I therefore venture to propose that at 2 a.m. on each of four Sunday mornings in April, standard time shall advance 20 minutes; and on each of four Sundays in September, shall recede 20 minutes, ...
This sounds a bit closer to the ideal of just working with respect to sun rise. It eliminates the abrupt transition twice a year and allows the adjustment to be greater or less than an hour which is good because the useful adjustment changes with latitude.
These days we could just skip the discrete changes all together and just use solar time directly.
Computationally it's not a problem, but other issues abound.
We want time zones to coordinate across large areas. A similar issue exists because the day has different length the further you are from the north pole. With southern summers and northern winters occurring at the same time.
If we can assume that clocks have decent computing power, and if we can assume that people that need to do time calculations across significant longitude differences have access to decent interactive computing power (smart phones, tablets, desktops), then I think some kind of dual time system might be workable.
Clocks would display both a "World Time" and a "Zone Time".
The World Time would probably simply be UTC.
For coordinating across significant longitude differences World Time would be used.
The Zone Time applies to a region similar to a current time zone, and is an idealized solar time, adjusted so that sunrise each day is around 6:00 AM LZT (Local Zone Time).
Schools, stores, most offices, mass transit routes that are mostly in the zone, theaters, meals, and similar things would base their schedules on LZT. So office hours, for example, would typically be 9:00 AM to 5:00 PM LZT, year around.
This would let most people work on a schedule that syncs well to the Sun, which should have tangible benefits for many people. There would be no going to work or school in the dark, and in summer the longer daylight time would automatically come after work/school.
This needs to assume people have decent access to interactive computing power because it would complicate planning things that involve interactions between LZT and World Time, especially when planning events weeks or months ahead.
(And yes, under this system 6:00 AM LZT from one day to 6:00 AM LZT the next day would not be exactly 24 hours. So? Things for which that matters would be scheduled on World Time, not on LZT).
It's an interesting proposal, but it seems like it would work best for more equatorial regions. In my not too northern location, that would mean in the summer I get up 5 hours earlier than in the winter.
The idea isn't that everyone gets up at sunrise. People in equatorial regions would be the last people to care about this stuff. DST is not an issue there. People in more polar regions tend to not care that much about daylight in the winter. There is so little of it and it is so cold that it makes no difference. There is so much light in the summer that it makes little difference when you get up. People in polar regions might actually like a sort of anti-DST offset so they can get some darkness in the evening in the summer.
It is people in the temperate latitudes that tend to care.
Our circadian rhythms sync to the day/night cycle, which is sensed by changes in light levels. Sunrise is a major driver of that, and so syncing our schedule to sunrise works out well biologically for us.
If the day is synced to solar noon, so that the time we need to wake up relative to sunrise changes throughout the year, then we may need to work to keep circadian rhythms synced to our schedule. For parts of the year where our wake up time is significantly after sunrise, we may need to block sunlight from our sleeping rooms. For parts of the year where we need to well before sunrise, we may need to have bright lights with the appropriate spectrum go off when it is time to wake up.
Comments
This sounds a bit closer to the ideal of just working with respect to sun rise. It eliminates the abrupt transition twice a year and allows the adjustment to be greater or less than an hour which is good because the useful adjustment changes with latitude.
These days we could just skip the discrete changes all together and just use solar time directly.
Or just shift to DST permanently. Nearly everyone is awake 4-7 so why not just have sun then and have mornings and nights sometimes sunny?
Computationally it's not a problem, but other issues abound.
We want time zones to coordinate across large areas. A similar issue exists because the day has different length the further you are from the north pole. With southern summers and northern winters occurring at the same time.
If we can assume that clocks have decent computing power, and if we can assume that people that need to do time calculations across significant longitude differences have access to decent interactive computing power (smart phones, tablets, desktops), then I think some kind of dual time system might be workable.
Clocks would display both a "World Time" and a "Zone Time".
The World Time would probably simply be UTC.
For coordinating across significant longitude differences World Time would be used.
The Zone Time applies to a region similar to a current time zone, and is an idealized solar time, adjusted so that sunrise each day is around 6:00 AM LZT (Local Zone Time).
Schools, stores, most offices, mass transit routes that are mostly in the zone, theaters, meals, and similar things would base their schedules on LZT. So office hours, for example, would typically be 9:00 AM to 5:00 PM LZT, year around.
This would let most people work on a schedule that syncs well to the Sun, which should have tangible benefits for many people. There would be no going to work or school in the dark, and in summer the longer daylight time would automatically come after work/school.
This needs to assume people have decent access to interactive computing power because it would complicate planning things that involve interactions between LZT and World Time, especially when planning events weeks or months ahead.
(And yes, under this system 6:00 AM LZT from one day to 6:00 AM LZT the next day would not be exactly 24 hours. So? Things for which that matters would be scheduled on World Time, not on LZT).
It's an interesting proposal, but it seems like it would work best for more equatorial regions. In my not too northern location, that would mean in the summer I get up 5 hours earlier than in the winter.
The idea isn't that everyone gets up at sunrise. People in equatorial regions would be the last people to care about this stuff. DST is not an issue there. People in more polar regions tend to not care that much about daylight in the winter. There is so little of it and it is so cold that it makes no difference. There is so much light in the summer that it makes little difference when you get up. People in polar regions might actually like a sort of anti-DST offset so they can get some darkness in the evening in the summer.
It is people in the temperate latitudes that tend to care.
Why anchor it to sunrise instead of solar noon?
Under the sunrise scheme, Chicago, for example, would have sunset at 3:08 PM LZT on the winter solstice and 9:14 PM LZT on the summer solstice.
Our circadian rhythms sync to the day/night cycle, which is sensed by changes in light levels. Sunrise is a major driver of that, and so syncing our schedule to sunrise works out well biologically for us.
If the day is synced to solar noon, so that the time we need to wake up relative to sunrise changes throughout the year, then we may need to work to keep circadian rhythms synced to our schedule. For parts of the year where our wake up time is significantly after sunrise, we may need to block sunlight from our sleeping rooms. For parts of the year where we need to well before sunrise, we may need to have bright lights with the appropriate spectrum go off when it is time to wake up.