On Tue, 1 Mar 2011, Brent wrote: > > I suppose that with people around the world chatting on the the internet maybe > we will see a cyber time or a virtual time common to the entire earth.
We have had Universal Time for over a century. People still prefer to use their own local time. There is a weird psychology in the way we interact with clocks. People like to have a common point on the clock at which they do things like get up and go to work. We get uncomfortable if this cultural anchor is removed, which is why we have DST rather than just telling people to get up earlier in the summer. David Prerau's book variously titled "Saving the Daylight" or "Sieze the Daylight" is enlightening. > Or I suppose we could end up with a decimal time like we discussed earlier. It isn't just the revolutionary French that tried to decimalize time. Swatch Internet Time is an interesting example that is both a universal time and decimalized. It is UTC+1 where each day is divided into 1000 beats. It didn't catch on, of course. > Maybe it would make more sense for each of us to get in touch with our own > biological circadian clocks and keep time according to those individual > rhythms. Some people have tried sleeping 6 times a week, with a 28 hour cycle. It can work OK but it's rather anti-social. But it seems that our natural cycle is only slightly longer than 24h, though varies between people and is easily extended by exposure to artificial light in the evenings. We don't have a solid innate circadian rhythm that could give us a better length of day, even if it made sense to decouple ourselves from the Earth's diurnal cycle. > However, these would express time as a fraction of one earth day. > I wonder if time has to be based on the rotation of the earth? No, it doesn't: there are a number of alternatives. You can mark time using any regular oscillator attached to a counter. For a simple long-running timescale you need a long-running oscillator. For long time periods we used to use the earth's orbit around the sun and the moon's orbit around the earth as the oscillators. But observational calendars are awkward for planning: how many days until Ramadan? So now we generally use the rotation of the earth for long periods of time as well as short ones. In the mid 20th century it became clear that the Earth was not a sufficiently regular oscillator for precision timekeeping purposes. (It varies unpredictably because of earthquakes, volcanoes, sea currents, weather and suchlike.) As a result in 1956 the second was re-defined in terms of Newcomb's ephemeris, as a fraction of a year. In practice this meant that the time as measured by accurate clocks was compared to astronomical observations, in particular the occlusion of stars by the moon, which provided retrospective corrections for those clocks. These observations were very time consuming so there was a long delay between measurements being taken and corrections being published. At about the same time, atomic clocks were invented. These provide much more immediate access to accurate frequency and hence to accurate time. Hence in 1967 the second was re-defined in terms of the frequency of radiation emitted by the hyperfine transition of caesium - after several years work to calibrate it to the ephemeris second. An atomic clock by itself is not necessarily a particularly long-running device, especially if it is one of the most accurate primary frequency standards. So TAI (international atomic time) is based on measurements and comparisons between many clocks in many locations in an ongoing scientific metrological collaboration. It requires a lot more effort to maintain than solar time! There is a serious difficulty with atomic time. The Earth's rotation is slowing down because of tidal friction. Because of this the length of the day is increasing by about 1.4ms per century. Since the atomic second was calibrated to the Earth's rotation speed in 1900, over the course of a year an atomic clock now runs about half a second fast compared to a clock running on mean solar time. In the 1960s, radio time broadcast services used a complicated system of frequency adjustments and 100ms steps to closely track UT2 (which is the most uniform timescale based on earth rotation). This was simplified in 1972 to create UTC which we use to this day. Time broadcasts use a fixed frequency (so they can be used as a frequency reference) and count time at the same rate as TAI. Every so often (as determined by observations of the rotation of the Earth) a leap second is inserted to keep UTC within 0.9s of UT1. UT1 is less uniform than UT2 but it is more directly related to the Earth rotation angle, so it is more useful for astronomers and others who want to point an instrument at a particular place in the sky. For the last 12 years or so there has been an ongoing argument about whether UTC should become a purely atomic timescale, that is whether to stop inserting leap seconds. The reason is that leap seconds break the millenia-old rule of 60 seconds per minute. Most computing systems have a model of time that fundamentally assumes there are always 86400 seconds in a day, and it is very difficult to get them to handle leap seonds gracefully. The risk of disruption matters because, unlike DST changes, leap seconds are not scheduled to avoid the working day. Many of the common reasons for needing a precise (15 arc second) measure of earth rotation angle have disappeared in the last 20 years as astronavigation has been replaced by satellite navigation. On the other hand, some systems (telescopes, satellite dishes) that rely on UTC for an initial estimate of earth rotation angle will need to be changed so that they can obtain and use the value of UTC-UT1. A lot of the difficulties could be reduced if leap seconds were more predictable. At the moment the presence or absence of a leap second is announced about 6 months in advance. Increasing that to several years would make systems easier to test. It isn't yet clear whether leap seconds will be stopped or left alone or announced further in advance or what. The decision is in the hands of the rather opaque International Telecommunications Union Radiocommunication Sector, and it is due to be made at the World Radiocommunication Conference next year. Tony. -- f.anthony.n.finch <d...@dotat.at> http://dotat.at/ South Utsire, Forties, Cromarty, Forth: Southerly, veering westerly or northwesterly later, 5 or 6, decreasing 3 or 4 later. Slight or moderate, occasionally rough in South Utsire, Forties and Cromarty. Rain later. Moderate or good. --------------------------------------------------- https://lists.uni-koeln.de/mailman/listinfo/sundial
