Poul-Henning Kamp wrote: >By extension, that is why most calendar reform proprosals fall flat >before they even get talked about: they tinker with details. If >you want to reform calendars, do something radical so that people >can see the difference clearly.
I wonder if this is why TAI hasn't caught on properly: we use time-of-day notation for it. I prefer to deal with TAI (and, when appropriate, TT) in the form of a linear count of seconds since the 1958 epoch. My computer clock display has a field in that format (currently showing 1.546_464_427_5 Gs since the epoch). I'd probably get confused if I had a time-of-day TAI display. (For much the same reason, I don't have a UTC display. The only time-of-day display I have is the regional civil time, presently equal to UTC.) On the other hand, there's often a big resistance to accepting things that look different. Basically, people are a problem. -zefram