[LEAPSECS] Consensus building 2

Warner Losh imp at bsdimp.com
Wed Feb 2 21:29:45 EST 2011


On 02/02/2011 18:50, Stephen Colebourne wrote:

> OK, so we've got a little bogged down in redefining what appear to be

> well defined things, and whether a list like this should define things

> anyway. I'll give it one more go, but sadly I don't have the "patience

> of Job" if others don't also want consensus.

>

> Remember, I'm not an expert to the same degree as others on this list.

> Thus statements I make will be simplistic, but are the question is

> whether they are good enough for the problem at hand? Writing a

> clarification is only helpful if you propose a better statement to

> replace that you are objecting to.


Thanks for the update. It is a lot closer...

> A star is used for a new or amended line.

>

> General:

> * these points of consensus exist to aid the understanding of leap

> seconds not time in general

> * the terms seconds, minutes, hours and days are overloaded

> * relativistic effects do not significantly impact the understanding

> of leap seconds

> * definition: a time-line is the general passage of time

> * definition: instant - an instantaneous point on the time-line

> * definition: duration - the length of a portion of the time-line

> * definition: time-scale - a set of rules giving meaning to an instant

> * the length of time between two instants is a duration

>

> SI

> * definition: SI-second - a standardised unit of measurement for

> durations (defined in detail elsewhere)

> * the accurate measurement of an SI second is complex and typically

> achieved via cooperation

> - the SI-second forms the basis for many other fundamental units of measure

>

> TAI

> * definition: TAI-2008 - a time-scale commonly named TAI last revised

> in 2008 (defined in detail elsewhere)

> * the accurate measurement of TAI is complex and typically achieved

> via cooperation


Accurate creation of TAI is compled and achieved via cooperative
measurements. TAI isn't measured directly, but created after the fact
based on the measurements input. A subtle point, but one worth remembering.


> * the TAI-2008 time-scale is defined as a uniformly increasing count

> of TAI-seconds from a fixed epoch

> * definition: TAI-2008-second - the same as SI-second for the purposes

> of this discussion

> * TAI-2008 does not definitively imply a definition for minutes, hours and days

>

> Solar

> * definition: solar-time - time kept or measured by the Sun

> * apparent-solar-time and mean-solar-time are two forms of solar-time

> * the accurate measurement of solar-time is complex and typically

> achieved via cooperation

> * definition: mean-solar-day - the descriptive subdivision used when

> describing mean-solar-time

> * the length of a mean-solar-day in in SI-seconds varies over time

> * the length of a mean-solar-day in in SI-seconds is on average

> increasing with time

> * the length of a mean-solar-day is not a fixed number of SI-seconds

>

> UT

> * definition: UT - a time scale based on the rotation of the Earth

> (defined in detail elsewhere)

> * UT has a number of variants, the most common are UT1 and UTC

> * the accurate measurement of variants of UT is complex and typically

> achieved via cooperation

> * definition: UT1 - a smoothed variant of UT (defined in detail elsewhere)

> * definition: UT1-day - the subdivision used when expressing UT1

> * the length of a UT1-day in in SI-seconds varies over time

> * the length of a UT1-day in in SI-seconds is on average increasing with time

> * the length of a UT1-day is not a fixed number of SI-seconds

> * definition: UT1-second - a fraction 1/86400 of a UT1-day

>

> UTC-1972

> * definition: UTC-1972 - a time-scale commonly named UTC last revised

> in 1972 (defined in detail elsewhere)

> * the accurate measurement of UTC is complex and typically achieved

> via cooperation

> * the UTC-1972 time-scale is a continuous count of SI-seconds

> * the UTC-1972 time-scale defines UTC-1972-days

> * definition: UTC-1972-day - a duration either 86400 SI-seconds or

> 86401 SI-seconds long

> - the additional SI-second in a UTC-1972-day is a leap-second

> * the presence or absence of a leap-second is currently determined up

> to 6 months in advance

> * leap-seconds are added to UTC-1972 with the aim of keeping UT1 and

> UTC no more than 0.9 SI-seconds apart

> * leap-seconds are currently added about once every 18 to 24 months

> * UTC-1972 is an integral number of SI-seconds different from TAI-2008

>

> Humanity

> * definition: humanity-day - a non-scientific, commonly used term

> understood by 6bn humans

> - a humanity-day is interpreted in line with the rising and setting of the Sun

> * the legal definition of a day varies by country

> * the legal definition of a day may be based on UTC-day,

> mean-solar-day or humanity-day

>

> Maybe more controversial

> * UT1 is the most commonly recognised form of mean-solar-time

> * a UT1-day is the most commonly recognised form of mean-solar-day

Agreed up through here.

> * a UT1-day is the most commonly recognised form of a humanity-day

Not sure I'd agree on this point... We've had 40-odd years where the
humanity day has been a UTC day not a UT1 day. But the delta between
these is small. A humanity-day works well with anything that
approximates a UT1 day, since most people's humanity-day is local time
based not UT1 based. Local time is determined by timezones and is
generally close to an integral number of hours off of UT1 (or 1/2 hours
in the case of a few zones). Also some percentage of humanity lives
where the civil time differs from the local solar time by more than an
hour, sometimes for the entire year.


> * the duration of an SI-second is considered to be constant

> * the duration of a UT1-second is close to, but not equal to an SI-second

>


Agreed.

I think is is really good progress: one minor tweak and one point you
thought would be contentious that I'm contending...


> Stephen

> _______________________________________________

> LEAPSECS mailing list

> LEAPSECS at leapsecond.com

> http://six.pairlist.net/mailman/listinfo/leapsecs

>

>

>




More information about the LEAPSECS mailing list