Pax Calendar | Hacker News Reader

Pax Calendar

en.wikipedia.org

48 pointsrfreytag24 comments

At this point, I doubt there will ever be any proposal as easy to adopt and as radically-improved as Symmetry454 [1].

One day, I hope to live in a world where I get to use this calendar and the average response isn't for people to look at me funny…

For anyone looking for a tl;dr, Symmetry454…

- is perpetual; one printed calendar will work, more or less, indefinitely

- preserves the 7-day sabbatical week cycle (no intercalary days like IFC)

- simplifies the leap rule to a formula which should work for a very long time (at least a few millenia from now)

- preserves 3 months in a quarter, and 4 equal quarters making up a year

- results in every week, month, quarter, year, etc. beginning on a Monday and ending on a Sunday; this also means knowing just the day number tells you exactly what day-of-week it is

- ensures all months now consist of a predictable, whole-number of weeks

… it's just… so much better…

All the best,

[1]: http://individual.utoronto.ca/kalendis/symmetry.htm

I'm not a fan of having some months be 25% longer than most, but it is predictable. One thing to consider in calendar reform is monthly payments. They could easily be transferred to weekly payments in this system, or increased 25% for extended months.

What about this idea for a calendar: every year is 1 month, every month is 1 week, and every week is 1 day?

- Every year and month starts on a Monday

- No leap days needed at all

- You can use the same calendar every year (better yet, you might already have a blank piece of paper you can use)

- Every day is your birthday

You might be responding tongue-in-cheek, but you have essentially described an underspecified variant of FDN Rata Die [1].

FDN R.D. does have a place in calendrical calculations; namely, it serves as the narrow waist for converting between calendars.

All the best,

[1]: https://en.wikipedia.org/wiki/Rata_Die

This certainly isn't a great reason to avoid perennial calendars, but I honestly very much like that anniversaries fall on different days of the week each year. As a child a 'weekend' birthday gave me something to look forward to for multiple years to be able to party like Elton John or the Bay City Rollers. With a perennial calendar, if I were born on a dull workday, I'd be stuck celebrating in the middle of the school or work week for life! "No, cindy, I don't want to celebrate with the rest of the staff in the kitchenette with any more of your muffins."

My coworkers don't know my birthday lol.

Also due to infectious disease, celebrated my birthday with my family 3 months late recently.

I’ve always been fascinated by alternative calendars (also alternative timekeeping). However, barring an apocalyptic event, I don’t for see any alternative calendars taking off.

For most of human history, we didn’t need a standardized calendar, you just needed to agree with people that you interacted with. Religion pushed most of the standardization and later reinforced during the Industrial Revolution—helped along by colonialism.

Now that we are a globalized society, it would require parties from every nation and every industry to agree to the change and coordinate a simultaneous migration. Given we can’t get close to that level of cooperation when faced with potentially existential crises, it’s doubtful we could do it because it’s a “better” system than the current one.

I’d love to be proven wrong though!

Just imagine the state of the world in 1582, and the amount of cooperation, agreement, and coordination that it took to implement the Gregorian Calendar, which was an incremental iteration of the Julian one. Take the chaos and confusion created by that (and rippling into present times) and multiply it by ten or so, and then you may perhaps approach what it'd be like to change calendars today.

Pretty unpleasant to have an extra week once in a while, IMO. Seems this calendar would lead to confusing seasonal drift in more stable (pre-industrial) climate, but I guess there's no seasonal predictability going forward .

As a fun thought experiment, I once devised my own Equal Terrestrial Calendar, intended to be both predictable and actually secular.

- Weeks are 5 days long.

- Months are 6 weeks long (exactly).

- Thus, a month is 30 days, and as close as possible to an average lunar month.

- Every week/month/year begins on the same day of the week

- Work weeks are 3 days long.

- 3 intercalary holidays at the winter solstice, and 2 intercalary holidays at the summer solstice (make it 3 every leap year)

- 30 days/month * 12 months + 5.25 days (intercalary) = 365.25, but in a fairly even spread

Not possible for a calendar to be perfect, but I'd really enjoy living in this world.

What if instead we were to go with non-24-hour days? We could fix even more problems with such a system, like:

* Days starting at the same hour:minute all the time

* The current impossibility of a decimal clock. We're constantly told how using base-10 makes math so much easier, who wouldn't want a day being 10 hours long and a week 10 days long? If the answer is 168, you've screwed up somewhere

* The conceit that ancient deities like Apollo and Ra are so important that even after we've abandoned those religions, we must continue to allow our calendars revolve around their golden chariots. I say "no more!", we will spurn the astronomical bodies as well

* The relegation of agriculture to the primitive past, where it belongs. I only eat Hot Pockets anyway. The chicken milkers and tomato tree people can go live in yurts or something if they believe caveman calendars are important

* The ability to insert special 5-minute days into the calendar in emergencies, like when I have a new game on pre-order that is scheduled to come out in 60 days' time. We could easily reduce that to an effective 3 hours or so

Sounds fun. But the day is evenly divided already, so there's less pressure to change it.

> unpleasant to have an extra week once in a while

> 3 intercalary holidays at the winter solstice, and 2 intercalary holidays at the summer solstice

Isn't the latter an example of the former - ie. a 'extra few days' somewhere?

I'm curious about the winter solstice vs summer solstice there - it sounds like if you're in the 'wrong' hemisphere, these are inverted, yeah? Or do you mean people in Australia would have '3 intercalary holidays' (I guess you mean 3 days) around our summer solstice?

I'd also note that the solstice can be on any of 3 days, typically.

Time is tricky.

> Isn't the latter an example of the former - ie. a 'extra few days' somewhere?

Technically, yes, but that was a poor choice of words on my part. But "1 extra week every few years" is not the same as "3-day-holiday at the same time every year."

> I'm curious about the winter solstice vs summer solstice there

I was referring to Northern solstices for brevity. The intercalary days would happen at the same time, so folks in the Southern Hemisphere would get the longer-on-average set at their summer solstice.

> I'd also note that the solstice can be on any of 3 days, typically.

Yes, and a lunar month isn't a whole number of days, and a solar year isn't a whole number of days, either. We make do. Perhaps by setting aside a set of 3 days that the solstice could land on :)

Sure, the further problem there is the December solstice can occur[0] on any of 20, 21, 22, or 23rd of December, timezone of the observer notwithstanding.

Trying to be tricky with fantasy calendars will often catch you up.

You mentioned this was a secular calendar, but appear to have designated the two current / conventional Sabbaths (once again Islam gets the rough end of the stick).

Obviously religious people will oppose a shift from the 7 day week structure, as it will almost definitely break their brand of mysticism.

The design also locks in a 3-day work week, and 2 days where 'normal' people are seemingly not expected to work.

Right now we're tantalisingly close to reducing the conventional 5-day work week down to 4, so I'd be wary about trying to implement a calendaring system (or a social structure) that locked us into 3-day work weeks. It just burdens future-us yet another road block in several years when we realise working 60% of the time is as silly as the equally arbitrary 71%.

The nice thing about the four-day work week discussion is that it encourages people (though not everyone assumes or understands this feature) to think more fluidly in terms of days-of-the-week that you work, and days-of-the-week that you don't. Basically I think it'd be a massive shame to retard that aspect of the thought experiment all over again.

Anyway, cultural baggage is hard to shift when trying to come up with a Brand New Calendar, and it's enticing to try to reject arbitrary dates.

F.e. rather than try to follow some earth / sun configuration events, I'd probably just stick with the 1st of January as the western world's 1st day of the year - people's DOB and age will continue to track accurately that way. (Look at the mess for genealogists created by the ~1752 shuffle).

Other examples are assuming we need 12 months -- we very much don't -- and your proposed design effectively has 14 months anyway, just two of them are very short & variable length.

You've settled on solstice rather than equinox for these mini-breaks, which may indicate that one or both of the solstices is when your culture currently/commonly has a holiday?

Season commencements, of course, are not consistently observed around the world. In sane countries like Australia we consistently start summer on the 1st of December, so that summer spans the 3 hottest months, and winter on the 1st of June, etc. Our seasons kinda nearly center themselves around the season's equinox / solstice.

Similarly the 30 day month -- I can't work out any advantage to that, other than to mostly retain the 12-month calendar you're more familiar with.

I'd argue trying to have a calendar month length be near, but very much not the same as, the lunar cycle, is actually worse. With a half day length delta between the two, you're guaranteeing ~60 unique permutations of lunar : calendar, so not at all "predictable" for the casual citizen.

[0] https://en.wikipedia.org/wiki/Solstice#Solstice_determinatio...

Or we could go with a real classic from Ye Olde Days: The Babylonian calendar[1]. A visual explanation is here [2]. I rather like that they use the physical properties of a pendulum to link length and time.

1: https://en.wikipedia.org/wiki/Babylonian_calendar 2: https://upload.wikimedia.org/wikipedia/commons/7/71/Sumerian...

I don't think this reformed calendar goes far enough. Many of our services are billed monthly - Internet, phone, electricity, property tax, etc. I don't know if the pax week would be considered its own super-short month or fused into another month, but it massively complicates financial calculations and issues of fairness.

Other people have pointed out issues like with 13 months, you can't divide the year into quarters.

I don’t like leap weeks. I’d much rather have 1-2 leap days. Let them exist outside the months: year starts with 0 and every 4th year 00 then 0.

For anyone thinking that will be a pain to code, side benefit, it’ll hasten the adoption of sum types.

I recently ended up in a wikipedia rathole looking at various alternative calendar systems, at least in part to be able to give a true but unintelligible answer to the question of when my birthday is. Among the oddballs were the Discordian calendar which divides the year into five seasons of 73 days each with 5-day weeks and the Pataphysical calendar which has 13 months each of 29 days, but for most months, the 29th day is imaginary.

Will make things rather hard for financial businesses that expect 3 months per quarter and 4 quarters per year.

finance is even weirder. for example the US/360 bond calendar assumes 12 months of 30 days each. to make various calculations easier, I guess.

cf https://qntm.org/calendar

Why not have the 13th month be a 29 day month and every four years have a leap day?

Shire Calendar beats it.

Like other simplifying calendar proposals, Pax Calendar is faced with an almost impossible technical integration challenge if human beings have to achieve the transition, recoding all existing systems to conform to the new calendar requirements rather than the Julian/Gregorian calendar support we've built up over hundreds of years.

But now, with AI emerging, it might be possible to foresee a day, a half-century from now, when the transition to Pax or Synchrony or another such calendar is much less laborious and far more feasible, with a transition and switching cost that is actually palatable, given that AI and advanced robotics may largely be able to reprogram itself, and other systems AI can be given access to, rather than consuming a bajillion hours of human engineering and maintenance time.

Maybe in 75 or 100 years a transition will become more crystal clear as an achievement we can complete with the help of ubiquitous, universal, super-efficient, sudo-trusted ML.

But anyone hoping for a transition in the next 25 or 30 years is probably suffering from delusion.

Reply on news.ycombinator.com