[LEAPSECS] Following an open source process

Rob Seaman seaman at noao.edu
Sun Mar 6 12:45:48 EST 2011


Poul-Henning Kamp wrote:


> The most likely standard group to sort out trouble, should the Doctor-Who-on-psycodelics timescale continue, is a very good question. My bet: it will be a cabal of open source people.


Most everyone on this list is part of the open source community, though some of us prefer a bit more mature development process than "cabalistics".

If you believe that universal time all by itself is psychedelic, well, you might want to take your programmer's pasty complexion out to bake a little in the Sun.

If it is the notion of a single civil time scale that combines both universal time and atomic time that bothers you, then we have something to work with. Then this becomes a classic case of engineering trade-offs. We have requirements such as:

- the civil day is the synodic day

- the unit of time is the SI-second

- a non-UT timescale should be called something other than "UTx"

And we have options to debate such as:

- preserving the current time scale to permit an orderly transition

- immediately lengthening the leap second schedule to the 2-3 years that appears possible without loosening the DUT1 constraint

- building a plan to prudently increase DUT1 in stages to permit lengthening the schedule to 10 years or beyond

However, pretending the entire world can ignore the synodic day is not an option.

At the same time we can consider options for replacing UTC with some other mechanism. The trade-offs should be explicit in a well conceived plan that seeks consensus in advance. "Cease issuing leap seconds. Period." is not a plan of any sort. And it is patently obvious that there is no consensus on the issues.

Without a consensus on a solution (or range of possible solutions), implementing the draft in front of the ITU will result in widespread pernicious - dangerous - confusion immediately, as well as a massive mess to clean up - whether in a decade, a century or a millennium.

Rob Seaman
National Optical Astronomy Observatory


More information about the LEAPSECS mailing list