Although it takes longer to say, “Do we really need a leap second?” than to allow one to come and go, the ongoing wrangle over these troublesome if fleeting moments has spanned a decade now. Most of the debate has occurred off the clock, at least as far as everyday living goes. Alarms continue to wake us, Thanksgiving turkeys get cooked, and trains run on time. Who’s going to miss a second, leaping or otherwise, and why does it matter?

It’s framed more seriously, but that’s one of the questions a committee of the International Telecommunication Union (ITU) has been formally considering since 1999. The committee has now recommended that Coordinated Universal Time (UTC), which is the alarm-triggering and turkey-roasting sort we’re all familiar with, part ways with mean solar time (or “UT1” as the timekeeping elite call it). This is astronomical time, based on the Earth’s rotation and the sun doing what it’s appeared to have done for millennia—rise in the east and set in the west. By dispensing with the leap second, UTC will be free to tick away with its celebrated atomic precision, while the wobbling and less predictable Earth will have to muddle along as best it can.

I’m not making this up. Here’s an excerpt from the ITU committee’s summary: “The issue in the draft revision… is whether or not to define UTC as continuous uninterrupted time scale by eliminating the provision for leap seconds…. The major technical consequence of eliminating leap seconds would be a gradual divergence between UT1 and UTC on the order of about 1 second per year.”

The committee is now asking member states to weigh in yea or nay. The deadline for comments is Oct. 31, 2010, and to aid the discussion, the committee has drafted a brief questionnaire. I find the last query particularly civilized:

“If it is agreed to eliminate leap second within five years after approval of the revision of Recommendation IT-R TFR.460.6, would that create technical difficulties for your administration? (If YES, could you explain your reasons?)”

It’s ironic that civilization, having outgrown the usefulness of solar time, must now decide if it might have created more than it can handle with Coordinated Universal Time. The world has had the benefit of UTC since 1961, when national metrology institutes began using atomic time scales to measure our golden hours and diamond minutes. Early versions consisted of quartz clocks whose frequencies were calibrated by a single atomic clock. This system of International Atomic Time (or TAI, based on a French acronym) became more refined as our technological wizardry advanced. Now TAI is computed by taking the weighted average of more than 300 atomic clocks located at more than 60 timing labs around the world.

Atomic clocks use caesium-133 atoms as a reference point because all but the outermost of their 55 electrons have very stable orbits within an electromagnetic force. Atomic time is more predictable than solar time because the Earth doesn’t rotate with the same precision as caesium-133’s revolving electrons. The difference between the two systems works out to about a second annually, give or take the Earth’s mood during that year. So every year and a half, on average, the International Bureau of Weights and Measures outside Paris, where the world’s atomic time is coordinated, adds an extra second—the infamous leap second—to the world’s calendar. This adjusted time is what we use as UTC, and it’s what keeps UTC and solar time in a working partnership. (For those interested in such things, a leap second caught in midleap can be viewed here.)

Aeronautics experts are among those interested in dispensing with the leap second as a way of reducing risk. It seems that GPS devices ignore all leap seconds added after 1980, so these navigational aids are becoming increasingly out of sync with UTC. At the AIAA Guidance, Navigation, and Control Conference held in Toronto in August, a white paper was presented that outlined some of the problems of coordinating between GPS- and UTC-centric worldviews.

Leap seconds can also be pesky for software designers. As Joab Jackson reports in a recent issue of Broadband, “The leap second added on to the end of 2008, for instance, caused Oracle cluster software to reboot unexpectedly in some cases.” And because leap seconds can’t be predicted too far ahead, based as they are on the Earth’s unpredictable gyrations, new and time-sensitive software can’t be proactively safeguarded in a way that also addresses the necessities of a design project’s timeline. Computer systems already up and running, such as some data-logging applications, telecommunications systems, and time-distributions services, must make adjustments every time a leap second is added.

Those dubious about severing time’s relationship with an observable reference to the natural world point out that the ITU’s recommendation doesn’t offer any practical solution for what would become an ever-widening gap between solar and atomic time. The notion of adding a “leap hour” every 700 years or so, though intriguing (How would you spend it, for instance?), would still leave several centuries’ worth of our descendants coping with the slack. Ignoring it completely would leave them with the eventual problem of sitting down to breakfast at high noon. And how would our own biological clocks factor into this scenario? Humans might be smart, but we’re still affected by things like daylight, moonlight, and the seasons.

Astronomers, understandably, aren’t too happy about the idea, either. Their costly and sensitive telescopes rely on solar-based time. Reconfiguring to TAI would be expensive and time-consuming. Besides, they already have enough on their hands contributing to a scientific discipline that is tied to light years.

Perhaps we’d be better served by standardizing our various relationships with the leap second rather than eliminating it. Sounds like a job for metrologists. Luckily, we have a couple of years to consider (or forget about) the problem. The ITU committee charged with solving it won’t submit its findings for final approval until 2012.