[LEAPSECS] UTC is dispredictable

[Tom Van Baak]

Funny how we go around and around about this: is UTC
continuous or discontinuous? Well, it's continuous until it
isn't. It's a predictable offset, until it isn't. It's just plain old
24x60x60 mechanical gear sexagesimal notation, except
when it isn't.

We all know what UTC is, but struggle with a word to describe
what it is. I and others often use the word discontinuous but
some take offense to this. I understand. I know it's not really
discontinuous in terms of a count of SI seconds. It may be
unpredictable, but that word too has connotations that go too
far and offense is taken again. It's smooth, but it has jumps;
it's linear, but has steps...

There are only so many words to use. And each word you pick
is already used by multiple disciplines. So when we adopt words
to describe timekeeping or time scales, we're very late to the
game; the words we choose already have deep connotations
to readers, especially those with PhD's.

UTC is a little weird (how's that for a technical word). The rare
leap second spoils an otherwise clean sequencing of digits.

So if using the word discontinuous is a problem how about we
invent a *new word* that exactly describes the odd way that UTC
counts seconds. A word that no one else has used yet; a word
that belongs exclusively to those who characterize timescales?

I vote for: dispredictable.


Google for it. It's open for the taking. By definition it perfectly
describes the sort of discontinuously continuous, predictably
unpredictable, the almost-always-but-not-quite 86400 nature of
UTC. And then we no longer have to have any arguments about
what UTC is (perhaps PHK can provide a Danish translation).

If you like this, I'll move on to inventing a word for time_t so we
don't have to go over that again and again.

/tvb
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://six.pairlist.net/pipermail/leapsecs/attachments/20110307/9304d732/attachment.htm>