Leap Days and Stuff
Posted: Mon Jul 22, 2013 10:30 am
This is going to be more of a thinking out loud exercise for me, to try to see where I can get to, and what I think I know about "Leap Days and Stuff". Hope one or two of you guys will join in for the ride and set me right in the places I hit a duff note.
The world goes around the sun and takes approximately (more of that word later) 365 and a quater days to complete one revolution. It's tilted on it's axis in relation to the plane of it's passage around the sun (by about 23 degrees?) and this tilting and how it effects how the rays of sunlight fall on the earth, brings about our seasons. Now, because of this quater day, if we didn't make adjustments to our calander every now and again the seasons would soon become out of synch with the given time of year that we are used to seeing them. If I'm right, the result of not having this adjustment would be that in each sucessive year the seasons would lag behind the dates a little more and a little more. If your birthday was in dummer [typo I just had to leave in] you would gradually see it shift into spring and the winter. This would cause untold problems in the Christmas card industry and so cannot be tolerated.
Ok, why not take the bull by the horns and extend the length of a day so that that 1/4 day is spread out over the 365 days of the year. Time measurement is after all as arbitrary as any other human imposed measurement system on nature. Why not do away with the need for all this leap year stuff by making a new standard length of time for a second, so that what is now 24 hours becomes just short of 24 hours; in that way the quater day could be divy'd up between all the current days of the year and it would be 'job done'. But hang on - if you did that would you not just shift the problem from the years, over to the days. The day is measured by a complete revolution of the earth on it's axis - it's not an arbitrary thing at all. If you add more time into it then surely you are going to finish up with the times of day wandering around in respect to day and night. By adding more time into each day you are going to finish up with each day starting a bit further into the next - soon the new day will start at dawn, then at the suns highest point for the day, then in the evening. It would be chaos - it just wouldn't do.
Ok - so we're stuck with adding days in. Now as the earths revolution around the sun takes approximately365 and a quater days then if we add one extra day every four years this puts it right(ish). But heres where the 'approximately' starts to count. It isn't exactly one quater day - it's approximately and over the years all these approximatlies start to add up. So how do you deal with them. In fact the error is such that over time the seasonal shift is over compensated for by the addition of one day every four years, and this starts to stack up in the opposit direction ie the seasons start to gain and your summer birthday moves into autumn and then winter. This is dealt with by ignoring the leap year rule (that a day is added into each year number that is divisanle by 4) on one occasion each century - and that is in the first [ie the '00'] year. However again this tips the balance in the other direction (ie back into a seasonal lag - allbeit over a much more extended time) and so for the last trick [I'm aware off anyway]. At the turn of a new millenium you ignore the ignoreing the leap year rule at the start of new centuries, and stick another day in anyway. And this I think puts our calander into pretty good shape, keeping the seasons in place wise, for the next ten or twenty thousand years (might even be more than this). There may be yet a further rule taking us even deeper into the future, but if so I'm not aware of this.
Couple of questions i) What is the point in space by which we decide upon a revolution of the sun having been completed. I don't know if suns revolve or not (I'd guess they do) but if so how do you judge when a revolution has been completed - there must be a point, say the crossing of the earths eqitorial plane with the suns or something, by which you can fix the relative positions of sun and earth.
ii) Every day I bundle up the newspapers from the day before and send them back to the depot. I date the returns slip and I'm always amazed how often there seems to be an symetry or a sequence in the date numbers. Here's a few:- 10/11/12, 09/11/13, 11/12/13 [poor examples I'm afraid - some of them are quite beautiful but alas I can't remember them and never write them down]. I'd guess that as the year number rises the occurence would be less, but not necessarily so. As I say it astounds me how often this happens - and I've often wondered if it would be a predictable thing by algorithm, or whether it is just too random an occurence to lend itself to prediction in this manner.
The world goes around the sun and takes approximately (more of that word later) 365 and a quater days to complete one revolution. It's tilted on it's axis in relation to the plane of it's passage around the sun (by about 23 degrees?) and this tilting and how it effects how the rays of sunlight fall on the earth, brings about our seasons. Now, because of this quater day, if we didn't make adjustments to our calander every now and again the seasons would soon become out of synch with the given time of year that we are used to seeing them. If I'm right, the result of not having this adjustment would be that in each sucessive year the seasons would lag behind the dates a little more and a little more. If your birthday was in dummer [typo I just had to leave in] you would gradually see it shift into spring and the winter. This would cause untold problems in the Christmas card industry and so cannot be tolerated.
Ok, why not take the bull by the horns and extend the length of a day so that that 1/4 day is spread out over the 365 days of the year. Time measurement is after all as arbitrary as any other human imposed measurement system on nature. Why not do away with the need for all this leap year stuff by making a new standard length of time for a second, so that what is now 24 hours becomes just short of 24 hours; in that way the quater day could be divy'd up between all the current days of the year and it would be 'job done'. But hang on - if you did that would you not just shift the problem from the years, over to the days. The day is measured by a complete revolution of the earth on it's axis - it's not an arbitrary thing at all. If you add more time into it then surely you are going to finish up with the times of day wandering around in respect to day and night. By adding more time into each day you are going to finish up with each day starting a bit further into the next - soon the new day will start at dawn, then at the suns highest point for the day, then in the evening. It would be chaos - it just wouldn't do.
Ok - so we're stuck with adding days in. Now as the earths revolution around the sun takes approximately365 and a quater days then if we add one extra day every four years this puts it right(ish). But heres where the 'approximately' starts to count. It isn't exactly one quater day - it's approximately and over the years all these approximatlies start to add up. So how do you deal with them. In fact the error is such that over time the seasonal shift is over compensated for by the addition of one day every four years, and this starts to stack up in the opposit direction ie the seasons start to gain and your summer birthday moves into autumn and then winter. This is dealt with by ignoring the leap year rule (that a day is added into each year number that is divisanle by 4) on one occasion each century - and that is in the first [ie the '00'] year. However again this tips the balance in the other direction (ie back into a seasonal lag - allbeit over a much more extended time) and so for the last trick [I'm aware off anyway]. At the turn of a new millenium you ignore the ignoreing the leap year rule at the start of new centuries, and stick another day in anyway. And this I think puts our calander into pretty good shape, keeping the seasons in place wise, for the next ten or twenty thousand years (might even be more than this). There may be yet a further rule taking us even deeper into the future, but if so I'm not aware of this.
Couple of questions i) What is the point in space by which we decide upon a revolution of the sun having been completed. I don't know if suns revolve or not (I'd guess they do) but if so how do you judge when a revolution has been completed - there must be a point, say the crossing of the earths eqitorial plane with the suns or something, by which you can fix the relative positions of sun and earth.
ii) Every day I bundle up the newspapers from the day before and send them back to the depot. I date the returns slip and I'm always amazed how often there seems to be an symetry or a sequence in the date numbers. Here's a few:- 10/11/12, 09/11/13, 11/12/13 [poor examples I'm afraid - some of them are quite beautiful but alas I can't remember them and never write them down]. I'd guess that as the year number rises the occurence would be less, but not necessarily so. As I say it astounds me how often this happens - and I've often wondered if it would be a predictable thing by algorithm, or whether it is just too random an occurence to lend itself to prediction in this manner.
!
}