Re: [LEAPSECS] Precise time over time

From: Markus Kuhn <>
Date: Thu, 11 Aug 2005 13:29:47 +0100

Poul-Henning Kamp wrote on 2005-08-10 18:26 UTC:
> If you want a really disturbing experience, visit a modern robotical
> slaughterhouse, and while you are there, observe and think about
> what a one second difference could do the the tightly coordinated
> choreography of the robots.
> The problem with leap seconds is when the systems are in touch with
> each other, have synchronized clocks and know it, and then suddenly
> some of the systems insert an extra second, and some of them don't...

You seem to imply that all aspects of the timing and synchronization of
these systems are best synchronized to one single global coordinate
system. Why would you want to do that?

When I design and build a house, I would be a fool if I tried to specify
and determine the coordinates of every single brick in the International
Terrestrial Reference System or one of its friends. There is no
advantage of doing that, therefore nobody does it. Instead, I set up my
own local coordinate system and synchronize and align everything nicely
within that. At the start or end of the project, I may bother finding
out how my local coordinate system relates to one or more of the global
coordinate systems. But if WGS84(GPS) has a brief 15 cm hickup this
afternoon, or if South Cambridgeshire sinks next week another 42 mm in
the ITRS because of the low locale water table, there is no need why
this should cause the robots that build my house to collide or place
their bricks wrongly on top of each other.

Using global coordinate systems, by their very nature, is fraught with
far more risks than using local coordinate systems. That is valid in
both space and time. And that is why we use lots of different coordinate
systems in real life and engineering applications.

My attack may have seen ad hominem, simply because the way you argue,
you really come over to me as an obsessive enthusiast for the use of
just one single reference frame everywhere. Sorry if I lack the English
language skills to say this in a more polite way.

May be, it is because I'm not a big beefeater, but TAI synchronized
slaughter robots sound to me as detached from reality as slaughter
robots that know their orientation to within a few ┬Árad relative to
hundreds of quasars. All they really need is to run of the same master
clock as their neighbors. That's what cuts the meat. There is no need to
attempt to continuously resynchronize them, or the rest of the factory,
with UTC at all.


Markus Kuhn, Computer Laboratory, University of Cambridge || CB3 0FD, Great Britain
Received on Thu Aug 11 2005 - 05:30:01 PDT

This archive was generated by hypermail 2.3.0 : Sat Sep 04 2010 - 09:44:54 PDT