On Sat, 22 May 2021 at 03:09, Philip Paeps via tz <tz@iana.org> wrote:
I think the first part of this change (merging timezones with no changes
post-1970) makes a lot of sense.  I'm less convinced about the second
part of the change (merging timezones into Etc).  That feels like
indirection for the sake of indirection.

I'm inclined to agree here.  In particular, the latter group (things like America/La_Paz -> Etc/GMT+4) seems to encourage things or behaviors which historically cause confusion, especially for novices:
  1. using fixed-offset zones for civil time with the blind expectation that it won't need to change
  2. the "backwards" offset signs from POSIX
  3. the outdated GMT-vs-UT(C) solecism
Arguably, helping to prevent (1) is the entire point of this project.  The others, while less of a concern, have been continually de-emphasized enough over the years here that I'm not sure that giving them increased prominence is a good thing.

That said, I do broadly support the consolidation of purely geographical zones like Europe/Monaco -> Europe/Paris (retaining historical work in backzone).  I think, in most cases, a stricter 1970 principle can be reasonably explained to passers-by.  Especially since we deal with cities/communities and not borders, contiguity need not even be a requirement — though if any of the resulting pairings are particularly far-flung (I haven't checked thoroughly yet), a small amount of duplication may be advisable and otherwise unavoidable.

--
Tim Parenti