On 2026-01-16 17:45, Benjamin via subsurface wrote:
Dirk, you forgot the other corner case

Argh :)  You're killing me here.  I *really* want to let this part of
this discussion die for now because I'm ok with Dirk's call that it's
not the most pressing thing we need to address right now, and it
clearly ignites passions at least as much as the political arguments
over whether there should even *be* daylight saving does.

What I need to know to deal with the code I'm currently working on
for now *has been answered*.

And I can agree to disagree about how serious a problem it might be,
and how many people it might affect, and whether or not we care about
those people's problem with this.

But it really doesn't help the goal of trying to help everybody
properly *understand* the problem, which is Important even if we
still totally disagree on the Best Solution. If the basic *facts*
of the problem are misunderstood or misrepresented by some of the
participants then now we have *two problems*.

So:

some countries don't have fixed DST transitions.

This is perfectly true.  I'd even go so far as to say *most*
countries, and occasionally these political decisions are made
with very little forward notice.

But this:

then you get the lovely corner case of needing to keep these
historical changes, so that you can see when exactly

Is absolutely not.  *We* do not need to deal with that at all.
That's what this is for: https://en.wikipedia.org/wiki/Tz_database

And *we* do not need to do anything special to use it, the underlying
system functions that *we already use* take care of that for us.

The only problems *we* have, if misunderstanding isn't part of the
brew, is "Do we care?" and if we do, how do we best handle the
already existing records with no timezone data.  Every other
problem related to this was long ago solved and we don't need
to do anything special at all to use those solutions.

And I think we have a pretty firm answer to "Do we care" for now.
QED :)
_______________________________________________
subsurface mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to