[LEAPSECS] Conversational caffeine

Warner Losh imp at bsdimp.com
Fri Jan 28 18:17:44 EST 2011

On 01/28/2011 15:41, Poul-Henning Kamp wrote:

> In message<4E42655D-5E59-4A8F-8051-3138C987C0FA at noao.edu>, Rob Seaman writes:


>> Leap seconds are a means to an end. Start at the end and work backwards.

> And by "start at the end", we of course, using proper system engineering

> practices[1], start by determining if "the end" is a good idea in the first

> place.


> Poul-Henning


> [1] The definitive optimization of any task, is realizing that it doesn't

> need to be carried out.

Unfortunately, "the end" is not well defined or universally agreed up.
If it were, most of the arguments here wouldn't happen, most of the
systems would implement something sane and we'd have little to do with
the waning hours of our Friday Afternoons. Or at least find Something
Else to argue about :)

Sometimes I think we should define a second that's 1e-11 or so shorter
so that the problems of leap seconds disappear into the noise for a
generation or three, but I doubt that would fare any better than any of
the other suggestions made...


More information about the LEAPSECS mailing list