PDA

View Full Version : Life Memorials, Holidays, and Math: A Poll For Your Thoughtful Reflection.


Rain Man
09-11-2010, 12:18 PM
Okay, we've got Christmas. And Thanksgiving. And the 4th of July.

Then we've got Labor Day, President's Day, New Year's Day, Memorial Day, President's Day, Columbus Day, and so on.

Presumably we'll have other great people and events in the future. President Jonas in 2056. The great liberator Timmy Lorfman in 2113. The long-overdue Britney Spears Day in 2140. The Victory over Mexico in the Great North American War of 2247. The Benevolent Alien Arrival in 2251. Victory in the Slime Alien Wars of 2310. And so on.

At some point, the math is obvious. If the United States survives long enough, every day will be a holiday. It's difficult to envision how we can have 365 days of holidays if we want to keep the fast food restaurants running and the hover trains operating and the alien slime removal facilities going. So what do you think should happen?

Should we:

A. Set a threshold that no more than XX percent of days will be holidays, and at some point if we want a new holiday we have to vote out an old holiday. For example, no more Labor Day if we're going to celebrate Microsoft Vistas Remembrance Day.

B. Consolidate holidays to celebrate more than one event on a day. For example, the last Thursday of November is now Thanksgiving and Exposed Nipple Appreciation Day.

C. Stop creating new holidays once we're full. For example, everyone might appreciate how Julius McDonald singlehandedly destroyed the Islamic Terrorist Movement and solved World Hunger through his brilliant orations, but we're full up with holidays, so he'll just get a mention on the news once a year.

D. Stop having holidays at all, since they're a capitalistic construct of major corporations.


The same applies with moments of remembrance, too. It might just take longer, but the math says that if we survive long enough, eventually every minute of every day will be a moment where we're supposed to have a moment of silence. With those, you've got another option, which is to have an expiration date where you stop having the moment of silence at some point. For example, when Sweden is hit by the meteor, Congress authorizes 25 years of moments of silence, but it then sunsets afterwards.

I look forward to hearing your opinions.

Psyko Tek
09-11-2010, 01:01 PM
12 holidays
one per month
and everybody gets the day off not just government and banks
consolidate within month as necessary

Groves
09-11-2010, 01:33 PM
12 holidays
one per month
and everybody gets the day off not just government and banks
consolidate within month as necessary

You work for a calendar company, don't you.

cdcox
09-11-2010, 01:58 PM
I was told there would be math in this thread? Looks like I am going to have to bring my own.

Here is my stystem:

We fix the big 4: New Year's Day, the 4th of July, Thanksgiving and Christmas.

Then, on January 1st we stochastically determine the number of additional holidays for that year. We set the mean at some predetermined level (say 8) and then give in a nice fat standard deviation (like 6). Any negative numbers drawn from the distribution would get rounded to zero, so you would still have 4 holidays that year. Most years you would get extra holidays but some years you would not. Some years you would get a whole bunch of extra holidays, like 20! It would make the whole year great!

Once you knew how many extra holidays you would have, we would hold a lottery to determine what dates they actually fell on. The lottery would be heavily weighted so that if there were only a few extra holidays, they would tend to fall on Memorial day, Labor day, and MLK day. So most years you would get those days off, but not every year. The lesser holidays, like Columbus day, ear wax day, and Arbor day, would be have fewer votes, but every so often, you would get those days off too.

Here is a possible outcome of this proposal, showing the total number of holidays each year:

2011 12
2012 16
2013 12
2014 12
2015 7
2016 4
2017 11
2018 14
2019 26
2020 10
2021 4
2022 22
2023 4
2024 8
2025 10
2026 11
2027 19
2028 19
2029 9
2030 4

T-post Tom
09-11-2010, 03:08 PM
Everyday is a holiday if you're Matt Cassel and under contract. (Boom. This thread is roasted.)

Rain Man
09-11-2010, 03:22 PM
I was told there would be math in this thread? Looks like I am going to have to bring my own.

Here is my system:



I really like this idea. Instead of, or in addition to, weighting the original holidays when picking dates, I'm wondering if we could add a formula for the days that would overweight days that are adjacent to already selected holidays, as well as days that are the most distant from any of the already selected holidays. That way, you'd be more likely to get a four-day weekend in February or October, and less likely to go six months without a holiday, while also not having a day off on Tuesday and then on Friday.

Calcountry
09-11-2010, 03:35 PM
I was told there would be math in this thread? Looks like I am going to have to bring my own.

Here is my stystem:

We fix the big 4: New Year's Day, the 4th of July, Thanksgiving and Christmas.

Then, on January 1st we stochastically determine the number of additional holidays for that year. We set the mean at some predetermined level (say 8) and then give in a nice fat standard deviation (like 6). Any negative numbers drawn from the distribution would get rounded to zero, so you would still have 4 holidays that year. Most years you would get extra holidays but some years you would not. Some years you would get a whole bunch of extra holidays, like 20! It would make the whole year great!

Once you knew how many extra holidays you would have, we would hold a lottery to determine what dates they actually fell on. The lottery would be heavily weighted so that if there were only a few extra holidays, they would tend to fall on Memorial day, Labor day, and MLK day. So most years you would get those days off, but not every year. The lesser holidays, like Columbus day, ear wax day, and Arbor day, would be have fewer votes, but every so often, you would get those days off too.

Here is a possible outcome of this proposal, showing the total number of holidays each year:

2011 12
2012 16
2013 12
2014 12
2015 7
2016 4
2017 11
2018 14
2019 26
2020 10
2021 4
2022 22
2023 4
2024 8
2025 10
2026 11
2027 19
2028 19
2029 9
2030 4What? No linear Algebra? What is the rank of the matrix?

seclark
09-11-2010, 04:34 PM
i'll keep what i've got, unless 2022 can cover every year.
sec

alnorth
09-11-2010, 04:42 PM
Here is a possible outcome of this proposal, showing the total number of holidays each year:

One weakness of this totally random system is that a period like 2021 through 2023 could spark mass rioting, wars, starvation, and eventually the end of our civilization.

Think about it. It is bad enough that you begin the year with an announcement that this year is going to be a crappy 4-holiday period. You trudge through the year angry at not being able to go to Vegas on a whim during Arbor day like last year, but then... a light at the end of the tunnel. They announce 22 holidays next year! Twenty-Two holidays. My God, think of the possibilities, the frivolity, the joy of being paid almost 10% of the time to not work. But then disaster strikes with the announcement of another 4-holiday year. What? No! I mean, I could handle going back to something reasonable like 15, but 4? Two years ago sucked so bad! I will not do it! Rabble! Rabble Rabble Rabble!

I think instead to ease people into and out of lean and rich years, the mean should be whatever the number was last year, and have a random walk of say up to +/- 3 or 4 holidays, maybe a maximum of 25 and a minimum of 4.

edit: here's an example of that random walk

2011: 12
2012: 10
2013: 10
2014: 14
2015: 11
2016: 9
2017: 11
2018: 13
2019: 15
2020: 18
2021: 14
2022: 16
2023: 15
2024: 16
2025: 16
2026: 20
2027: 24
2028: 23
2029: 22
2030: 21
2031: 19
2032: 18
2033: 15
2034: 11
2035: 7
2036: 9
2037: 8
2038: 7
2039: 4
2040: 4

Rain Man
09-11-2010, 05:05 PM
I think instead to ease people into and out of lean and rich years, the mean should be whatever the number was last year, and have a random walk of say up to +/- 3 or 4 holidays, maybe a maximum of 25 and a minimum of 4.



But if you did that, you'd end up with longer cycles, whether good or bad. In good cycles, you'd end up with lower productivity and depressions, which would lead to unemployment that would make more days off useless to large proportions of the population. In bad cycles, you'd have people getting worked too hard for too long, leading to lower productivity. I think you need the opposite, perhaps some sort of mathematical forcing toward the mean with occasional outliers. For example, you might have a correction factor on the X+1 year's calculation that is the difference between the past three years and the expected long-term average.

alnorth
09-11-2010, 05:51 PM
But if you did that, you'd end up with longer cycles, whether good or bad. In good cycles, you'd end up with lower productivity and depressions, which would lead to unemployment that would make more days off useless to large proportions of the population. In bad cycles, you'd have people getting worked too hard for too long, leading to lower productivity. I think you need the opposite, perhaps some sort of mathematical forcing toward the mean with occasional outliers. For example, you might have a correction factor on the X+1 year's calculation that is the difference between the past three years and the expected long-term average.

hey, thats a cool idea. I revised the formula to add the suggested forcing back to 12. Beginning in 2014, the average of the last 3 years is subtracted from 12 and added to the result. There's lots of annual variation, but we very rarely hit the edges now. The random walk is preserved near the mean, but when you drift too far you swing back.

In this example, we have 3 rich years 2022-2024 so the forcing formula applied a -4 holiday penalty for 2025. (and we are unlucky enough to roll a -2, so we fall from 14 holidays to 8). Then in the following 3 poor years 2025-2027 where we actually hit the minimum at one point, the forcing formula tacks on an extra 6 holidays for 2028, plus we are lucky enough to roll a +3, which is why we swing up from 6 to 15 for 2028.

2011: 12
2012: 13
2013: 15
2014: 13
2015: 13
2016: 16
2017: 10
2018: 7
2019: 5
2020: 12
2021: 17
2022: 18
2023: 17
2024: 14
2025: 8
2026: 4
2027: 6
2028: 15
2029: 16
2030: 15
2031: 10
2032: 10
2033: 14
2034: 17
2035: 14
2036: 12
2037: 14
2038: 16
2039: 12
2040: 6