Deciyear Calendar | Hacker News Reader

Deciyear Calendar

deciyear.com

36 pointsjoaomsa57 comments

My favorite part is that they designed a new calendar system where the months are named starting at one (Primilis, Secundilis, etc.) but numbered starting at 0. So Primilis is month 0, Secundilis is month 1, and Decilis is month 9. This somehow feels even worse than our current system where there’s a shift of 2 in the other direction for September to December.

They also claim that this is useful when dealing with quarters despite their number of month not being divisible by 4. (Admittedly, it works on the level of weeks.)

Why even name the months?

For conversation, "last October" is easier than "last month 10"

No it isn’t. If they were named Blorple and Frotz it would be harder than 10. Only because you’ve had your lifetime to learn the order of the Gregorian months is it in any way easier.

Day:Month::Number:Word makes for an easier pattern recognition. It’s the same DD/MM ambiguity problem, but using DD/OCT solves it quite well in conversation and otherwise.

Only because you’ve been trained your life to know the order. Plus the words don’t even sort.

They don't sort... in your language, or how you call them. So what?

Having dates (month names) that sort is great if you are a programmer.

Otherwise, so what indeed.

Funny you say that since October started off as meaning the 8th month.

october is still a month "name", like decilis, not just "the 8th month"

It is a name now, but originates from an adjectival form of the latin number eight "octo"

Japanese gets away with doing that. So does Chinese - They even give the weekdays numbers. People will adapt and start to think of it as nouns, if you introduce it wide enough.

When it comes to the number of months in the year, 10 is an entirely arbitrary number because, unlike the metric system, there's no opportunity to leverage the convenient zero-shifting property of a base-10 numeral system (the concept of "kilomonths" and "megamonths" is incoherent as a unit of measure given that the number of days in a year is not divisible by 10). Might as well just smooth out the current 12 months (12 being a more highly composite number than 10) to 30 days each and then save the leftover days for a worldwide party outside of time.

you're right of course, but man I wish we had megamonths, the name is great.

you're describing the French Republican calendar btw. I blame Christianity for the travesty of stopping it's adoption. https://en.wikipedia.org/wiki/French_Republican_calendar?wpr...

Ya, those pesky Christians. If only Robespierre wasn't so reluctant to use the guillotine. Too few Vendees I tell you!

I don't completely agree; when using a year as base unit, it makes somewhat sense both directions. Deka- and hectoyears are equal to common decades and centuries, and megayears would be relevant for geology and astronomy etc. In other direction deci- and centiyears could be sensible, equaling a decimal "month" and "week" respectively.

Playing devils advocate here, the problem with this proposal is that it didn't go far enough with its decimalization. A "better" (more decimal) approach would be to divide year to 100 "weeks" (10 "months" with 10 "weeks" each), and have each "week" vary in number of days. This would open up to fun ways to notate it, for example now (2023-12-17T21:03:31Z) could be "2023.95/2.87744Z", i.e. "week" 95 (or month 9 week 5) of 2023, day 2 of the "week", and 877.44 millidays ("minutes") since midnight. It's only slight abuse of decimal notation :)

All the problems listed aren't real world problems. No one has a problem understanding the calendar and there are no engineering problems with it either unlike the inch. For the few folks in astronomy we have computers.

Why would there be a problem? Half the world's population prays according to a base 7 week.

Nor is this proposal new. Its been tried (at least) twice before failed spectacularly embittering people. Based 10 (french) or 5 (soviet) weeks were introduced at the blood thirsty peak of thise anti-religious regime with the explicit goal of also erasing worship to the God of Abraham.

About understanding, we're used to it but you can't trivially know whether 02/05/2026 is a Monday. Even remembering whether July has 30 or 31 days can be a chore. Irregular months also can have some small annoying effects since "in a month" is not always the same amount of days. There's even the 360 day calendar for that https://en.wikipedia.org/wiki/360-day_calendar

Just one more in the long history of https://en.wikipedia.org/wiki/Calendar_reform#Proposals

Actually there are trivially easy ways to calculate DOW off the top of your head. No one bothers because it was never a real problem. People used to carry their calendar books on them to keep track of their engagements. Now we carry a smartphone.

Nevertheless, my colleague bothered to do it and he could tell you the DOW without pausing. He is of average intelligence maybe a little more intelligent than I (though I was better at mental math)

Actually there are trivially easy ways to calculate DOW off the top of your head.

Then what are they?

Year 2, month 5, day 2026?

I guess we could solve the "is it DD/MM/YY or MM/DD/YY?" problem while we're at it, but ISO8601 already did.

You appreciate that no calendar can get around the mod 365 and leap year adjustments? That's baked into the Earth orbit around the Sun

This is denial. There are absolutely problematic side-effects with the existing calendar system, and the obvious use-cases that jumped out to me are on the business/financial side. Yes, the changing length of months makes financial planning slightly more difficult. The arbitrary nature of how many non-business days occurs in a month cause issues (some months can have many more 'business' days than other months, skewing financial data for those months and the months surrounding them). Don't get me started on the fact that different years have, in the US, different lengths of time between Thanksgiving and Christmas, skewing consumer behavior.

I'm not saying this is a good enough reason to even begin trying to upend the existing calendar system and replace it. But the somewhat random nature of how weekends and certain holidays are distributed throughout the calendar can have effects on business, and presumably other areas. One might suggest it is all awash because the bad effects of one year/week/month will erase the good effects of a future one, but I don't find that comforting if it contributes to, say, the company I work for having an under-performing year, resulting in layoffs (granted, there would be other factors here as well). Regardless, the inconsistency is a problem.

I recall working at a FinTech company and hearing some team gloat about something like an 8% increase in MoM revenue, only to be shut down by someone in finance pointing out that there were 8% more bank days that month. Imagine someone wasn't thinking of this and people got promoted (or fired) because of a total nothing-signal being misinterpreted as a hugely positive (or negative) performance. The existing calendar allows this.

> But the somewhat random nature of how weekends and certain holidays are distributed throughout the calendar can have effects on business, and presumably other areas.

Yes, and? Isn't it to be expected? Calendars are long-living standards, that adjust over long periods of time, because of the practical impact they have on societies.

Why should those commons adjust themselves in order to satisfy self-imposed, often antisocial business requirements? (how MoM revenue performance is calculated and rewarded is strictly a concern for said businesses).

If it should contribute to an underperforming year for your company, it would be a way lesser factor than the lack of preparation or perspective from its owners/executives, because the calendar-related discrepancies are already known in advance (and it's not a particularly recent phenomenon either).

Would even such a deciyear (or any other new) calendar be adopted, it would "diverge" again in the next 10/20 years by the simple sheer force of society cultural evolution: "we will have more bank days here", "we need more working days there", "ho, we need to mark this particular date because of this event", "ho, these partner countries follow a new tradition, we need to adjust somehow".

Ok. Perhaps a new calendaring system could make sense for businesses. The only way forward would be for businesses first to design and adopt it, in parallel with the public calendar system, and show how it works better. Would there be a favorable business case? Especially in relation with financial and regulatory requirements which are still based on civil years? (and even different across countries)

The calendar proposed by the article has the same problem, see the part about the yearend period and leap years. Yes, the issue occurs less often, but the question "what date is one month from today" still doesn't have a simple answer. Do we want 36 days from today? Today's date next month? What if today is the 2nd of Decilis, are we talking about the 2nd day of yearend or the 2nd day of Primilis?

These questions are a pain in the ass with the Gregorian calendar and they would be a pain in the ass with this one, and basically any conceivable calendar. No calendar can change the fact that there isn't an integer number of days in a year, and that the closest integer (365) doesn't have nice factors for making a calendar out of.

Earth has 4 corner simultaneous 4-day Time Cube within single rotation.

I only use calendars that acknowledge the 96-hour Cubic Day.

https://timecube.2enp.com/

LMFAO thanks that was hilarious

Six weeks of six days (4 work days and 2 rest days) in a month would be better than four weeks of nine days (6 work days and 3 rest days) in my opinion.

Proposals like thus are so unrealistic that it's difficult to take then seriously. This one has lots of problems, but perhaps the worst (and yet, easiest to fix) are the millidays. People do not naturally think in thousands. Put a separator in there.

More fundamentally, the "second" is embedded everywhere in culture, science and engineering. Trying to replace the second with microdays is...let's be polite...suboptimal.

I’m somewhat partial towards the centiday (roughly 15 minutes, 14m 24s to be precise).

> People do not naturally think in thousands

A lot of us don't have trouble understanding 1,000 millimeters equals 1 meter.

Same here: 1,000 millidays equals 1 day.

This is worse than my 365 month calendar since it still has that ugly 5-day “year end”. In my calendar, every month is exactly one day long, no exceptions. Leap years simply add an extra month.

I think the 5 day "year end" is actually fine. So many people take lots of holidays around the end of the year so that period is kinda messed up anyways. It seems that it would be nice to just give everyone 5 days off and sort of unify it.

Since unfortunately 365 isn't a round number it seems like the best compromise. The other top approach would be counting days.

My mind blown moment back in middle school was the example of a ballerina spinning in the back of a flat bed truck going round and round on a race track.

It is obvious to anyone here I guess but to my child mind, this was a revelation. The fact is the speed at which the ballerina spins in her own dance has nothing to do with the speed at which the truck goes around the track means. Similarly, Earth's day of 24 hours and 365 days (I guess 365.25 something) a year are unrelated.

Back to the topic, I think it is useless to try to square the circle of a scientific time measurement goal while trying to guarantee that a day has 24 hours while also trying to guarantee a year has integer number of days.

We can try to say that we have defined a meter as something or a kilogram as something and pretend they are not arbitrary but how can any self-respecting physicist / astronomer pretend that our concept of time is anything but arbitrary?

If we really want to, we could shift the Earth almost exactly 7000 km closer to the Sun to permanently destroy February the 29th.

Calendar reformations are fun perennial thought experiment topic, even if at this point it seems very unlikely that we'd change away from Gregorian. That being said, this seems particularly weak proposal. If we were to accept this "yearend" thing (which I think is unpalatable), then 12 months of 5 weeks with 6 days each (i.e. 30 day months) would be massive improvement over this 10 × 4 × 9 structure. You'd get even quarters (3 months vs 2.5 months), more palatable work-week (4 days vs 6 days), and best of all the yearend could be a full 6 day week which keeps the year and week cycles in sync. You could simply drop the yearend completely roughly every 8th year to keep in sync with tropical year ((7×366 + 360)/8 = 365.25).

Not a single mention of the topic/reason calendars were invented in the first place: keeping track of the 4 seasons (for agriculture purposes). Though since each month has 4 weeks, I can see a season being 2.5 months making sense. At least much more than the 13-month year.

There are not 4 seasons near the equator; we use lunar calendars in Islam and Buddhism both in Indonesia

Does it make use of the equinoxes and solstices or is it strictly lunar-based?

I like the variety and history reminders of the Gregorian. Why do we have to make a homogenous suburb out of everything?

Efficiency. Its always efficiency.

I know, right? I'd rather eat a good meal than take a pill or drink soylent.

6-day work week?

Through sheer fatigue, you'll get nothing done on the 5th day (just like now), and certainly not the 6th.

Maybe it would be better to do something like 3 on 1 off 3 on 2 off.

Seems like a combination of the Kodak Calendar (1) and Swatch Internet Time (2):

1. https://en.wikipedia.org/wiki/International_Fixed_Calendar

2. https://en.wikipedia.org/wiki/Swatch_Internet_Time

Didn't the French try this during the revolutions (1792?) like Vendémiaire, Brumaire, Frimaire, etc.

Found it : https://en.wikipedia.org/wiki/French_Republican_calendar

Calendars are a prime example of Metcalf's Law. (The influence of a network is the square of the number of elements participating.)

https://hankehenryontime.com is another alternative calendar, but doesn't attempt to also "decimalize" time units shorter than the day.

What benefit does this have besides just being a thought experiment?

Nobody would adopt this, so why bother?

Wrong approach. We should change our numbering system to base-12 instead.

You’re a Dozenalist. There are dozens of us!

What makes it especially appealing to me is that we don’t need to invent new names for the digits A and B (though I’d like to come up with better glyphs than reusing those letters).

Ten and Eleven work just fine, and can be used for larger numbers like “forty-ten” (₁₀58) and “eleventy-four” (₁₀136)

Another bonus is that when we transition to this superior numbering system, I can sneer at the metric world, what with their archaic finger-based measuring system. The Foot will return to a place of honor.

A more sensible proposal than changing the calendar.

But consider base 12 has few advantages over base 6 but base6 greatly simplifies math since you only need to remember 24 entries for the commutative pair tables (multiplication)

This is a proposal to change the clock, change the calendar, and abolish timezones in one fell swoop. This would be harder to introduce than introducing each of these three things separately. But I think it is satire.

You should see a real hoot, decimal time: https://en.wikipedia.org/wiki/French_Republican_calendar

What real problem does it solve?

Obligatory link to XKCD's "Standards" https://xkcd.com/927/

Obligatory xkcd 320 "28-hour day"

https://xkcd.com/320/

And xkcd 1061 "EST"

https://xkcd.com/1061/

Reply on news.ycombinator.com