On Fri, 28 Oct 2022 at 11:08, gera via tz <
tz@iana.org> wrote:
Thing is, there are certain regions which will get a change on its DST as soon as October 30th, 2022.
This is the case for Chihuahua (America/Chihuahua) which seems wont change its time this sunday, as it'll be aligned to Mexico/General, again, this same October 30th, 2022, according to this same law.
That is good to know. So far, our understanding had been that ALL regions would fall back on 30 October (or 6 November, for those that follow US rules) as originally scheduled, and therefore that we would have until 2 April 2023 before the change affected any timestamps. But if any regions are indeed changing zones by NOT falling back this weekend, that is a different story and we should work to get that into a release as soon as possible.
Do you have a clear reference for which regions will be doing this? Perhaps a comparison of the regions listed in the recently approved law and the previous law it replaces?
This sets what I think are really hard implications for most of the services running on this timezone.
If the said law is published between now and next Saturday, I guess it'll be havoc among users on this timezone.
I still don't think there's an easy fix from tz to this, but I think it's important for somebody to notice this.
Yes, and it underscores the importance of allowing sufficient time to allow changes to be clearly communicated, encoded, tested, published, and propagated to end-users. This issue is well familiar to frequent readers of this list, but remains somewhat less familiar to the governments of the world.
Even if we're able to get a release out in the next ~40 hours before this happens, by this point, it's unlikely to get to most end-users in time.