BSD (I think...), Linux, OSX, and Windows, when fully updated, already have up to date automatic compensation for DST. Updating a profile should NOT be required...
phpbb didn't read DST, but instead had the user configure an absolute offset from UTC in their profiles. SMF is probably better, I just need to check.
What sucks are all the stupid things in my house, that were made before congress thought it would be Quite a lovely idea to screw with DST's starting and stopping time, instead of just removing it and having us all use a single time all year in the first place.
The NIST WWV broadcast sends the DST schedule. Compliant devices should read it, rather than using anything hardcoded. Indeed, one of my radio set clocks actually correctly updated at the new date, even though it was made before the DST schedule change was contemplated.
However, agreed on abolishing DST. Here's the funny thing. Our country was obsessed with the disasters that might occur on Y2K, not realizing that the maintainers of embedded and other critical systems are always
on alert for a multitude
of possible software and design bugs that could otherwise, if not addressed, cause sudden failures, which the public remains unaware of. Yet they will happily, on a whim, support a law to change the schedule a couple years out which causes the need to redesign and support changes in an enormous array of embedded software.
So they make these estimates on how much energy will be saved by extending DST a few measly weeks. Did the resources wasted by reprogramming computer systems, and supporting problems caused by systems which were not fixed in time (or neglected) even factor in? Does anyone think we came out ahead?
Aren't the problems caused by having daylight time changes twice every single year
worse than the one-time event of a few Y2K bugs?
Software developers and especially those who deal with databases and transactional or distributed systems (those resembling state machines, etc.) will know what I'm talking about here: The fact that government authorities can change a scientific scale
on a whim and for the future makes human notations of time, a supposedly scientific measurement, non-deterministic
. I thought the whole direction of science was finding eternal, constant definitions for bases of measure, such as the one remaining one, mass (the kilogram). One could argue that the need for leap seconds causes that anyway, but there's no reason to make things more complicated than need be.