That's why my time system https://kybernetikos.github.io/UIT/ rotates the clock depending on your GPS location. The idea is that the numeric time is the same anywhere in the world but your clock is displayed with solar midday directly up and solar midnight directly down.

It explicitly intends that each solar day is split into 10 which is then subdivided into 100 and then again. This makes an ergonomic system for human usage although the fact that its time periods vary according to the day mean its not appropriate for science and engineering where consistency and accuracy are more important.

Doing away with time zones means arranging international meetings and travel is easier. Putting a full day on one click face lets you draw on the sun rise and set times and also lets you draw a days worth of appointments on the face. Since it's base ten addition and subtraction are trivial.

This seems to ignore latitude shifts through the year. Why should solar days affect your day? I'm quite OK to wake up at 8am in the dark, and not so much at 4.30am in light. We were not all born on the equator.

If you set your alarm for 2.80 you'll wake at the same time whether it's dark or light. Or perhaps I've misunderstood your objection?

Anyway the numeric value is the same everywhere so it doesn't affect your time, it just affects the way your clock face is drawn.

365.25 is a ratio- the number of rotations of the earth per revolution. In a real sense, it comes from our solar system.

If you shy away from it, any given point in your year's calendar will drift in terms of seasons. You see this in the around 354 day Hijri calendar, for instance.

(Edited: removed claim 365.25 is unitless. The unit is rotations/revolution)

Well, it's hard to change the number of solar days it takes to go around the sun one full revolution. Anything different and we'd see the seasons shift.

As for number of hours in a day. 24 is very divisible which is better than say 10. You can divide it into 1, 2, 3, 4, 6, 8, 12, 24 without fractions.

As for 60 minutes and 60 seconds, these again are very divisible: 1, 2, 3, 4, 5, 6, 10, 12, 15, 20, 30, and 60.

The more interesting question, should we have chosen base 12 instead of base 10. It would of course be harder to count using your fingers.

> It would of course be harder to count using your fingers.

No necessarily. 4 fingers * 3 knuckles per finger make 12, so you can count in base-12 using only 1 hand. I think the Sumerians (?) used that * 5 fingers on the other hand to count in base-60, resulting in our base-60 time system.

I read a good argument in favour of senary on the basis of divisibility. It has nicer properties than base 12 for that. Personally for time I don't find the convenience of its divisibility to remotely outweigh the inconvenience for addition and subtraction of using a base different to the way we count.

Whatever base we use for counting, and I can imagine arguments for senary decimal duodecimal binary or trinary, we should use the same base for counting time.