Re: Types of time

From: Rob Seaman <seaman_at_noao.edu>
Date: Fri, 8 Sep 2000 12:45:31 -0700 (MST)

I like Steve Allen's taxonomy and think a lot of beneficial discussion
can come out of an analysis like that. In particular this might be a way
to begin to quantify the relative costs of all the different proposals.
(Should any detailed proposals actually ever be made.)

But this message will comment a bit further on this observation:

> The current UTC system is biased in favor of your applications
> and other astronomically based ones, and against the rest of us.

Yes - this is probably correct (although I suspect that folks might be
surprised at some of the ways that astronomy creeps into our lives).
But note that it isn't that UTC is astronomically rooted that causes
problems - it's that UTC is not an unsegmented timescale.

The reality is that there are technical problems that require
unsegmented time scales - and there are technical problems that
require a time scale that is stationary with respect to the Earth's
rotation. Until and unless Steve's taxonomy is completed we won't
know what fraction of the applications fall into each category, but
I don't hear anybody suggesting that both classes of problems will
not remain important indefinitely.

Since both of these type of time (as well as many other types) will
remain necessary, we will continue to generate both. And as folks
have mentioned, it is eminently possible for the various professional
constituencies to convert at will between the two types of time.
(Although there may be costs associated with that conversion.)

Whatever the results of the taxonomy, we have already demonstrated
just by postulating that a taxonomy exists in the first place that
future engineers will continue to have to worry about the best choice
of a time scale for each project. This design choice will place
constraints on implementation. Each implementation will result in
technical and economic trade-offs.

All this is something that professionals should expect to have to
deal with on any given work day.

But what should civilians have to deal with? There are a heck of a
lot more of them than of us. They are much less capable of handling
design trade-offs.

The suggestion that under no circumstances will future generations
of the civilian population ever need to worry about a monotonic and
accelerating drift of the world's clocks relative to midnight or noon -
well, that suggestion seems to me to be a non-starter.

For just one thing, at a time when the quality of science education and
of the resulting level of public scientific understanding is receiving
intense scrutiny - does it seem likely that a proposal to downgrade
the quality of civil time will be politically supportable?

Someone suggested simply issuing a leap *minute* every 50 years
or so. Note that this is a different proposal than the survey and
this list were initiated to discuss. I doubt I'm alone in welcoming
the chance to comment on something other than the most extreme
"Leap Seconds? Who needs 'em?" proposal.

Rob Seaman
National Optical Astronomy Observatory
Received on Fri Sep 08 2000 - 12:45:40 PDT