What's your take on this "leap second" business? Should we keep it or let it expire? http://www.bbc.co.uk/news/science-environment-16597191
In the short haul it's a pain in the nice person, since it potentially jiggers the clocks every year. On the other hand, it gets to be a problem if left uncorrected. We need a correction system, but perhaps over a longer period. There isn't an easy answer. Tom
I haven't been following this issue closely in recent years, so am not familiar with all the issues. But this seems more a case of botched system planning by some designers than a real fundamental problem. So now some folks are trying to change time definitions so they don't have to properly fix their systems. And they won't be around to worry about it when their proposed redefinition inevitably causes other problems. Because of the well known feature of leap seconds, anyone in the past four decades who was designing a system needing a continuous time scale, simply f'ed up if they selected UTC. They should have selected a continuous (no leap second) time scale from which UTC is defined -- e.g. TAI in the early years, TT more recently. Buck up. The tool you need existed before your systems were designed, so fix your systems. Bad planning on your part does not constitute an emergency on my part.
Change to leap-hour. By the time adjustment is needed, this whole discussion is probably obsolete. Whether due to using another timescale, no more humans on the planet or other reason, I do not know, but chances are quite big that I am correct in this assumption.
One of the driving factors is a great many computer and automated systems do not have provisions for making time changes while in operation. What surprises me, is that if they don't address these provisions now, then the issue could be quite a bit nastier when the errors get really big.
The leap second keeps things in sync. But it requires too much maintenance for too many systems. Doing away with it could save a lot of money, at least for a while (a few decades, perhaps). A leap hour was suggested years ago but was not accepted. There appears to be no painless solution. Either way we go, a price has to be paid. Question is: who will pay, when, and how much?
We're looking at this problem entirely the wrong way. Just put together a team with Bruce Willis & friends to use some of the world's unused atomic arsenal to redistribute regions of crustal mass to set the earth's spin where it belongs. I gotta say, our inability to see the obvious causes a lot of unnecessary travail.
Looks like they decided to kick the can down the road and reconvene in three years to... discuss it some more. Then wash, rinse and repeat.
I'm still trying to get us to abandon daylight savings time. Also, we need to do something about the way the East always gets the new day ahead of us. How are we ever going to compete if we're always behind?
From my perspective, a delay is a step in the right direction: don't change UTC, and make system designers who need a continuous time scale move to an existing continuous time scale, which they should have used in the first place. Loran Time is probably dead, but [ame="http://en.wikipedia.org/wiki/International_Atomic_Time"]TAI[/ame], [ame="http://en.wikipedia.org/wiki/Terrestrial_Time"]TT[/ame], or [ame="http://en.wikipedia.org/wiki/GPS_time"]GPS time[/ame] should fit the bill for most uses. And there are more for very serious applications. Changing UTC to fix some system problems is like trying to solve the Y2K problem by changing the calendar to fit the faulty legacy software.
Is the discussion really about leap seconds or about the "purpose" of time. If your machine needs some sort of uninterrupted time, then give it one and the machine problem is solved. If humans want the clock to match what the sun and stars are doing, then what we have now is the best answer. So making machines happy or humans happy seems to be the real issue.
