[tz] [PROPOSED] Avoid backward links in zone.tab

Michael H Deckers michael.h.deckers at googlemail.com
Tue May 11 12:32:58 UTC 2021


   On 2021-05-10 18:04, Paul Eggert wrote:

> Perhaps someone could add systematically-formatted comments to 
> 'backzone' giving geographical coordinates for names like Asia/Hanoi 
> that are not already mentioned in 'zone1970.tab'.



      Yes, I understand.
      Just to make sure: my point is not a question only for backzone
      data. What happens with the proposed change is that

         any location name of a timezone in the mainline tzdb data can
         no longer be (easily) connected with geographical coordinates as
         soon as the location name (is a link and) also occurs in
         the backzone file.

      Rather, the corresponding geographic coordinates in zone.tab
      are no longer connected with the timezone name, only with
      an ISO country code. The same holds for several proposed lines
      in zone1070.tab such as
            AQ,KW,SA,YE  +2438+04643   Etc/GMT-3   Arabia, Syowa
      What are these geographical coordinates supposed to indicate?

      Of course, there is no error (all the advertised functions of tzdb
      will continue to work, except perhaps choosing Syowa with tzselect)
      -- it is only my failure to understand the conceptual schema
      of the tzdb data. One of the successful design choices of the tzdb
      database was to identify timezones by locations (rather than
      by areas or countries), so that location names and location
      coordinates are equivalent means of identifying timezones
      of locations. Is this still true?

      Michael Deckers.




More information about the tz mailing list