Send sundial mailing list submissions to
sundial@uni-koeln.de
To subscribe or unsubscribe via the World Wide Web, visit
https://lists.uni-koeln.de/mailman/listinfo/sundial
or, via email, send a message with subject or body 'help' to
sundial-requ...@uni-koeln.de
You can reach the person managing the list at
sundial-ow...@uni-koeln.de
When replying, please edit your Subject line so it is more specific
than "Re: Contents of sundial digest..."
Today's Topics:
1. Re: Why we should reform the Calendar (Michael Ossipoff)
----------------------------------------------------------------------
Message: 1
Date: Sun, 29 Jan 2017 12:27:56 -0500
From: Michael Ossipoff <email9648...@gmail.com>
To: Dan-George Uza <cerculdest...@gmail.com>
Cc: sundial list <sundial@uni-koeln.de>
Subject: Re: Why we should reform the Calendar
Message-ID:
<caokdy5aes1j6nfkjxxvq_wxw1x1mfggq9+f7wfjx72graio...@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"
Here are two (unimportant) objections to the Nearest-Monday year-start
system:
1. It's based on the Gregorian leapyear-rule, meaning that it isn't
self-contained & free-standing. Mostly an aesthetic objection, and I don't
consider it important.
2. It inherits certain properties of the Gregorian Calendar, which could
otherwise be adjustable, choose-able. This, too, I consider only an
aesthetic objection.
Here are the properties that I refer to:
The Gregorian leapyear-rule was designed to minimize the date's variation
at the (northern) Vernal Equinox, the March equinox.
We often hear it said that the mean tropical year is the time from one
March equinox to the next. Not so. That's because the length of a tropical
year depends on at what point of the ecliptic it's measured.
A tropical year is a seasonal year, the time during which the center of the
sun leaves & returns to some particular point of the ecliptic. That
duration is different, depending on the point of the ecliptic at which it's
measured.
That's because the Earth's orbit is continuously changing, due to
perturbation by other planets, mostly Jupiter.
So, for example, the following tropical years have different durations: The
March Equinox year, the September Equinox year, the June Solstice year, and
the December Solstice year.
Leapyear-rules try to achieve some desired "mean year". The Gregorian
leapyear-rule's mean-year is 365.2425 days. Probably mean solar days, I
assume. As I said, the Gregorian's mean-year is intended to approximate the
March Equinox year.
A mean solar day is the duration between meridian-transits of the mean sun.
The mean sun is a fictitious sun that goes around the celestial equator at
a constant rate, coinciding at the equinoxes with another fictitious sun
that goes around the ecliptic at a constant rate and coincides with the
real sun at aphelion & perihelion.
Often the length of (various kinds of) a tropical year is given in
ephemeris days (defined in terms of planetary motions), also callled atomic
days.(when defined in terms of atomic clock measurements). The length of an
ephemeris day was fixed in the early 19th century. But, since then, the
length of the day has increased a bit, and so the ephemeris day no longer
matches the mean solar day.
For example, we often hear it said that a mean tropical year is 365.24219
days. But that's ephemeris days. According to a Wikipedia article
(calendarists that I've spoken with haven't expressed disagreement with
it), a mean tropical year is actually currently about 365.24217 mean solar
days.
The length of the mean tropical year is the (current value of the)
arithmetic mean of the lengths of the tropical year measured at all the
points of the ecliptic.
So, if you calculated two Earth orbits, accounting for planetary
perturbations. recording the time at many different points of the ecliptic,
and then used those times to calculate the tropical year with respect to
those many points of the ecliptic, and then, over one circuit around the
ecliptic, numerically integrated the tropical-year-length, with respect to
ecliptic longitude, and then divided by 2 pi radians ( = 360 degrees), that
would give you the length of a mean tropical year.
As I said, according to Wikipedia, it's currently about 364.24217 days.
So, anyway, the Gregorian leapyear rule's mean-year, of 365.2425 days is
intended as an approximation of the March equinox year of about (it seems
to me) 365.24239 days.
It's understandable that they chose to favor the Vernal Equinox year. The
equinoxes are the time when the solar declination is changing fastest, and
when the season is changing fastest.
Problem: The Northern Hemisphere's Vernal Equinox isn't the Southern
Hemisphere's Vernal Equinox. The Vernal Equinox, near the beginning of
Spring, is a revered, honored & celebrated time. But why should the people
of the Southern Hemisphere celebrate the *northern* Vernal Equinox?
In 1582 that wasn't a problem for Europeans. But this is a different
century now. In this century, choosing the northern Vernal Equinox as the
basis for the calendar's mean year is more than a little north-chauvinistic
and inegalitarian.
So I'd prefer to use an approximation to the mean tropical year, instead of
the March equinox year, as a leapyear-rule's mean year. That's what my
leapyear-rule proposal does. (I'll get to that soon).
Another fair choice, another good compromise between North & South, would
be a mean-year length that's the arithmetic average of the March &
September Equinox years.
Some calendarists like the June Solstice tropical year, as a mean-year for
a leapyear-rule.
That's because the we're now only about a millennium into a roughly 10,000
year period during which the length of the June Solstice year will change
remarkably little--not enough to cause any significant
calendar-displacement with respect to the seasons..
To quote one calendarist: "Welcome to the 1st millennium of the Age of the
June Solstice Year!"
So the current period of remarkable stability of the length of the June
Solstice year has only been in effect since roughly the time of the Battle
of Hastings.
So of course the June Solstice tropical year has great appeal as the
mean-year for a leapyear-rule. And that wouldn't be unfair to the South,
because the Winter Solstice is celebrated as much as the Summer Solstice.
Though my calendar-proposal is to use the mean tropical year, or the
arithmetic average of the March & September Equinox years as the mean year
for a leapyear-rule, I'd have no objection at all to the use of the June
Solstice year, which has great appeal.
The point is that we can choose what tropical year we use for a mean year
for a calendar's leapyear-rule. But if we use the Nearest-Monday year-start
rule, we're inheriting the Gregorian's use of the March Equinox year as the
tropical year that the calendar's mean year approximates.
That isn't really a problem, but it would be nice to make that choice for
ourselves--as my Minimum-Displacement leapyear-rule (defined later) does.
The other things is that the Gregorian's 365.2425 day mean-year, being more
approximate, results in more drift (with respect to its intended tropical
year length) than would a more precise approximation. And when the
calendar's relation between date & ecliptic longitude oscillates, in the
leapyear-system, about what central date/season relation does it
oscillate? Wilth the Gregorian, and hence with Nearest-Monday, that's out
of our hands, decided for us.
I'm not saying that that's a problem either.
It's just that it would be *nice* to have the luxury of choosing, for
ourselves, 1) what tropical year we want the calendar's mean-year to
approximate; and 2) what date/season relation we want for the calendar's
center of oscillation.
The Minimum-Displacement leapyear-rule allows the luxury of making our own
choice of those two adjustment-parameters.
This posting is already very long, and so I'll save the
Minimum-Displacement leapyear-rule for a (immediately subsequent) next
posting.
Michael Ossipoff
I seems to me that the March Equinox tropical year is something like
365.24239
On Sat, Jan 28, 2017 at 8:54 PM, Michael Ossipoff <email9648...@gmail.com>
wrote:
I don't think it's really off-topic, because, with sundials, we're
interested in the EqT, which is given in terms of the calendar's dates.
Though Gorman is a comedian, he's obviously given the matter some serious
consideration, and I perceive some serious interest in calendar-reform.
But I have a few disagreements with his proposal:
*1. Blank Days:*
Gorman proposes a "fixed calendar", a calendar that will be the same for
every year. I have no objection to that. After all, so far as we know
(except for each year setting a new record for increasing global warming)
what we can expect from each year, nature-wise, is really the same. So, why
should two successive years have different calendars, with different dates
having different days-of-the-week?
So far so good. There are two ways proposed for achieving a fixed
calendar:
*1. Blank Days:*
A fixed calendar must have a number of days that's a multiple of 7, That's
what enables each calendar to start on the same day of the week, allowing
every date to have a day-of-the-week that doesn't change from year to year.
So Gorman would make one of the 365 days a "blank day", a day that isn't a
day of the week. Then the days-of-thes-week would resume after that day. so
the year would have only 364 days that are days of the week. That being a
multiple of 7, each year will start on the same day of the week, as desired.
Problem: I'm sorry, but it doesn't make any sense for the day after a
Saturday to be anything other than a Sunday. ...or for there to be an
intervening day between a Saturday & a Sunday.
Speaking for myself, I completely reject "blank-days". And I'm not the
only one. Elizabeth Achellis, over several decades, up to around 1955,
proposed a fixed calendar with blank-days. The League of Nations, and later
the U.N. were giving serious consideration to it, and it might have been
accepted, except for the strong opposition to the blank-days,
A compromise was offered to Achellis: A leap-week (described in the next
section below), to achieve a fixed calendar. She wouldn't accept that
compromise, and her proposal was indefinitely tabled around 1955, and never
got anywhere since. You could say that the blank-days were the Achilles'
heel of Achellis' calendar proposal.
*Leap-Week:*
So a 364 day common (non-leap) year achieves a fixed calendar, because 364
is divisible by 7. What about the 365th day? Well, we could deal with it
the same way we deal with the fact that the 365 day year is shorter than
the 365.24217 day Mean Tropical Year (MTY)...by occasionally lengthening a
year, to periodically compensate for the length-mismatch. So we'd deal with
the short common year just as we do now.
So, what we do is have a 364-day common year, and (by using a leap-year
rule that I'll talk about later), when that 364-day common year gets about
half a week out-of-step with the seasons, we add a leapweek, to set that
displacement back.
Gorman didn't talk about the leapyear-system, and we can presume that he
meant to use the existing Gregorian leapyear system, which would be fine,
for a leapday calendar such as he proposes. But for a leapweek calendar,
which is what I (and many others) propose, a new leapyear system is
required. No problem. I'll get to that after I discuss my disagreements
with Gorman's proposal.
Summary: A fixed calendar should be achieved via a leapweek, instead of by
blank-days. If Achellis had agreed to that, we might be using her calendar
right now.
*2. Thirteen Months:*
Really, the only reason for a reform calendar to have months, is for
continuity & familiarity with our current Roman-Gregorian Calendar.
For example, Elizabeth Achellis's *World Calendar *had, in each quarter,
months with the following lengths: 31,30,30. Having 12 months, with 30 or
30 or 31 days, means that the calendar is familiar, looks familiar, and it
means that the dates in the new calendar have really the same seasonal
meaning as the dates in the old calendar.
Achellis' 31,30,30 quarters achieves that. But there are other proposals
of a calendar with
30,30,31 quarters. The advantage?:
1. The 30,30,31 calendar's months' start-days never differ by more than a
day, from those of our current Roman months, when both month-systems start
on the same day. Achellis' 31,30,30 quarter system can differ by at least
twice as much.
2. The 30,30,31 quarters divide the weekdays most equally between the
months of the quarter.
So, if you're going to have months at all (and that's for continuity &
familiarity), then you want 12 months, of 30 & 31 days. Preferably the
30,30,31 quarters.
With 13 months of 28 days, the dates wouldn't have anything like the
seasonal meaning that they do now. Continuity, familiarity, and the
justification for having months at all, would be lost.
The 30,30,31 quarter system is an improvement over our current Roman
months, because the months are much more uniform. That allows much
meaningful & accurate monthly statistics.
But suppose you want something more radical (as is Gorman's 28X13 system):
In that case, just don't have months, because their continuity &
familiarity purpose would be lost anyway. Use the WeekDate system.
No months.
Weeks are numbered.
Here's today's date in the (currently internationally widely-used) ISO
WeekDate calendar:
4 Saturday
That means Saturday of the 4th week.
Actually, because not all countries and languages call the da
ys of the week by the same names, here is how the ISO (International
Standards Organization) words today's date.
2017W046
The "W" indicates that the WeekDate system is being used.
The "04" denotes the 4th week.
The "6" denotes the 6th day of that week.
(The ISO WeekDate Calendar uses a week (and therefore a year) that begins
on a Monday, probably so that the weekend won't be split in half.)
The ISO WeekDate Calendar is, as I said, widely used internationally, by
Companies & Governments, for their planning of business & governmental
dates & events. ...making it easy to plan them in advance once, and then
leave them, because it's a fixed calendar. Of course the resulting dates
then have to be eventually translated into Roman-Gregorian dates.
...but they wouldn't have to, if we adopted the ISO WeekDate calendar as
our civil calendar, worldwide.
ISO WeekDate has the great advantage of use-precedent. ...lots of it.
I personally like the ISO WeekDate as the best calendar-reform proposal.
But, recognizing that many people wouldn't want to give up the months, and
would want to keep them for familiarity & continuity, the 30,30,31
month-system could be a good alternative proposal, if ISO WeekDate isn't
accepted.
But it has been argued that ISO WeekDate is so convenient, and already so
widely-used, that it could easily edge-out the Roman-Gregorian Calendar,
from the bottom up, by increasingly wide use, if companies & government
start using it so much that the public start finding it convenient to use
it too.
So those are my two disagreements with Gorman's proposal, and my
alternative suggestions.
But I should comment on the leapyear rule. Actually, the ISO WeekDate
Calendar deals with that in a really easy, natural, simple & obvious way.
Each ISO WeekDate year starts on whatever Monday is closest to the
Gregorian January 1st of that year. So, for example, this year, 2017, the
Gregorian year started on a Sunday. So the nearest Monday to Gregorian
January 1st was January 2nd. That Monday, Gregorian January 2nd, is the day
on which ISO WeekDate 2017 started.
As I said, today, in the ISO WeekDate Calendar, is:
4 Saturday
(or 2017W046)
That way of defining the start of the ISO WeekDate year (the Monday closes
to Gregorian January 1st) is called the Nearest-Monday year-start system.
Note that the Nearest-Monday year-start system doen't have to mention
leapyears or leapweeks at all. It's *effectively* a leapweek calendar,
because some of the years have 53 weeks instead of 52. But the simple
Nearest-Monday year-start rule doesn't need to mention leapyears or
leapweeks.
Not only is it used with the ISO WeekDate Calendar, but of course it could
also be used with a 30,30,31 quarters calendar too.
Calendar reform advocates propose all manner of different leapyear
systems. But there's nothing wrong with the Nearest-Monday year-start
system, and conversations have suggested to me that Nearest-Monday would be
the favorite way to make a fixed calendar.
In fact, with Nearest-Monday, the maximum displacement of dates with
respect to seasons, is barely more than the ideal minimum that could be
achieved by the fanciest leapyear system.
I also propose a fancier, deluxely-adjustable system, but I won't try your
patience with that here, because Nearest-Monday is entirely good enough,
and is the system with obviously by far the best acceptance-potential.
Michael Ossipoff.
On Sat, Jan 28, 2017 at 3:38 PM, Dan-George Uza <cerculdest...@gmail.com>
wrote:
A bit off topic, but I enjoyed this quite a lot!
https://youtu.be/EcMTHr3TqA0
Dan
---------------------------------------------------
https://lists.uni-koeln.de/mailman/listinfo/sundial
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<https://lists.uni-koeln.de/mailman/private/sundial/attachments/20170129/9cea9958/attachment.html>
------------------------------
Subject: Digest Footer
_______________________________________________
sundial mailing list
sundial@uni-koeln.de
https://lists.uni-koeln.de/mailman/listinfo/sundial
------------------------------
End of sundial Digest, Vol 133, Issue 27
****************************************
