[LEAPSECS] Bloomberg announced its smear

[Peter Vince]

On 27 September 2016 at 14:40, Warner Losh <imp at bsdimp.com> wrote:
>
>...
> Eliminating leap seconds would be a great way to unify all these
approaches :)
> And it too is compatible with JTS.


Smearing seems like a clever way of papering over the problem of the
leap-second, but as others have said, surely a standard needs to be
defined, and that way everyone can compare and match their time - surely
the whole point?

I am interested that Google have chosen to linearly smear over 20 hours,
thus increasing each second in that period by 13.8888... microseconds -
surely such an irrational step is difficult to achieve?  Could it be that
they actually step 125 microseconds every 9 seconds, or maybe 25
microseconds every 1.8 seconds?  I wonder if Christopher has any inside
knowledge on the details?

I'm also curious about the change from cosine to linear smear: the linear
smear results in a very sharp step change in frequency, exactly what we
DON'T do, whereas the cosine smear has a very smooth and gentle change,
surely more easily followed.  Could it be that this would need some very
small (way sub 100ns) steps at the start and end which would be almost
impossible to achieve accurately?  I guess that if we know the time is
changing at a fixed rate, we can easily allow for that when trying to hold
a frequency?  Again, does Christopher have any inside info on the thought
processes?

However, these new problems, and the Azure systems disagreeing between
countries, all comes back to what Warner said - scrapping leap seconds
solves all these problems, at the expense of - with due respect to Rob,
Steve. etc. - the astronomers having to increase the range of DUT1 on their
software.

Peter
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://pairlist6.pair.net/pipermail/leapsecs/attachments/20160927/995d0c7d/attachment-0001.html>