tzdb data for North and Central America
Hello. You might want to note that the America/Halifax timezone also applies to Îles-de-la-Madeleine and the Listuguj reserve in Quebec. Officially, this came into effect on January 1, 2007 (*Legal Time Act*, CQLR c T-5.1), but the legislative debates surrounding that bill say that it is "accommodating the customs and practices" of those regions, which suggests that they have always been in-line with Halifax. Also, you call America/Whitehorse "south Yukon" and America/Dawson "north Yukon", but the divide is actually between east and west, at the 138th meridian. -- *Jeffery Nichols*, JD, MPA, MA
Thank you for those corrections. I installed the attached patch to tzdb's development version and these fixes should appear in the next release.
Great! I was also wondering whether anyone on your team has more data about what towns were part of America/Nipigon. I'm trying to map all of the Canadian IANA zones on OpenStreetMap. Your tzdb mentions that Nipigon was the largest town in its zone, which suggests that the zone wasn't limited to Nipigon town boundaries, but it doesn't give any clues about what other towns were found to follow it. Thanks in advance. On Fri, 10 Jan 2020 at 16:51, Paul Eggert <eggert@cs.ucla.edu> wrote:
Thank you for those corrections. I installed the attached patch to tzdb's development version and these fixes should appear in the next release.
-- *Jeffery Nichols*, JD, MPA, MA
On 2020-01-13 09:15, Jeffery Nichols wrote:
On Fri, 10 Jan 2020 at 16:51, Paul Eggert wrote:
Thank you for those corrections. I installed the attached patch to tzdb's development version and these fixes should appear in the next release.> Great! I was also wondering whether anyone on your team has more data about what towns were part of America/Nipigon. I'm trying to map all of the Canadian IANA zones on OpenStreetMap. Your tzdb mentions that Nipigon was the largest town in its zone, which suggests that the zone wasn't limited to Nipigon town boundaries, but it doesn't give any clues about what other towns were found to follow it. Look at the sites http://efele.net/maps/tz/canada/ and https://github.com/evansiroky/timezone-boundary-builder for Canadian boundary info and background; search in the mailing list archive from 2006 back for Nipigon and Ontario, and look in zone.tab and northamerica for comments about NW Ontario in context:
$ egrep '^CA\s+.*\s-\sON' /usr/share/zoneinfo/zone.tab CA +4339-07923 America/Toronto Eastern - ON, QC (most areas) CA +4901-08816 America/Nipigon Eastern - ON, QC (no DST 1967-73) CA +4823-08915 America/Thunder_Bay Eastern - ON (Thunder Bay) CA +484531-0913718 America/Atikokan EST - ON (Atikokan); NU (Coral H) CA +4953-09709 America/Winnipeg Central - ON (west); Manitoba CA +4843-09434 America/Rainy_River Central - ON (Rainy R, Ft Frances) $ less '+/Ontario|Nipigon|Atikokan|Rainy_River|Thunder_Bay' northamerica If you can't find an online searchable tz mailing list archive, you can download about 18MB+ of archives starting from https://mm.icann.org/pipermail/tz/1986-November.txt.gz to the current month, gunzip to about 70MB+ of posts, import them into a folder in your mail client for convenient searching, and subscribe to the mailing list(s) to keep your local archive up to date. -- Take care. Thanks, Brian Inglis, Calgary, Alberta, Canada This email may be disturbing to some readers as it contains too much technical detail. Reader discretion is advised.
Jeffery Nichols, If you haven't already seen it, please check out the Timezone Boundary Builder project here: https://github.com/evansiroky/timezone-boundary-builder. I have made a script that extracts data from OpenStreetMap to create shapefiles of the world's timezone boundaries. I'd be very interested in using some of your research and OpenStreetMap edits to produce better data. Evan On Tuesday, 14 January 2020, 06:19:49 pm PST, Jeffery Nichols <jeffery.nichols@gmail.com> wrote: Great! I was also wondering whether anyone on your team has more data about what towns were part of America/Nipigon. I'm trying to map all of the Canadian IANA zones on OpenStreetMap. Your tzdb mentions that Nipigon was the largest town in its zone, which suggests that the zone wasn't limited to Nipigon town boundaries, but it doesn't give any clues about what other towns were found to follow it. Thanks in advance. On Fri, 10 Jan 2020 at 16:51, Paul Eggert <eggert@cs.ucla.edu> wrote: Thank you for those corrections. I installed the attached patch to tzdb's development version and these fixes should appear in the next release. -- Jeffery Nichols, JD, MPA, MA
On 2020-01-13 09:15, Jeffery Nichols wrote:
On Fri, 10 Jan 2020 at 16:51, Paul Eggert wrote:
Thank you for those corrections. I installed the attached patch to tzdb's development version and these fixes should appear in the next release.> Great! I was also wondering whether anyone on your team has more data about what towns were part of America/Nipigon. I'm trying to map all of the Canadian IANA zones on OpenStreetMap. Your tzdb mentions that Nipigon was the largest town in its zone, which suggests that the zone wasn't limited to Nipigon town boundaries, but it doesn't give any clues about what other towns were found to follow it.
Look at the sites http://efele.net/maps/tz/canada/ and https://github.com/evansiroky/timezone-boundary-builder for Canadian boundary info and background; search in the mailing list archive from 2006 back for Nipigon and Ontario, and look in zone.tab and northamerica for comments about NW Ontario in context: $ egrep '^CA\s+.*\s-\sON' /usr/share/zoneinfo/zone.tab CA +4339-07923 America/Toronto Eastern - ON, QC (most areas) CA +4901-08816 America/Nipigon Eastern - ON, QC (no DST 1967-73) CA +4823-08915 America/Thunder_Bay Eastern - ON (Thunder Bay) CA +484531-0913718 America/Atikokan EST - ON (Atikokan); NU (Coral H) CA +4953-09709 America/Winnipeg Central - ON (west); Manitoba CA +4843-09434 America/Rainy_River Central - ON (Rainy R, Ft Frances) $ less '+/Ontario|Nipigon|Atikokan|Rainy_River|Thunder_Bay' northamerica If you can't find an online searchable tz mailing list archive, you can download about 18MB+ of archives starting from https://mm.icann.org/pipermail/tz/1986-November.txt.gz to the current month, gunzip to about 70MB+ of posts, import them into a folder in your mail client for convenient searching, and subscribe to the mailing list(s) to keep your local archive up to date. -- Take care. Thanks, Brian Inglis, Calgary, Alberta, Canada This email may be disturbing to some readers as it contains too much technical detail. Reader discretion is advised.
Brian Inglis wrote:
If you can't find an online searchable tz mailing list archive, you can download about 18MB+ of archives starting from
https://mm.icann.org/pipermail/tz/1986-November.txt.gz
to the current month, gunzip to about 70MB+ of posts, import them into a folder in your mail client for convenient searching, and subscribe to the mailing list(s) to keep your local archive up to date.
Hm, I just visited https://mm.icann.org/mailman/listinfo/tz followed the link "Overview of all iana.org mailing lists" in the footer of that page, which points to https://mm.icann.org/mailman/listinfo but didn't see the tz list on that page. Is that intentional, or am I missing something, or should the mailing list admins be asked to add a reference to the tz list? Martin -- Martin Burnicki Senior Software Engineer MEINBERG Funkuhren GmbH & Co. KG Email: martin.burnicki@meinberg.de Phone: +49 5281 9309-414 Linkedin: https://www.linkedin.com/in/martinburnicki/ Lange Wand 9, 31812 Bad Pyrmont, Germany Amtsgericht Hannover 17HRA 100322 Geschäftsführer/Managing Directors: Günter Meinberg, Werner Meinberg, Andre Hartmann, Heiko Gerstung Websites: https://www.meinberg.de https://www.meinbergglobal.com Training: https://www.meinberg.academy
Howdy, Quoting Martin Burnicki on Friday January 17, 2020:
Hm, I just visited
https://mm.icann.org/mailman/listinfo/tz
followed the link "Overview of all iana.org mailing lists" in the footer of that page, which points to
https://mm.icann.org/mailman/listinfo
but didn't see the tz list on that page.
Is that intentional, or am I missing something, or should the mailing list admins be asked to add a reference to the tz list?
That index page is not very useful, it contains only a small fraction of hundreds of mailing lists hosted on mm.icann.org, and those that are listed are probably inadvertantly listed based on one of the raft of configuration toggles in the Mailman interface. IANA leverages ICANN's mailing list server for the few dozen mailing lists we administer and none are listed on that index page, we link to them from the individual topic area or registry the list supports (for those that are open to the public). I've dropped a note to ICANN's Engineering and IT folks to see if they want to do anything about cleaning up that index page, but for completeness, we operate two lists to support the Time Zone Database: Discussion List: https://mm.icann.org/mailman/listinfo/tz Announcement List: https://mm.icann.org/mailman/listinfo/tz-announce kim
On 1/13/20 8:15 AM, Jeffery Nichols wrote:
I was also wondering whether anyone on your team has more data about what towns were part of America/Nipigon.
I got that data from my printed copy of the Shanks International Atlas. It contains hundreds of place names in Canada but does not sort them by time zone region. I just now looked in Shanks (4th edition) for place names beginning with "A" with the same time zone history as Nipigon (Shanks calls this Canada's region 35), and found the following: Alfred Algoma Mills Algonquin Almonte Alvinston Appleton Apsley Arden Armstrong Armstrong Station Aroland Attawapiskat Auburn The book has latitude and longitude for each of these entries. It would take some time to transcribe all this info (including B-Z) for all of Canada. Also, Shanks partitions the world into many more regions than tzdb does, because he doesn't have the 1970 cutoff that tzdb does. So he partitions Canada into 253 regions and there could be other Shankian regions that share Nipigon's post-1970 time zone history and would thus fall under America/Nipigon in tzdb. With the above in mind, you could grab a copy of the book and get the data you need. Although the book is out of print, copies of the final (6th) edition can be found in libraries: https://www.worldcat.org/oclc/76950459 https://www.worldcat.org/oclc/1135266594 and used copies are for sale in various locations. I misplaced my copy of the 6th edition during the 2011 astrologer lawsuit ruckus (it probably ended up with one of the attorneys I consulted), but I just now ordered a used copy so that I can more easily answer historical questions like yours in the future.
This is great, thank you! It looks like mapping the Nipigon zone will be trickier than I thought, because some of those towns are on the far side of places bigger than Nipigon, like Sault Ste Marie. On Fri, 17 Jan 2020 at 02:01, Paul Eggert <eggert@cs.ucla.edu> wrote:
On 1/13/20 8:15 AM, Jeffery Nichols wrote:
I was also wondering whether anyone on your team has more data about what towns were part of America/Nipigon.
I got that data from my printed copy of the Shanks International Atlas. It contains hundreds of place names in Canada but does not sort them by time zone region. I just now looked in Shanks (4th edition) for place names beginning with "A" with the same time zone history as Nipigon (Shanks calls this Canada's region 35), and found the following:
Alfred Algoma Mills Algonquin Almonte Alvinston Appleton Apsley Arden Armstrong Armstrong Station Aroland Attawapiskat Auburn
The book has latitude and longitude for each of these entries. It would take some time to transcribe all this info (including B-Z) for all of Canada. Also, Shanks partitions the world into many more regions than tzdb does, because he doesn't have the 1970 cutoff that tzdb does. So he partitions Canada into 253 regions and there could be other Shankian regions that share Nipigon's post-1970 time zone history and would thus fall under America/Nipigon in tzdb.
With the above in mind, you could grab a copy of the book and get the data you need. Although the book is out of print, copies of the final (6th) edition can be found in libraries:
https://www.worldcat.org/oclc/76950459
https://www.worldcat.org/oclc/1135266594
and used copies are for sale in various locations.
I misplaced my copy of the 6th edition during the 2011 astrologer lawsuit ruckus (it probably ended up with one of the attorneys I consulted), but I just now ordered a used copy so that I can more easily answer historical questions like yours in the future.
-- *Jeffery Nichols*, JD, MPA, MA
I believe the Shanks data for Ontario are rather messy and not fully reliable. He lists a lot of places all over Ontario (about 250) which according to him follow the same time zone history as tz zone America/Nipigon I think considerable local research is needed to sort out pre-1970 Ontario time zone history. Shanks describes 116 different rule sets for Ontario. I attach screen shots from klm files on Google Earth showing how the towns listed for Ontario by Shanks are locally distributed. Label (5) represents the Shanks rule set which corresponds to TZ rule set America/Nipigon, while according to TZ the zone America/Nipigon is a tiny area. On 17.01.20 10:01, Paul Eggert wrote:
On 1/13/20 8:15 AM, Jeffery Nichols wrote:
I was also wondering whether anyone on your team has more data about what towns were part of America/Nipigon.
I got that data from my printed copy of the Shanks International Atlas. It contains hundreds of place names in Canada but does not sort them by time zone region. I just now looked in Shanks (4th edition) for place names beginning with "A" with the same time zone history as Nipigon (Shanks calls this Canada's region 35), and found the following:
Alfred Algoma Mills Algonquin Almonte Alvinston Appleton Apsley Arden Armstrong Armstrong Station Aroland Attawapiskat Auburn
The book has latitude and longitude for each of these entries. It would take some time to transcribe all this info (including B-Z) for all of Canada. Also, Shanks partitions the world into many more regions than tzdb does, because he doesn't have the 1970 cutoff that tzdb does. So he partitions Canada into 253 regions and there could be other Shankian regions that share Nipigon's post-1970 time zone history and would thus fall under America/Nipigon in tzdb.
On 1/24/20 5:24 AM, Alois Treindl wrote:
I believe the Shanks data for Ontario are rather messy and not fully reliable.
Thanks for looking into that, and for the images. Maybe Shanks's region 5 (which I labeled "America/Nipigon") was just a catchall for Ontario locations that Shanks had no further data for? At any rate, I agree that the current tzdata is dubious for older Canadian timestamps, since much of this stuff was derived from Shanks.
participants (8)
-
Alois Treindl -
Brian Inglis -
Brian Inglis -
Evan Siroky -
Jeffery Nichols -
Kim Davies -
Martin Burnicki -
Paul Eggert