[HN Gopher] Exploring 120 years of timezones (2021)
       ___________________________________________________________________
        
       Exploring 120 years of timezones (2021)
        
       Author : colinprince
       Score  : 77 points
       Date   : 2024-10-14 05:44 UTC (17 hours ago)
        
 (HTM) web link (blog.scottlogic.com)
 (TXT) w3m dump (blog.scottlogic.com)
        
       | docdeek wrote:
       | > 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!
        
       | captn3m0 wrote:
       | 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
        
       | thomasdziedzic wrote:
       | posted 2 days ago: https://news.ycombinator.com/item?id=41818394
        
         | cf100clunk wrote:
         | Originally in 2021:
         | 
         | https://news.ycombinator.com/item?id=28537438
        
       | ape4 wrote:
       | I imagine the Linux timezone database has grown a lot over the
       | years
        
       | ck2 wrote:
       | 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...
        
         | rogerrogerr wrote:
         | 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.
        
           | ck2 wrote:
           | US government has given spacex like $20 billion, I think you
           | misunderstand who controls who and how
        
             | euroderf wrote:
             | Are you ready for Trump Lunar Standard Time ?
        
       | thecosas wrote:
       | Makes my head hurt thinking about long timeframe datasets across
       | geographies and accounting for all of that as these continue to
       | shift.
        
         | cf100clunk wrote:
         | 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.
        
           | ghaff wrote:
           | 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
        
             | limitedfrom wrote:
             | 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/2013/11/30/24499
             | 5264/... [1] https://imgur.com/source-http-wp-me-poxsk-
             | pq-8IFLFoJ
        
           | growse wrote:
           | 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.
        
           | mattpallissard wrote:
           | 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.
        
       | romanhn wrote:
       | 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.
        
       | AStonesThrow wrote:
       | 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!
        
         | mtsolitary wrote:
         | 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.
        
           | jasomill wrote:
           | 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)
           | 
           | [1] https://www.wolframalpha.com/input?i=time+in+London+at+1%
           | 3A3...
        
       | dang wrote:
       | Discussed at the time:
       | 
       |  _Exploring 120 Years of Timezones_ -
       | https://news.ycombinator.com/item?id=28537438 - Sept 2021 (9
       | comments)
        
       | taeric wrote:
       | 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.
        
       | TypicalHog wrote:
       | Timezones are just so, so, so nasty. Why couldn't we just somehow
       | make out shit function on UTC or something?
        
       ___________________________________________________________________
       (page generated 2024-10-14 23:01 UTC)