Addressing Common Concerns with a World Without Timezones | Hacker News Reader

Addressing Common Concerns with a World Without Timezones

connorcallahan.info

33 pointsTheSenate45 comments

One problem will be deciding on when tomorrow is. If your work day starts at 9pm, you have lunch at midnight, do you then go home the next day? We could lose lots of complexity of timezones, but swap in the complexity of dates. Financial transactions and trading sessions could get really complicated fast.

A recent XKCD panel made that joke actually

https://xkcd.com/2542/

so just to play out the thought experiment here, what you could get is a new vocabulary for time abstraction. AM/Morning, PM/Night, mid-day, Mid-night are all relative to the observer not consistent to the standard clock time. Your early morning in NYC is still my evening in Japan, but they both happen at 09:00MST (Metaverse Standard Time)on Earth Cycle 245 of 365. The concept of days of the week and months of the year would also be subjective to the location of the observer - not part of the official Meta-calander which is just the same 365 units of 24hrs wherever you happen to be. What they call day-time Tuesday in Botswana is part of two date units, for some people hour 17 is always in the morning, and these local facts would get normalized as habit over time. An interesting thought experiment perhaps for some future inter-galactic post space expansion civilization's standards body (ISO Technical Committee 2.34 x 10^14)

Earth cycle 245 is one of the big problems in the modern world. When does it change over to 246? Do I wake up on the 245th but the date changes before lunch? What happens with billing systems, calendars, and what not?

Left/right, port/starboard, north/south are all in common use because we need to be able to talk about directions relative to different frames of reference. It’s just as true that we need to talk about time in multiple different frames of reference.

I think that where you live you will become accustomed to when the date changes. You'll know around what time of day it happens and you'll be fine.

It still seems less complex than tracking dates across time zones.

Or maybe you're just overestimating how much the average person needs to deal with cross-timezone events vs. things happening in the local timezone and things would be worse after all.

Although I get the appeal of getting rid of time zones and agreeing on one universal time zone, I am not sure if the pros outweigh the cons here.

Having it be the next day when you wake up is really intuitive and kinda practical. Having 9.00 mean "some time during the morning" all around the world is also quite comfortable.

I think from the perspective of a travelling human time zones are more practical than not having time zones, especially since your modern devices automagically change them anyways. Such an "automagic" change could not happen when 9.00 suddenly means "middle of the night".

For non-traveling folks that coordinate via the net it could surely have it's merrits (which is why we have UTC on servers?), but even there I could imagine software could help.

E.g. calender tools could just have a tool that allows scheduling for international participants. You enter the locations and it shows you the different day cycles next to each other — because after all isn't this the issue? Finding a time where it is roughly day-like for all participants of the event?

Using a universal time zone wouldn't really help you there, because you would still need to know what 9.00 means in some other part of the world.

> What they call day-time Tuesday in Botswana is part of two date units, for some people hour 17 is always in the morning, and these local facts would get normalized as habit over time.

Isn't that in some ways just reintroducing time zones through the back doors?

I don't understand what problem he's trying to solve.

When I need to call my overseas colleague, I'll either need to look up his timezone to see what time it is in localtime to know if he's sleeping or eating lunch, or I'll have to look up a chart to know that it's 19:00UTC so it's typical lunchtime for him.

But when I travel, it's nice to look at any clock and know that it's almost dinner time without consulting a chart to figure it out based on the UTC time. Or when I'm watching a movie and the actor is in bed and the clock turns 06:00, I know he's been laying awake for most of the night, I don't need other cues to tell me what his normal sleeping time is.

I can understand the utility of UTC for coordinating events across the globe (which I do regularly when I look at logfiles), but using one global time seems like it would make things more complicated for more people than those that have things simplified.

He's trying to gain a minor benefit of cross-timezone synchronization at the expense of a major cost of disconnecting your lifelong sense of time from your daily activities. You would have to relearn what 3am meant in a practical sense (mid-day?, mid-of-night?) in your part of the world, which would be a big pain, but you could eventually do it. What would the big payoff be for that effort? 3am for you would be the same "instant" as 3am for someone in another timezone. Great, but you wouldn't know whether that instant would be their mid-day or mid-of-night, so you would have to look it up anyway to schedule the meeting, just as you would if you were on different timezones.

Then you move to another timezone and have to learn the meaning of 3am all over again.

Lots of pain, no net gain.

This solution was posted recently:

https://news.ycombinator.com/item?id=28928550

It maps GMT hours to A-Z letters (omitting I and O). The advantage is that there isn't any confusing overlap between common 'folk' understanding of what hours are what, and the new time values are easily distinguishable from traditional timestamps without need for a TZ marker.

In a way it would carve out a separate, global sense of time in our mental space for use when we do need to coordinate (or just want to opt out of DST shenanigans), and yet let notions like "12:00 means lunch" live on alongside each other.

I am on the US East Coast and have frequent meetings with London. They are sensitive around not scheduling meetings too early in my day - and likewise I am sensitive to scheduling meetings too late in their day.

The mental model would have to shift:

Old Way: Add/Subtract X hours and map it to a "9 to 5" schedule to determine whether its "early" or "late"

New Way: Map "early" or "late" to the new timezone and location of the person.

Seems like the new way is too much of a hassle.

That's always been my problem with this. For everything that timezones currently solve, they suggest using solar time instead. So time zones, with a new name.

Currently we reference local time for most things, and UTC when coordination is required. The suggestion appears to boil down to using UTC (or equivalent) as the default, and time zones for everything that requires coordination. Except renaming timezones so we can pretend they were solved.

Are these not strictly equivalent? The only argument I can see against it is inertia - the new way (by virtue of that same strict equivalence) is no better and we're all already used to the old way.

There are several things you have to take into account when designing a good timekeeping system for use by the general public for their ordinary personal and business timekeeping.

The ones relevant to timezones are that we live on an approximately spherical world whose rotation is not tidal locked to the star it orbits, its rotational axis is significantly out of its orbital plane, and we evolved to synchronize our biological rhythms to local sunrise and sunset.

As the article notes, timezones does not save us from having to take into account differences in longitude when tying to coordinate with someone who is not local. The properties of our world mentioned earlier ensure that.

But it would also mean that when we are visiting someplace at a significantly different longitude, we'd have to keep taking into account that difference when coordinating with the people there.

For example if where I'm from in "everybody uses the same time" world we have breakfast at 13:00 and lunch at 19:00 and I'm visiting a place where breakfast is at 3:00 and lunch is at at 9:00, and the people I'm visiting tell me we are going to a meeting at 10:00, I've either got to remember that lunch as at 9:00 here and so I should eat before the meeting, or I've got to remember that the place I'm at is 10 hours ahead of my home on its day/night cycle, so when they say something is at 10:00 it is in the part of their day/night cycle that corresponds to the day/night cycle at 20:00 at my home which I recognize as early afternoon so I should have lunch before the meeting.

With the current timezone system they have breakfast at 6:00 and lunch at 12:00 just like we do at my home. When I'm visiting and they want an after lunch meeting they will say 13:00 and I know that's just after lunch.

With the current system then I only have to worry about timezones when I'm coordinating something across timezones. With the no-timezone system I have to worry about time differences both when I'm coordinating something across a significant longitude different and when I'm visiting someplace a significant longitude away and trying to simple trying to coordinate things local to that area. That's more work than with the current timezone system.

A shorter version is that (fixed) timezones perhaps address some problems and create some new ones if everyone is sitting at their own local computer and/or making phone calls.

But eliminating timezones make it much more confusing for someone traveling. Yes, some local norms differ. Spanish dinnertimes is a commonly quoted example. But, in general, if you're in a city in a different timezone you have a pretty good idea about how the flow of the day goes with some adjustment for local norms.

“it feels weird to wake up at 3 pm and work from 5 pm to 1 am.“

It feels weird to start work on Monday and get off on Tuesday. idk how we’ll deal with words like “tomorrow”.

Edit: i guess night shift people deal with it. But having worked night shift I can’t help but remember that time really revolves around the dayshift workers. “Tomorrow” to your night shift peers meant one thing, but to the world at large I always was explicit with which day I was referring to.

I read the article and my concerns are still unaddressed. Like what's the point?

are you trying to say you don’t enjoy overengineering?

I don't enjoy overengineering when a (seemingly) simpler solution exists and is already deployed

This author must live on the internet and never have to go to the local pool, fly on a jet (“it’ll be 3 am when we arrive!!”), or do anything in the physical space. The number of local times transactions outweighs the number of remote transactions, within a locale, by far.

> These are simply numbers that we have collectively settled > upon.

Many a Utopian project began with a premise like this. You can say the same thing about spelling or the calendar, both of which "logically" need reforming.

Can we get new commentary from a person effectively living in a world without timezones, ie, China?

In practice the west of the country uses their own local time. Even if not official. But China is not so “wide”, so even then, the difference is just two hours.

One caveat with that is that a huge percentage of the Han Chinese population lives in the east of the country. The west makes up only 28% of the population, and a big percentage of that is minority populations.

So any response you get is likely going to come from people who would probably live in the same time zone either way.

Why stop there, we could adopt Unix millis…

https://www.reddit.com/r/linuxmemes/comments/qjo7vp/the_unix...

It seems much more reasonable to highlight the UTC offset more often in normal everyday clocks, so that everyone knows both the local & UTC time all the time - so when deciding on e.g. a meeting, everyone is ready to work with UTC instead of local.

I predict we'll drift into a dual time zone world, where we use both our local time zone and UTC, when each makes sense.

As the world globalizes, UTC will slowly increase market share for the foreseeable future.

For anyone having trouble understanding why this won’t take off, imagine if where you live, we called down “up” and up “down”, because on the other side of the world, up was called up and that’s the opposite of our down, but we want to remain globally consistent.

What possible kind of utility would this serve? And how much pain would it cause?

We DO have compasses with a "North" and "South" heading. And they are pretty useful sometimes.

Right, and those mean something already in a global sense. Nothing is being redefined there, changing it from a local coordinate system into a global coordinate system. Time is a local coordinate.

For nostalgia fans, some Swatch watches had this "internet time" gimmick at the end of the 90's: https://en.wikipedia.org/wiki/Swatch_Internet_Time

I'd be happy if daylight savings time was terminated. DST is fine if you're a kid, as a lifetime goes by before you gotta reset the clocks. But as an older person, changing the clocks is a constant nuisance.

Relevant reading as well: https://qntm.org/abolish

Getting rid of timezones would be a massive, international, almost fruitless undertaking.

Think about the effort it would take to make the US use metric coloquially, times 195, just to gain what, the certainty that there is a single timezone instead of several? Also, who gets to decide what time it "actually" is, and why? UTC has roots in 19th century british colonialism, which doesn't hold that much weight today.

If we can't coordinate a worldwide vaccination campaign to combat a demonstrably deadly virus, exactly what absolutely massive event would cause the world to agree on getting rid of a perfectly serviceable, if annoying, timekeeping strategy?

I've read a proposal for a system which has four timezones. It seems to be gaining traction, being the most widely known option that isn't associated with colonialism or anything similar. I was skeptical at first, but the guy behind it lays the whole plan out on his website in detail and I must admit he makes one hell of a case for it.

https://web.archive.org/web/20050829214306/http://www.timecu...

> UTC has roots in 19th century british colonialism

Being a Yank who should hate this by identity, I feel reasonably qualified to say I don’t care. UTC is a standard already, so if we’re standardizing (which I agree we shouldn’t) there is no point choosing another hub.

This is silly. We don’t need fewer time zones, we need more of them: one for each major metro. It’s not 1870 anymore. Time should jump by a few seconds every night at 3 a.m. so that sunrise is always at seven a.m.—no daylight savings nonsense. Humans would once again wake with the dawn, the way nature intended, which would be a boon to health and mental well-being. If you need to know the time in some other area, just ask Siri, Google, or Alexa.

> sunrise is always at seven a.m

Why is fixed time for sunrise more important than fixed time for sunset?

I don't care if it's dark in the morning, I'm just gonna sit in an office anyway. I'd rather have more sunlight in the evening when I can go outside.

If you look at .beat which is pretty much the same idea most common concerns are not addressed here.

How do you chose a meridian people think is fair? Swatches choice pissed people off.

How do you celebrate New Years Eve? You think people will support a system that tries to kill one of the year's biggest parties?

There's answers and counter answers, it's hard to model. Could Facebook just do it and become a defacto companion to standard time.... maybe.

He doesn’t address the inevitable objection that UTC is “western” centered.

I don't see how that can be an objection to using a single timezone globally? We don't have to choose UTC.

As a person who preemptively complains about other people theoretically complaining that something that hasn't happened is "western centered," I personally can't imagine not making this western centered.

Practically speaking, you're picking GMT or probably ES(D)T (NYC). And GMT is probably a slightly better choice for both historical reasons and because it's closer to more land timezones today.

So? This would be an objection no matter where it was centered. Center it on Antarctica and someone else will find that objectionable.

Reply on news.ycombinator.com