I just discover your submission and - if you can read french - I wonder what you could think of this paper : http://marty.alain.free.fr/confs_2019/?view=grain where I try to use the imperial system [foot,inch]] inside the metric system.
People like what they are used to, so they can be simply defending the habit. US should be pushing more strongly to switch to metric system instead of dragging this for ages. Speed signs should use kilometers per hour, and weather reports should use Celsius.
For one thing, as mentioned in the article, not exposing hospital patients to a completely unnecessary risk of death due to incorrect dosage of medication?
When you're living inside a system, it seems perfectly normal to you, I get that. To me, when someone explains the American measurements system to me, here's what I hear:
> Mr. Zabladowski: If you must know, the currency here is a little strange. First of all, in the small money comes the drakmids. At the regular, normal rate of exchange, there are 59 drakmids to one silver zlobeck.
> Bart Collins: "Zlobeck"?
> Mr. Zabladowski: Three silver zlobecks make one golden kratchmuk. A pastoola normally is, uh, 44,000 kratchmuks. But these, they tell me, are not normal times...
It's 50 years, maybe a little more, since Britain required all education to be in SI units. That's two generations taught entirely metric. Weather has been in Celsius only for almost as long.
People still mostly refer to their height in feet & inches, buy a pint of beer, but petrol's in litres, they drive at 40 mph, and calculate mpg. When it's cold most people use Celsius, but when it's hot, people always remember Fahrenheit and it's 80s, 90s, etc. Never quite worked that one out. :)
It's going to take a bloody long while, whenever the switch starts, whether it includes road signage or not.
Interesting. At least Britain pushed for it for real and as you say, even with that it takes a long time. In US it's very lax. Government basically gave up finishing the job, so it's stalled.
There's other silly ones. Wood has been metric almost as long as I can remember, but 50x25mm is still commonly referred to as 2x1 etc. If it's PAR it'll be smaller than either. :) plumbing pipes and fittings are still mostly really BSP - British Standard Pipe uses inside diameter, metric outside. 15mm pipe and fitting is really a 1/2" BSP with imperial thread.
UK was probably too lax about it to be honest. They didn't require all packaging to use metric only until the last 20 years, so plenty of shops kept on putting up signs in pounds even if the till worked the price in kilos.
Since metric packaging was required there's been one clear downside - you're far, far more likely to have manufacturers hide price rises by dropping 5 or 10 ml or grams from a product, in an apparently identical pack. Repeat a few times and much is now randomly sized, where once all brands had comparable sizes. Choc bars, jam, used to be 1lb, 1/2, 1/4. Now some random number of g.
> but when it's hot, people always remember Fahrenheit and it's 80s, 90s, etc
The US is the only country using Fahrenheit as far as I know, I never heard anyone in the UK using anything other than Celsius. You don't even say "Celsius", just "degree", it's implied it's Celsius.
I have the same experience. I'm an American expat so sometimes people will explicitly call out celsius for my benefit or as a joke, but yeah, nobody ever says celsius when they talk about the temperature here.
The tabloids still seem to like Fahrenheit for heatwaves, and I've heard it used reasonably often in summer - in offices etc. I can only remember hearing someone use deg F when it's 70+. It's certainly declining and not as common as say 30 years ago, but it's definitely one of the holdouts.
No one ever adds C or F, or even degrees very often, unless writing. It's -5; it's 30; it's 92, which are all self defining in UK climate.
I don't live in the UK but nearby in Europe. I know a handful of Brits but have never heard them express temperatures in Fahrenheit (and they know I'm American, so if they switched I always figured it would be for my benefit). Is that really true about the hot weather thing?
Just this past week for example, my colleagues were complaining about the 33 degree heat wave, were excited for the weather to drop back down to the 20s on Friday, and were complaining about the random 31 degree heat wave coming on Saturday.
The only times I've heard reference to Fahrenheit in the UK is my grandma in the 1990s, and in tabloid newspapers when the temperature reaches 100°F.
These are the same tabloids that will do anything anti-European they can think of. I don't think it matters that people don't really know what 100°F is.
This is not really a problem I think. It's become an object in itself and you don't really use this as a reference when buying milk or water I imagine. In France you can also get a "pinte" of beer (pronounced as "paint"), it's probably a relic from before the metric system and probably not the same as the English pint though.
Celsius vs Fahrenheit is tricky because Fahrenheit does have some nice properties and several sciences use Kelvin anyway. I think the best one would be a new one, rescaled to match with 0°C and 100°F.
Calculating mpg is an interesting one because even if you moved to calculating km per liters it would still doesn't match the most common usage, we use liters per 100 km.
One thing where almost everyone fails in my opinion is shoe sizes. Why can't we just use centimeters? Instead we have four competing standards and the EU one doesn't relate to anything.
In France you can also buy vegetables by the pound (une livre), but if course both of these are just shorthands for half a litre and half a kilo, respectively, they're not actually the old measures.
The main niceties are the increased granularity (nobody would use half °C in practice) and 100°F being the normal body temperature and a good landmark to relate to hot weather.
0°C is absolutely super practical and is imperative to have as a landmark. 100°C boiling water... it isn't really that practical in day to day use in my opinion (when was the last time you saw 100°C written on a device?). In the high temperatures the ones I care about in a practical sense are on the oven for cooking and on my soldering station, both much higher than 100°C anyway. The only time I care about boiling water temperature is when I am boiling water in which case I don't need a thermometer.
So I think 100°F is more interesting as a landmark than 100°C for intuiting temperatures. You can think of it as the percentage of how hot it is. But 0°C is more interesting than 0°F.
I don't understand why we want round numbers here. I know 40 C is "very hot" and for Americans that's 100 F. Using a rounder number makes absolutely no difference, since I'm never going to divide temperatures.
It doesn't even matter much for meters and kilometers, since I very rarely need to talk about things that are kilometers away and meters away in the same context. It's handy to have sane units for those, though, since there you can multiply and divide in your calculations.
Temperature arguments never seemed particularly compelling to me. They assume 100 is more convenient, which I find wrong, since you're usually talking about 102, 86, 95, etc anyway.
> when was the last time you saw 100°C written on a device?
Daily. Ovens in Europe typically will have a temperature range that's something like 50-400 C. I've never seen oven without a 100 C setting. Sauna thermometers will typically go from ~20 degrees to either 120 or 140 Celsius.
I was surprised, when I found out, how comparable many of those relics are. Many countries had a variation on a pint, a pound etc, that were in the same ballpark but never quite the same. I'd always presumed pre-metric Europe would have much wider variation in units. Maybe they stem back to the Romans, like £sd did.
You'd be surprised on milk - I'd say the commonest supermarket sizes are still really pints. Usually 2.272l (4.0 pints), and 2 and 6 pint equivalents. Posh variants like some organic, filtered etc are in 2.0 litre packs - mostly opaque so you don't notice the 4 pint capable pack with lower fill level. :)
Shoe sizes are infuriating. Would be nice to find a way to standardise, and have manufacturers agree on what size x means, as they've diverged wildly from reality.
The real problem is that it's bad for US exports. Outside the US, almost nobody has inch tools.
US defense, aviation, and automotive have been metric for decades now. Electronics still has some inch stuff in through-hole parts.
US construction, though, is still inch-oriented.
Home Depot stocks few metric fasteners. I had a conversation there with someone from a restaurant who was desperately trying to get the fasteners needed to fix some piece of metric equipment on a weekend.
Home Depot didn't even have basic metric screws.
US Aviation in the sense of the "aero" part of aerospace is still very much US Customary Units. I don't have enough experience with the "space" part of aerospace to comment on that. We still use pounds of thrust, slug and pound masses, psi for pressures and so on for things related to turbines and engines and so on. Strangely, we use kelvins instead of Rankines but that's the only exception I can think of.
Edit: Note that this isn't to say that aerospace engineers prefer the US units or that we don't know how to use the metric system or anything, it's just how it is. In this particular field, it's just the accepted standard.
Iso fasteners and metrics blueprints are favored for all NATO activities, but it's not all military stuff. Moreover, a lot of iso hydraulics stuff is imperial.
In Aviation there are nautical miles for ground distance, knots for speed and feet for altitude. But in US aviation there are also statute miles for airport visibility and feet for runway visibility.
Customary units are bigger mess than imperial vs. metric.
Latitude is measured along a Great Circle of Longitude - confusing isn't it? So the distance between (say) 10 and 30 degrees North is the same as the distance between 50 and 70 degrees North.
The Latitude circles (in the 'horizontal' plane) get smaller as you go from Equator to the Poles, but all the Longitude Circles (in the 'vertical' plane') are all Great Circles and go through both the North and South Poles and so are all (pretty much) equal in size.
One place where the metric/American mixup is a regular pain is paper sizes. American scientific journals are generally published on Letter paper, while all other scientific journals are generally published on A4 paper. The difference in aspect ratio is just enough to cause the bottoms or sides of pages to be cut off if you don't pay attention.
One nice thing about the international paper standard is its subdivision property: if you fold a piece of paper in half (along the correct axis), it retains its aspect ratio. That's pretty useful for printing pages 2-up, and for reducing paper waste. It's a similar argument to the "argument of twelves," but it goes in the opposite direction: the metric standard is favored.
TBF you've got to be slightly obsessed to see the metrication in A-series: most people will interact with A3 and A4 papers which are respectively 420x297mm and 297x210mm, if you go up you end up with A0's 1189x841mm which doesn't look very metric either. You've got to realise that it's 1 sqm in a sqrt(2) ratio to see the metrication (the aspect ratio being chosen so it stays constant when you halve the page through its larger edge).
The B series actually looks "more metric" at first glance: B0 is 1414x1000mm.
Optimally humanity should have chosen base 12 instead of base 10, then the metric system would have used it :)
But anyway, to me it doesn't actually matter.
Admittedly I don't do workshop stuff that much, but when I do, I measure with a ruler, and never needed fractions for that, numbers like 3.333 work fine.
Where do you need fractions if you've got floating point notation, plus if you'd like to use fractions nothing stops you from saying "1 third of a meter".
And the imperial system doesn't really use base 12 much anyway? The only ratio of 12 I see in there is inch and foot. All the rest is various random different ratios (e.g. a mile is 5280 foot? really? and gallon vs ounce is a ratio of 160?), and Fahrenheit is its own separate mess.
> All the rest is various random different ratios (e.g. a mile is 5280 foot? really? and gallon vs ounce is a ratio of 160?)
TBF you're skipping intermediate units. After all 12^3 would be 1728 so non-decimal units quick get to weird ratios.
Of course you're also right that imperial and customary measurements use seemingly random ratios even within the same unit of measure: the mile is 5280 feet because a statute mile is 8 furlong, a furlong is 10 chains, a chain is 22 yards (though technically there's the intermediate "rod" unit which is 1/4th of a chain, and thus 5 ½ yards) and a yard is 3ft.
Those 8/10/22 ratios are still quite random, and I've never heard anyone use furlong or chain as units (at least in movies).
To be fair, I should have said 1760 yards, which is equally weird (its prime factorization is 2 * 2 * 2 * 2 * 2 * 5 * 11, especially the 11 in there is weird, plus the 5 in there makes it more lean towards base 10 than base 12)
There are 22 yards to a chain, 10 chains to a furlong and 8 furlongs to the mile. The intermediate units have fallen out of use, but it's not like the mile is a completely wacky number.
> There are 22 yards to a chain, 10 chains to a furlong and 8 furlongs to the mile […] it's not like the mile is a completely wacky number.
Have you read what you wrote? And you forgot the 4 rods to a chain (and thus 5 ½ yards to a rod).
That's 4 different ratios (none of which is 12) to go from kilometer-scale to meter-scale.
> I see you've never cooked anything. Ever.
I fail to see how finding "there are 3 tsp to a tbsp, 2 tbsp to an fl oz, 8 fl oz to a cup, 2 cups to a pint, 2 pints to a quart and 4 quarts to a gallon" odd translate to "you have never cooked anything" compared to "this is a millilitre, it works fine under the entire range, for graduation simplicity there are 10mL to a cL, 10cL to a dL and 10 dL to a L, a liter is a convenience designation for 1dm³".
> Optimally humanity should have chosen base 12 instead of base 10, then the metric system would have used it.
If humans had evolved to have twelve fingers and twelve toes, then this could’ve perhaps happened. Even if humanity had chosen base 12 for whatever reason, I’d imagine we’d have different single digit symbols for numbers “A” and “B” that we use in hexadecimal now, with something like “10” representing what we call as 12 in decimal today.
Of course we'd have had 12 unique symbols if humanity had adopted base 12 instead of base 10, say if that would have happened historically with arabic numerals (but it's a bit more complicated historically of course because the roman numerals and other systems before it were already decimal, even the babylonean base 60 was actually decimal based anyway, their notation uses 6x 10 symbols, not 5x 12 symbols).
Obviously it didn't and history won't change :)
And yep, no matter in what numeric base you are, "10" will represent the amount of symbols of that base. "10" in binary is 2 in decimal, "10" in hex is 16 in decimal, etc...
I think one thing that's often glossed over in the imperial vs metric discussion is that users of metric almost universally stick to decimal numbers, while fractions are a lot more commonly used with imperial.
For many Europeans, dealing with values like 3/8 or 5/16 seems completely unintuitive and even deciding which of them is larger will take us some thinking. So being able to elegantly converting fractions seems like a non-issue since you would never use them in the first place.
52 comments
[ 5.4 ms ] story [ 817 ms ] threadon Chrome
You should be long past asking about why, and should be asking why it didn't happen years ago.
When you're living inside a system, it seems perfectly normal to you, I get that. To me, when someone explains the American measurements system to me, here's what I hear:
> Mr. Zabladowski: If you must know, the currency here is a little strange. First of all, in the small money comes the drakmids. At the regular, normal rate of exchange, there are 59 drakmids to one silver zlobeck.
> Bart Collins: "Zlobeck"?
> Mr. Zabladowski: Three silver zlobecks make one golden kratchmuk. A pastoola normally is, uh, 44,000 kratchmuks. But these, they tell me, are not normal times...
(From the movie "The 5,000 Fingers of Dr. T".)
People still mostly refer to their height in feet & inches, buy a pint of beer, but petrol's in litres, they drive at 40 mph, and calculate mpg. When it's cold most people use Celsius, but when it's hot, people always remember Fahrenheit and it's 80s, 90s, etc. Never quite worked that one out. :)
It's going to take a bloody long while, whenever the switch starts, whether it includes road signage or not.
UK was probably too lax about it to be honest. They didn't require all packaging to use metric only until the last 20 years, so plenty of shops kept on putting up signs in pounds even if the till worked the price in kilos.
Since metric packaging was required there's been one clear downside - you're far, far more likely to have manufacturers hide price rises by dropping 5 or 10 ml or grams from a product, in an apparently identical pack. Repeat a few times and much is now randomly sized, where once all brands had comparable sizes. Choc bars, jam, used to be 1lb, 1/2, 1/4. Now some random number of g.
The US is the only country using Fahrenheit as far as I know, I never heard anyone in the UK using anything other than Celsius. You don't even say "Celsius", just "degree", it's implied it's Celsius.
"What's the temperature?"
"It's gonna be 21."
"Got it, thanks."
That's how it usually goes.
No one ever adds C or F, or even degrees very often, unless writing. It's -5; it's 30; it's 92, which are all self defining in UK climate.
Just this past week for example, my colleagues were complaining about the 33 degree heat wave, were excited for the weather to drop back down to the 20s on Friday, and were complaining about the random 31 degree heat wave coming on Saturday.
These are the same tabloids that will do anything anti-European they can think of. I don't think it matters that people don't really know what 100°F is.
This is not really a problem I think. It's become an object in itself and you don't really use this as a reference when buying milk or water I imagine. In France you can also get a "pinte" of beer (pronounced as "paint"), it's probably a relic from before the metric system and probably not the same as the English pint though.
Celsius vs Fahrenheit is tricky because Fahrenheit does have some nice properties and several sciences use Kelvin anyway. I think the best one would be a new one, rescaled to match with 0°C and 100°F.
Calculating mpg is an interesting one because even if you moved to calculating km per liters it would still doesn't match the most common usage, we use liters per 100 km.
One thing where almost everyone fails in my opinion is shoe sizes. Why can't we just use centimeters? Instead we have four competing standards and the EU one doesn't relate to anything.
Which exactly? Walter melting at 0°C and boiling at 100°C seems the most practical.
0°C is absolutely super practical and is imperative to have as a landmark. 100°C boiling water... it isn't really that practical in day to day use in my opinion (when was the last time you saw 100°C written on a device?). In the high temperatures the ones I care about in a practical sense are on the oven for cooking and on my soldering station, both much higher than 100°C anyway. The only time I care about boiling water temperature is when I am boiling water in which case I don't need a thermometer.
So I think 100°F is more interesting as a landmark than 100°C for intuiting temperatures. You can think of it as the percentage of how hot it is. But 0°C is more interesting than 0°F.
We use 0.1°C for even better granularity. E.g. weather sites, all my digital thermometers, etc.
It doesn't even matter much for meters and kilometers, since I very rarely need to talk about things that are kilometers away and meters away in the same context. It's handy to have sane units for those, though, since there you can multiply and divide in your calculations.
Temperature arguments never seemed particularly compelling to me. They assume 100 is more convenient, which I find wrong, since you're usually talking about 102, 86, 95, etc anyway.
One decimal place is commonly used.
Daily. Ovens in Europe typically will have a temperature range that's something like 50-400 C. I've never seen oven without a 100 C setting. Sauna thermometers will typically go from ~20 degrees to either 120 or 140 Celsius.
You'd be surprised on milk - I'd say the commonest supermarket sizes are still really pints. Usually 2.272l (4.0 pints), and 2 and 6 pint equivalents. Posh variants like some organic, filtered etc are in 2.0 litre packs - mostly opaque so you don't notice the 4 pint capable pack with lower fill level. :)
Shoe sizes are infuriating. Would be nice to find a way to standardise, and have manufacturers agree on what size x means, as they've diverged wildly from reality.
I moved to Finland and I'm still not used to the idea of giving my height in cm/m, or weight in kg. Until now I've never needed to know either.
US defense, aviation, and automotive have been metric for decades now. Electronics still has some inch stuff in through-hole parts.
US construction, though, is still inch-oriented.
Home Depot stocks few metric fasteners. I had a conversation there with someone from a restaurant who was desperately trying to get the fasteners needed to fix some piece of metric equipment on a weekend. Home Depot didn't even have basic metric screws.
Edit: Note that this isn't to say that aerospace engineers prefer the US units or that we don't know how to use the metric system or anything, it's just how it is. In this particular field, it's just the accepted standard.
Customary units are bigger mess than imperial vs. metric.
The Latitude circles (in the 'horizontal' plane) get smaller as you go from Equator to the Poles, but all the Longitude Circles (in the 'vertical' plane') are all Great Circles and go through both the North and South Poles and so are all (pretty much) equal in size.
One nice thing about the international paper standard is its subdivision property: if you fold a piece of paper in half (along the correct axis), it retains its aspect ratio. That's pretty useful for printing pages 2-up, and for reducing paper waste. It's a similar argument to the "argument of twelves," but it goes in the opposite direction: the metric standard is favored.
The B series actually looks "more metric" at first glance: B0 is 1414x1000mm.
Obviously now, with global standards and decimals, not using metric is foolish.
But anyway, to me it doesn't actually matter.
Admittedly I don't do workshop stuff that much, but when I do, I measure with a ruler, and never needed fractions for that, numbers like 3.333 work fine.
Where do you need fractions if you've got floating point notation, plus if you'd like to use fractions nothing stops you from saying "1 third of a meter".
And the imperial system doesn't really use base 12 much anyway? The only ratio of 12 I see in there is inch and foot. All the rest is various random different ratios (e.g. a mile is 5280 foot? really? and gallon vs ounce is a ratio of 160?), and Fahrenheit is its own separate mess.
TBF you're skipping intermediate units. After all 12^3 would be 1728 so non-decimal units quick get to weird ratios.
Of course you're also right that imperial and customary measurements use seemingly random ratios even within the same unit of measure: the mile is 5280 feet because a statute mile is 8 furlong, a furlong is 10 chains, a chain is 22 yards (though technically there's the intermediate "rod" unit which is 1/4th of a chain, and thus 5 ½ yards) and a yard is 3ft.
To be fair, I should have said 1760 yards, which is equally weird (its prime factorization is 2 * 2 * 2 * 2 * 2 * 5 * 11, especially the 11 in there is weird, plus the 5 in there makes it more lean towards base 10 than base 12)
There are 22 yards to a chain, 10 chains to a furlong and 8 furlongs to the mile. The intermediate units have fallen out of use, but it's not like the mile is a completely wacky number.
> gallon vs ounce is a ratio of 160?
I see you've never cooked anything. Ever.
Have you read what you wrote? And you forgot the 4 rods to a chain (and thus 5 ½ yards to a rod).
That's 4 different ratios (none of which is 12) to go from kilometer-scale to meter-scale.
> I see you've never cooked anything. Ever.
I fail to see how finding "there are 3 tsp to a tbsp, 2 tbsp to an fl oz, 8 fl oz to a cup, 2 cups to a pint, 2 pints to a quart and 4 quarts to a gallon" odd translate to "you have never cooked anything" compared to "this is a millilitre, it works fine under the entire range, for graduation simplicity there are 10mL to a cL, 10cL to a dL and 10 dL to a L, a liter is a convenience designation for 1dm³".
If humans had evolved to have twelve fingers and twelve toes, then this could’ve perhaps happened. Even if humanity had chosen base 12 for whatever reason, I’d imagine we’d have different single digit symbols for numbers “A” and “B” that we use in hexadecimal now, with something like “10” representing what we call as 12 in decimal today.
Obviously it didn't and history won't change :)
And yep, no matter in what numeric base you are, "10" will represent the amount of symbols of that base. "10" in binary is 2 in decimal, "10" in hex is 16 in decimal, etc...
I would not go as far as to say it probably means something but here we are... :)