Knots are one Nautical mile per hour, and 1 Nmi == 1 minute of latitude. Nautical miles are based on the shape of the earth and are thus independent from the rest of the US unit tragedy. And quite useful in navigation.
Of course, the meter was originally based on the shape of the earth as well - the distance from the pole to the equator was defined as 10 Mm in 1793.
As far as the US unit tragedy goes, I think it's important to remember that SI units are full of compromises in exactly the same way that US units are, because they need to accommodate practical, human scale concerns. For example, degrees Celsius is a ridiculous concept from a scientific point of view (ask any beginning chemistry student who just messed up their test because they forgot to convert to Celsius to Kelvin before applying PV=nRT). If you believe the point of a unit system is to be useful for science, it's hard to defend a unit where 0 doesn't actually mean 0, since it introduces all sorts of confusion in the same way as traditional units do.
And why does the whole world still measure days with 24 hours in them, each with 60 minutes, each with 60 seconds? Because decimal time (although France tried to impose it during the transition to SI units) is just too darn inconvenient for humans to use. The failure of decimal time shows why humans prefer non-decimal units: because the resulting units are human-scale. Converting from seconds to hours or days is not as easy as multiplying by 1000, but kiloseconds is an inconvenient unit - too short for an hour, while 100 kiloseconds is significantly longer than a day. Somehow the metric world manages these conversions just fine (in the same way the US manages conversions between inches, feet and yards).
The traditional units evolved over hundreds of years, and I'm glad the US keeps them around - like nautical miles, they are quite useful.
> The failure of decimal time shows why humans prefer non-decimal units: because the resulting units are human-scale.
I really don't agree though. The failure of revolutionary France to switch to decimal time like it switched to decimal units, is that time was already well standardised back then, while a single standard of decimal units was a real improvement over the mess of local units that was in use before.
I think you're making the same mistake every argument against metric always makes: "inches are more convenient than centimeters, because otherwise how can we call 19" racks!". Well otherwise, racks would be 50 centimeters and not 19".
The same way, people would work just fine with 10-hour days, 100-minute hours and 100-second minutes. Maybe people would use half-hours more often than today to divide the day, but I can assure you that it would just work as conveniently.
There is nothing inherently natural in dividing the day in 24/60/60 parts, we're just used to it. Also, if you have ever worked with time keeping software (or worse, hardware) you would know the metric world actually doesn't manage these conversions just fine.
The reason 12, 24, 60 are so useful is how many factors they have. You can divide the first two in half and thirds, and 60 additionally by 5ths and 10ths without resorting to fractions.
Being able to more simply divide by thirds and sixths might not be important most of the time, but it's not completely useless, and you can't do it in base-10.
Actually while they're useful, they're not that useful.
The point is that being able to divide the day up into 10 chunks, which is just about the only thing we can't yet evenly divide it up into[+], is even less useful.
It's certainly less useful than the gargantuan international effort it would take to change one arbitrary system for another.
The reason decimal time did and always will fail is that no matter how you slice it, 365.24ish is not divisible by 10.
Moreover there is nothing inherently natural about 10 as a base. Like many of our modern conveniences it is entirely arbitrary and about the worst possible option.
That is the point. There is noway to cleanly divide up the time taken by the earth's orbit by the time taken by the earth's rotation. They are simply incompatible. It is impossible to say objectively that any one scheme is better than another, with one exception: the current scheme is always better than all the others by virtue of being current.
The advantage of decimalisation is the geometric expansion it engenders which cannot happen because the year is not 10, 100, 1000 or any other multiple of 10 days. It is not an integer multiple of anything days. It is evidence that no matter how much we try and break the universe up into nice coherent chunks, the universe has other ideas. Fighting reality will fail.
The advantage to breaking up the day into units less than 1 day is the factorisation that can be applied. 24 factors much better than 10. Pretty much everything factors much better than 10.
There is no advantage to decimalising the day except to give basement-dwellers a boner.
24 and 60 equally fail when you try to turn days into years. That doesn't mean you give up on picking a good system for fractions of days and multiples of years.
There is no[1] advantage to having a smattering of small factors. People make much too big a deal about dividing into thirds.
[1] "no" in the same way that decimalization has no advantage, in that the advantages of either system are small and unnecessary.
We didn't give up. We found one and we stuck to it. I don't care about factoring, that's a minor cool hack.
The point is simple: Adding or removing dots from the face of a clock is a complete waste of (heh) time.
If you people really need to wank over a calender, pick one with jugs in it or get rid of timezones and daylight savings. That would be actually helpful.
> Adding or removing dots from the face of a clock is a complete waste of (heh) time.
Nobody was advocating for doing that. You started an argument about doing so, in reply to a post merely saying that it would be "as convenient" as the current system, and that decimal time failed because we already had a unified time system.
> If you people really need to wank over a calender
Nobody had even mentioned calendars until you showed up. "you people" = "ChoHag".
So a month will be 36.524ish days. Where in the metric regularity does 36.524ish fit? I can see that it's somewhere between the 10s and the 100s column...
A year isn't 100 days, it isn't 1000 days, and the whims of naked apes with a penchant for euclidean geometry isn't going to change that.
Well, the decimal calendar had 12 months of 30 days each, with either 5 or 6 extra days depending on leap years.
Months had three ten-day "weeks".
As you see, nobody is trying to fight physical reality, since day and year are hard lengths that you can't redefine.
But decimal time, decimal calendar and more generally decimal units are all about trying to eliminate as much as possible the historical peculiarities that are not grounded in physical reality:
The irregular length of months, the weird 24 hours in a day, and the base 60 which has been obsolete for millennia now. All of these do not give any advantage to the users of these units today.
Divisibility by 3, 6 or 12 are useless today because we now master decimal numbers easily. I know Imperial units users still often use fractions, but metric units users don't, and it just works. Not once in my life have I been thinking "oh, how I wish I could express 0.41 centimeters in fractions". In the same way, your now 1-hour task would take you something like half a decimal hour instead, or 50 decimal minutes, and you wouldn't even think about it.
> Well, the decimal calendar had 12 months of 30 days each, with either 5 or 6 extra days depending on leap years.
So not 10^x, 10^y and 10^z consistently without the constant need for special-case exceptions?
> Months had three ten-day "weeks".
So not 10^x-day weeks then?
> decimal [time-measuring] units are all about trying to eliminate as much as possible the historical peculiarities
Eliminate? I think you mean change.
> The irregular length of months, the weird 24 hours in a day, and the base 60 which has been obsolete for millennia now. All of these do not give any advantage to the users of these units today.
There is nothing more right about 24 than 10, unless you need to take your socks off to count.
There is nothing more right about 60 than 10, unless you need to take your friends' socks off to count.
10 gives advantages to people who count on their fingers. But, amusingly, less advantage to finger-counters than just about any other even-numbered base that isn't 2.
> Divisibility by 3, 6 or 12 are useless today because we now master decimal numbers easily.
I see you're a maths teacher! I, also, love the pure enthusiasm of the classroom. You can smell it a mile away.
I think that's enough from me. The rest is just imposition anyway.
We can't not consider current timekeeping. There is exactly 0 chance we can get everybody to change in the same instant and even if they did history is still there. Education, and thus civilisation's shared consciousness, will include the old timekeeping until the renaissance, industrial revolution, the inventions of computers and space travel and the near-ELE cataclysm of the final death knell of religion are all historical footnotes. Computers will need to understand the old timekeeping forever. We won't reduce the amount of bullshit humans deal with, we will have increased it. And with all that, the earth's orbit still won't be 1000 days.
And all for what? So I can count the hours in a day without getting my feet cold? Woo!
Well we use to use cubits and chains for measuring everything too. Science is slowly moving to si, we should probably all start cobaidering a proper system for time that works around metric.
I don't like the idea of explaining to my children that we use this system, because it's the way it has always been done. That's a horrible idea .
I look forward to teaching my child that we use this system and using its peculiarities to spawn discussions on number theory, astronomy, history, even art.
It is a 1-D measurement. It's the pressure exerted by the atmosphere that would raise a column of mercury that 1-D length up a vacuum at 1g of gravity.
Even if not taken literally, the particular unit of "inches" is largly meaningless in this case. All we know (assuming the obvious mechanism of measurement) is that the air pressure in <choice of force units> is linear to the reported value. The conversion factor is dependent on the instrument itself (surface area, internal resistance, etc) although there may be a standard that is known specialists.
Two square inches of mercury an inch tall weights twice as much as one square inch of mercury an inch tall. But there's twice as much area, so it's applying the same amount of force to each square inch.
The point is you care about the force, but don't care about the area, and a tube of mercury factors out the area.
