Terran Calendar, Slashdot Comments, and Calendar Reform

Oh those geeks know a thing or two about everything.

Well I came across this article at slashdot.org, Introducing a Calendar System for the Information Age, wherein the Terran Computational Calendar (TCC) makes its case.

First, the original article from slashdot.org:

Introducing a Calendar System For the Information Age

Posted by timothy on Thursday March 27, 2014 @03:03PM
from the might-not-last-a-whole-week dept.
First time accepted submitter chimeraha (3594169) writes “Synchronized with the northern winter solstice and the UNIX Epoch, the terran computational calendar contains 13 identical months of 28 days each in addition to a short Month Zero containing only new year’s day and a single leap year day every four years (with the exception of every 128 years). The beginning of this zero-based numbering calendar, denoted as 0.0.0.0.0.0 TC, is on the solstice, exactly 10 days before the UNIX Epoch (effectively, December 22nd, 1969 00:00:00 UTC in the Gregorian Calendar). It’s “terran” inception and unit durations reflect the human biological clock and align with astronomical cycles and epochs. Its “computational” notation, start date, and algorithm are tailored towards the mathematicians & scientists tasked with calendrical programming and precise time calculation.There’s a lot more information at terrancalendar.com including a date conversion form and a handfull of code-snipits & apps for implementing the terran computational calendar.”
In many ways, I find that the TCC is similar to theAbysmal Calendar’s perpetual year. Although the 13-month year is hardly new (Positivist, International Fixed, Pax Calendar, Dreamspell, New Universal Perpetual), nor is aligning the intercalary day with the Southern Solstice (Aquarian Calendar Calendar 13, Kluznickian Calendar), however, it is the number of distinct features that TCC incorporates that make it more similar to theAbysmal than others.
  • 13-month year
  • 28-day month
  • 1 intercalary day (on Dec 21)
  • 128-year leap year cycle
  • numbering from 0
  • numbering weekdays, months, etc. without attaching names
  • existing weekdays are not incorporated into the calendar

Certainly none of these were original ideas to me, so I’m not accusing anyone of anything. Just noting the similarities.

There are a number of issues with any reform proposal. This one is designed for coders (I’m guessing), whereas it leaves out certain cultural references (like the cycle of the moon), as it is outside of the TCC’s creator’s purvue.

~  ~  ~  ~  ~  ~  ~  ~  ~

However, what I’d rather look at are the problematic issues and so forth that came up in the insightful and funny comments on slashdot.org.

Here are two key comments:

And time in .beats? (Score:3)
by oneiros27 (46144) on Thursday March 27, 2014 @03:18PM (#46595151) Homepage

Are we going to have to use Swatch Time [swatch.com] with this calendar?

All kidding aside, they mention:

MINUTES, SECONDS, & FRACTIONS OF A SECOND
Both minutes and seconds have a range from 0 to 59. If including a fraction of a second, write it as a decimal at the end: 41.13.27.23.59.59.999 TC .

… so no handling of leap seconds. I know some people would be happy about this, but if you’re not going to care about solar noon, why deal with leap days and such, too?

(and for those who complain that UTC shouldn’t have leap seconds … I say go and use TAI or GPS, but don’t change UTC because you don’t want to deal with the complexity)

From the above, I looked into 3 things I hadn’t spent a lot of time considering – the second. So, looked up leap seconds, difference between UTC and TAI. So, complex to say the least. I’ll post about that when I’ve properly researched and considered what there is to consider.

Not bad (Score:3)
by jdavidb (449077) on Thursday March 27, 2014 @03:41PM (#46595429) Homepage Journal

As far as calendars go, this is not a bad effort. I don’t think I would personally use it, but I’ve seen (and created) far, far worse. It is very regular; the rules have few exceptions, and the exceptions are well-defined. There aren’t too many decisions in it that stand out as glaringly unjustified or confusing, other than of course by definition, when you create a new calendar, the very decision to do so stands out as glaringly unjustified. :)

Nice.

/sigh (Score:5, Interesting)
by Guppy06 (410832) on Thursday March 27, 2014 @04:05PM (#46595699)

…Epagomenal days wreak havoc on “monthly” billing cycles (see: Coptic calendar, Mayan calendar, et al.). This is why the Julian and Gregorian bissextile day is explicitly a part of February.

and a single leap year day every four years (with the exception of every 128 years).

The Gregorian calendar design explicitly rejected more precise intercalation cycles in favor of numbers that were easier to remember (i.e. more user friendly). Hell, the quadrennial bissextile cycle introduced by the Julian calendar got screwed up in Augustus Caesar’s own lifetime. Never underestimate the need for simplicity.

The beginning of this zero-based numbering calendar, denoted as 0.0.0.0.0.0 TC

We can’t even get all programming languages to start their arrays at 0. What makes you think it’ll be easier for non-programmers to accept this?

is on the solstice, exactly 10 days before the UNIX Epoch (effectively, December 22nd, 1969 00:00:00 UTC in the Gregorian Calendar).

The solstice is an instant; the date it occurs on depends entirely on your meridian/time zone (e.g. the Chinese calendar explicitly specifies Beijing time). So “exactly ten days” is a meaningless descriptor.

Besides, since you’re adopting a quadrennial intercalation cycle, that instant will drift back about six hours every year, further screwing up your “exactness.”

Last but not least: the solstice is a fundamentally difficult astronomical phenomena to measure. The instant it occurs is somewhere in the window where the sun’s north-south motion is too small to measure. Equinoxes have historically been measured with far greater precision.

It’s “terran” inception and unit durations reflect the human biological clock

Then where the heck are your 28-day months coming from? The billions of people who live under a lunar or luni-solar calendar already know that the average synodic month is about 29.5 days, and that’s the “month” that affects tides and human fertility cycles.

and align with astronomical cycles and epochs.

Really?

There is no integer number or integer ratio of days (mean solar or otherwise) in a tropical year
There is no integer number or integer ratio of days (mean solar or otherwise) in a synodic month
There is no integer number or integer ratio of months (synodic or otherwise) in a tropical year

Days, months and years have nothing to do with each other; there is nothing to “align” to.

Its “computational” notation, start date, and algorithm are tailored towards the mathematicians & scientists tasked with calendrical programming and precise time calculation.

Days, months and years aren’t SI units, and the one true SI unit of time has jack shit to do with any of them.

This one’s going to require a bit of thought. I’ll break it down point by point, see how theAbysmal measures up to the poster Guppy06’s criticism.

Point 1

Epagomenal days wreak havoc on “monthly” billing cycles (see: Coptic calendar, Mayan calendar, et al.). This is why the Julian and Gregorian bissextile day is explicitly a part of February.

I don’t doubt that having a day outside of the monthly, weekly, or annual cycle can “wreak havoc” etc. I think that may have to do with the nature of such a large change to anyone, rather than the nature of the system itself. I may be misunderstanding Guppy06’s argument here. The division of the year into 52 weeks (not weekdays, that’s something else entirely – 52 7-day periods) being evenly divisible into 13 months of 28 days each, or 4 quarters of 91 days each, that’s months of 4 weeks and quarters of 13 weeks strikes me as having more regular periods than the Gregorian currently does.

With the Gregorian, “monthly” billing can be 28, 29, 30, or 31 days, depending on the month and year. In the case of flat rate “monthly” billing, a single fee is the same over a disparity 4 lengths of month.

Quarters in the Gregorian calendar are 3 months long, ranging from 90 to 92 days.

What the epagomenal or intercalary day represents is a day outside of these cycles. A day where there is no commerce. A day where no money exchange happens. Everything shuts down for one day. (Lucky bastards in the southern hemisphere get the summer solstice. Congratulations!)

The argument against a day untouchable by commerce doesn’t interest me. Commerce has permeated so much that keeping something sacred would do us well.

Point 2

and a single leap year day every four years (with the exception of every 128 years).

The Gregorian calendar design explicitly rejected more precise intercalation cycles in favor of numbers that were easier to remember (i.e. more user friendly). Hell, the quadrennial bissextile cycle introduced by the Julian calendar got screwed up in Augustus Caesar’s own lifetime. Never underestimate the need for simplicity.

Well, “easier to remember,” “more user friendly,” can mean different things to different people under different circumstances. Hard point to argue either way. I agree that the need for simplicity is important – however, having something that fits together mathematically can also make it easier to remember, which in and of itself is more important than mere simplicity.

In any case, the rule “every four years with an exception every 128 years” strikes me as more simple than the Gregorian leap year rule “every four years except – in years divisible by 100 there is no leap year – except except in years divisible by 400 there is a leap year.”

1/4 -1/128

vs

1/4 – 3/400

At the very least, 128 is a binary number, so counting the years in binary might also help – especially as the I Ching a divination tool based on a binary system is explicitly associated with the calendar, and so a precedent has been established.

Using the I Ching as  a model, a leap year counter using 7 vertical lines would provide the means of measuring the leap year day. Begin with 7 dark lines (year 0), and count from the bottom up.

7 black lines  = 0000000, 1 white 6 black lines = 1000000, 1 black 1 white 4 black = 0100000, etc.

i-ching-7-linesThere it is 128 in perfectly elegant fashion. Every time you hit seven white lines, skip a leap year. Then start over with the seven black lines and voila.

Point 3

The beginning of this zero-based numbering calendar, denoted as 0.0.0.0.0.0 TC

We can’t even get all programming languages to start their arrays at 0. What makes you think it’ll be easier for non-programmers to accept this?

I have no idea what the programming languages and arrays have to do with anything (i.e. I’m very ignorant), however, the most compelling argument I’ve heard to get people (unsure whether any of them were programmers, or programming languages for that matter) is that we already do count this way, when we use digital clocks. Midnight doesn’t roll around to 12:00 a.m. or 24:00 a.m. – it rolls to 0:00.

I think that there are different numbering systems going on with our time system, and they aren’t all compatible. Such that it would take a simplifying means of organizing everything along similar lines. Good luck. Nevertheless many have tried. As have I.

We number the seconds, minutes, hours of the day by counting them. 1 o’clock a.m. means that 1 hour has passed since midnight. 1 o’clock = 1 hour

We number days of the month with ordinal numbers: 1st, 2nd, 3rd, 4th, etc. Ordinal numbers determine the order in which something occurs. This doesn’t count the days, it simply tells you which one follows which.

We number years (in a lot of calendars, not just the Gregorian) by sequential numbers 1, 2, 3, 4, etc.

So, any system, such as the TCC, theAbysmal, and others, which seeks to harmonize the numbering systems of existing time measure.

We measure clock time as: 00:00:00.0 Hour:minute:second.fractions of a second

We can measure calendar time the same way 00:00:00:00:00:00 Millennium:Century:Decade:Year:Month:Day

It may help to think of it as counting the seconds, days, years, whatever.

So if you see a date 14:05:13
it’s year 14, month 5, day 13 – that’s how much time has passed since the calendar start date.
14 years x 365 days + 5 months x 28 days + 13 days + 3 leap year days
= 14×365 + 5×28 + 13 + 3
= 5110 + 140 + 16
= 5266 days full days have passed since the calendar start date.

The current day is not yet finished, so it is not yet counted, but we count its Hours, minutes, and seconds until midnight.

Point 4

is on the solstice, exactly 10 days before the UNIX Epoch (effectively, December 22nd, 1969 00:00:00 UTC in the Gregorian Calendar).

The solstice is an instant; the date it occurs on depends entirely on your meridian/time zone (e.g. the Chinese calendar explicitly specifies Beijing time). So “exactly ten days” is a meaningless descriptor.

Besides, since you’re adopting a quadrennial intercalation cycle, that instant will drift back about six hours every year, further screwing up your “exactness.”

Last but not least: the solstice is a fundamentally difficult astronomical phenomena to measure. The instant it occurs is somewhere in the window where the sun’s north-south motion is too small to measure. Equinoxes have historically been measured with far greater precision.

All good points. Although theAbysmal calendar places its intercalary day on the date equivalent to December 21st, it is meant to fall on or about the southern solstice. It is not meant to be used as a point by which to establish the measure of the year. theAbysmal Calendar has no opinion one way or the other about that, and since the equinox seems to be working out so well,  why not continue to do so.

At any rate, I don’t think the above argument applies to theAbysmal Calendar as it seems to be sloppy writing on the part of the TCC’s creator.

Point 5

It’s “terran” inception and unit durations reflect the human biological clock

Then where the heck are your 28-day months coming from? The billions of people who live under a lunar or luni-solar calendar already know that the average synodic month is about 29.5 days, and that’s the “month” that affects tides and human fertility cycles.

theAbysmal puts the ends/beginnings of its quarters on or about the solstices and equinoxes, to align the calendar with the earth’s seasons. Granted, each region has its own particular climate and weather, however, regardless of whether it’s the high Arctic or the equatorial jungles, there are changes through the year. Temperature, weather, animal and human migration, etc…

Further, theAbysmal does have a lunar component that follows the lunar cycle as defined by NASA (in UTC). The lunar month (lunation) begins at the new moon.

Point 6

and align with astronomical cycles and epochs.

Really?

There is no integer number or integer ratio of days (mean solar or otherwise) in a tropical year
There is no integer number or integer ratio of days (mean solar or otherwise) in a synodic month
There is no integer number or integer ratio of months (synodic or otherwise) in a tropical year

Days, months and years have nothing to do with each other; there is nothing to “align” to.

Its “computational” notation, start date, and algorithm are tailored towards the mathematicians & scientists tasked with calendrical programming and precise time calculation.

Days, months and years aren’t SI units, and the one true SI unit of time has jack shit to do with any of them.

I think the OP here is talking out of his or her backside. The arguments are all valid, so what else is there to add?

The last line regarding days, months, and years not being SI units (only the second is) is particularly interesting. For the digital world of programmers and the languages in which they work base their time on seconds.

I’m sure one could define the measures of the year in terms of seconds – but I also believe that there are two sets of numbers one needs to look at – 1) the hard facts, 2) the elegant approximation. That’s one of the reasons that theAbysmal Calendar is divided into the 13-month calendar as well as the lunar calendar – the lunar months track the astronomical phenomena (new moon, eclipses, solstices, equinoxes, etc), whereas the 13-month calendar is rules-based and approximates (hence the equinox can be out from the end of the quarter by a day or two).

In the end, this makes me realize how little attention I’ve been giving to the second.

~  ~  ~  ~  ~  ~  ~  ~  ~

 

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: