I think I've changed my mind; UTC leap seconds should *not* be abolished.


There's *a lot* to unpack and explain, but basically,

This thread on leap minutes and hours got me thinking

And this post made me think and disagree

Resolving ambiguous or nonexistent UTC timestamps is a much easier and encapsulable problem when the shifts and deltas are within 1 second. The domain of applications which handle UTC leapsecs are niche (scientific, military, fintech), and when they do, it's fairly simple, or can be with good design.

The @hare standard library separates the concerns of 'civil' timeshifts (timezones, DST) and timescale timeshifts (leap seconds), going as far as to include a 'timescale' type.


The proposed alternative leap-minutes or leap-hours make a lot civil time software needlessly complicated, blurring 'civil' and 'timescale'. The same clusterduck of problems with timezone shifts and DST would now be part of UTC itself.

Then there's also the Earth's phase and velocity of rotation, drifting noons and midnights, politicians and dictators chiming in, forward compatibility, etc. It's a charged topic for sure.

@torresjrjr @hare Amazing, I never thought about this.

Thank you very much for sharing it :ablobcatwave:
Sign in to participate in the conversation
Qoto Mastodon

QOTO: Question Others to Teach Ourselves
An inclusive, Academic Freedom, instance
All cultures welcome.
Hate speech and harassment strictly forbidden.