# What day of the week is Jan 1st 1,000,000AD ?

Discussion in 'Computer Support' started by Trevor Smith, Nov 4, 2003.

1. ### Ralph MannGuest

Trent C said:
Oh bugger.
Back to the drawing board ;-)

I'm wondering if those sites that show the day as Friday instead of Saturday
have based their calculations on the first day being Sunday instead of Monday.
People do tend to think that Sunday is the first day of the week, as can be
shown on calendars.

Ralph Mann, Nov 5, 2003

2. ### Trent CGuest

Let's not give up completely. Look again at my calculations - we start with
1,000,000 years and subtract 2,000 because we're starting from 2,000 to give
us a base line (and, incidentally, so we're staying in the Gregorian
calendar) [998,000]

Then we multiply by 365 [364,270,000] and add a full complement - less
2,000 - of an additional 249,500 days - that's 250,000 less 500 for the
2,000 years we're skipping - to account for the 1 in 4 leap years
[364,020,500] and for god's sake make sure you're following this with your
own calculator!

Then we readjust for leap century years - only 1 in 4 century years is also
a leap year, so we must add back the 3 in 4 century years that weren't leap
years. This is where I'm struggling with fatigue - it's been a long day in
the office. Am I right in thinking that this calculation is
((998,000/100)x3), giving us 29,940? [364,050,440]

Then we just divide by 7 [52,007,205.71], which now suggests Wednesday.

Any thoughts?

Trent C, Nov 5, 2003

3. ### Ralph MannGuest

Trent C said:
You forgot to add 1 for the fact that 2000 is a leap year ;-)
Why are you deducting anything when you are already not including the first 1999
(not 2000) years in your count anyway?
No, you are skipping the first 1999 years.
You are using 1st Jan 2000 as a starting point.

Ralph Mann, Nov 5, 2003
4. ### rednoseGuest

Do you mean Monday December 31st 999,999. ?

SJ

rednose, Nov 5, 2003
5. ### Trent CGuest

I don't think we need to add one for the year 2000 being a leap year, as it
all balances out by the end of 999,999.
I'm only deducting the 500 from my calculation of the total number of
additional leap days in 1,000,000 years - 1,000,000 years, 1 leap year in 4
is 250,000 days, less 500 for the 2,000 years we're skipping in the
calculation.
No, my calculation is designed to get from January 1st 2000 to December
31st, 999,999, so I am counting from the beginning of 2,000 to the end of
999,999: 998,000 years.

Trent C, Nov 5, 2003
6. ### Ralph MannGuest

Trent C said:
Ofcourse you do, as it *is* a leap year, and *does* contain 366 days.
Don't you have to deduct days seeing as the centuries non-divisible by 400
*don't* have 366 days?

Ralph Mann, Nov 5, 2003
7. ### Trent CGuest

I take your point, Ralph, but surely this is dealt with at the end of the
998,000 years, as the last 3 years aren't leap years. If the starting point
was a leap year out of sequence, I'd agree with you.
Yes, we do that later on. I separated the processes for extra visibility.
By the way, I emailed the boffins at the National Physical Laboratory in
Rugby with this query. They run one of the atomic clocks and hopefully
it'll tickle their intellectual interest. I'll post back to this thread if
I get a response, but it would be great if we could get a resolution by
ourselves.

Trent C, Nov 5, 2003
8. ### Ralph MannGuest

Trent C said:
I have to concede on this point as the total number of years is wholly divisible
by 4. ;-)
Don't forget that you are going up to 999,999 and not 1,000,000, so you need to
drop 1.
That will be interesting, did you ask what weekday they start off with on 1st
Jan 0001 ?

Ralph Mann, Nov 5, 2003
9. ### Trent CGuest

Yes and no. I'm going from the beginning of 2000 to the end of 999,999 (see
my earlier comment below)
Don't even go there!

Trent C, Nov 5, 2003
10. ### Ralph MannGuest

Trent C said:

Why not?
You will need confirmation of whether Sunday or Monday is officially the first
day of the week to base the decimal calculation on.

Ralph Mann, Nov 5, 2003
11. ### Trent CGuest

'cos we're starting at January 1st, 2000, not 0001. That way we don't have
to account for the 18th century move to the Gregorian calendar.

Trent C, Nov 5, 2003
12. ### Bri.Guest

The year has commeced in January, since 1752 (UK) or 1582 (Europe),
depending on your location.

Bri., Nov 5, 2003
13. ### Ralph MannGuest

Bri. said:
So what do these sites use to calculate dates ranging from 5000+BC to 10
I doubt they use a mixture of calendars and just use one formula to calculate
what the date "would have been" based on todays calendar.
The same goes for what the date "might be" by 1million AD, as no-one knows what
system will/may replace the present system.

Ralph Mann, Nov 5, 2003
14. ### Trent CGuest

I agree completely. That's why I think it's best to start form a known
point, like 01/01/2000 - a Saturday.

Trent C, Nov 5, 2003
15. ### Ralph MannGuest

Trent C said:
Actually, I think it would be simpler to forget about the first 2000 years
anyway and start from 1st Jan 2001.
Most of them contain bad memories for me anyway ;-)

Ralph Mann, Nov 5, 2003
16. ### Trevor SmithGuest

I went here... http://www.geocities.com/eu84/frtop_files/calendar.htm
Thanks for the link. I would agree with Saturday for the Gregorian calendar,
because in 400 years there are 365*400+97 days or 146,097 days. As this
figure divides *exactly* by 7, it follows that as January 1st 2000 was a
Saturday, so will be January 1st 2400, January 1st 2800, January 1st 3200
.... , January 1st 1,000,000.

So much for the Gregorian calendar. I have seen the tropical year (period
for the repetition of the seasons) given as 365.242190 days, whereas the
Gregorian calendar approximates a year as 365.2425. The difference between
these two numbers, 0.00031 amounts to an error of around 310 days over the
next million years! In other words Jan 1st will be mid-summer (for the
northern hemisphere) around 500,000AD (Gregorian) and would be heading back
towards mid winter by 1,000,000AD! Making a lot of assumptions, this
probably would be prevented by new leap year rules, which would affect the
day of week for 01/01/1,000,000.

It gets worse. I have also seen that the Earth is slowing down by 0.005
seconds per year, per year. Over a million years -
0.005+0.010+0.015+0.020+...+4999.995+5000.000 seconds or about 79 years.
ASSUMING the rate of slowdown is constant. So now I'm not even certain what
YEAR it will be in 1,000,000 AD.

Trevor

Trevor Smith, Nov 5, 2003
17. ### Just TaylorGuest

I've just decided to use the ATM, it doesn't matter if the bank is open.

Just Taylor, Nov 5, 2003
18. ### EdGuest

.... and we thought we had troubles with Y2K.

(excellent thread guys, keep it up)

Ed, Nov 5, 2003
19. ### SteveGuest

Except that on 1st jan 1,000,000 AD all the ATM's will crash with the year
1M bug.

Steve, Nov 5, 2003
20. ### Ralph MannGuest

Trevor Smith said:
I don't know where you get that reasoning from, just because it is divisible by
400, doesn't mean it is divisible by 7.

Ralph Mann, Nov 5, 2003