[LEAPSECS] The relation between calendars and leap seconds.

[Tony Finch]

On Mon, 10 Nov 2008, Rob Seaman wrote:

> Tony Finch wrote:

>

> > While we are taking a historical view of calendars, it's probably worth

> > observing how past problems similar to the current situation have been

> > resolved. UTC is an observational calendar, and over history these have

> > almost always been replaced with arithmetic calendars: this eliminates

> > problems of communication from the observers who determine the calendar to

> > its users, gives users more independence and allows them to make more

> > accurate plans for the future, at the cost of a smallish error that makes

> > future dates drift away from the events they were previously attached to.

>

> Words like "smallish" are to be avoided when writing requirements.

> Fundamentally your assertion is still that nobody but astronomers will

> notice, right?  So how might this be characterized in specific terms?

> That will give us something to grapple with.


I deliberately chose a vague word there because the size of the error is
an important point of discussion. It's another area where requirements
change over time. 365.25 days in a year was good enough for Julius Caesar,
but not for Pope Gregory.

For civil time we have some weird double-think going on, with subsecond
precision for universal time, but hours of error for most locations'
civil time. Clearly the "smallish" error (of civil time vs. local mean
or apparent solar time) for civil purposes can be comparatively large,
but for navigation and astronomy it must be as small as possible.

Halsbury's Laws of England has some interesting (bizarre, self-
contradictory) paragraphs on the meaning of civil time, which I'll
append as a postscript because of their length.


> I like your observational/arithmetic formalism -


Not mine! I got it from a post to this list by Zefram on 1st June 2006.


> - although I might draw different conclusions.  The thing about

> calendars is that the observational aspect is quantized - today is

> today, tomorrow is tomorrow.


Yes, but it is only precise when averaged over long periods of time, that
is, you can count days this way but not minutes. This problem also applies
to other observations such as the new moon or the spring equinox. The lack
of precision (and, for many observational calendars, variability depending
on location) leads to unpredictability: you can count months or years
accurately but your day numbering may be off. Hence we used mean time in
preference to apparent time, and we use arithmetic calendars in preference
to observational ones.


> What is this human factors requirement?


Predictability.


> Adi Stav wrote:

> >

> > Have there been suggestions, indeed, for such a predictable SI-second-

> > based calendar that synchronizes with the Earth's rotation?

>

> Well, that is the UTC that we have.


No, it isn't predictable.

Tony.


PS.

215. Local time.

Apart from statute [1] or special convention, the hour of the day has to
be ascertained by reference to the sun in the particular place. At a given
moment, therefore, the time is different in different places. The hour at
which a court is fixed to sit means prima facie the hour at the locality
where the particular court is to sit, and not Greenwich mean time [2].

[1] See para 216 post.

[2] Curtis v March (1858) 3 H & N 866, where failure to appear in court
in Dorchester, when the judge took his seat at 10 a.m. (according to
Greenwich mean time) constituted no default, since according to the local
time it was some minutes short of 10 a.m. This situation could not arise
today, since Greenwich mean time is now applied by statute: see para 216
post.

216. Greenwich mean time.

For the purpose of statutes, subordinate legislation, deeds or other legal
instruments, it is provided by statute that, unless the contrary is
expressed (and subject to the provisions regarding Summer Time [1]),
expressions of time are to be taken to refer to Greenwich mean time and
not to local time [2]. Regard must be had to this rule in applying the
numerous statutes [3] in which certain hours of the day are specified
within which acts may or may not be done. It is apprehended that
Greenwich, and not local, time must be considered in fixing the hour or
day of an event with regard to which provision is made in an instrument
such as a policy of insurance, and that on the other hand the statutory
rule should not be applied in a case where the instrument was executed or
the event was expected to happen or did happen in a foreign country [4].

It has been held that 'sunset' in certain enactments is not such an
expression of time as previously mentioned [5], but refers to local time
[6].

[1] Ie subject to the Summer Time Act 1972 s 3: see para 217 post.

[2] See the Interpretation Act 1978 ss 9, 23(3), Sch 2 paras 1, 6; and
evidence vol 17(1) (Reissue) para 569; but see also R v Logan [1957] 2 QB
589, [1957] 2 All ER 688, C-MAC, where it was held that a statute said to
commence on 1 January came into force on the day which was 1 January in
the particular place where the Act had to be applied.

[3] Eg the Marriage Act 1949 s 4 (hours for solemnisation of marriages:
see matrimonial law vol 29(3) (Reissue) para 85); the Licensing Act 1964 s
59 (as amended), s 60 (as amended) (sale and consumption of liquor except
during permitted hours: see intoxicating liquor vol 26 (2004 Reissue)
paras 289 et seq).

[4] See R v Logan [1957] 2 QB 589, [1957] 2 All ER 688, C-MAC.

[5] See para 213 ante. [which says a day is midnight-to-midnight, or any
period of 24h, or sunrise-to-sunset, depending on context]

[6] Gordon v Cann (1899) 80 LT 20, DC, where the obligation imposed by
what is now the Road Vehicle Lighting Regulations 1989, SI 1989/1796, reg
25 (see road traffic vol 40(1) (2007 Reissue) para 393), to light a
carriage used on the road half-an-hour after sunset was in question, and
it was held that regard must be had to the actual hour of sunset at the
particular place; MacKinnon v Nicolson 1916 JC 6, where 'sunset' and
'sunrise' in the Salmon Fisheries (Scotland) Act 1862 s 27 (repealed),
which made it an offence to fish for salmon between those times in certain
circumstances, were held in Scotland to mean the times at which the sun
sets and rises at the locus of the alleged offence. See also the Night
Poaching Act 1828 s 12 (as amended); and animals vol 2 (2008) para 791.

217. Summer time.

[gives the start and end of the period of summer time]

218. Reckoning of time during summer time.

During the period of summer time, the time for general purposes in Great
Britain [1] is one hour, or during any part of that period for which it is
so directed by Order in Council [2], two hours, in advance of Greenwich
mean time [3].

Wherever any reference to a point of time occurs in any enactment, Order
in Council, order, regulation, rule, byelaw, deed, notice or other
document whatsoever, the time referred to is, during the period of summer
time, deemed, with certain exceptions, to be the time fixed for general
purposes under the Summer Time Act 1972 [4].

[1] 'Great Britain' means England, Wales and Scotland: see the Union with
Scotland Act 1706, art 1; and constitutional law and human rights vol 8(2)
(Reissue) para 3. The Summer Time Act 1972, however, applies also to
Northern Ireland, the Channel Islands and the Isle of Man (see ss 4(1),
5(1)), but Orders in Council under s 2 may make different provisions with
respect to the islands (see s 5(2); and para 217 note 2 ante).

[2] The power to direct by Order in Council that the time for general
purposes is, during any part of the period of summer time, to be two hours
instead of one hour in advance of Greenwich mean time is conferred by ibid
s 2(1)(b). At the date at which this volume states the law no Order in
Council under s 2(1)(b) for any subsequent year had been made. As to
Orders in Council under the Summer Time Act 1972 see para 217 note 2 ante.

[3] See ibid ss 1(1), 2(1)(b). As to Greenwich mean time see para 216 ante.

[4] Ibid s 3(1). A print-out from a Lion Intoximeter device recording
results of a breath test for excess alcohol was not admissible for failure
to comply with the Summer Time Act 1972 ss 1, 3 even though timed
according to Greenwich mean time when British summer time was in effect at
the time of the offence: Parker v DPP [1993] RTR 283, (1993) 157 JP 218.
For the purposes of the Police and Criminal Evidence Act 1984 s 69, errors
in the time or date of such print-outs caused by the Intoximeter device
recording Greenwich mean time during British summer time which do not
affect any material aspect of the document produced do not render the
document inadmissible as evidence: DPP v McKeown [1997] 1 All ER 737,
[1997] 1 WLR 295; DPP v Horswill (2 July 1997, unreported). Nothing in the
Summer Time Act 1972 affects the use of Greenwich mean time for the
purposes of astronomy, meteorology or navigation, or affects the
construction of any document mentioning or referring to a point of time in
connection with any of those purposes: s 3(2).

["by Order in Council" is the way the British Government issues
regulations under the authority of an existing statute. Statutes
are enacted by Parliament, but Orders in Council are not.]


-- 
f.anthony.n.finch  <dot at dotat.at>  http://dotat.at/
FAIR ISLE FAEROES: NORTHEASTERLY BACKING NORTHWESTERLY 5 TO 7, OCCASIONALLY
GALE 8 AT FIRST, DECREASING 4 IN FAEROES LATER. ROUGH OR VERY ROUGH DECREASING
MODERATE OR ROUGH. RAIN OR SQUALLY SHOWERS. MODERATE OR GOOD.