[LEAPSECS] Schedule for success

[Poul-Henning Kamp]

In message <774B1BAC-3AA2-454E-A141-7FA77D31488B at pipe.nl>, Nero Imhard writes:


>You don't need such a reason. It's up to the people (ITU?) who want to  

>change an existing definition to make their case. "Although we hear  

>protests, we don't see any objection we deem valid" doesn't quite cut  

>it, does it?


That's how the ITU works.

Look at how the V.90 was ratified for instance.


>I still don't get why you are insisting that UTC could be changed.


It was changed in 1958 and 1972, I see nothing preventing it from
being changed in 2008, -9, -10 or any other year.


>switching to something like TI [...]


We have millions of documents which mandate, directly or indirectly,
the usage of UTC.  Starting with legislation about local timezones
over POSIX to international treaties about transport, communication
and power-generation.

Taking leap seconds out of UTC has a clear implementation plan, it
is one document that needs changed and ratified in a UN forum and
we are done with the problem.

Changing all the other documents and teaching people about proper
choice of timescales is a proven path to failure.


>What keeps bothering me is that the prospect of *changing a  

>definition* (of UTC) doesn't seem to make you (phk) blink at all.


Not only does it not make me blink, I shake my head in wonder at
people like you, who seem to think that UTC is sacred in some
sense.

Did you also complain when the length of the meter was redefined ?
wasnt that equally an insult to the users of the meter, who
relied on it being a metal bar i Paris ?

UTC is a definition for the advancement of international communications
not one of the 10 commandments.

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.