[LEAPSECS] The leap second, deep space and how we keep time -Brooks
alexcurrant911 at yahoo.com
Wed Jan 28 10:10:09 EST 2015
My company got burned by mis-programming for Y2K. It's hard to test things that happen every few years.
It might be a good debating point to say that in one specific incident the future ITU decision would not help, but it's a diversion. I have trouble believing people who keep misleading and outdated information on their web pages, so maybe somehow an ITU decision to abolish leap seconds would stop the one-second jumps on New Year's eves. I admit I don't know anything about how the jumps got mis-programmed. Sometimes there is a step where humans type.
From: Poul-Henning Kamp <phk at phk.freebsd.dk>
To: Leap Second Discussion List <leapsecs at leapsecond.com>; Martin Burnicki <martin.burnicki at meinberg.de>
Sent: Wednesday, January 28, 2015 4:45 AM
Subject: Re: [LEAPSECS] The leap second, deep space and how we keep time -Brooks
Ideally, all software should be flawless and bugfree.
This lofty goal would come at a cost, I don't think I need to convince
you that it would be a very high cost, for civilization as such.
We can argue if it would be worth the investment.
I personally think it could be, but on the balance of probabilities
it probably will not be.
We do not need to argue if it is going to happen, because it is not,
it would require draconian restrictions on free speech ("You cannot
share a program with anybody else until it has been reviewed,
formally tested etc. etc. etc.")
So taking it is a given that pedestrians will always be allowed to
program, no matter what we do, the question really is: Do we try
to make it easier for them to not make mistakes ?
Abolising leap-seconds would make programming easier for all the
pedestrians, at the cost of minor extra effort for the time-nuts and
If leap-seconds cannnot be abolished, announcing them 10 years ahead
of time would still make it a lot easier for the pedestrians to not
mess up, at the cost of extra effort for operating system programmers.
Retaining leap-seconds in the hope, or even demanding, that the
pedestrians upgrade their programming skills is not credible however.
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.
LEAPSECS mailing list
LEAPSECS at leapsecond.com
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the LEAPSECS