## Thursday, July 15, 2010

### Time Units

To be honest, many of our most commonly used time units are quite rubbish. Consider, for instance, the hour, the minute, and the second; these divisions of time, in units of 24 and 60, render many computations of time difficult. A small saving grace is that they are more suited for fractional computations, but then again the utility of such fractional convenience is limited since most people think in whole units, if one scale down. Beyond this small saving grace, these units of time are almost entirely arbitrary.

On the other hand, the day, month, and year are elegant and logical measures of time, in that they can be easily (in terms of required technology) calibrated against natural phenomena. Hence, these units of time have been independently invented by almost every civilization.

Take the day, for instance. It requires no explanation as to how it is defined. It is measured by the length of time for the Sun to move from the apex of the sky to the apex of the sky again (i.e., noon to noon). However, taking this fact into account, it seems inherently more sensible for the day to start from noon rather than from midnight, since one requires additional tools to accurately determine the onset of midnight.

The month, though a nebulous construct of varying lengths, also has a natural basis. This is the moon phase cycle, which can be easily observed to wax and wane with regularity. Unfortunately, while the lunar month can be accurately calibrated, calendars cannot be constructed based solely on it as it does not exactly cover the length of a year. Attempts to reconcile the two have led to many workarounds, such as variable length months, leap days, months, etc. Very messy work, not at all elegant.

The final regular unit of time, and quite possibly the one with the greatest importance, is the year. While we all know the year to be the length of time for the Earth to complete 1 orbit, it does seem quite difficult to measure this. The seasons do indeed repeat on a yearly basis, but variations make it impossible to use this fact to measure the length of a year. The easiest way is to observe the positions of the stars in the night sky; the patterns should change as the Earth orbits, and match again once the Earth has moved to the exact location it occupied one year ago. However, even this method is not simple, requiring some manner of astronomical technology. It may be easier to automate the process by constructing special structures that are only aligned with certain stars or constellations on a particular day of the year; many such structures exist. If I were to have the time and resources, I might find it interesting to design and construct a simple version of such a year-measuring device.

The worst time unit is, in my opinion, the week, which is entirely arbitrary and illogical. It is neither correlated to any naturally repeating occurrence, nor does it allow for simple computation, fractional or otherwise. It boggles the mind as to how it came to be adopted. Still, as with many things, it is so deeply entrenched in our societal model that it may be impossible to rectify this aberration.