Slashdot Mirror
Leap Second May Be On the Chopping Block (ieee.org)
szotz writes: The days of 61-second minutes may be coming to an end. The World Radiocommunication Conference is meeting for nearly the entire month of November, and one of the hot-button issues is what to do about the leap second. The addition to UTC is supposed to keep atomic time aligned with Earth's rotation, but past leap seconds have caused server crashes, and some are worried that future problems could be even worse. Going into the conference, it doesn't look like there's much of a consensus on what to do. One official is expecting weeks of debate.
So people write crappy code and rely on it without adequate fault tolerance strategies and that's somehow the problem of people who measure time? I think not.
John
I like to hang out at Area 15. Less security and you glimpse an occasional dyslexic alien.
Table-ized A.I.
In my 35 years, I've always seen time as a counted measure of how much time has passed since I started counting. I seem to be forever learning that's my novel idea.
Why should I care where the sun is, where the moon is, where the earth is, with respect to time? If it's winter, I can start work at 9, I can start at 10, I can start an hour after sunrise. I don't need to adjust my clock to start work at the same clock-display every day. I see nothing wrong with a company that has different business-hours by the season.
Similarly, since I'm not in the old west, I don't care if "high noon" is an noon, or 1, or 1 second after noon. I've never determined the time based on the sun.
Last I checked, we have perfectly wonderful time-keeping and gps devices these days. So ocean ships and submarines no longer need a sextant and a chronometer to figure out when and where they are.
So here's my petition. I'd like time to always move forward, and the same rate of 1 second per second. I'd like it to not jump, leap, crawl, rewind, fast forward, restart, end , or eject.
Shit, I just realized that I'm not 35 years old. Or I am 35 years old, but not when measured in seconds. Wait for it...ok, now I'm 35 years old. Phew.
The problem with leap seconds vs. leap days is that leap seconds do not follow a regular pattern. It's fairly simple to write code that follows the leap year rules: February 29 occurs when (year mod 4 == 0 and year mod 100 != 0) or year mod 400 == 0
The timing of leap seconds, on the other hand, are chosen on a case-by-case basis by some standards body (IERS) and announced, usually with only about six month's notice. Thus hardcoded rules for leap seconds are not a good idea, and you have to have some means of distributing leap second announcements to all systems where clock accuracy is critical. Either that or you accept a one-second margin of error, don't count the leap second and correct the clock later.
Leap-seconds were properly handled in computer software before most of today's software engineers and programmers were born.
Back in 1969, I started working on a software system that already handled leap-seconds quite smoothly. At that time, keeping UTC aligned with the rotation of the earth involved introducing fractional seconds and also having UTC seconds NOT the same duration as atomic seconds (TAI). In 1972, this was simplified by having UTC seconds exactly the same duration as TAI seconds and (after an initial fractional leap) introducing only leaps that were full seconds. The software in the system on which I was working DID NOT HAVE TO CHANGE!!.
Internally, the system on which I was working -- which evolved and continued in use to operate military space satellites for over 20 years -- kept all time in TAI, which never has leap-seconds. A relatively small routine converted in either direction between UTC for displays and TAI for internal time. Another small routine converted from UTC to UT! to sidereal time, the latter more closely reflecting the rotation of the earth, which is gradually slowing and also has predictable periodic fluctuations. The purpose of all this was that we needed to know very accurately the spot on the rotating earth directly under the orbiting space satellite. The position of the satellite was known in TAI while the surface of the earth was rotating very closely to sidereal time.
Also note that the network time protocol (NTP) also accounts for leap-seconds and has done so for decades.
I can only conclude that the current attempt to do away with leap-seconds is a result of lazy software "professionals" trying to shift blame for their ignorance about leap-seconds.
We let "one year" drift enough that it only needs to be corrected once it's off by a whole day. Why not let [86,400] seconds accumulate and have a second leap day every hundred thousand years or so?
Because sunlight levels over a day control more recurring human biological processes, such as alertness, than any natural phenomenon with a year's period.