This is always a fun discussion, as the reality is the local time makes enough sense for people to just use it. Having to deal with times across the globe is the anomaly, not the norm. Such that it is unsurprising that it is difficult to do.
Even more amusing, the time being fixed length days is definitionally not the norm. Is why the Pacific Northwest is miserable right now. Sun is barely up by 8, and largely down by 5 in the evening. Moving the agreed offset for when we should be awake, as we will do in a couple of weeks, does little to help.
And I want to double down on how I worded that. I often see it proposed that we should all move to UTC or some such, but just change when working hours are. But, largely, that is exactly what daylight saving's time is. Yes, it is couched in silly phrasing of "saving daylight," but it is functionally the same as everyone agreeing to move school hours by an hour.
I will also double down on my silliest of hot takes. With computers in control of most clocks, I think the problem with DST is that it moves everything by an hour. We could, with today's technology, move to something where the clock moves +10 minutes for 6 months and then -10 minutes for 6 months every year and nobody would even really notice.
This is unlikely. I work with people all over the world and generally we communicate our own time zone plus the top local time zone for our audience.
In an organization where people are accustomed to indexing their activities to other people who live in a different time zone, it’s actually easier to use the most common time zones than it is to switch everyone to UTC. When using UTC, you have to do the mental gymnastics twice. You first have to relate your own time zone to UTC, and second relate your audience time zone to it. But what happens in practice is that you quickly learn what times in your local time zone correspond to movements of the day in your peer time zones. And because everybody has the same mental model, it’s easier (and less error prone) to just to use local time zones.
I find I'll often try and use relative times in conversation, e.g. "after the stand-up" or "at the top of the hour" rather than specifying a wall clock time.
I'm actually surprised more chat programs don't automatically "internationalize and localize" dates. Seems the same effort they put into spotting a username could look for dates in the user's locale, and then render it in the appropriate one everywhere. We'd need some conventions to indicate it should not change the date, maybe. Most uses would probably be fine with not having that.
> In 1900 this dataset of 282 named timezones indicated 220 different offsets (from UTC), and while some of these were integers, (e.g. Europe/Prague, Europe/Rome), the majority were not - for example Moscow was 2 hours, 30 minutes and 17 seconds ahead.
Moscow changed after that to 2 hours, 31 minutes and 19 seconds for a few years before aligning with a ‘whole hour’ timezone. Fascinating - thanks for sharing!
My favorite video about the madness of timezones: https://youtu.be/-5wpm-gesOY. Many great examples, but the one that blows my mind is Israelis and Palestinians living in West Bank observing different timezones in the same geographic location.
Why does it require international agreement? This seems like the kind of thing even a private entity like SpaceX could just declare, and everyone else would fall in line naturally out of convenience and not-really-caring.
In the name of logic it seems to me that a worldwide paradigm shift away from human-(un)friendly time zones to UTC with 24 hour clocks would do wonders. We're stuck with our present system, and instransigence means we probably always will be.
I said for years that we should just use UTC everywhere and that people should wake up at offsets. One day my wife told me she was tired of hearing it and it was a stupid idea. That she could walk into nearly any town anywhere and know that at 9am she could walk into any bank, grocery, or diner.
Sure there are other ways we could communicate offsets, days, whatever, but having everyone on the same daily schedule and just adjusting the schedule itself keeps it simple for most of the population.
Those of us who travel in particular like to know that, absent some cultural anomalies like eating dinner in Spain, local time roughly corresponds to activities where we’re from without constantly doing mental conversions.
Yes, there are a lot of weird often politically-motivated time zones out there but for the most part it works
Except the perception of anomalous dinner time in Spain is also due to time zones. Spain switched to CET in 1940 to be in solidarity with Germany[0] (nowadays it will still translate to easier coordination with France, Germany, and Italy), meaning that despite much of the country being west of London, it shares the same time zone as Serbia and Poland. This means that if you live your normal life according to the sun, the clock time will 1 to 1.5 hours behind depending on where you are in Spain[1].
Ah, yes, the evergreen comment that suggests it would be better if people living far from the meridian experience the joy of Wednesday turning into Thursday at some point whilst they're awake.
I just had a huge and surreal argument with Google's Gemini about the current time in other time zones.
It was Sunday, 10/13. I asked what time it was in London:
It is currently 1:39 PM MST. As GMT timezone is 7 hours behind MST, the current time in London is 6:39 AM GMT on Sunday, October 13, 2024.
Well, no, it's not.
So I started grilling it on simple math ("What is 13:39 plus seven hours") and the real offsets of the time zones, and it would respond correctly, and then I'd ask for the current time again, it'd apologize profusely, and give me some B.S. calculations right on the same line!
I started inquiries about other places, like NZ and across the International Date Line, and it was spectacularly wrong. Some places like PDT it was OK. And it was repeatedly getting tripped up and apologizing and immediately supplying the wrong info again.
I was not trying to trick it; I was not going into ambiguous situations or half-hour zones. My locale doesn't observe DST, but it was clearly adapting for MST vs. DST zones.
The other comical thing was that Gemini would sometimes refuse outright to give the current time, referring me to other methods, as if there were some security blocks on that query. But only sometimes!
I simply must conclude that time zones are one of the most difficult programming problems. And I don't know how LLMs do time/date calculations. But Gemini clearly has a long way to go from these simple math confabulations!
WolframAlpha is my go-to for one-off calculations involving either durations and units.
For your example, the WolframAlpha query "time in London at 1:39 pm MST on October 13" returns the correct answer, "9:39:00 pm BST | Sunday, October 13, 2024"[1].
While WolframAlpha's natural language processing is hit-or-miss, its responses clearly state assumptions made in the face of ambiguity. In this example:
Assuming "MST" is a named time zone
Assuming "time" is referring to a calendar computation
Assuming month/day
Assuming Mountain Time (United States; no observed DST rule)
Assuming London (United Kingdom)
UTC is awful with its arbitrary, unpredictable leap seconds which basically no one cares about unless doing astronomy or navigation. Let’s just switch to TAI and be done with it!
Until we colonize Mars obviously… actually even the Moon has different G field so a different flow of time.
This is always a fun discussion, as the reality is the local time makes enough sense for people to just use it. Having to deal with times across the globe is the anomaly, not the norm. Such that it is unsurprising that it is difficult to do.
Even more amusing, the time being fixed length days is definitionally not the norm. Is why the Pacific Northwest is miserable right now. Sun is barely up by 8, and largely down by 5 in the evening. Moving the agreed offset for when we should be awake, as we will do in a couple of weeks, does little to help.
And I want to double down on how I worded that. I often see it proposed that we should all move to UTC or some such, but just change when working hours are. But, largely, that is exactly what daylight saving's time is. Yes, it is couched in silly phrasing of "saving daylight," but it is functionally the same as everyone agreeing to move school hours by an hour.
I will also double down on my silliest of hot takes. With computers in control of most clocks, I think the problem with DST is that it moves everything by an hour. We could, with today's technology, move to something where the clock moves +10 minutes for 6 months and then -10 minutes for 6 months every year and nobody would even really notice.
> Having to deal with times across the globe is the anomaly, not the norm
True, so far. But the world becomes a little more global every year, which makes the time zone friction grow.
My guess is that using both local and UTC time will become very common within 1-2 decades. As in "The raid starts at 11:00 PST/20:00 UTC".
This is unlikely. I work with people all over the world and generally we communicate our own time zone plus the top local time zone for our audience.
In an organization where people are accustomed to indexing their activities to other people who live in a different time zone, it’s actually easier to use the most common time zones than it is to switch everyone to UTC. When using UTC, you have to do the mental gymnastics twice. You first have to relate your own time zone to UTC, and second relate your audience time zone to it. But what happens in practice is that you quickly learn what times in your local time zone correspond to movements of the day in your peer time zones. And because everybody has the same mental model, it’s easier (and less error prone) to just to use local time zones.
Our team has members in every US timezone.
I find I'll often try and use relative times in conversation, e.g. "after the stand-up" or "at the top of the hour" rather than specifying a wall clock time.
I agree for simple cases like this.
But the more timezones the participants are in, the harder this approach gets.
I'm actually surprised more chat programs don't automatically "internationalize and localize" dates. Seems the same effort they put into spotting a username could look for dates in the user's locale, and then render it in the appropriate one everywhere. We'd need some conventions to indicate it should not change the date, maybe. Most uses would probably be fine with not having that.
> In 1900 this dataset of 282 named timezones indicated 220 different offsets (from UTC), and while some of these were integers, (e.g. Europe/Prague, Europe/Rome), the majority were not - for example Moscow was 2 hours, 30 minutes and 17 seconds ahead.
Moscow changed after that to 2 hours, 31 minutes and 19 seconds for a few years before aligning with a ‘whole hour’ timezone. Fascinating - thanks for sharing!
My favorite video about the madness of timezones: https://youtu.be/-5wpm-gesOY. Many great examples, but the one that blows my mind is Israelis and Palestinians living in West Bank observing different timezones in the same geographic location.
Needs a 2021 in the title. I was curious if it took into account the controversial changes in the tzdb from 2021 https://www.theregister.com/2021/09/28/time_zone_database_co... since those would have changed the pre-1970 data.
But the visualisation was made around Jul 2021, so it avoids those changes (but also the more recent ones): https://github.com/colineberhardt/timezone-viz
we are getting a moon time zone, maybe, eventually, requires international agreement so maybe never
https://www.npr.org/2024/04/09/1243405460/space-news-moon-na...
Why does it require international agreement? This seems like the kind of thing even a private entity like SpaceX could just declare, and everyone else would fall in line naturally out of convenience and not-really-caring.
US government has given spacex like $20 billion, I think you misunderstand who controls who and how
Are you ready for Trump Lunar Standard Time ?
Celebrating 120 years, plus or minus four hours, of timezones
Discussed at the time:
Exploring 120 Years of Timezones - https://news.ycombinator.com/item?id=28537438 - Sept 2021 (9 comments)
Makes my head hurt thinking about long timeframe datasets across geographies and accounting for all of that as these continue to shift.
In the name of logic it seems to me that a worldwide paradigm shift away from human-(un)friendly time zones to UTC with 24 hour clocks would do wonders. We're stuck with our present system, and instransigence means we probably always will be.
I said for years that we should just use UTC everywhere and that people should wake up at offsets. One day my wife told me she was tired of hearing it and it was a stupid idea. That she could walk into nearly any town anywhere and know that at 9am she could walk into any bank, grocery, or diner.
Sure there are other ways we could communicate offsets, days, whatever, but having everyone on the same daily schedule and just adjusting the schedule itself keeps it simple for most of the population.
Those of us who travel in particular like to know that, absent some cultural anomalies like eating dinner in Spain, local time roughly corresponds to activities where we’re from without constantly doing mental conversions.
Yes, there are a lot of weird often politically-motivated time zones out there but for the most part it works
Except the perception of anomalous dinner time in Spain is also due to time zones. Spain switched to CET in 1940 to be in solidarity with Germany[0] (nowadays it will still translate to easier coordination with France, Germany, and Italy), meaning that despite much of the country being west of London, it shares the same time zone as Serbia and Poland. This means that if you live your normal life according to the sun, the clock time will 1 to 1.5 hours behind depending on where you are in Spain[1].
[0] https://www.npr.org/sections/parallels/2013/11/30/244995264/... [1] https://imgur.com/source-http-wp-me-poxsk-pq-8IFLFoJ
Opinion seems divided on whether that’s the sole reason but it’s very probable it’s a major contributor.
Ah, yes, the evergreen comment that suggests it would be better if people living far from the meridian experience the joy of Wednesday turning into Thursday at some point whilst they're awake.
Easily solved by everyone in the world going to bed at 10pm UTC /s
https://qntm.org/abolish explores this idea in a fair amount of detail.
I just had a huge and surreal argument with Google's Gemini about the current time in other time zones.
It was Sunday, 10/13. I asked what time it was in London:
Well, no, it's not.So I started grilling it on simple math ("What is 13:39 plus seven hours") and the real offsets of the time zones, and it would respond correctly, and then I'd ask for the current time again, it'd apologize profusely, and give me some B.S. calculations right on the same line!
I started inquiries about other places, like NZ and across the International Date Line, and it was spectacularly wrong. Some places like PDT it was OK. And it was repeatedly getting tripped up and apologizing and immediately supplying the wrong info again.
I was not trying to trick it; I was not going into ambiguous situations or half-hour zones. My locale doesn't observe DST, but it was clearly adapting for MST vs. DST zones.
The other comical thing was that Gemini would sometimes refuse outright to give the current time, referring me to other methods, as if there were some security blocks on that query. But only sometimes!
I simply must conclude that time zones are one of the most difficult programming problems. And I don't know how LLMs do time/date calculations. But Gemini clearly has a long way to go from these simple math confabulations!
LLMs don’t do time/date calculations at all. They’re just good enough at predicting the next word to trick us into thinking they can.
WolframAlpha is my go-to for one-off calculations involving either durations and units.
For your example, the WolframAlpha query "time in London at 1:39 pm MST on October 13" returns the correct answer, "9:39:00 pm BST | Sunday, October 13, 2024"[1].
While WolframAlpha's natural language processing is hit-or-miss, its responses clearly state assumptions made in the face of ambiguity. In this example:
[1] https://www.wolframalpha.com/input?i=time+in+London+at+1%3A3...Thank you kindly!
It seems that WolframAlpha is venerable and stable. I first heard of it long, long ago. (Was it only released in 2009? Seems older.)
It must have amazing staying power. It seemed like a quite powerful tool. Why is it such a well-kept secret?
I imagine the Linux timezone database has grown a lot over the years
Timezones are just so, so, so nasty. Why couldn't we just somehow make out shit function on UTC or something?
UTC is awful with its arbitrary, unpredictable leap seconds which basically no one cares about unless doing astronomy or navigation. Let’s just switch to TAI and be done with it!
Until we colonize Mars obviously… actually even the Moon has different G field so a different flow of time.
posted 2 days ago: https://news.ycombinator.com/item?id=41818394
Originally in 2021:
https://news.ycombinator.com/item?id=28537438