7 ms·
One of the main purported benefits of DST is that it in effect moves some of the extra sunlight from the longer days of summer from the morning to the evening,
by tzs 9y ago
One of the main purported benefits of DST is that it in effect moves some of the extra sunlight from the longer days of summer from the morning to the evening, where most of us are in a better position to use it.
That's because without DST as the day lengthens the extra daylight is split evenly between morning and evening.
But WHY does it split evenly? It's because we historically use noon as our reference. We think of noon on day N+1 as being exactly one day after noon on day N. Noon is a sensible reference, fairly easy to determine with ancient technology.
Suppose that instead we had historically used sunrise as our reference. By definition, then, sunrise would be 8 AM every day. As the days lengthened in the summer, all of the extra daylight would then have shown up in the evening automatically.
It's interesting to ponder whether or not such a reference could have worked. I think that in the days before clocks, when the sundial was the most sophisticated timepiece, this could have worked quite well. It would also fit in well with work schedules, meaning people who needed to wait for light to work would be going to work at about the same time every day. It would also fit in with animal sleep schedules.
With clocks, though, a noon reference is easier. A sunrise reference requires a much more sophisticated clock, or it requires frequent manual adjustment.
Nowadays, though, clocks easily have the sophistication to handle that. We probably could actually reasonably handle using sunrise as our reference, and have our clocks adjust daily to keep that so.
It would mean that a given clock time on day N+1 is no longer exactly 24 hours later than that same clock time on day N, but I don't think that would matter for ordinary life. I'd be willing to put up with that if it means I get the benefits of DST (more evening light in summer) without the two big discontinuities a year.
Edit: to be clear, I am assuming that if we did a sunrise-based system instead of a noon-based system we would still do time zones. 8 AM would be simultaneous everywhere in the zone.
[1] In this comment "day" refers to sunrise to sunset.
- Retric 9y agoCalibrating a sundial to noon is a one time effort. Calibrating a sun dial to dawn requires an update every single day. Further, dawn depends on local geographic conditions making it an ambiguous term.
- tzs 9y agoI'd expect that a sunrise-based system would use the calculated time that sunrise would occur for an observer at the middle of the time zone on perfectly spherical planet. Well, if we did one today, anyway. If it had done it in ancient times it would probably be based on sunrise in the capital city as observed each day by some designated religious office such as a high priest at the main temple of the dominant religion.
- dredmorbius 9y agoLargely true, though at fine enough resolution, noon shifts a bit as well, for various reasons (I believe it's precession, inclination, etc.). https://physics.stackexchange.com/questions/38270/why-is-the-solar-noon-time-different-every-day https://physics.stackexchange.com/questions/38270/why-is-the...
- Mz 9y agoThere is one flaw with your theory here: Historically, noon was defined locally by the sun being directly overhead. All time was local time. Then we invented trains and railroads and you couldn't schedule anything with such a system. Thus was born standardized time zones where it is the same time across large swaths of land. Noon is no longer strictly correlated to the sun being directly overhead. I think there are all kinds of things we could do, including promoting the gig economy. If I work when, where and as much as I choose, it makes little difference what time it is (as long as I meet deadlines). The world population is both high enough and mobile enough that you can now find 24 hour establishments in a lot of places, even relatively small towns. The small town I am in has enough 24 hour places that I have quite a lot of latitude on when I shop, when I get lunch, etc. I don't care all that much when the sun rises or sets. I care much more about things like if it is raining, and that is because I live without a car, so I walk everywhere.
- Swizec 9y ago> (as long as I meet deadlines). But how do you meet deadlines if you can't agree on what time it is?
- Mz 9y agoMy remarks in no way suggest we do away with standardized time zones. So, I don't have any idea how you are making that leap.
- m0llusk 9y agoMy experience with the gig economy is that my schedule ends up being whatever is convenient for my clients.
- Mz 9y agoThat hasn't been my experience. I do what I can to try to promote the idea of gig work done well such that it is a net positive for the worker in metrics like quality of life and control over their schedule. Granted, some things are inherently more flexible than others. An Uber driver can choose what hours to work, but is at the beck and call of people who need rides. If you want to make good money, you realistically need to work during busy times. But a lot of knowledge work can realistically be done whenever works best for the employee, so long as they get it done in time for the due date. You do need to arrange for such expectations though.
- pointytrees 9y agoWhat you have written here is what I came to logically myself as well. However, it would then be more difficult for someone in Sacramento to ask someone in San Francisco what time the train leaves. Plus, all sorts of similar issues. Mental math of "train arrives at 3pm, takes me 30 minutes to walk there, fifteen minutes to pack. Gotta be sure to be out of my meeting by 2:15" but if you live 30 sunrise minutes west, you and your coworker would have different 2:15 times. You are 45 sun minutes west, but you go out to a job site with coworkers which is another 15 sun minutes west. You think you're able to leave on time to get home for your daughter's birthday, but forgot your clock that is calibrated to your current gps location, is going to make you 15 minutes late when you get home. Good luck.
- comicjk 9y agoYou could still have sunrise-based time zones the same way we have noon-based timezones now.
- grzm 9y agoWould you elaborate on how this would work? I feel like I'm missing something obvious. The sun is roughly in the same position in the sky at noon each day—at least it's going to be midday for any human approximation of it. Sunrise (and sunset) vary throughout the year, and increasingly so with latitude along the same longitude. If the idea is to be sunrise-based, sunrise isn't going to be at roughly the same time along longitude lines, while midday sun is at roughly noon.
- photojosh 9y agoYour sunrise wouldn't be exactly 6AM (or 7AM or whatever), it would be between 6-7 depending on where you are in the time zone.
- grzm 9y agoSure. We have that now. But it also wouldn't be the same along the same longitude. Yes, 6AM would be the same, but now sunrise, midday, and sunset would all vary. At least with noon-based time zones, midday occurs at the same time along longitudes. People have an expectation that noon occurs midday, which is why the system is set up the way it is. This is how it was before we had time zones. It's a human-interface thing, more than anything else. Am I missing something? Sure, we could have all sorts of different schemes. Most of them are not going to match human intuitions.
- thaumasiotes 9y ago> Suppose that instead we had historically used sunrise as our reference. By definition, then, sunrise would be 8 AM every day. Where'd you get this? Reference sunrise is 6 AM, halfway between midnight (the middle of the night) and noon (the middle of the day).
- kobeya 9y ago> Suppose that instead we had historically used sunrise as our reference. No need to suppose. That’s how many cultures have marked the day. That’s why Jewish sabbath is sunset to sunset. However your plan does suffer one drawback—time zones would need to change with latitude too. And what about the polar regions?
- romanovtexas 9y agoThis is a good suggestion, however, it might not be useful globally. People in the tropics/around the equator get so much sunlight that they would not want to trade the convenience of our current time systems with having to adjust it based on the sun just for that extra daylight.
- bryanlarsen 9y agoPeople on the equator get no more sun time then anybody else. It's just that theirs is evenly distributed, they get ~12 hours of sun every single day of the year. They'd be completely indifferent, their current clocks would still work, they'd just be annoyed at the other time zones moving around on them.
- romanovtexas 9y agoFair enough, although my statement was more about the amount of sunlight they get per day, for all days of the year versus someone closer to the poles having to adjust to constant fluctuations. You're right when you say that they get ~12 hours every single day, but their clocks would still have to account for the slight changes (by a few minutes) in time for the sunrise. Which means changing their clocks everyday/every few days, based on the sunrise. But for what? They anyway get ~12 hours sunlight, so it's not a worthy tradeoff for them, compared to someone in Massachusetts. If only humans worked like batteries, soaking up the summer sun to use it in winters..
- bryanlarsen 9y agoWow, this is a really cool idea. It does mean that we'd have a lot more time zones because time zones would change both along longitude and latitude. It'd also mean that the time delta between two time zones north and south of each other would have a delta that changes every day, and wouldn't be a multiple of a single hour. To illustrate, think of two time zones, one on the equator, and another straight north enough that the day length changes by an hour between the solstices. On the equinox their clocks are identical and on each solstice they are 30 minutes apart in either direction. Halfway between an equinox and a solstice, they'd be 15 minutes apart. And so on. If you try and adjust the rules so that each zone is always an integral multiple of an hour apart from any other zone, then you end up with clocks jumping forward & back an hour which is basically what we were trying to avoid in the first place! So you'd always need a computer when trying to arrange a meeting across time zones. I imagine we'd get used to it, though. Another question is: what time would you pick for sunrise on really northern latitudes below the Arctic circle? For example Anchorage AK has a shortest day and shortest night of 5.5 hours. 9AM might be a good choice for the winter solstice, but it'd really suck waking up in the dark on the summer solstice when there's 18.5 hours of daylight! So maybe they'd set their time zone to smoothly change from a sunrise of 6AM to a sunrise of 9AM and back with the rest of the variance in the sunset times. North of the Arctic circle they're screwed no matter what they do, so I imagine they'd just join the time zone of their neighbors to the immediate south just for commercial convenience. :) It gets complicated, but time zones are complicated no matter what you do, and one hour clock shifts really suck. And sticking with DST also sucks, see minikites' comment (https://news.ycombinator.com/item?id=15564160 https://news.ycombinator.com/item?id=15564160). So this really does seem to be the best of bost worlds. I'd vote for it.
- spiznnx 9y agoIf you want to have constant day lengths, you can't make sunrise your reference. It's physically not possible. There are (almost) exactly 24 hours between each solar noon. There aren't 24 hours between each sunrise, that's why it moves every day. You'd have to change your day length at least twice a year. Perhaps in the past this would have been ok, but today it's IMO definitely not worth the additional complication that comes from changing day lengths. edit: I didn't read parent comment properly. Updated my comment.
- bryanlarsen 9y agoQuoting the OP: "It would mean that a given clock time on day N+1 is no longer exactly 24 hours later than that same clock time on day N, but I don't think that would matter for ordinary life. I'd be willing to put up with that if it means I get the benefits of DST (more evening light in summer) without the two big discontinuities a year."
- SiVal 9y agoYou say you would still have timezones, and timezones are approximately the area (all points) of the globe where it is approximately the same time, say all points where it is closer to 8am than to 7am or 9am when the centerline is 8am. With the current system, that makes the edges of time zones perfect north-south meridiens of longitude (which are then tweaked, of course for local reasons.) Sun rises at approx 6am and sets at 6pm every day on the equator. Go straight north to another point, and in the northern summer, it's sun rises before it does at the equator, and in northern winter it rises after the equator. So, with your system calling sunrise 8am, the edges of a timezone would drift with the seasons and would only be north-south meridiens twice a year at the equinoxes. If you decided to define the timezones with fixed (non-drifting) boundaries, your system might require a 2-dimensional grid of small timezones with boundaries not just on the east and west but on the north and south, with behaviors relative to their neighbors that would make timing a phone call to another time zone pretty much require computer assistance.
- JackMorgan 9y agoI like that idea a lot, but obligatory: (You advocate a ________ approach to calendar reform https://qntm.org/calendar https://qntm.org/calendar)
- tjdalaska 9y agoLive in Fairbanks, Alaska. There is no way this system would be feasible at northern latitudes. The days get longer/shorter at a rate of ~7 min. a day. Instead of an hour time change twice a year, it would be equivalent to an hour time change once a week. (and in december, it would get dark at 11:30am)
- bostik 9y agoGo north enough (or south enough) and you get this as part of your winter time weather forecast: Sunrise: January 29th
- tjdalaska 9y agoOh, and November 19th in Barrow would be ~1,560 hours long, and you would you just skip those 65 days? Or you could have November 20th in Barrow while it is different dates in other parts of the world?
- wirrbel 9y agoYour idea has precedent. "Babylonian hours" or greek hours (don't know if there is a common english name for them, "Babylonische Stunden".
- TeMPOraL 9y agoInteresting, but holy shit, how it would break everything. Having a day defined as a period of constant length of 24 hours is, I feel, a feature, not a bug. It allows for scheduling, for easily mapping between hours and days in your head. The system being anchored at the point more-less in the middle between sunrise and sunset is elegant, IMO, because - as you say - "the extra daylight is split evenly between morning and evening". This synchronizes the day period with our circadian rhythm (based on day/night cycle), which allows for easier dealing with other people. Variable-length day sounds like an ongoing chaos full of manual corrections and constant recomputations. I feel it would be a step back for our civilization.
- gumby 9y agoYour theory is inconsistent with the history of timekeeping: > I think that in the days before clocks, when the sundial was the most sophisticated timepiece, this could have worked quite well. Actually it's quite hard to figure out when "sunrise" occurs: it's not an instantaneous event, plus it's hard to calculate as an abstract (when would the sun peek over the horizon on an ideal billiard-ball planet); in the real world there are atmospherics, mountains etc. Plus for all these reasons it's hard to measure when you're close. On the other hand noon is easy with a sundial as that's when the shadow "disappears" (really: "is shortest" on most of the planet) except of course on rainy days. You can see the shadow shorten, see noon, and see it expand again. It's literally a 0 in the curve. The use of a nighttime sextant (or astrolabe that was on HN recently) is a way of transforming that calculation (known fixed angle) to other stars or times. So it was the clock that made the time of sunrise work: if you could synchronise your clock to noon on D0 you could then use the clock to index into your ephemeris to figure out times of other celestial events post D0-noon. Also, pre-clock, hours were not fixed duration, they were 12 divisions of the day. Minutes and seconds were post-clock inventions.