6 ms·
The problem is mostly (or so I’ve heard) that the drift is relevant for astronomical applications, and they rely on time dissemination which is done in UTC. If
by layer8 4y ago
The problem is mostly (or so I’ve heard) that the drift is relevant for astronomical applications, and they rely on time dissemination which is done in UTC. If UTC decides to start deviating from TAI by dropping leap seconds, those applications will be in trouble. I’m sure that the problem is overblown, but this is the reasoning that was put forward in the past.
- ronsor 4y agoPeople who are using time for astronomical applications will simply have to track the offset if they need to. I'm not sure how that's much more problematic than the current situation.
- agrajag 4y agoAgreed. _Someone_ has to deal with the drift between astronomical time and earth time, better the class of software that deals with space than every time critical system in the world.
- myself248 4y agoBut leap seconds hose up astronomical applications too. They all start with TAI and then apply corrections as needed. There are lots more corrections than leap seconds, but they're applied proportionally, not as leaps. Leap seconds are a solution in search of a problem.