[tz] Proposal to use Asia/Tel_Aviv for Israel - Jerusalem is not internationally recognized as part of Israel

random832 at fastmail.us random832 at fastmail.us
Tue May 7 15:47:32 UTC 2013


On Sat, May 4, 2013, at 13:37, Jonathan Leffler wrote:
> The cost isn't the cost of generating the patch file.  It's all the other
> things that have to change that make it expensive.  Most of those changes
> are not visible to the tz database; they are costs to the consumers of
> the
> tx database.  The database changes when necessary, not to be cosmetically
> more attractive.

At the risk of annoying people by continuing this discussion, I'm
confused what the supposed cost of this change actually is. Both names
already exist in the database, which makes it even less of a cost than
renaming a timezone (and leaving an alias behind) usually has. The
original change made in 1996 seems best described as "to be cosmetically
more attractive" anyway, and doesn't seem to have caused any [technical]
problems.

Unlike Mr. Conradi, I'm willing to assume the original change was based
on a simple mistake rather than some malicious plot, but I don't see why
there's such a big objection to changing it back.

High-minded arguments like the one made in
http://mm.icann.org/pipermail/tz/1998-August/010224.html ignore the fact
that A) the database explicitly identifies what political boundary each
timezone falls within in zone.tab and B) even if not, the policy
implicitly depends on this by requiring that at least one timezone exist
per region (and per difference since 1970 within a region), despite (for
example) Europe/Oslo, Europe/Copenhagen, and Europe/Stockholm not having
actually differed since 1970 - all having no DST between 1970 and 1980,
and on EU after 1980. Actually, it's unclear to me why those shouldn't
be aliases when (for example) Europe/Busingen is one. But I digress.

"This naming regime survived the demise of the Soviet Union without
having to rename `Europe/Moscow' or `Asia/Tashkent'; it survived the
recent revolution in the Congo without having to worry about its
country-code change " - Yes, but zone.tab did have to change. Not in
1991, since it has only existed since 1996, but it did in fact change
for the Zaire/Congo change.

The database itself wouldn't even have to change at all. There's no
problem having a zone.tab line that names an alias - Europe/Busingen is
one such example.

The only "cost" left is the perceived cost of being seen to give into
political pressure. But that ignores the fact that the database is
taking a political position _now_, by mentioning a disputed territory in
zone.tab. And the fiction that the city names used as timezone
identifiers aren't seen by humans is just that, a fiction.


More information about the tz mailing list