Ask Slashdot: How Many Time Standards Are There? 214
Posted
by
timothy
from the didn't-suffer-slept-right-through-it dept.
from the didn't-suffer-slept-right-through-it dept.
jjoelc writes "Being one of those 'suffering' through the time change last night, the optimist in me reminded me that it could be much worse. That's when I started wondering how many different time/date standards there really are. Wikipedia is a good starting point, but is sorely lacking in the various formats used by e.g. Unix, Windows, TRS-80, etc. And that is without even getting into the various calendars that have been in and out of use throughout the ages. So how about it? How many different time/date 'standards' can we come up with? I'm betting there are more than a few horror stories of having to translate between them..."
Time Standards vs. Time Formats, and Y10K problem (Score:5, Informative)
.
Time Formats, again IMHO, would be the "standard" (ha, I heard it [that word] both ways!) used for displaying, communicating, or storing "time data values" on paper, verbally, or in a computerized (or book-keeping) record. One example: "yYYYY-MM-DD-HH-mm-ss.{fractional value of second}" [note I added an extra "y" digit to allow for the Y-10K problem!!!). Floppy disks and TRS-DOS and Apple DOS and MS-DOS and CPM and UNIX and so many others use different formats for this. They also use different "loci" for the "origin point" of time (the "epoch", e.g. time elapsed since point $x$ in time. Gregorian year 1904 for old macs, 1970 for the unix epoch, etc.
Critical Dates (Score:5, Informative)
on a side note, i love this website:
http://www.merlyn.demon.co.uk/critdate.htm [demon.co.uk]
it is a huge list of important dates relevant to computer programs, algorithms, and O/Ses.
ISO 8601 (Score:4, Informative)
http://xkcd.com/1179/
This subject has already been discussed.
Re:GPS Time (Score:5, Informative)
What is rather annoying is that GPS time is UTC without leap seconds; but(for some reason) is different than TAI, which is also UTC without leap seconds.
Re:Total (Score:5, Informative)
http://xkcd.com/1179/ [xkcd.com]
ISO 8601
But, since you mention the overabundance of standards...
http://xkcd.com/927/ [xkcd.com]
Re:Really two very broad subjects (Score:5, Informative)
Wrong. Only if you're in a part of the world that observes Daylight Saving. Which puts you in the minority. [wikipedia.org]
Re:Excel's year 1900 bug (Score:5, Informative)
Re:Time Standards vs. Time Formats, and Y10K probl (Score:4, Informative)
The length of day is not stable
http://www.nasa.gov/topics/earth/features/japanquake/earth20110314.html [nasa.gov]
Over the long (looooooong) term days are getting longer
http://en.wikipedia.org/wiki/Day_length#Historical_variation_of_day_length_because_of_tidal_acceleration [wikipedia.org]
Re:Time Standards vs. Time Formats, and Y10K probl (Score:5, Informative)
Dude, how on earth can you get it all so badly wrong? Is this a subtle postmodernist troll or something? Sorry if it isn't, starting a discussion on a scientific subject with "IMHO" sets off my bullshit alarm.
Historically, a day was defined as 1 earth rotation. An hour was 1/24th of a day, a minute as 1/60th of an hour, and a second as 1/60th of a minute. Hardly very arbitrary, is it? Problem is, turns out that there are constant fluctuations and drift on the length of a celestial day (and year). This is very impractical because there are no known clock mechanisms (bar the solar system itself) that can catch these fluctuations, so humanity needed a more solid definition of time for entirely irrelevant tasks such as performing precise scientific measurements and keeping GPS sattellites in sync. Common off-the-shelf clock mechanisms couldn't be used for this purpose because they also fluctuate too much. Instead, we redefined the lenght of a second based on an immutable physical property: the duration of 9,192,631,770 periods of the radiation corresponding to the transition between the two hyperfine levels of the ground state of the caesium 133 atom. Cesium was "arbitrarily" chosen because of atomic clock engineering and accuracy considerations, the hyperfine transition was "arbitrarily" chosen because it is not too easy to skew, directly relates to a physical constant and can be measured in a reasonably convenient way, and the number 9,192,631,770 was "arbitrarily" chosen to make the SI second as close as possible to the then-best estimate of the average duration of the celestial second (1/86400 of a day). Then we redefined the SI minute, hour, day,... based on that physically immutable(*) second. Problem is, that didn't stop the earth rotation and solar system from fluctuating. And that's why we have leap seconds now and then; to keep our non-arbitrary SI-based time in sync with the arbitrary vagaries(+) of the solar system.
(*) Well, pretty immutable for the practical applications you're going to care about. There's always room for improvement [wikipedia.org].
(+) To be precise, orbital mechanics are well-understood so in that sense they're not random.(#) Problem is, it's a chaotic system, so no matter how precise we measure all the boundary conditions, a simulation of the solar system will over time deviate more and more from reality. That's why leap seconds are based om measurements.
(#) If you look even deeper into the subject, there are solar winds and weather-dependent tidal effects contaminating the whole shebang with fluctuations we can't even predict a few weeks in advance because they're complex. So yeah, arbitrary.
Re:Time Standards vs. Time Formats, and Y10K probl (Score:5, Informative)
The standardization of the second took place 46 years ago. It is now the basis of SI, the international system of standards. The following is from NIST, the Federal Government agency in charge of standards [nist.gov]:
"The unit of time, the second, was defined originally as the fraction 1/86 400 of the mean solar day. The exact definition of "mean solar day" was left to astronomical theories. However, measurement showed that irregularities in the rotation of the Earth could not be taken into account by the theory and have the effect that this definition does not allow the required accuracy to be achieved. ... Experimental work had, however, already shown that an atomic standard of time-interval, based on a transition between two energy levels of an atom or a molecule, could be realized and reproduced much more precisely. Considering that a very precise definition of the unit of time is indispensable for the International System, the 13th CGPM (1967) decided to replace the definition of the second by the following ...:
"The second is the duration of 9 192 631 770 periods of the radiation corresponding to the transition between the two hyperfine levels of the ground state of the cesium 133 atom."
Not only that but, length is now defined in terms of the second:
"In turn, to further reduce the uncertainty, in 1983 the CGPM replaced this latter definition by the following definition:
"The meter is the length of the path travelled by light in vacuum during a time interval of 1/299 792 458 of a second.
"Note that the effect of this definition is to fix the speed of light in vacuum at exactly 299 792 458 mÂs-1. The original international prototype of the meter, which was sanctioned by the 1st CGPM in 1889, is still kept at the BIPM under the conditions specified in 1889."
Re:Total (Score:2, Informative)
YYYYMMDD is a poor choice, IMO, for two reasons.
1st, morons get halfway through (having input the 20 and 12 into day and month fields respectively) and then become baffled by the "year" part being, say, 0326. If you include a separator character, it compels them to realize it's not DDMMYYYY before they start abusing it. Fortunately, this no longer applies strongly, since they'll now trip over the month 13.
2nd. YYYY-MM-DDTHH:MM:SS is the motherfucking standard because it's reducible -- yes, YYYYMMDDTHHMMSS is permitted in ISO8601, but I'll show you why it sucks...
So all these (extended formats) are valid:
2013-03-11T05:36:45
2013-03-11T05:36
2013-03-11T05
2013-03-11
2013-03
2013
Perfect and elegant, yes?
Use basic formats, and all but one are valid:
20130311T053645
20130311T0536
20130311T05
20130311
201303
2013
YYYYMM is forbidden because it's confusable with YYMMDD (which was formerly permitted under ISO8601), whereas YYYY-MM is distinct from YY-MM-DD -- so if you just have the good sense to use the extended formats, you don't have to worry about it.