of an uninhabited island has defined a local time, the TZ database should have a zone designation, whether the island is uninhabited or not. The central place cannot be the largest inhabited place, but could be the geographic center of the island or a landmark of some kind (mountain top, abandoned station, etc.). The mere existence of an ISO code, a FIPS code, internet TLD, whatever, should be of no concern to TZ. Oscar van Vlijmen

2 1

Re: [tz] Theory - proposal to delete the reference to population
by Tobias Conradi May 21, 2012

May 21, 2012

On Sun, May 13, 2012 at 1:46 AM, Paul Eggert <eggert(a)cs.ucla.edu> wrote: > On 05/12/2012 04:08 AM, Tobias Conradi wrote: >> >> If two ships enter territorial waters of HM > > > They'll continue to do what they've always done, > which is keep their own times at the discretion > of their captains. Do you have any source, international treaty or so, to support your claim? If they get on land, would that change your claim, or would they observe what has been reported as official time for HM by Australian agencies? -- Tobias Conradi Rheinsberger Str. 18 10115 Berlin Germany http://tobiasconradi.com/

13 29

Bug in tzdata2012c/northamerica 2x Idaho County in America/Los_Angeles
by Tobias Conradi May 21, 2012

May 21, 2012

tzdata2012c/northamerica ftp://ftp.iana.org/tz/data/northamerica https://github.com/valodzka/tzdata/blame/master/northamerica 2007-08-20 tzdata2007g https://github.com/valodzka/tzdata/commit/4cd595118d4f623edd792ec341ac1bd60… traced back to http://mm.icann.org/pipermail/tz/2007-August/014494.html Paul Eggert eggert at CS.UCLA.EDU Fri Aug 17 22:56:11 UTC 2007 -------cite------- # US Pacific time, represented by Los Angeles ... # ... northern Idaho (Benewah,... # Idaho, ... , and Shoshone counties, # and the northern three-quarters of Idaho county) ... Zone America/Los_Angeles --------cite end----- -- Tobias Conradi Rheinsberger Str. 18 10115 Berlin Germany http://tobiasconradi.com/

3 5

Uzbekistan - insufficient location desciption
by Tobias Conradi May 18, 2012

May 18, 2012

tzdata2012c/asia no location comment except # Uzbekistan for Asia/Samarkand Asia/Tashkent Also no other comment, telling where the Uzbekistan info comes from. tzdata2012c/zone.tab Asia/Samarkand west Uzbekistan Asia/Tashkent east Uzbekistan No definition for "West Uzbekistan" or "East Uzbekistan" found. Trying to improve: Asia/Samarkand per definition includes Samarkand Province, where Samarkand is located. "west Uzbekistan" if only contrasted by "east Uzbekistan" and using the map at http://en.wikipedia.org/wiki/Provinces_of_Uzbekistan likely additionally refers to Karakalpakstan, Xorazm, Buxoro, Navoiy Uncertain remain Jizzax (north east), Qashqadaryo (south) and Surxondaryo (south). But since the two southern country subdivisions, if looked at as one unit, border Samarkand, they may be included in Asia/Samarkand. This leaves five provinces, with Jizzax six, for the Asia/Tashkent. If that is the case zone.tab could be improve by Asia/Tashkent east Uzbekistan (Andijon, Fergana, Namangan, Tashkent, Sirdaryo [, Jizzax]) -- Tobias Conradi Rheinsberger Str. 18 10115 Berlin Germany http://tobiasconradi.com/

1 1

Theory - proposal to delete the reference to population
by Tobias Conradi May 18, 2012

May 18, 2012

References to population in tzcode2012b/Theory 1) "However, uninhabited ISO 3166 regions like Bouvet Island do not need locations, since local time is not defined there." 2) "Use UTC (with time zone abbreviation "zzz") for locations while uninhabited." ------- 1a) http://wikitravel.org/en/Bouvet_Island reports Olav Peak exists on Bouvet Island, without discussing a need for the peak. It is out of the scope of the tzdb as defined in the Theory file to decide on whether regions need locations. 1b) there is no definition of "uninhabited" German WP reports a team of people has been there after the 1970-01-01 cutoff point: http://de.wikipedia.org/wiki/Bouvetinsel#cite_note-6 "Vom 24. Dezember 1978 bis zum 8. März 1979 hielt sich ein norwegisches Forschungsteam auf der Insel auf und führte biologische und geologische Forschungsarbeiten durch." There are more sources supporting the claim that it had been populated. 1c) There is no evidence that local time is not defined if a population is absent. http://en.wikipedia.org/wiki/Yukon-Koyukuk_Census_Area,_Alaska is mostly uninhabited but the whole area is observing UTC-09:00 according to http://commons.wikimedia.org/wiki/File:Timezoneswest.PNG maybe covered by America/Anchorage Alaska Time but no further geographic description found HM as no zone, but http://www.timegenie.com/country.time/hm "To ensure accuracy of information, timegenie.com contacted the Australian Antarctic Division [http://www.antarctica.gov.au/] . On January 3, 2005, a senior official confirmed that Heard Island and McDonald Islands are UTC/GMT +5 hours." 2a) A location changes its UTC offset when people move out or in? http://en.wikipedia.org/wiki/Mount_Chamberlin_(Alaska) on UTC most of the year? A ship landing on an island, changes the UTC offset observed on that island? The time a meteorite lands on island depends on whether people have been there or not? 1+2) Violate tzcode2012b/Theory: ----- Scope of the tz database ----- The tz database attempts to record the history and predicted future of all computer-based clocks that track civil time. To represent this data, the world is partitioned into regions whose clocks all agree about time stamps that occur after the somewhat-arbitrary cutoff point of the POSIX Epoch (1970-01-01 00:00:00 UTC). 1+2) Are inconsistently applied. Uninhabited locations are treated differently, depending on whether they are located in countries that have been inhabited during all time between 1970-01-01 and 2012-05-12 or not. If Bouvet Island would be part of TF, it could be included in a zone but since it is covered by BV it is not. -- Tobias Conradi Rheinsberger Str. 18 10115 Berlin Germany http://tobiasconradi.com/

3 6

Theory - typo - simplifes -> simplyfies
by Tobias Conradi May 18, 2012

May 18, 2012

tzcode2012b/Theory Indicate to humans as to where that region is. This simplifes use. simplifes -> simplyfies -- Tobias Conradi Rheinsberger Str. 18 10115 Berlin Germany http://tobiasconradi.com/

5 8

Infrastructure - ftp timestamps for .gz not showing creation time
by Tobias Conradi May 18, 2012

May 18, 2012

ftp://ftp.iana.org/tz/releases/ failing to show the timestamps for archive creation time, showing 2011-10-13 00:00:00 instead until release of tzcode2011i.tar.gz and tzdata2011m.tar.gz inclusive. -- Tobias Conradi Rheinsberger Str. 18 10115 Berlin Germany http://tobiasconradi.com/

2 2

Terminology mismatch in tzcode file format manpage / header
by Andrew Paprocki May 18, 2012

May 18, 2012

The file format is described with these fields: .I tzh_ttisgmtcnt The number of UTC/local indicators stored in the file. .I tzh_ttisgmtcnt UTC/local indicators, each stored as a one-byte value; they tell whether the transition times associated with local time types were specified as UTC or local time, and are used when a time zone file is used in handling POSIX-style time zone environment variables. A number of variables in localtime.c, tzfile.h, and zic.c dealing with these fields have variations of this "isgmt" substring. This presents a naming conflict because technically GMT and UTC are two different things. I find the inconsistent naming creeping into functions which deal with the file format. I would like to prevent variations of the string "GMT" from winding up in API names since the definition of the field refers to UTC. I propose wholesale replacing of the "isgmt" substring in tzcode files with "isutc" so that the file format fields are named consistently with their definition. If absolutely necessary, a #define can be used in tzfile.h so that code referencing tzh_ttisgmtcnt will still compile. -Andrew

4 7