FW: Who ist the maintainer of tzdata?
Since sending this message Alois Treindl has joined the time zone mailing list. --ado -----Original Message----- From: Alois Treindl [mailto:alois@astro.ch] Sent: Thursday, March 10, 2005 5:24 AM To: tz@lecserver.nci.nih.gov Subject: Who ist the maintainer of tzdata? I have a project to expand tzdata. Summarizing shortly, the rough idea is to - add boundary descriptions to each area with a different timezone history so that with given coordinates it can automatically be determined to which timezone history 'table' a given location belongs. I am aware that many non-trivial issues of geo-information processing arise with such a goal. - add more details for the complicated timezone history of some countries, e.g. Canada, where other sources have identified more than 250 regions with different timezone history, whereas current tzdata has only 15 regions. I am aware that this is a large project which will require resources measured in man-years. I may be in a position to contribute/sponsor such resources. I would like to communicate with the current maintainers of tzdata about these ideas, and about the best approach for such an expansion. My motivation background is the field of serious astrology, where reliable timezone history information is an essential requirement. tzdata is doing an excellent job to provide reliable information for current timzones, but it lacks sadly (and understandably, given the complexity of the problem) in historical detail. Alois Treindl
These days I do most of the work in coordinating the maintenance of the tz data. Alois Treindl <alois@astro.ch> writes:
- add boundary descriptions to each area with a different timezone history so that with given coordinates it can automatically be determined to which timezone history 'table' a given location belongs.
- add more details for the complicated timezone history of some countries, e.g. Canada, where other sources have identified more than 250 regions with different timezone history, whereas current tzdata has only 15 regions.
These are both quite worthy projects. Yay! To my mind, the second is easier. The main thing is to identify each region, give it a name suitable for the tz scheme (given the number of regions, I'd suggest 3-level names like "America/Ontario/Nipigon", much as we already do for Indiana), and to record the data in tz format (i.e., format suitable for the zic program <http://www.linuxinfor.com/english/man8/zic.html>). For the former, we'd need to settle on a data format. I'm by no means an expert in the GIS area, but if you look at the "Time zone boundaries" section of <http://www.twinsun.com/tz/tz-link.htm> you can find previous work in this area (none of it sufficient for tz's needs, alas, but it's better to start from something than from nothing). I'd prefer a format that is freely available, and that can be processed by free GIS tools like GRASS <http://grass.itc.it/>. We also would need to address the copyright issues. Currently the tz database is public domain. If this might be a problem for you please let us know.
On Thu, Mar 10, 2005 at 02:25:04PM -0800, Paul Eggert wrote:
These days I do most of the work in coordinating the maintenance of the tz data.
Alois Treindl <alois@astro.ch> writes:
- add more details for the complicated timezone history of some countries, e.g. Canada, where other sources have identified more than 250 regions with different timezone history, whereas current tzdata has only 15 regions.
These are both quite worthy projects. Yay!
To my mind, the second is easier. The main thing is to identify each region, give it a name suitable for the tz scheme (given the number of regions, I'd suggest 3-level names like "America/Ontario/Nipigon", much as we already do for Indiana), and to record the data in tz format (i.e., format suitable for the zic program <http://www.linuxinfor.com/english/man8/zic.html>).
Paul (and interested others), Assuming you had these timezone histories for places which differed pre-1970 but have not differed port-1970, and assuming they were in tz format and in the public domain, what would you do with them? Is the idea is to integrate them into the tz database? If so, would they be put in existing source files like northamerica, europe, australasia as new zones? Or would they be zones in a separate source file? (Maybe called something like "historical"?) Or just have the information as comments? I think there is something to be said for a separate source file, which need only be zic'ed by users who really want those zones. I can imagine lots of Canadians being confused by a choice of hundreds of different time zones, for example. ___________________________________________________________________________ David Keegel <djk@cybersource.com.au> http://www.cyber.com.au/users/djk/ Cybersource P/L: Linux/Unix Systems Administration Consulting/Contracting
participants (3)
-
David Keegel -
Olson, Arthur David (NIH/NCI) -
Paul Eggert