Re: [Ntp] CLOCK_TAI (was NTPv5: big picture)

[Philip Prindeville <[e-mail]>]

> On Jan 5, 2021, at 8:51 AM, Magnus Danielson <[e-mail]> wrote:
> 
> So I think we need to choose between two different scenarios, either one
> where we allow multiple timescales being sent, without TAI-UTC
> difference at all times, or we go for a unified timescale meeting the
> ideal model properties, and say that we need to be able to resolve the
> issue. There is no real way around that central question.


It might help to stop conflating issues.

Clock synchronization is a simple problem requiring linear, non-leaping time.

Calendaring is an orthogonal issue of breaking groups of seconds into human-friendly units of minutes, hours, days, years, etc.

NTPv4 tried to resolve both issues because it seemed convenient at the time: in retrospect, it’s not clear that it didn’t create more problems.

I say that we drop the unnecessary encumbrances, and shift attention to the necessities that have since emerged that NTPv4 didn’t address (like clarity, good security, better extensibility, etc).

-Philip