Someone I know linked me to this mailing list archive thread:
https://mm.icann.org/pipermail/tz/2021-September/thread.html#30503
Seems like they're fucking around with timezone data, but he says it's mostly for muh diversity reasons, but I'm too stupid to follow this thread.
It looks like a sane person is trying to fork the modified time DB away into a separate project and leaving the original alone but I'm having a hard time following this entire thread. Anyone know of what's going on?
Data has never been more bulky and cumbersome. We may have beefier hardware but that's just allowed everything to bloat up. Scale it out by the number of idiots and how many devices each of them has that connect to a thousand servers a day and suddenly efficiency becomes an issue again. Doubly so now that everything is headed back to mainframe style computing with "the cloud".
That and for those who program everything is just one big optimization puzzle and those who are really good at it tend to be obsessives to begin with, so it's just a natural extension of what they already were.
I'll be the first to agree with you on that. But this is an old library that already has the data structures to handle this sort of complex dataset. To my knowledge no one is complaining that the time zone structure is more complex than it needs to be, because the domain itself is complex. And they haven't removed any of the complexity, they've just shuffled it around to some other part of the library.
Yes. The trick is to know where to fight your battles. But still the obsessives also tend to value correctness. If you're such a time zone nerd that you're maintaining an open source time zone library in your spare time, I would imagine you care about the details. Otherwise you wouldn't be doing it at all.