[tz] [PROPOSED] Merge timezones that are alike since 1970

Stephen Colebourne scolebourne at joda.org
Fri May 28 20:44:14 UTC 2021


This change would make sense if no pre-1970 data was available for any
zone (including LMT). But instead what users get is the data of
somewhere else.

For example, Norway's and Sweden's time zone history is being wiped
out in favour of that of Germany. Can no-one here see the political
sensitivity in that?

This has a very serious impact on Joda-Time because it normalizes
time-zone IDs. (It treats a Link as the key to the normalization, so
anything at the weak end of a Link is replaced by the ID at the strong
end. You might complain that it shouldn't do that, but it has operated
that way for 20 years...)

This code:
  DateTimeZone zone = DateTimeZone.forID("Europe/Stockholm");
  System.out.println(zone);
will print "Europe/Berlin" if this change is not reverted. I consider
this to be catastrophic.

I maintain, as I always have, that the bare minimum tzdb should
provide is a full time-zone with history (not a Link) for each region
identified by an ISO-3166-1 code.

Stephen


On Fri, 28 May 2021 at 20:52, Michael H Deckers
<michael.h.deckers at googlemail.com> wrote:
>
>
>     On 2021-05-28 17:10, Magnus Fromreide via tz wrote:
>
> > The remaining superzones should also be stripped of any historical data
> > since it would be incorrect for a whole lot more places in the zone than
> > it is now.
>
> *
>     "Stripping" historical data is not possible in the tzif format.
>
>     A timezone with name N will give, before 1970, the correct values
>     for some location, but not necessarily for the location named N
>     (if the location named N does not appear in the file backzone
>     then they are guaranteed to be for the location N, but this cannot
>     be detected from the tzif files).
>
>     In my opinion, this is better than the previous choice (where the
>     data before 1970 could be incorrect for every location agreeing
>     with N after 1970).
> *
>
>
> > What will happen with all that data? It is quite interesting to read the
> > tzdb sources as of now because of all the extra history.
>
>
>     Yes, it is! The historical date are not (all) lost; most are kept in
>     the file backzone and can be made to reappear with appropriate zic
>     option (PACKRAT).
>
>     Michael Deckers.
>


More information about the tz mailing list