Time was when time could be measured only in relation to the rotation of the earth, as evidenced by the rise and fall of the Sun. Then along came atomic clocks which measure time relative to the vibrations of caesium atoms. A few zillion of these occur every second, unfailingly and invariably. An atomic clock won't loose or gain more than a couple of seconds in three million years.

From this absolute is derived all the time signals used for every timing mechanism in the world: from time pips on the radio to the time server that keeps your PC time displayed with great accuracy.

There is just one small problem: the agreed definition of a 24 hour day derived from the vibrations of the caesium atom does not exactly equal a real day: the time it takes for the earth to complete one rotation on its axis. It's short by a fraction of a second. That's no big deal to you or, left uncorrected it would simply mean that in about six hundred years sunrise and sunset would each be about an hour early.

But that difference would soon have significant impact on systems that need to know the precise point of the earth on its daily rotation: astronomical telescopes, sextants, geostationary communications satellites etc. So, in 1972 international agreement was reached to insert 'leap seconds' every few years to keep atomic-clock derived time in sync with real world time. We've had 21 of them so far.

However computer systems are not geared up to handle 61 second minutes, and the results of these leap seconds can be unpredictable. According to a report in the Wall Street Journal "On Jan. 1, 1996, the addition of a leap second made computers at Associated Press Radio crash and start broadcasting the wrong taped programs. In 1997, the Russian global positioning system, known as Glonass, was broken for 20 hours after a transmission to the country's satellites to add a leap second went awry. And in 2003, a leap-second bug made GPS receivers from Motorola briefly show customers the time as half past 62 o'clock."

The worry is that more dire and disastrous consequences could occur, for example if an air traffic control system or a plane's navigation system malfunctioned. So there is pressure to scrap leap seconds in favour of a leap hour the first of which would be inserted about 500 years hence. And this is where the ITU comes in. It controls the document (ITU-R TF.460) that defines Co-ordinated Universal Time (UTC) - the time used by every system from cellphone networks to satellites that needs accurate clock time - and the relationship between UTC and Absolute Atomic Time (TAI).

ITU working committee 7A is due to debate in November a US proposal to scrap leap seconds. That proposal was submitted in September 2004 but made public less than two weeks ago with only four weeks allowed for comment. It has set huge swathes of the scientific community at odds with the ITU. They are incensed at the idea and the covert way in which it has been introduced. Back in July the WSJ quoted a US spokesperson saying the matter was private between the US and the ITU.

Aside from outright opposition to the move, its opponents are calling for at least much broader public debate on the issue.

In a statement the UK's Royal Astronomical Society (which for several hundred years until the introduction of atomic time administered the global time standard, Greenwich Mean Time, based on sunrise at Greenwich in London) accused the proponents of the change of "seeking to solve their problem by exporting problems to those who use clock time as a measure of mean solar time. These include astronomers, satellite operators and potentially all who study environmental phenomena related to the rising and setting of Sun."

To put it bluntly: they reckon programmers should figure out how their systems can handle 61 second minutes without crashing, instead of foisting the problem onto others.

The astronomers argue that "The idea that clock time follows solar time is deeply embedded in contemporary technical culture through a wealth of literature (text books, web pages etc) and in the skills of working scientists and engineers around the world."

It is also fundamental to all whose work schedule is governed by the rotation of the earth. Ever cognisant of the power of the farming lobby the Queensland Government has steadfastly refused to follow NSW into daylight saving. They need not worry about this change: it will not affect their children or those of several generations to come.

Those that will be affected will be the operators of satellites whose orbits have to be kept synchronised to the rotation of the earth and telescopes which use precise timing to determine where in the heavens they should point. The upgrade costs per telescope could be as high as $500,000.

There are many more air traffic control systems communications networks and other systems whose programmer would no longer have to worry about leap seconds so I would not be surprised if expediency wins out not just over scientists but traditionalists who argue that time always has been and always should be tied into the rotation of the earth.

Oh, and by the way, 2006 will be late in coming. The next leap second is due to be inserted at midnight 31 December this year. If you plan to fly that night, you might like to re-consider.