So, I'm curious, though this maybe off topic. Assuming we reached space, how do you think we solve the time problem? I can't think of anything that doesn't multiply the above insanity by 10.
Time zones are great-if you have one planet. But now you've also got varying day lengths, years, etc. I'm curious if you've given any thought to this; my immediate thought is that we all use a standard frame of reference, but that'd necessitate going against the light cycle in order for it to actually be sane. Which is probably a problem, biologically.
Maybe we'll all end up on software to manage figuring it out for us, or something. But I've never been able to find a good solution, and it seems like kind of a certainty that we'll reach space at this point.

The whole thing about trying to find the time in another city is just a semantics thing. It will stop being difficult to say when people start having a need to say it.

If I want to talk about how much $1000 was worth twenty years ago, does wolfram alpha just tell me it was worth $1000? Of course not. It says it was $1579.62: http://www.wolframalpha.com/input/?i=%241000+%281995+money%29

The same sort of thing applies here. If you want to know what time it is in Melbourne, they'll tell you what time that time of day corresponds to your present time. Asking how many hours off they are will still be a perfectly valid question. It's just that the answer won't be an integer number of hours.

Midday and midnight won't mean the middle of the day and the middle of the night, but unless you go to bed at 8:00 pm and wake up at 4:00 am, they already don't.

The only problem is the changing of the day. But that's already confusing. Is the appropriate greeting at 1:00 am "good night" or "good morning"? 12:00 am almost always refers to midnight, and almost always the one at the beginning of the day, but it's still confusing enough that it's common for things to happen at 11:59 pm.

I've got it! What if we instituted time zones in 1-second intervals instead of 1-hour, thereby ensuring that people allegedly in the same time zone don't experience days that are nearly a whole hour out of sync?

The logical conclusion is continuous time but that's probably technologically and/or physically impractical, and 1 second is probably a good enough approximation for everyday use. This will surely have exactly zero drawbacks.

I have a mental image of the person who successfully eradicated time zones, but somehow failed to eradicate daylight saving, falling to their knees screaming NOOOOOOOOOO

Bauglir that was called "Railway Time", and the problems it caused are why we have large "time zones" in the first place - zones where all the clocks are synchronised regardless of the position of the Sun.

Listen, the important thing about mistakes is not that we avoid making them, it's that we avoid learning from them.

Sadly your wise formatting policies prevent me from making this sentence transparent, but I'm not being even slightly serious.

Surprising to learn that anybody ever thought that was a good idea, though.

This essay does a great job of showing why the abolition of time zones is a bad idea, but is there actually anyone who needed convincing of that?

Yes, pretty much everybody who has experience with time zones entertains the idea of abolishing them now and then. I see it come up in many discussions of date/time software problems. The sad fact is that modern timekeeping, while horrendous on a technical level, is about as good as it gets.

Interesting that you bring up China since that country has only one time zone. Russia has 9. Both systems seem to work.

It saddens me to say this but I am guilty of wishing the world was on one time zone. Planning cross time zone meetings can be such a pain. Not nearly as painfully as planning them without though.

Well, I have to say, that given I live in Perth, Western Australia and so I am fortunate enough to share the GMT+8 time zone, I'm all for this new system. ;)

You missed the obvious first step central to any plan to abolish time zones :

1) Upload the entire population to a digital earth with a worldwide Minecraft-style day/night cycle.

Surely this is what all the proponents of time zone abolition intend.

That a change is difficult or requires adaptation does not indicate that it is not an improvement.

But I never asked for time zones to be abolished, only daylight saving time. Spontaneously adding/removing hours is absurd. At least we have UTC.

Bauglir: People thought Railway Time was a good idea when rail transit was the fastest way to get anywhere. And when telegraphs were the fastest way to communicate -- nearly instant in transmission, true, but with long latency from the time sender begins writing to the time receiver begins reading.

Synchronized time DIDN'T MATTER back then. You couldn't get to another place fast enough to experience "jet lag" (because there were no jets), so why care?

[I thought the sqrt(-1) test accepted j as an answer?)

Also note that in some equatorial countries the sun rises and sets pretty much exactly at what we would call 6a.m. and 6p.m. Thus it’s often convenient to count not from midnight, but instead count 12 daylight and 12 night hours. E.g. school starts at 1 in the day, meaning get up at 11 at night, have breakfast at sunrise, then hop on the bus.

Not really about time zones per se, but it underscores the point that talking about time should be above all convenient for humans in everyday situations.

The entire argument of the post, though, applies exactly as well to dates. We all use exactly the same calendar, all around the world, and agree on the date at all times, similar to what we'd experience if we were all on China Time.

But that means I can't tell what the weather is like if I want to go visit Uncle Steve! I mean, January is pretty cold here, but when I google for the date in Melbourne, it just says it's January over there too! You can't even say that Melbourne is "6 months ahead", because it's still January there, that's not right.

We've somehow developed language around the lack of "date zones", and can reasonably talk about Melbourne being in summer in January while we're in winter in January. We can talk about places closer to the equator having longer summers and shorter winters than places closer to the poles, despite them having the same dates. And the benefit of all this is that we can schedule events really easily, with a single date meaning the same thing to everyone around the world (except for the complication of time zones...), while simultaneously having a decent grasp on what the weather is like by describing the date over there by its "season name".

We could do the same with time. Use a single numeric global time, just like the single numeric global date, and use time words to refer to segments of the day according to our schedule - morning, afternoon, etc - just as we use season words to describe segments of the year according to the weather. It's totally reasonable, in such a world, for Google to answer your "what time is it in Melbourne" query with "early morning", letting you know it's probably good to wait a bit before calling.

The day naming issue is the only really annoying bit. We're okay with naming a year by a span of 365 days from a certain date, rather than from a certain season (so the southern hemisphere year doesn't start in July), but we're attached, with good reason, to thinking of days according to our schedule, rather than as a span of 24 hours from an arbitrary point. But I'm sure we could figure out something for it.

Tab Atkins, did you skip reading the article and go straight to the comments? I thought we decided time zones are a good thing. I know that I'm convinced!

Two comments...

#1: Don't assume folks actually check the time before calling. If its important I suspect they'll call anyway...

#2: Who still calls people??? With text messages, tweets, email, Facebook, pokes, etc. It doesn't really matter when communication is sent.

Brilliant. And it's simply amazing to me that there exist people who think abolishing timezones would be a good idea. Proof if any were needed that being "smart" has only a loose correlation with talking sense!

Camllorn: We already do this. The Curiosity team operates on a Mars day (they call it a Sol), which is of course about forty minutes longer than an Earth day. So their work hours slip by forty minutes a day and slowly go out of alignment with local days.

(Sauce: http://www.space.com/16975-curiosity-mars-rover-mission-control.html)

I guess what I'm saying is that we'll always find a way to make it work. In fact, we already have. For a given value of work.

As a software developer and international meeting organiser, I hate time zones.

I built http://fuck.timezon.es to work around the problem, but still I dreamt of a day when time zones would be abolished; how naive I was.

I understand now - planetary rotation (on a tilted axis), elliptical orbits and their resultant cycles of light, dark and seasons mean life without time zones is more awkward even than life with them.

I'll endure them for now, but at the first opportunity I'm leaving Earth to live in a Dyson Sphere.

And in that Dyson Sphere we'll use a single currency, unicode and the metric system :P

The ISS uses UTC+0 which is a nice clean time, midway between USA and Russia, and would be especially convenient for Europeans.
NASA used to use EST, and recently was using Central time (for Houston).

I think abolishing time zones would be simpler than you describe. You could look up/memorize an offset to see what level of daylight there corresponds to what level of daylight here. Uncle lives in Melbourne? "Hmm, my offset's +0, his offset's +10, so when it's 4 here, it's as though it's 14 there. I wouldn't be awake at 14 here, so it's best not to call him." The question "What time is it where you are?" becomes useless/meaningless because you already know the time; instead, it's replaced with the question "What offset is it where you are?", the answer to which is constant for a particular location and is therefore simpler. That said, I wouldn't like to abolish time zones, as I do a lot of traveling, and not having to get used to different time schemes is convenient.

If/when we start living on multiple planets, it won't get much harder, except that you need to remember how many hours in a day there are on each planet. The question "what time is it there" will get harder to answer, because not only would Mars have time zones, but the offset between those time zones and Earth time zones would change constantly. However, that should be nearly a non-issue at that point, with the advent of computers.

MichaelSzegedy, I think the system you described, with the system of integer offsets for different regions, has already been proposed and attempted. They're called "time zones" if memory serves, and it has the added bonus of being able to use the same numbers to refer to roughly the same part of the diurnal cycle in different regions!

Very funny, TroyHurts. The system I described would be qualitatively different from current time zones. People would use different numbers to think about time than currently. You'd wake up in the morning in Indonesia, and it'd be 20 o'clock. The offsets are only relevant when you care about how daily life differs from region to region; otherwise, it's universal time, which is desirable because it simplifies life for transportation agencies and programmers. Sam's issues only occurred because there was no central source that told the time differences between when morning etc. occurred between different regions.

After abolishing time zones, take 2:

I want to call my Uncle Steve in Melbourne. What part of the day are they in?

Google tells me it's "just after sunrise on Saturday" there. Probably best not to call him right now. The end.

At present, Google understands queries about local time because they are common and useful. In a world without time zones there would no longer be queries about local time, but some terminology would evolve for describing local variations in when-in-the-day-it-is and Google would understand queries about that. For exactly the same reason as, in the present with-time-zones world, it understand queries like "time in melbourne".

The terminology might end up being awfully similar to the notation we now use for local times, as MichaelSzegedy suggests, and we might then end up using exactly as we now use local times, as Sam suggests, in which case indeed we'd be back where we started. But I don't think it's obvious that that would happen.

(Would it be "just after sunrise"? Quite possibly not. It might be "two hours before the start of a typical working day" or something. Exactly what Google would find it appropriate to say would depend on what conventions we developed after the abolition of time zones, which I don't feel comfortable predicting with any confidence.)

The issue of knowing where the sun is for your friends around the world is a problem I've been dealing with for a few years now. Since they all travel, seemingly work odd hours, and generally don't have the concept of a sleep scheduled, I've had trouble keeping track of when it's appropriate to call or not.

Of course, the simplest solution is just send a text and have them get back to me whenever, but that's not a solution which is really worth typing out a few paragraphs to explain. Especially when pretty much everyone can think of it for themselves.

The way I try and manage it all was by replacing my desktop wallpaper with a map of the Earth with the light of the sun projected on it. At a glance, I can see where noon, morning, night, evening, and all the little solar time keeping names for anywhere in the world provided I know enough geography to approximate city locations.

While it works for me, putting such a projection along side your usual clocks probably isn't the best 'solution' to the 'solution'. Just a quick shorthand that works for phone etiquette about as well as asking for the time in Lolipoluza Land on Google.

Which really only means that the coders get out of this for the better (at least for the adjustment period). Time zones are a nightmare to keep track of if you're writing your own program, but that's only for enthusiasts and those who don't know the rule of checking to see if someone else has already solved your problem first.

It would interesting if nothing else to see a world where people don't talk with time-notation and just looked to a map of the whole Earth every time they wanted to make a call.

@Camllorn:

Most likely, we would use calendars and time zones synchronized to the local day on planets, "ship's time" aboard ships, and TAI for any kind of important business of potentially interplanetary significance.

Bodies without reasonably human-scale day/night cycles would have to get more creative, but the point, I think, is that everyone would use some kind of "local time" except when dealing with people far away. Like now, only moreso.

Ellie, have you ever seen xkcd.com/now?

This article is a good example for the absurdity how a lot of armchair self-appointed "savers of the world" think, who suddenly come up with "great ideas" how to reform society to eliminate poverty, racism, wars, or other negative things like time zones, and have absolutely no clue that their great ideas, while maybe looking great in an inspirational cartoon or blog post, would horribly fail in real life.

In some airports they have a world map showing where is in darkness and where has daylight. Although calling someone in the Arctic circle in June (we're still allowed months right?) when it happens to be daylight is no real guarantee.

@Val: I am suddenly curious as to why you mentioned an inspirational cartoon. Got any examples?

What are you talking about, Uncle Steve? Of course there's a lookup table.

@Ellie Kesselman:
Of course I read the article. My comment specifically called out several parts of Sam's article, used parallel naming and structure, answered specific problems from the post, etc.

I, like the later post by g, was pointing out that the problems Sam brings up agent as fatal as he makes them out to be. We solve the same problem with dates already, and it's easy to see how to apply the solutions from dates to answer some of the main problems that Sam brought up. The only string remaining issue is day naming, as I said.

I don't mind time zones. I just don't understand why in the world there are places that will use "daylight saving time". Adjusting everyone's clocks is stupid. In Hong Kong here, we just got schools hours start earlier in Summer and latter in Winter. No need to adjust clocks.

Sometimes you remind me of Tom Scott, especially when you cover similar topics, though in this case from very different angles.

https://www.youtube.com/watch?v=-5wpm-gesOY

Wait, hang on. With respect to the long-dead talk of Railway Time, I think maybe we were talking about different things. I was proposing a system that would have you adjusting your watch by 1 second every 1000 feet or so at the equator. That's a small enough distance, especially scaled down as you approach the poles, that it would easily have been an issue even when rails were the primary mode of long-distance travel. Instead of 24 time zones, you'd have 86400.

Railway Time looks like exactly the opposite of that.

I'm starting to think I'm bad at jokes.

@Tab Atkins: The "problem" you point out with dates isn't a problem that would be better solved by dividing the earth into "date zones," for the simple reason that weather and the date are only loosely correlated and are tied up with a billion other variables.

For instance, take Cincinnati and Cleveland. They're in the same state, it would seem natural to put them in the same date zone, and yet thanks to the "lake effect", a chilly day in Cincinnati could be two feet of snow in Cleveland! Or take Australia, the go-to example for "winter there, summer here," and yet average temperatures in Melbourne in July are well above freezing.

Any system of "date zones" that is fine-grained enough to answer the question of "What's the weather like where Uncle Steve is?" would be so complex that you'd basically have to look it up on Google every time.

...Which, incidentally, is what people already do. Before you go on a trip, look up the next week's forecast for your destination. More accurate than guessing based on latitude, and with the added advantage that we can still all use the same date for measuring time. You know, the thing it was designed for.

@Aegeus
Sure, it's definitely not exactly the same and climate is complicated, but I still think it's a reasonable analogy. We *do* have season words which we apply differently based on latitude, and they're useful - saying that Alaska has a "brief summer" and is "mostly winter" is useful, despite summer/winter technically being defined by dates. Similarly, Australian summer is American winter, etc. It just happens that longitude<->time is a more stable/direct relationship than latitude<->season. ^_^

I liked Swatch's internet time, where each day has 1000 "beats" (1 beat = 86.4 seconds), it's the same internet time everywhere, and there is no am or pm. For example: @765. No, it doesn't solve all the issues, and its not perfect, but people would quickly become used to, and aware of the "beat times" that they sleep, work, eat, play, and call people on the other side of the world. Also, there is no reason why we can't continue use time zones for the concept of the local "day", but use a single "world time" (be it UTC, Swatch's internet time, or whatever) for things like international calls and international travel etc, and even local time reference.

@Camllorn
<br/>I don't really see the use for a universal time; when it comes to keeping and comparing time on different planets, it'd probably be easier (at least when it comes to consistency and at-a-glance understanding, if nothing else) to just keep local time with a system for giving enough context to figure the difference. For example:<br/><br/>Say local time goes from 0 to 100 (0 for when the sun is as far from azimuth as it can get, 100 for when it's at azimuth), with prefixes - for when it's going down and + for when it's going up. So +100 would be immediately before the solar noon at your position and -100 would be immediately after. After that, you could place your line of longitude so people can figure what your time is relative to their's - again, with a + or -, except this time + for east and - for west. So if you wanted to find out what time it was in Greenwich relative to +53 -41 (So around the area of New york if the International Date Line was 100 and not 180), you can pretty easily figure that it'd be +94 0. If it also happened to be +53 at -41 on the moon, and your friend was at the moon's prime meridian, it'd be +94 there, too. 10 east from the prime meridian, it'd be -96. If you were aboard a spacecraft or artificial satellite, the time would probably be calculated by whatever longitude and rotation period the intended occupants were comfortable with.<br/><br/>Doesn't really help with the fact that planets rotate at different rates, but eh - you'd need to memorize rotation periods and what time it was last time you checked at what longitude and how long it's been since you last checked. You could probably just look up the times on the InterPlaNet anyways.

I agree with Tom's post from earlier this year. Refreshing that someone out there shares my opinion.

google current day night world map. problem solved.

Another argument against abolishing time zones: Santa Claus would have to visit every house on Earth simultaneously, many during broad daylight.

"I sometimes go to bed as late as 7pm!

...That, uh, that's very late in this country. It's the equivalent of 2pm where you live. I think. Just trust me, I'm hardcore."

As a former night-shift worker, I think it's rude to assume someone must be awake just because the sun is shining at their longitude -- although there's a good chance they are because the phone keeps ringing off the hook and it's hard enough to sleep when the sun is out. (Yes, I suppose one could just turn off the ringer and sleep through any true emergencies.)

@BGK I very much like the idea of Internet Time. That unfortunately means that we'd need two clocks: one that tracks local solar time and one that tracks Internet Time. However, it's more practical than arithmetic with time zones (Russia and China being a prime example of time conversion hell).

One possible solution is to remove some of the time zones instead of all of them while continuing to use a 24-hour day. The beneficial aspect of this is that our timekeeping, sleep schedules, etc. are not tampered with, and it provides a more general indication of when it's safe to call. For example, the eastern and western hemispheres could have 2 time zones each, resulting in 4 global time zones.

Supposing this was adopted by the world, the question of whether it's safe to call Uncle Steve is simplified because you could literally add 6 hours to your time and get a reasonable approximation of what time it was. If 6 hours is too much, perhaps dividing into 6 time zones of 4 hours each would be better. Theoretically the entire contiguous 48 states of the USA and all of Canada could be on one time zone under the six-zone scheme, but that would imply that South America would require two zones. Perhaps a mapping scheme would work where the western and eastern hemispheres would be separated into 12-hour zones and further subdivided into 6 sub-zones that are made up of 2 hours each (instead of 12 sub-zones made up of 1 hour each). Times would be specified using the following scheme:

2W0500 would represent 5am in the 3rd Western sub-zone (e.g. New York, UTC-5). The corresponding time in London (UTC) would be 0E1100, 11am in the 1st Eastern sub-zone. The math is done as follows: 05 + 2((2 + 0) + 1) = 05 + 6 = 11. Likewise, to handle 0900 in Tokyo to San Francisco, it would be 4E0900 to a time in 4W: 09 - 2((4 + 4) + 1) = 09 - 18 = 24-09 = 15, so 4W1500 is the answer. If you got the zone wrong, you'd be off by ~2 hours, which is a lot better than being off by 4 zones.

Somebody please think of the farmers? Well, they get up with the sun typically, so the time issue doesn't matter as much as it does to someone who needs to be at work at 8am. However, they make a living selling food to people, so market times are relevant. If the market is open from 8am to 4pm across another time zone, that means it would be 6am (or 10am) to 2pm (or 6pm) where the farmer resides, which results in the need to leave even earlier for the market (or leave the market "late" by the farmer's usual standards). Of course, if we didn't have time zones, this wouldn't be an issue.

So the real problem seems to be that time zones make planning work-related events hell, but they are convenient for personal reasons. With that in mind, is there a solution?

Much more important:
Adopt time-intervals to the decimal system.