[ntp:hackers] ITU and Leap second elimination

juergen perlinger juergen.perlinger at t-online.de
Mon Sep 30 18:14:15 UTC 2013


On 09/30/2013 09:00 AM, Poul-Henning Kamp wrote:
> In message <5248D087.5010804 at flatsurface.com>, Mike S writes:
>
>> Implying it is not monotonic is simply admitting that one does not 
>> understand the proper representation of UTC.
> In a mathematical sense your are right, and the correct thing to
> curse UTC for is that it has a variant radix.
>
> Very few people know what that means, and the effect of the variant
> radix on all the systems which do not or can not implement it, is
> a discontinuity at the moment the radix varies.
>
> There fore, calling it non-monotonic is not entirely wrong:  That is
> the practical effect people experience.
>
First, I'm not a mathematician, and I'm no native English speaker,
either. I know the term 'radix' only from a different context, though I
clearly understand the meaning here. And a bit of googling around
helped, too ;-)

I have to agree that using the term 'non-monotonic' is probably not the
best way to describe the effect of leap seconds. Would 'piecewise
defined in terms of TAI' a better expression to describe the nature of
UTC? That's how I would translate the 'variant radix'. Or would 'leap
second era' a better way to term the segments? I don't know. Variant
Radix is spot on, though...

UTC is monotonic per se, of course. I guess Mike knew what I meant and
just wanted to show superiority.

After a while, every discussion ends in a cloud of semantic nit-picking.
It's important to agree on the meaning of words, but it can be overdone.



More information about the hackers mailing list