It also enables common culture. If someone says they got up at 5am for a flight then it is widely understood that that is pretty early for a typical person, without having to explain as much in words.
If someone says they didn't get home until late, you know they probably mean 9 to some small hour of the morning etc.
Yes! Whenever people say that we should have one timezone I always bring this up. Either you have to look up what time it is in another country or you have to look up what time of day it is. Either way you still have to look it up.
You may have already read it, but in case you haven’t, I think this post illustrates the challenges with a single timezone rather nicely: https://qntm.org/abolish
In fairness, that's not what they're arguing for. They're advocating using UTC for yourself, and since no one else does that you'll inevitably get good at conversions to other time zones.
If you're scheduling something with someone non-local to yourself, you need not only know their time zone, but you also need also consider their latitude/longitude and what time of year it is so you account for the potential use of daylight saving time and their current sunrise/sunset times.
For example: if you schedule something at 7PM - is that daytime or nighttime? Well, if it's the first of October, and the person is in Kansas City, the answer is daytime. However, if they're in Phoenix, the answer is nighttime. And here's the kicker: the fact that Kansas City uses daylight saving time is irrelevant, as it would still be mostly light out even if you used standard time due to its higher latitude.
If they happen to work the "average" schedule for their location, and you know what location they will be in at the time of the meeting, and you don't mess up the conversion.
But if you try to schedule a meeting with someone they will immediately know if the time is reasonable for them and let you know.
Plus I don't see how it actually makes a difference, you just look up "what are business hours for $city" rather than "what time is it in $city".
Why does it matter for your normalized baseline to be UTC? Why not your local time and convert everything into that in your head. The reference time zone seems arbitrary and your primary residence's local time makes the most sense IMO.
Because it's "Universal". It becomes fairly easy to calculate time of an unknown city, instantly once you know the offset. One less mental calculation.
I don’t buy it at all. If you are that used to conversions, you probably remember the direct offsets by heart. I know I do for the three or four time zones I care about.
That said the whole conversion thing is not an immaterial burden unless you’re really up in the sky longer than walking on earth.
The whole point of UTC is to have a shared reference baseline to communicate with the world; however the author is obviously not concerned about that, I don’t think, as they have to do the conversion into the third party’s local time in their head if they want to have any friends, so that point is moot.
FWIW, most phones allow dual time display, and GMT watches would be extra helpful here as you can essentially track both at once, using either as your 'main.'
> After an eight-hour flight, if you went to bed at 21:00 yesterday, you’ll do the same today.
The problem is, depending on where you landed, 21:00 could be typical start of the workday, and you'd be showing up to an empty office.
> If you had a call at 13:00 yesterday, it’s still at 13:00 today
This part sounds nice, until DST comes along to sabotage you. For only being twice a year though, it's probably still a net gain, one I hadn't really considered for someone who travels a lot.
That's how I have conky setup on my desktop. "Year : Julian Day : HH:MM:SS" in both local and UTC. With a fixed width font and the two right above each other it's exceptionally useful. It makes looking at logs captured in UTC much easier to parse.
Many years ago I committed to setting everything to 24hr clocks. Even when I still had a standalone digital alarm clock that plugged into the wall; in fact that was the impetus to change.
Because it was very easy to miss a little "PM" pip in the corner, when reading or setting the time; and setting an alarm depended on getting the pip correct, so eliminating the pip for me meant eliminating ambiguity in the digits.
It's been a fairly successful transition, although sometimes I look at "16:15" and think it's 6pm so the "mentally subtracting 12" needs practice. And there are still special snowflake UI widgets that ignore my i18n settings, and constrain how I enter or specify a time.
Using continious letters for addititives values is a good idea (F+3) can be predicted easily.
Yet i agree that confusing « symbols » should be avoided.
Forcing UPPERCASE formeront plus using a prefix (Z might be a good one)
And if your language doesn’t use the Latin alphabet?
I like the concept of a global time that maps to local time, and while I am unconvinced that Latin letters are the best way to do that, I can’t think of a better way. It’s either letters of some sort, symbols, or names, and none of those sound good.
You should live by the sun, regardless of what the clock says. The clock is useful for measuring time, and UTC, TAI, etc can be useful for that, but if you want to wake up, etc, you can do by the sun light. For local time, you can use the solar time, which can be used with a sundial.
That’s still at midnight locally though. What’s maybe odder is that your UTC date changes in the middle of the local day. So “tomorrow” can be “this afternoon”.
That does seem odd on paper, but in practice we already deal with that anyway from midnight being kind of arbitrary and not necessarily while we're asleep (almost never, personally) so we do things like scare-quote 'tomorrow' because really what we mean is 'my next waking period'. It'd be weird at first for sure, but once accustomed to it I don't think that would be a drawback personally, just a different arbitrary.
That's great to have an alarm to 11am which is 7am at your timezone and be woken up at 3am after cross-atlantic flight next day.
Some people just like to add complexity to their lives and keep convincing themselves that it's better.
Anecdotally, I worked on a military facility where all clocks were in "master timezone" of the country without daylight saving offset. 6 months a year we put "+1" sign made of electric tape to not forget about that it's summer and 0400 is actually 0400+TZ+1. I worked there for 5 years and never got used to it.
When dealing with anything across timezones the reality is you ARE dealing with 2 time contexts.
1) global time. i.e. UTC
2) local “sun” time. i.e. local timezone
When scheduling a meeting you are trying to find an intersection of global and local time that works.
It’s messy but also our physical reality. The conversion still needs to be done, just in the article the person is living in global time and having to convert to local time when needed. Vs living in local time and converting to another different local time.
Nope, no. I appreciate the experiment and I did do this myself for about 4 months when doing some intense cross-Atlantic business consulting traveling.
Unlike DST today, time zones do still have economic utility in the modern world since there are many more personal and industrial activities that exclusively happen during day or night in a single locale and far fewer that happen 24x7 and routinely traverse multiple locales.
The article focuses a lot on how this reduces friction when working across time zones but glazes over how it makes interacting with everything local require a conversion.
5 days a week I am working with people in Seattle, NYC, Munich, and often London or India. I have to coordinate meetings across time zones a lot. I travel a lot.
And yet at home or on the road I have to process things like "this store closes at 18:00" or "this train is at 16:45" infinitely more. Putting my devices in UTC requires me to think for all of those interactions which is way worse.
My third job was medical telemetry and that was sufficient for me to not only put everything in utc but also correct for client server clock skew.
Problem is though with distributed offices and remote workers. “The servers started crashing at 10:31 can you look at it?” Doesn’t mean anything without more context. If it’s 10:50 local does that mean it started 20 minutes ago or 7:20 ago? 3:20 ago because they’re on the east coast?
> As a programmer, I’ve always been annoyed by the concept of administrative time zones.
> Five years ago, I decided time zones should be abolished, and everyone should use one coordinated time.
This is called many things: utopianism, dogmatism, authoritarianism, absolutism. Seems common in programmers. I don't understand the mechanism for it, but it would be great if we could figure out a way to prevent it, and teach that in schools. Too many people like this in the world today.
Eh, I think it's more a drive to simplify things that exist because of inertia/"that's how we always did it". Time zones are arbitrary and silly (sun is most likely not halfway at noon for most of the zone), daylight savings is even worse. Imperial units too, get rid of them.
> sun is most likely not halfway at noon for most of the zone
Being able to express where you are in the 24-hour cycle is useful. That's what local time is: an approximation of the time since the last solar midnight.
To the extent that the approximation is poor, it's an argument for improving it (Spain for example should really set their clocks back an hour), not for getting rid of it entirely.
For the same reason I still like temperature in Fahrenheit, despite knowing that it is inferior in all contexts other than weather and most of the world doesn't use it. Around 100F is dangerously hot and 0F is dangerously cold.
Besides, Kelvin is the most true temperature scale, not Celsius.
Imo the best system would've been Celsius with double the precision, e.g. water freezing still at 0ºC but water boiling at 200ºC.
That way you have the "human livable" range kinda between 0-100, which feels very intuitive. Anything above 100ºC becomes effectively unlivable. It also means that it is much easier to distinguish between certain "zones". e.g. saying "70s" or "80s" is easier and more clear than with celsius where you typically are staying within a sliding 10º range from day-to-day.
I agree this is an improvement, but you'd still have a bunch of negative number days in the populated northern and southern hemispheres, which Farenheit further improves upon.
Yes that is marginally better in Farenheit, but centering on the freezing point of water seems more useful overall. Then you'd also end up with the coldest spots on earth never being less than -200C and the hottest spots never being more than 200C which is nice symmetry.
> but centering on the freezing point of water seems more useful overall
But why? Knowing the freezing point of water (at sea level, mind you) is only rarely useful to me in daily life. I think the eutectic point of salt and water (that is, the lowest point at which it's possible to cause ice on the roads to melt by adding salt to them) is marginally more useful; this happens to be about -6F or about -21 C. Perhaps it's a bit more important for farmers who need to know whether their crops will freeze overnight, but as a lifelong city/suburbs dweller who knows next to nothing about agriculture I couldn't really say. At any rate, most people are not farmers.
I'm asking this somewhat rhetorically: I don't really care about Celsius vs. Fahrenheit. The pros and cons between the two are so vastly outweighed by the utility of having a global standard that I really wish the US would switch. I'm just pointing out that I don't think that one particular physical property of one particular substance makes for a very strong argument either way.
I think water freezing without salt is much more common than freezing with it. For example knowing the freezing point of water is useful for setting your fridge/freezer temperature.
in the temperate places most people live, the temperatures rarely drop below 0°F and rarely exceed 100°F and humans can sense a change of about 1°F so setting a thermostat in F is a simpler matter.
the 60's are warm in the sun and cool in the shade or breeze, the 70's are warm, the 80's are hot and the 90's are really hot, and over a hundred you're looking at death. The 50's you need a jacket, the 40's you need more jacket, the 30's are downright cold and you need an overcoat, and the 20's are frigid.
Celsius has approximately none of those properties. Celsius the 20's encompasses a vast change in comfort level.
As a celsius-user: you dont need more precision to judge how a temperature feels. The ranges you mention for Farenheit exist for Celsius as well. And if you need more precision you can use fractional numbers but I have rarely seen more precision in household thermometers than just rounding to 0.5 or the next whole number. I assure you that it is a completely usable scale.
To expand on your point: a difference of 0.5 degrees Celsius is almost exactly a difference of 1 degree Fahrenheit (really 0.9). So, indeed, a thermostat that can be set to half-degrees Celsius is roughly equally useful as one that can be set to full degrees Fahrenheit.
99% of the time that I communicate about temperature, it’s in relation to the weather. Fahrenheit optimizes for the right thing. “How does it feel on a scale of 0-100?” is the nr 1 thing I use temperature for, nothing else comes close.
omg! i've never seen that xkcd before, but on my own I do think switching to metric doesn't have much payoff for the average person, and I buy all the same socks! It's awesome for folding, for losing them one at a time, etc.
Just send him an email. Or politely ask him about his schedule and don't make assumptions about it, because I, for instance, can be awake at 5am a weekend and asleep at 9pm on a workday.
Exactly. By GP's logic, time zones are absolutist, and every town could & should observe solar noon independently.
I think folks are averse to "world time" (for reasons, largely inertial), so maybe the baby step is try 1 timezone per country (like China's done for .. 75 years ?).
I'd even argue Local Time has only ~4 useful times: dawn, daytime, dusk, & nighttime. Where I grew up, the parks closed at dusk every day. Nobody complained
If I take a walk at 7:05:35 PM every day, it seems very precise but doesn't indicate whether I need sunglasses or a flashlight. It's meaningless precision, like 0.6235 slices of pizza. If I'm coordinating a walk with you, I might as well use UTC: it still won't tell light from dark, but at least nobody'll be waiting for an hour due to DST. It'd make more sense to schedule our walk at `1 hour before dusk`, or "just" settle for UTC, IMO.
China did that for political reasons. The Republic of China used five, and Russia uses two timezones across those longitudes. Similarly, I also find it quite weird that France and Spain are in CET. France shifted when it was occupied in WWII, but maybe it's justified to remain in CET to reduce friction with the economies of its eastern neighbors. Whatever. Vive la weird. Countries do that to themselves.
A deviation of half an hour from noon is barely perceptible unless you use a sunclock. People can roughly estimate when the sun will set by just looking at local time and considering the current season. What throws a wrench in the works is daylight saving time. I fully agree that DST a huge annoyance and that its benefits were always rather situational.
Seems to me that standardizing the entire world on a single time is confronting and managing the complexity of the world.
The existing system is more inertial and political than practical.
"Noon is when the sun is at its zenith in your area" is significantly less useful in 2025 than being able to coordinate globally with ease. It's also just not true in many (most?) cases, as noted in the article, where places like China have a single timezone even though there is a 3-hour discrepancy in the sun's position from one side of China to the other.
Different parts of the world are at different times of day, that complexity doesn't disappear if you ignore it. The more distributed your team is, the more important this becomes. People keep saying timezones only exist because of inertia, but I do not buy it. There's definitely complexity caused by political boundaries that isn't inherent to the problem, but the primary reason timezones exist isn't inertia, it's that they continue to be relevant and useful. Because people continue to keep hours based on their local conditions. I'm skeptical that the juice is worth the squeeze as far as eliminating the small overhead caused by political boundaries.
3 hours is one thing, 12 hours is another (and "So you want to abolish timezones" notes that people in China commonly use lookup tables, creating implicit timezones).
When I coordinate across timezones, I say, "Hey does 9amPT/11amET work?" (Assuming those are the relevant timezones). I don't see the problem with that approach?
Let's say you abolish timezones. Okay. How do I know whether it's rude to call Uncle Steve in Melbourne? I'm going to consult some kind of table and implicitly reintroduce timezones, right?
Because timezones require conversion/coordination by either you or the other party every time.
Conversely learning when your Australian co-worker or uncle gets off work (in UTC) must only be done once.
e.g. You learn that Uncle Steve in Melbourne gets off work around 07:00 UTC. So a reasonable time to call him and chat about life is 08:00 UTC. You text him and say "hey, long time no chat, want to catch up at 08:00?". He doesn't have to look anything up because he is also using UTC (or whatever). He says "actually I have a dinner thing, what about 11:00 instead?". Know confusion can be created, because this is the same time frame of reference for you too.
Yep, you have to be explicit when coordinating across timezones. I never say "how about 11" in that context, I say "how about 11ET." If you want to do that by using UTC as a lingua franca and everyone does their own conversion, that makes a lot of sense. If you want to abolish timezones, that's just brushing the problem under the rug.
People keeping different hours in different locales is inherent complexity. This is the crux of it. If you want to abolish timezones, you need to explain why it isn't a problem that people keep different hours in different locales. If you can't, then it boils down to timezones being annoying. They are annoying, but that's not sufficient reason to eliminate them.
Couldn’t I just learn that Uncle Steve gets off work at 01:00 CST, by figuring out his offset from UTC and combining it with mine?
Your solution doesn’t change that I have to figure out some “how many hours different is Uncle Steve”, just that somehow I’ll magically figure that out without using the simple UTC offsets we use to denote those differences now. It just papers over them.
Both scenarios require learning, only one scenario requires constant back-and-forth conversion, which in turn introduces opportunities for mistakes and confusion.
Here is another example: I schedule a meeting with my co-worker in London. I say "how about 10am your time?" This is because I know he is X hours behind my timezone because I've learned that fact. He agrees and I show up an hour late because I forgot about DST and that he is only X-1 hours behind now.
If we both coordinated around UTC, he would've said "actually 11 is better for me" and I would've showed up an hour earlier.
I can't tell if your comment is intentionally ironic, as in you're saying we need to enforce your view on others that enforcing their view on others is bad.
If this person was vandalizing cultural treasures or something to pressure everyone to use UTC, then yes I agree his forced technocratic fix to a social issue is bad. Just writing about it and sharing an opinion is actually very constructive, even if you disagree with his take.
An opinion becomes something a little more deserving of criticism when it’s saying things like “it should be abolished and everyone should do what I say.”
This wasn’t phrased like “I prefer to use UTC time and here’s why I like it,” it’s more like if this guy becomes king of the world we’ll literally be forced to used UTC time.
This is why the person you replied to said:
> it would be great if we could figure out a way to prevent it, and teach that in schools. Too many people like this in the world today.
Too many people think that constructively sharing a personal opinion and sharing a desire to impose restrictions on others are of equal merit.
> An opinion becomes something a little more deserving of criticism when it’s saying things like “it should be abolished and everyone should do what I say.”
I must have missed that quote, I seem to be unable to find that text in the post.
> This wasn’t phrased like “I prefer to use UTC time and here’s why I like it,”
I don't know, it kind of seems like it was:
> Here’s Adam’s story of how living by UTC transformed his productivity, and why it might work for you, too.
and
> Give it a try, you might find it as liberating as I do.
They're making it sound awfully optional if their intent was to put you in jail if you didn't do it.
> Here’s Adam’s story of how living by UTC transformed his productivity, and why it might work for you, too.
> Give it a try, you might find it as liberating as I do.
Did you miss these quotes, or are they also part of the absolutism? Here are some ways I read it, which seem like fair interpretations and also don't make the author stand out as being particularly dogmatic or authoritarian:
Considering the quotes I provided, it seems likely that the author may have been suggesting that we stop using timezones as a society in order to mitigate the only two downsides that they listed, since those downsides only exist as a result of society still using timezones while they as an individual stopped using them.
Another way to read it could be that the author made a statement they believe is probably true ("timezones should be abolished") but they were not fully convinced. To gain confidence, they decided to gather further information: "I began with myself...To my surprise, it was easy" and then ended with two downsides of their experiment, suggesting that they realize it may not be a perfect solution, at least right now.
A third way one could read it is to consider that the author didn't propose fines, jail-time, or any other specific penalties for continued use of timezones in their hypothetical world. Without enforcing a penalty, what does it even mean to "abolish" a timezone? In the context of the whole post, it seems to be more of a thought experiment imagining what could be if society as a whole switched, rather than a prescription of what must be.
The unwillingness to engage with the author's thought experiment strikes me as more dogmatic than anything the author wrote, but perhaps I am not fully considering other possible interpretations.
Using UTC for everything strikes me as attempting to apply thinking that's convenient for machines to humans.
In the vast majority of cases, the most salient property of a timestamp, for humans, is what point it occurred in the 24-hour day/night cycle (and, secondarily, where it occurred in the 7-day week cycle).
Local time is an approximation of exactly that 24-hour cycle; the local time anywhere is approximately the elapsed time since the last solar midnight in that location. (Yes, it's not perfect, because in some places local time has a persistent offset from solar time, or even one that changes twice a year, but it's close enough).
UTC is an approximation of the elapsed time since the last solar midnight in England. Most people don't live in England, so this just isn't relevant most of the time.
If we all used UTC, sure, people living in one place would get used to the new correspondence to solar time; someone who spends all their time in Arizona would quickly get used to the fact that they now eat dinner at 12:00 UTC instead of 7pm. But it would make traveling more tedious: you'd have to re-learn the mapping every new place you went. It'd also make communicating with people abroad more difficult. If someone tells you "I went to bed at 08:00 last night", you have to know that in New Zealand that means they're an early riser or were sick; in New York it means they had a wild party and in Poland it means they worked the night shift.
There are a few cases where having a shared absolute time reference is useful; for example, scheduling meetings with people in many different countries, but in those cases people tend to spontaneously settle on a time standard and it doesn't cause many problems in practice. And even in many of those cases, the local solar time is still relevant (you wouldn't want to schedule a meeting for someone around their solar midnight), so you have to have some way of expressing it anyway.
Right. The 7-day week cycle is also completely arbitrary, perhaps the most arbitrary of all time duration units, but having something of that nature has proven useful.
The 7-day week originated in ancient Mesopotamia, particularly Babylon, around the 3rd millennium BCE. It was tied to the observation of seven celestial bodies visible to the naked eye: the Sun, Moon, Mercury, Venus, Mars, Jupiter, and Saturn. Each day was associated with one of these "planets," a system later adopted and refined by other cultures.
The Babylonians passed this concept to the Jews, who formalized it in their religious practices, linking it to the biblical creation story in Genesis, where God created the world in six days and rested on the seventh, the Sabbath. This Jewish tradition influenced the structure of the week in the Western world.
The Romans also adopted the 7-day cycle by the 1st century CE, replacing their earlier 8-day market cycle (nundinae). They named the days after their gods, aligned with the celestial bodies: dies Solis (Sun), dies Lunae (Moon), dies Martis (Mars), dies Mercurii (Mercury), dies Iovis (Jupiter), dies Veneris (Venus), and dies Saturni (Saturn). These names persist in Romance languages (e.g., French: lundi for Moon, mardi for Mars).
The 7-day week spread through the Roman Empire and was cemented by Christianity, especially after Emperor Constantine made Sunday a day of rest in 321 CE. Other cultures, like the ancient Egyptians and early Chinese, had different cycles, but the 7-day week became dominant globally due to Western influence, trade, and colonization.
It’s not perfectly aligned with natural cycles—365.25 days in a year don’t divide evenly by 7—but its cultural and religious roots have made it a near-universal standard.
The symptoms could be connected to emptyer wallets, because on workdays people earn, on weekends people spend. It would take time and shifting of minds to adapt countrys economics to, for example, three day weekend.
> someone who spends all their time in Arizona would quickly get used to the fact that they now eat dinner at 12:00 UTC instead of 7pm. But it would make traveling more tedious: you'd have to re-learn the mapping every new place you went.
Following local times won't spare you from having to learn local culture anyway, such as Spanish dinner times ranging around 21-23.
That’s true, but you still have to know much less information. In most cultures, most people eat the main evening meal somewhere between 18h and 23h; Southern Europe tends later in that range than most places, whereas the US tends earlier. This is still a lot simpler than each longitude having its own unique offset that can range anywhere from -12 to 12.
If you visit Spain as a tourist and eat at 19:00 local time every day, you’ll be slightly out of sync with the locals but you aren’t going to cause any huge drama. If you never even learned the fact that Spanish people eat later, you’d still basically be fine. So knowing that cultural fact about Spain isn’t absolutely essential the way it would be to learn the offset of each place in a world without time zones.
I think that if you want to use UTC, you should make it clear by adding "Z" (or "UTC") on the end of the numbers which tell you what time it is. This makes it more clearly, avoiding some of the confusion they mention in that article.
However, I think that local time should be used as well as UTC, for different purposes, but that the local time should just use solar time, rather than using time zones and DST.
You can’t have standard seconds with solar time though (or minutes or hours), unless you live at the equator [0]. So it will be interesting to calculate at what local time your CI build is expected to complete, or if you have enough time to watch that 120-minute movie before whatever the equivalent of 10 o’clock would be in solar time.
Or alternatively, each day would have to end at a different time, but then 12 o’clock wouldn’t be solar noon (or midnight).
Solar time doesn't involve different lengths of seconds. The length of day and all the pieces doesn't change. Local solar time is effectively time zone on longitude. It is mean solar time.
The downside of solar time is that 1 minute of time is 15 arc minutes which is roughly 17 mi east west. Which means need to deal with time zones in same metro area.
The railroads invented time zones because dealing with local solar time was a pain. Minute accuracy is needed for time tables. We might be able to do it with computers, but it would require knowing the location of everything.
When the minute accuracy is needed between different places, that is when UTC would be used. It is helpful for travel, and for many other purposes.
For local stuff that does not involve precise measurement of timing, you can use solar time (and you can use a sundial), and you can do by the sun light (even, in summer it is more light and in winter it is dark early, you can use that, e.g. to go outside when it is light and don't need to turn on light inside, and in night time is dark you can sleep), and by the moon light (less commonly, but sometimes it is also meaningful and useful).
Even for things that do use precise measurement of timing, UTC is not always appropriate; for some uses you will use TAI, or SI seconds instead of UTC seconds.
UTC seconds are the same as SI seconds. Or rather, there is no defined term "UTC second". You may be thinking of seconds-since-epoch counts that don’t include leap seconds, like in POSIX timestamps. That's not UTC though. UTC is a naming scheme for points in time, not a prescription for how to count seconds.
What's true is that the exact number of seconds between future UTC times isn't known, because leap seconds aren't known arbitrarily in advance. But that doesn’t usually affect measurements of time. It does affect adding (large) SI-second counts to the current time to compute a future time. The exact UTC designation of that future time may not be known in advance.
Nothing is more rad than checking the dot-beat on my translucent plastic watch while waiting for dial-up to connect so I can update my ICQ status to let my global friends know I’m listening to “Barbie Girl” from a 128k MP3 in WinAMP. Pre-emptive multitasking baby.
I like to think .beats might've caught on, if they were UTC-based instead of UTC+1, because Swatch is headquartered in Switzerland, and I think that little offset made it obviously tongue-in-cheek.
I'm not quite sure how you're reading it, but what they mean is 'the reason to have different offsets is so that morning/sunrise happens circa 8am everywhere in the world (in local times), but why is that a goal - why care if it's circa 8am or circa 9pm or anything else, it can be at different times in different places instead of them having different offsets to make it the same'.
After an eight-hour flight, if you went to bed at 21:00 yesterday, you’ll do the same today.
This will not work except if you are completely shifted with the local time when you travel.
If you change of timezones, the sun will not raise and go down at the same time in UTC obviously. So you will go to sleep in the middle of the day.
Also, when you are at home, everything will be in local time anyway, you will not see the difference, and going abroad you will constantly have to do conversion in your mind from local time to UTC otherwise you will not know when shop open and close, the train will leave, ...
You can complain and work on improving a tool that will covert badly, but for me it is kind of transparent with recent technology. Phone and watch update themselves automatically when you leave an airplane in a different time zone country and most tools look at your location to show you relevant time.
If you say that it will make it more convenient to work with people in other timezones, that's not even the case, because UTC will not tell you if a time is a good time or not regarding local time. So someone giving you his country/timezones is enough for you to select a right time.
Let's imagine I'm working with you that is located a quarter of the world away, how do I know in UTC what time is the work day morning proper time for a meeting? 11h UTC? 22h UTC? 4h UTC?
It's a little different when going to bed one day and waking up another doesn't happen anymore. And small numbers aren't morning, big numbers aren't night.
I hate this form of nerd clickbait. tech bro discovers UTC, thinks he’s cracked the code of the universe, and writes a navel-gazing manifesto about how changing his watch changed his life. of course it ends in a SaaS plug.
I’m convinced we need to move to a world time. The Internet is part of life and we have people coordinating around the globe. Time is time everywhere.
Sunlight is different everywhere. But not everything timed is tied to sunlight. There is no reason that the time should be 12:00 at peak sunlight locally.
UTC is an issue near the international date line. I could see residents of Pacific islands wanting to not have the date change mid-day. Should think about adding hours 24, 25, 26, etc. to Wednesday that are equivalent to hours 0, 1, 2, etc. on Thursday.
(And as a quant who did a lot of time programming for financial markets, fuck Daylight Saving Time and the leap second.)
The way I see it is you are deciding between making the instant very clear or making their "time of day" clear.
But IMHO "time of day" is less important because 1. People in the same time zone still have different schedules so it is an approximation anyways. 2. If you know what instant is being discussed it is very easy for each person to relate it to their personal schedule.
So I agree that switching costs aside using UTC would be better overall.
201 comments
[ 2.9 ms ] story [ 225 ms ] threadIf someone says they didn't get home until late, you know they probably mean 9 to some small hour of the morning etc.
And it's an idea that I see come up not too rarely on any post about time subtleties (time zones, leap seconds, etc.)
If you're scheduling something with someone non-local to yourself, you need not only know their time zone, but you also need also consider their latitude/longitude and what time of year it is so you account for the potential use of daylight saving time and their current sunrise/sunset times.
For example: if you schedule something at 7PM - is that daytime or nighttime? Well, if it's the first of October, and the person is in Kansas City, the answer is daytime. However, if they're in Phoenix, the answer is nighttime. And here's the kicker: the fact that Kansas City uses daylight saving time is irrelevant, as it would still be mostly light out even if you used standard time due to its higher latitude.
But if you try to schedule a meeting with someone they will immediately know if the time is reasonable for them and let you know.
Plus I don't see how it actually makes a difference, you just look up "what are business hours for $city" rather than "what time is it in $city".
Well said. Ill use this when explaining to my north american friends how 17:00 is 5pm in my head without doing any math.
That said the whole conversion thing is not an immaterial burden unless you’re really up in the sky longer than walking on earth.
The whole point of UTC is to have a shared reference baseline to communicate with the world; however the author is obviously not concerned about that, I don’t think, as they have to do the conversion into the third party’s local time in their head if they want to have any friends, so that point is moot.
> After an eight-hour flight, if you went to bed at 21:00 yesterday, you’ll do the same today.
The problem is, depending on where you landed, 21:00 could be typical start of the workday, and you'd be showing up to an empty office.
> If you had a call at 13:00 yesterday, it’s still at 13:00 today
This part sounds nice, until DST comes along to sabotage you. For only being twice a year though, it's probably still a net gain, one I hadn't really considered for someone who travels a lot.
That's how I have conky setup on my desktop. "Year : Julian Day : HH:MM:SS" in both local and UTC. With a fixed width font and the two right above each other it's exceptionally useful. It makes looking at logs captured in UTC much easier to parse.
Because it was very easy to miss a little "PM" pip in the corner, when reading or setting the time; and setting an alarm depended on getting the pip correct, so eliminating the pip for me meant eliminating ambiguity in the digits.
It's been a fairly successful transition, although sometimes I look at "16:15" and think it's 6pm so the "mentally subtracting 12" needs practice. And there are still special snowflake UI widgets that ignore my i18n settings, and constrain how I enter or specify a time.
https://web.archive.org/web/20211022024626/https://thehtime....
ZA:03 ZB:23 ZL:55 ZI:12 ZO:08
would work greatly for me
I like the concept of a global time that maps to local time, and while I am unconvinced that Latin letters are the best way to do that, I can’t think of a better way. It’s either letters of some sort, symbols, or names, and none of those sound good.
That's because you committed one of the capital sins for a programmer: you tried working with time. The old mantra "use a library" still rules.
> I decided time zones should be abolished, and everyone should use one coordinated time. Why care if the sun rises at 8 or 21?
You seriously need to go out more often.
Pretty much that. Most mammals I can think of live by what the sun, one way or another.
Source: https://pmc.ncbi.nlm.nih.gov/articles/PMC4183310/#:~:text=Gl...
Some people just like to add complexity to their lives and keep convincing themselves that it's better.
Anecdotally, I worked on a military facility where all clocks were in "master timezone" of the country without daylight saving offset. 6 months a year we put "+1" sign made of electric tape to not forget about that it's summer and 0400 is actually 0400+TZ+1. I worked there for 5 years and never got used to it.
1) global time. i.e. UTC
2) local “sun” time. i.e. local timezone
When scheduling a meeting you are trying to find an intersection of global and local time that works.
It’s messy but also our physical reality. The conversion still needs to be done, just in the article the person is living in global time and having to convert to local time when needed. Vs living in local time and converting to another different local time.
Unlike DST today, time zones do still have economic utility in the modern world since there are many more personal and industrial activities that exclusively happen during day or night in a single locale and far fewer that happen 24x7 and routinely traverse multiple locales.
5 days a week I am working with people in Seattle, NYC, Munich, and often London or India. I have to coordinate meetings across time zones a lot. I travel a lot.
And yet at home or on the road I have to process things like "this store closes at 18:00" or "this train is at 16:45" infinitely more. Putting my devices in UTC requires me to think for all of those interactions which is way worse.
Problem is though with distributed offices and remote workers. “The servers started crashing at 10:31 can you look at it?” Doesn’t mean anything without more context. If it’s 10:50 local does that mean it started 20 minutes ago or 7:20 ago? 3:20 ago because they’re on the east coast?
Being able to express where you are in the 24-hour cycle is useful. That's what local time is: an approximation of the time since the last solar midnight.
To the extent that the approximation is poor, it's an argument for improving it (Spain for example should really set their clocks back an hour), not for getting rid of it entirely.
Besides, Kelvin is the most true temperature scale, not Celsius.
I find this considerably more useful than the Fahrenheit equivalent.
That way you have the "human livable" range kinda between 0-100, which feels very intuitive. Anything above 100ºC becomes effectively unlivable. It also means that it is much easier to distinguish between certain "zones". e.g. saying "70s" or "80s" is easier and more clear than with celsius where you typically are staying within a sliding 10º range from day-to-day.
But why? Knowing the freezing point of water (at sea level, mind you) is only rarely useful to me in daily life. I think the eutectic point of salt and water (that is, the lowest point at which it's possible to cause ice on the roads to melt by adding salt to them) is marginally more useful; this happens to be about -6F or about -21 C. Perhaps it's a bit more important for farmers who need to know whether their crops will freeze overnight, but as a lifelong city/suburbs dweller who knows next to nothing about agriculture I couldn't really say. At any rate, most people are not farmers.
I'm asking this somewhat rhetorically: I don't really care about Celsius vs. Fahrenheit. The pros and cons between the two are so vastly outweighed by the utility of having a global standard that I really wish the US would switch. I'm just pointing out that I don't think that one particular physical property of one particular substance makes for a very strong argument either way.
Or: "below 0C my plants might die"
Or: "below 0C the pond will start to freeze over"
the 60's are warm in the sun and cool in the shade or breeze, the 70's are warm, the 80's are hot and the 90's are really hot, and over a hundred you're looking at death. The 50's you need a jacket, the 40's you need more jacket, the 30's are downright cold and you need an overcoat, and the 20's are frigid.
Celsius has approximately none of those properties. Celsius the 20's encompasses a vast change in comfort level.
https://xkcd.com/1982/ Had it entirely right.
- ex Celsius user
So You Want To Abolish Time Zones - https://qntm.org/abolish
I think folks are averse to "world time" (for reasons, largely inertial), so maybe the baby step is try 1 timezone per country (like China's done for .. 75 years ?).
I'd even argue Local Time has only ~4 useful times: dawn, daytime, dusk, & nighttime. Where I grew up, the parks closed at dusk every day. Nobody complained
If I take a walk at 7:05:35 PM every day, it seems very precise but doesn't indicate whether I need sunglasses or a flashlight. It's meaningless precision, like 0.6235 slices of pizza. If I'm coordinating a walk with you, I might as well use UTC: it still won't tell light from dark, but at least nobody'll be waiting for an hour due to DST. It'd make more sense to schedule our walk at `1 hour before dusk`, or "just" settle for UTC, IMO.
China did that for political reasons. The Republic of China used five, and Russia uses two timezones across those longitudes. Similarly, I also find it quite weird that France and Spain are in CET. France shifted when it was occupied in WWII, but maybe it's justified to remain in CET to reduce friction with the economies of its eastern neighbors. Whatever. Vive la weird. Countries do that to themselves.
A deviation of half an hour from noon is barely perceptible unless you use a sunclock. People can roughly estimate when the sun will set by just looking at local time and considering the current season. What throws a wrench in the works is daylight saving time. I fully agree that DST a huge annoyance and that its benefits were always rather situational.
It reads as tongue-in-cheek to me :)
The existing system is more inertial and political than practical.
"Noon is when the sun is at its zenith in your area" is significantly less useful in 2025 than being able to coordinate globally with ease. It's also just not true in many (most?) cases, as noted in the article, where places like China have a single timezone even though there is a 3-hour discrepancy in the sun's position from one side of China to the other.
3 hours is one thing, 12 hours is another (and "So you want to abolish timezones" notes that people in China commonly use lookup tables, creating implicit timezones).
When I coordinate across timezones, I say, "Hey does 9amPT/11amET work?" (Assuming those are the relevant timezones). I don't see the problem with that approach?
Let's say you abolish timezones. Okay. How do I know whether it's rude to call Uncle Steve in Melbourne? I'm going to consult some kind of table and implicitly reintroduce timezones, right?
Conversely learning when your Australian co-worker or uncle gets off work (in UTC) must only be done once.
e.g. You learn that Uncle Steve in Melbourne gets off work around 07:00 UTC. So a reasonable time to call him and chat about life is 08:00 UTC. You text him and say "hey, long time no chat, want to catch up at 08:00?". He doesn't have to look anything up because he is also using UTC (or whatever). He says "actually I have a dinner thing, what about 11:00 instead?". Know confusion can be created, because this is the same time frame of reference for you too.
People keeping different hours in different locales is inherent complexity. This is the crux of it. If you want to abolish timezones, you need to explain why it isn't a problem that people keep different hours in different locales. If you can't, then it boils down to timezones being annoying. They are annoying, but that's not sufficient reason to eliminate them.
Your solution doesn’t change that I have to figure out some “how many hours different is Uncle Steve”, just that somehow I’ll magically figure that out without using the simple UTC offsets we use to denote those differences now. It just papers over them.
Here is another example: I schedule a meeting with my co-worker in London. I say "how about 10am your time?" This is because I know he is X hours behind my timezone because I've learned that fact. He agrees and I show up an hour late because I forgot about DST and that he is only X-1 hours behind now.
If we both coordinated around UTC, he would've said "actually 11 is better for me" and I would've showed up an hour earlier.
If this person was vandalizing cultural treasures or something to pressure everyone to use UTC, then yes I agree his forced technocratic fix to a social issue is bad. Just writing about it and sharing an opinion is actually very constructive, even if you disagree with his take.
This wasn’t phrased like “I prefer to use UTC time and here’s why I like it,” it’s more like if this guy becomes king of the world we’ll literally be forced to used UTC time.
This is why the person you replied to said:
> it would be great if we could figure out a way to prevent it, and teach that in schools. Too many people like this in the world today.
Too many people think that constructively sharing a personal opinion and sharing a desire to impose restrictions on others are of equal merit.
I must have missed that quote, I seem to be unable to find that text in the post.
> This wasn’t phrased like “I prefer to use UTC time and here’s why I like it,”
I don't know, it kind of seems like it was:
> Here’s Adam’s story of how living by UTC transformed his productivity, and why it might work for you, too.
and
> Give it a try, you might find it as liberating as I do.
They're making it sound awfully optional if their intent was to put you in jail if you didn't do it.
“Five years ago, I decided time zones should be abolished, and everyone should use one coordinated time.”
That quote is also in the article.
> “it should be abolished and everyone should do what I say.”
The original and the (mis)quote are not the same thing at all.
> Give it a try, you might find it as liberating as I do.
Did you miss these quotes, or are they also part of the absolutism? Here are some ways I read it, which seem like fair interpretations and also don't make the author stand out as being particularly dogmatic or authoritarian:
Considering the quotes I provided, it seems likely that the author may have been suggesting that we stop using timezones as a society in order to mitigate the only two downsides that they listed, since those downsides only exist as a result of society still using timezones while they as an individual stopped using them.
Another way to read it could be that the author made a statement they believe is probably true ("timezones should be abolished") but they were not fully convinced. To gain confidence, they decided to gather further information: "I began with myself...To my surprise, it was easy" and then ended with two downsides of their experiment, suggesting that they realize it may not be a perfect solution, at least right now.
A third way one could read it is to consider that the author didn't propose fines, jail-time, or any other specific penalties for continued use of timezones in their hypothetical world. Without enforcing a penalty, what does it even mean to "abolish" a timezone? In the context of the whole post, it seems to be more of a thought experiment imagining what could be if society as a whole switched, rather than a prescription of what must be.
The unwillingness to engage with the author's thought experiment strikes me as more dogmatic than anything the author wrote, but perhaps I am not fully considering other possible interpretations.
https://en.wikipedia.org/wiki/Swatch_Internet_Time
https://en.wikipedia.org/wiki/International_Atomic_Time
In the vast majority of cases, the most salient property of a timestamp, for humans, is what point it occurred in the 24-hour day/night cycle (and, secondarily, where it occurred in the 7-day week cycle).
Local time is an approximation of exactly that 24-hour cycle; the local time anywhere is approximately the elapsed time since the last solar midnight in that location. (Yes, it's not perfect, because in some places local time has a persistent offset from solar time, or even one that changes twice a year, but it's close enough).
UTC is an approximation of the elapsed time since the last solar midnight in England. Most people don't live in England, so this just isn't relevant most of the time.
If we all used UTC, sure, people living in one place would get used to the new correspondence to solar time; someone who spends all their time in Arizona would quickly get used to the fact that they now eat dinner at 12:00 UTC instead of 7pm. But it would make traveling more tedious: you'd have to re-learn the mapping every new place you went. It'd also make communicating with people abroad more difficult. If someone tells you "I went to bed at 08:00 last night", you have to know that in New Zealand that means they're an early riser or were sick; in New York it means they had a wild party and in Poland it means they worked the night shift.
There are a few cases where having a shared absolute time reference is useful; for example, scheduling meetings with people in many different countries, but in those cases people tend to spontaneously settle on a time standard and it doesn't cause many problems in practice. And even in many of those cases, the local solar time is still relevant (you wouldn't want to schedule a meeting for someone around their solar midnight), so you have to have some way of expressing it anyway.
The 7-day week originated in ancient Mesopotamia, particularly Babylon, around the 3rd millennium BCE. It was tied to the observation of seven celestial bodies visible to the naked eye: the Sun, Moon, Mercury, Venus, Mars, Jupiter, and Saturn. Each day was associated with one of these "planets," a system later adopted and refined by other cultures.
The Babylonians passed this concept to the Jews, who formalized it in their religious practices, linking it to the biblical creation story in Genesis, where God created the world in six days and rested on the seventh, the Sabbath. This Jewish tradition influenced the structure of the week in the Western world.
The Romans also adopted the 7-day cycle by the 1st century CE, replacing their earlier 8-day market cycle (nundinae). They named the days after their gods, aligned with the celestial bodies: dies Solis (Sun), dies Lunae (Moon), dies Martis (Mars), dies Mercurii (Mercury), dies Iovis (Jupiter), dies Veneris (Venus), and dies Saturni (Saturn). These names persist in Romance languages (e.g., French: lundi for Moon, mardi for Mars).
The 7-day week spread through the Roman Empire and was cemented by Christianity, especially after Emperor Constantine made Sunday a day of rest in 321 CE. Other cultures, like the ancient Egyptians and early Chinese, had different cycles, but the 7-day week became dominant globally due to Western influence, trade, and colonization.
It’s not perfectly aligned with natural cycles—365.25 days in a year don’t divide evenly by 7—but its cultural and religious roots have made it a near-universal standard.
I've known multiple people in my life whose working shifts do not correspond to the seven-day week.
I know of no illness from which they suffer.
No other primate cares a whit. No other animal cares a whit.
Citations would be very useful here, otherwise it sounds like another random thing from minds unknown.
Following local times won't spare you from having to learn local culture anyway, such as Spanish dinner times ranging around 21-23.
If you visit Spain as a tourist and eat at 19:00 local time every day, you’ll be slightly out of sync with the locals but you aren’t going to cause any huge drama. If you never even learned the fact that Spanish people eat later, you’d still basically be fine. So knowing that cultural fact about Spain isn’t absolutely essential the way it would be to learn the offset of each place in a world without time zones.
However, I think that local time should be used as well as UTC, for different purposes, but that the local time should just use solar time, rather than using time zones and DST.
Or alternatively, each day would have to end at a different time, but then 12 o’clock wouldn’t be solar noon (or midnight).
[0] http://wordpress.mrreid.org/2014/10/19/rate-of-change-of-day...
The downside of solar time is that 1 minute of time is 15 arc minutes which is roughly 17 mi east west. Which means need to deal with time zones in same metro area.
The railroads invented time zones because dealing with local solar time was a pain. Minute accuracy is needed for time tables. We might be able to do it with computers, but it would require knowing the location of everything.
For local stuff that does not involve precise measurement of timing, you can use solar time (and you can use a sundial), and you can do by the sun light (even, in summer it is more light and in winter it is dark early, you can use that, e.g. to go outside when it is light and don't need to turn on light inside, and in night time is dark you can sleep), and by the moon light (less commonly, but sometimes it is also meaningful and useful).
Even for things that do use precise measurement of timing, UTC is not always appropriate; for some uses you will use TAI, or SI seconds instead of UTC seconds.
What's true is that the exact number of seconds between future UTC times isn't known, because leap seconds aren't known arbitrarily in advance. But that doesn’t usually affect measurements of time. It does affect adding (large) SI-second counts to the current time to compute a future time. The exact UTC designation of that future time may not be known in advance.
https://en.wikipedia.org/wiki/Swatch_Internet_Time
Nothing is more rad than checking the dot-beat on my translucent plastic watch while waiting for dial-up to connect so I can update my ICQ status to let my global friends know I’m listening to “Barbie Girl” from a 128k MP3 in WinAMP. Pre-emptive multitasking baby.
[1] https://en.wikipedia.org/wiki/The_Clock_Towers
For some reason no friends or family took me seriously when I brought it up at the time.
Isn't that sentence contradicting the entire thing? If you don't care, then use local time. Unless this is satire that I didn't get :P
Note I'm not saying I agree, just explaining.
You can complain and work on improving a tool that will covert badly, but for me it is kind of transparent with recent technology. Phone and watch update themselves automatically when you leave an airplane in a different time zone country and most tools look at your location to show you relevant time.
If you say that it will make it more convenient to work with people in other timezones, that's not even the case, because UTC will not tell you if a time is a good time or not regarding local time. So someone giving you his country/timezones is enough for you to select a right time.
Let's imagine I'm working with you that is located a quarter of the world away, how do I know in UTC what time is the work day morning proper time for a meeting? 11h UTC? 22h UTC? 4h UTC?
It's a little different when going to bed one day and waking up another doesn't happen anymore. And small numbers aren't morning, big numbers aren't night.
I hate this form of nerd clickbait. tech bro discovers UTC, thinks he’s cracked the code of the universe, and writes a navel-gazing manifesto about how changing his watch changed his life. of course it ends in a SaaS plug.
Sunlight is different everywhere. But not everything timed is tied to sunlight. There is no reason that the time should be 12:00 at peak sunlight locally.
UTC is an issue near the international date line. I could see residents of Pacific islands wanting to not have the date change mid-day. Should think about adding hours 24, 25, 26, etc. to Wednesday that are equivalent to hours 0, 1, 2, etc. on Thursday.
(And as a quant who did a lot of time programming for financial markets, fuck Daylight Saving Time and the leap second.)
But IMHO "time of day" is less important because 1. People in the same time zone still have different schedules so it is an approximation anyways. 2. If you know what instant is being discussed it is very easy for each person to relate it to their personal schedule.
So I agree that switching costs aside using UTC would be better overall.