[tz] What data should TZDB offer?

David Patte dpatte at relativedata.com
Sun Jun 6 19:14:26 UTC 2021

I agree that all historical data is murky, and gets murkier (?} the 
further you go back. This is to be expected, and users should be able to 
use the data as a 'best guess' until it is improved.

Until we all go back to local solar time, there will always be some 
inaccuracies, but tz should be the place to capture the current 'best 
guesses', and be the destination for capturing improvements.

On 2021-06-06 13:56, John Hawkinson via tz wrote:
> Tom Lane via tz <tz at iana.org> wrote on Sun,  6 Jun 2021
> at 13:31:06 EDT in <655997.1623000666 at sss.pgh.pa.us>:
>> I had a further thought about this: if we want to have both of these
>> principles (zone-per-country and stability of old data), then it would
>> make sense to insist that we don't create new per-country zones until
>> someone has done the research to fill in plausible old data back to
>> the LMT era for the proposed zone name.
> I don't think this is correct or fair.
> It's not correct because if there is a newly established zone (per-country or otherwise), early adopters of that zone can tolerate some flux in the data. It's not as if we're coming into an established zone with a lot of dependancies and expectations of stability. If the new zone goes in and today and the historical data comes a year later, that's probably OK. Not ideal, but OK. For sure we'd want to minimize adding it in fits and starts.
> I don't think it's fair because the tz database needs to turn on a dime to reflect political changes that happen on rapid timescales. If a zone into two countries (whether by one of two previously-aligned countries changing their time zone rules, or by other political or even military mechanism) overnight, we need to push a new release out ASAP, we can't say, "oh, sorry, you can't have working time on your computers, we have to research the history, just suffer along for 6 months."
> --
> jhawk at alum.mit.edu
> John Hawkinson

More information about the tz mailing list