1. 23
  1. 1

    My favorite is UTC with numeric timezone offset. I think doing that would buff the edges off the edge cases, giving us an unambiguous way to understand when an event will begin. Either that or just stay in local time because the UTC time is never applicable.

    That said, this is a fascinating theoretical case study in why just storing local future events in UTC Z might be very incorrect.

    1. 1

      As someone who supports scheduling software that’s used globally, timezone/DST issues are a big problem sometimes. The main problem is that many governments don’t give a flying fsck about letting the TZ database maintainers know in good time when they decide to change the DST changeover date, or if to implement DST at all.

      Russia for example decided to stay on summer time with 3 weeks notice a few years back, essentially creating around 10 entirely new timezone definitions.

      I have high hopes that the planned EU DST change will be announced in good time, but who knows what can happen in our current sublunar world.