Sorry, I still don't get it. The sun doesn't shift by an hour, it just shows itself for less time as we get into winter and more time as we get into summer. Why can't we just center our days so that noon is always when the sun is highest and be done with it? The shift makes no sense.
>Why can't we just center our days so that noon is always when the sun is highest and be done with it?
That's what abolishing the DST gets us closer to, and the answer is - nothing is stopping us. Arizona has it that way, and they're fine.
A longer answer: time zones are just an approximation to "sun is highest at noon" because it's awfully convenient to have clocks show the same time for people within a several hundred mile radius. The railroads made it pretty much a necessity, but even things like scheduled television programming would be very tedious with astronomical time: every city would end up with a slightly different schedule.
Think about it this way: the circumference of the Earth is about 25000 miles, and one revolution is 24 hours. That means that 1000 miles is roughly an hour difference, and 100 miles is about 5 minutes.
That means that, with astronomical time, if you have a 100-mile commute (sadly, not uncommon), you also have to account for the 5 minutes time difference between your home and work!
While this might not be the question you are asking, some threads on HN have considered the possibility of everyone being on astronomical time, since we can put a GPS chip in every clock (...heck, we pretty much do that already), and everyone gets their schedules from realtime systems.
But perhaps a more compelling alternative is just to give up and have everyone follow the same clock[1], extending the convenience of time zones (which are, after all, arbitrary and are an awful approximation for Solar time!) to the whole globe.
At my latitude (26°N) the zenith for noon changes about every 5 miles (8k) east or west of where I'm located, and the distance grows shorter the more north you go (at 45°N it's 3.8m/6.1k east/west). That's why time zones exist, to make this a tractable problem (although we somehow have managed to make a mess of time zones but that's another rant).
EDIT: my latitude was too precise for the distance mentioned. Simplified it.
I don't think it's pedantic at all, as far as I'm concerned I find that light in the evening is a lot more important and useful than in the morning and clearly our life cycles do not match that at all. Where I am in the summer (with DST) we still have daylight past 9pm while in winter it's dark at 7. DST would need to be several hours more to roughly match that (and in my subjective opinion it does it the wrong way around, I want more daylight in the winter evenings). I say good riddance.
I don't think what you're suggesting in line with OP. If that is not the case, your comment is worded much more effectively.
> . . . in my subjective opinion it does it the wrong way around, I want more daylight in the winter evenings . . .
I believe this is because DST originated as an energy saving measure to maximize usage of morning light. [1] I tend to agree with you as well. If we're not trying to conserve energy in the evenings, DST doesn't make much sense.
That may be true but certainly having morning light brutally change twice a year can't be a great thing then? I always found the switch from DST to winter time pretty jarring for instance, you go from very dark mornings to much brighter ones.
In the end no matter how you look at it trying to match any form of "sunlight cycle" by merely offsetting time is bound to fail. It's not a translation you need, it's a homothety.
OP seemed to be suggesting a gradual shifting of clocks each day instead of a single leap around each inflection point.
> It's an artificial construct with a vague purpose, and nothing else
A 24hr day is an artificial construct. DST is mostly an energy saving measure, and some also enjoy more daylight after leaving work.
> Please do basic research before accusing others of pedantry.
I certainly know the basics of DST and the earth's orbit. Thank you very much. These are the type of threads that tend to spawn from OP's style of comment, and I regret my reply.
Sorry, I need to learn to be more clear. I was not suggesting a gradual shifting of clocks each day. I was pointing out the silliness of shifting by an hour twice a year. My desired alternative is to not shift at all.
>OP seemed to be suggesting a gradual shifting of clocks each day
It's hard for me to see how it can be interpreted that way.
>These are the type of threads that tend to spawn from OP's style of comment,
Please don't blame OP for what you wrote, which is - to quote - "It’s called an approximation", and what you wrote now doesn't explain what you meant by that.
I responded in that fashion to mimic the OP's tone.
---
The fact that daylight saving time is an approximation for fixing sunrise to a specific time is common knowledge and should not require a thorough explanation. But for posterity [1] is a graph that illustrates the approximation.
Thank you for answering. Now let's discuss this answer.
The first question is: which specific time? This time will differ based on the timezone and geography. Because of timezones, sunrise time effectively makes a one-hour jump at the transition points, which makes this chart of yours look different for people living living close to timezone border vs. someone living in the center.
To that end, which time zone and which city was this chart made for? Because we'd need many charts like that to see the effect of DST on sunrise time. (Remember, timezones make sunrise/sunset times depend on longtitude (exact coordinates, actually) as well as latitude and time of year!).
Thankfully, some people have done just that[1][2] (note that [2] is interactive!).
As you can see by playing with [2], if your goal is to have sunrise before 7:00AM across the country on as many days as possible, then DST works against this goal.
Of course, one can always throw one's hands up in the air and say "it's an approximation". In the same sense, 0 is an approximation to any number -- just not a very good one.
The US foreign debt is approximately 0 dollars, if you don't talk about error margins.
Now, looking at the interactive chart at [2], do you really think that DST is a better approximation to "having the sun up at (your chosen time)" than doing nothing at all? Or shifting the whole timezone by an hour?
> The first question is: which specific time? This time will differ based on the timezone and geography.
That's not what it's about. The specific time is irrelevant - that's just an arbitrary number. The reasoning in the blog posts you link to is just based around making comparisons to another arbitrary number (7 AM).
Rather, the point is shape of the curve. DST makes the sunrise time approximate a constant line.
In other words, if you plotted the time difference between sunrise and "x o'clock" (for any x) for every day across the year, DST reduces the standard deviation of those values.
The argument is that circadian rhythm causes humans to naturally wake up near sunrise. Most people also wake up at some fixed time in order to start their workday (because businesses generally have fixed hours, and people wake up just before going to work). DST makes it so that the difference between this fixed time and 'sunrise' can be minimized over the year. Without DST, you will not be able to have a fixed wake-up time and also minimize the difference between that time and sunrise.
