I'm more perplexed that China decided it should only have 1 timezone. It also decided that timezone should be from the perspective of the eastern coast. When does the sun rise and fall for the western most people?
The sun rises and sets at different times according to both latitude and longitude. Timezones only handle longitudinal daylight adjustments. We already deal with plenty of incongruity around daylight.
Why not embrace the variability and let go of the idea that we can predict daylight based on the hour alone? Let people localize the time to discover where the sun is.
It does require a different way of thinking about time zones. It sounds complicated because it's unfamiliar, but if we ever make the transition, the system would be simpler, less ambiguous, and easier to reason about.
Your watch hands read '20230710T142847Z', don't they?
https://en.wikipedia.org/wiki/ISO_8601
One thing for which I would advocate is removing zero based numbering from hours and minutes. Clocks reading 12:15 for 00:15 is tremendously confusing. Most human oriented counting systems are not zero indexed. One indexed series align cardinal and ordinal numbering which reflects the passage of time rather than each hour being nominally disconnected from a sequence.
Apart from this being very funny and an interesting idea I have a question: how do you deal with leap years? Many important dates throughout the year (e.g. holidays, birthdays, business quarters etc.) are bound to specific days of months, do you just have to remember that all such days after February are one off sometimes in an ordinal date system?
Leap years: A sidereal year adjustment may occur on any day, there is not a requirement for it to occur on a specific day. Leap years merely have an additional day identifier in the series before returning to 001.
Anniversary events: Again, there is no requirement that the memorialization of a person or event occur on a specific month and day. Many occurrences such as Mardi Gras and Eid using a lunar calendar or the USA's Thanksgiving using the day of week.
If one wishes to commemorate an annual event at the same orbital position as when it occurred, then ordinal dates are more effective since the Gregorian calendar's fourth-year 02/29 throws off any anniversary after the first 1/6th of the year.
NB: ISO 8601 "leap week" instances occur at a different rate than leap years (a more parallel construction would be "leap week" and "leap day"). "[The] 53 week years occur on all years that have Thursday as the 1st of January and on leap years that start on Wednesday the 1st."
https://en.wikipedia.org/wiki/Leap_year
https://en.wikipedia.org/wiki/ISO_week_date
A higher order number than ordinal and cardinal is ratio. We can also express the current day as how far through the year we are. Since today is 192, it represents 52% of a year, the ratio of current day to days in the current year. Below is an example in bash.
$ date '+%j'
192
$ echo "scale=2; $(date '+%j') / $(date '+%j' -d $(date '+%Y')'-12-31')" | bc
.52Because it's a useful heuristic in knowing whether someone will be at work or not without having to localize every time.
having everyone roughly align their work day to 9am-5pm in their local time works a lot better than asking every single person in my company for their working hours in UTC.
the current solution is to set an explicit expectation that people work 9-5ish in their local offset.
is your proposal to do the same thing, but the expectation is that people work noon-4 to noon+4, or something like that? I don't see any major advantages to that, but I do see one major disadvantage: solar noon occurs at a different time every day. of all the arbitrary markers to align schedules with, this seems like a pretty bad one.
Yes. Rather than using solar noon at one day, we can look up an average solar noon time. It should give a better idea how people would plan their day. I understand that schedules will be different at different locations, but then 9 to 5 also has that much variation from location to location.
Just count the number of seconds since the GPS epoch in a galaxy-centered inertial frame. This post was made at 1,373,034,914.
I don't see anything referencing any planets?
"the duration of 9,192,631,770 periods of the radiation corresponding to the transition between the two hyperfine levels of the ground state of the caesium-133 atom taken as an average over 11 samples distributed throughout the earth in order to approximate the earth's inertial frame"
Seconds as used in Coordinated Universal Time are in fact defined based on planets, and earth specifically - because they use earth's inertial frame.
In principle this frequency can be measured in any inertial frame. Seconds are not defined on Earth specifically.
But it would definitely screw up cultural references, e.g.: a "5pm tea" then can only be understood with a broader cultural context: "UK, 5pm is late in the 2nd half of the working day for them".