What’s So Bad about the Imperial System Anyway?

As a Hackaday writer, you can never predict where the comments of your posts will go. Some posts seem to be ignored, while others have a good steady stream of useful feedback. But sometimes the comment threads just explode, heading off into seemingly uncharted territory only tangentially related to the original post.

Such was the case with [Steven Dufresne]’s recent post about decimal time, where the comments quickly became a heated debate about the relative merits of metric and imperial units. As I read the thread, I recalled any of the numerous and similarly tangential comments on various reddit threads bashing the imperial system, and decided that enough was enough. I find the hate for the imperial system largely unfounded, and so I want to rise to its defense.

Did you measure that room in 'feet', or in 'flip-flops'?
Did you measure that room in ‘feet’, or in ‘flip-flops’?

What is a system of units anyway? At its heart, is just a way to measure the world. I could very easily measure the length and width of a room using my feet, toe to heel. Most of us have probably done just that at some point, and despite the inconvenient and potentially painful problem of dealing with fractionalization of your lower appendage, it’s a totally valid if somewhat imprecise method. You could easily pace out the length of the room and replicate that measurement to cut a piece of carpet, for instance. It’s not even that much of a stretch to got to the home center and buy carpet off the roll using your personal units — you might get some strange looks, but you’ll have your personal measuring stick right with you.

The trouble comes when you try to relate your units to someone not in possession of your feet. Try to order carpet online and you’ll run into trouble. So above and beyond simply giving us the tools to measure the world, systems of units need to be standardized so that everyone is measuring the same thing. Expanding trade beyond the dominion where one could refer to the length of the king’s arm and have that make sense to the other party was a big driver of the imperial system first, and then the metric system. And it appears to be one of the big beefs people have regarding the United States’ stubborn insistence on sticking with our feet, gallons, and bushels.

How Ridiculous are We Talking?

quote-definition-of-a-meterThe argument that imperial units are based on ridiculous things like the aforementioned king’s arm? That’s not an argument when a meter was originally defined as one 10-millionth of the distance from the north pole to the equator. Even rigorously defined relative to the speed of light or the wavelength of krypton-86 emissions in a vacuum, the meter is based on phenomena that are completely inaccessible to the people who will use is, and unrelated to their daily lives. At least everyone has seen a foot that’s about a foot long.

Doing the conversions between imperial units and SI units is tedious and error prone, they say. Really? Perhaps I’d buy that argument a hundred years ago, or even fifty. But with pervasive technology that can handle millions of mathematical operations a second, there’s not much meat on that bone. I’ll grant you that it’s an extra step that wouldn’t be needed if everyone were on the same system, and that it could lead to rounding errors that would add up to quite a bit of money over lots of transactions. But even then, why is that not seen as an opportunity? Look at financial markets — billions are made every day on the “slop” in currency exchanges. I find it unlikely that someone hasn’t found a way to make money off unit conversions too.

Another point of contention I often see is that imperial units make no sense. Yes, it’s true that we have funny units like gills and hogshead and rods and chains. But so what? Most of the imperial system boils down to a few commonly used units, like feet and gallons and pounds, while the odder units that once supported specialized trades — surveyors had their rods and chains, apothecaries had their drams and grains — are largely deprecated from daily life now.

Deal with It

For the units that remain in common use, the complaint I hear frequently is, “Why should I be forced to remember that there are 5,280 feet in a statute mile? And why is there a different nautical mile? Why are there 12 inches in a foot anyway? A gallon has four quarts, why does that make sense?” And so on. My snappy retort to that is, again, “So what?” If you’re not a daily user of the imperial system, then don’t bother yourself with it. Stick to metric — we don’t care.

If you’re metrified and you’re forced to use imperial units for some reason, then do what a lot of us imperials have to do — deal with it. I’m a scientist by training, and therefore completely comfortable with the SI system. When I did bench work I had to sling around grams, liters, and meters daily. And when I drove home I saw (and largely obeyed) the speed limit signs posted in miles per hour. No problems, no awkward roadside conversations with a police officer explaining that I was still thinking in metric and thought that the 88 on my speedometer was really in km/h and I was really doing 55. If I stopped at the store to pick up a gallon of milk and a couple of pounds of ground beef for dinner, I wasn’t confused, even if I slipped a 2-liter bottle of soda into the order.

At the end of the day, I don’t really see what all the fuss is about. Imperial and metric both have their place, and each system seems to be doing its job just fine. If your argument is that imperial units are inelegant and awkward, even though you’re correct I don’t think that’s enough to sway the imperial holdouts. And if you’re just upset because we’re being stubborn and won’t join the enlightened metric masses, then I think you’re probably going to be upset for a long time to come.

883 thoughts on “What’s So Bad about the Imperial System Anyway?

      1. But it takes 2 hands to count to ten if you use your fingers. So 5 should be the base of metric then?

        BTW, you can actually count to 12 on one hand and to 144 if you use 2 hands. Take your thumb and index it to the top pad on your pointing finger. See the other 11 pads?
        So maybe imperial with a base of 12 is based on humans…..

        Each system has its pluses and if as an engineer you cant convert between the two then get out of engineering.

          1. Oh, yes I count in binary on my fingers :-) Glad I’m not the only one! Being left-handed, I start with 1 on my little finger on the left side, getting to 31 with my left hand before continuing with 32 on my right thumb! Yay 1023 with two hands :-)

        1. if you count each of three knuckles per finger plus one fingertip per finger using your thumb as position indexing, you can count up to 16 in one hand, and up to 256 if you index sets of 16 with the same system on the other hand.
          Happy hexing!

    1. Look, I just don’t see what is so hard to grasp about imperial:
      1,000 thous to an inch
      12 inches to a foot
      3 feet to a yard
      22 yards to a chain
      10 chains to a furlong
      8 furlongs to a mile.
      Its that simple!

        1. AWG isn’t a unit. It’s a table of standard sizes. It actually DOES make some sense, in that stocking a finite number of wire sizes is pretty much a requirement for a practical business. Every third AWG size is approximately a doubling of cross-sectional area (and therefore a doubling of current-carrying capacity). This puts the steps close enough together that you can always find a wire size that’s just a little bigger than what you need.

          Coincidentally, just as with ASA film speeds, this also means that an increase of 10 in AWG equates to ten times the cross-sectional area and weight. Roughly, because that’s how we roll in ‘murica.

          1. An increase of 10 in AWG means a _decrease_ of area to one-tenth, right? Logarithmic scales do have useful properties, and as a non-American, I never learned this one, so thanks. But the fact that bigger numbers means smaller wires is just confusing…

            (Okay, there’s something about thirty-ninth roots that hurts a bit, but I’ll live.)

          2. Yes, that’s correct – I skipped over that fact. Lower numbers = larger wire.

            The 39th root thing is the dumb part, and it’s why I had to say “approximately”. If they had been a little more careful, they could have made it four decades of cross-sectional area over 40 steps, which would have made these relationships exact. As with most standards, it’s not clear WHY this choice was made, but it doesn’t cause enough problems to be worth changing.

          3. According to Wikipedia (so take it with a grain of salt), the number is supposed to represent the number of dies the wire had to pass through to get to that size. I suppose this might also be the case for sheet metal, which similarly scales down in size for larger gauges.

    2. I’d just like to add that the article is in gross error:

      >” a meter was originally defined as one 10-millionth of the distance from the north pole to the equator. ”

      Not so – the metre was originally defined as the length of a seconds pendulum, which in its simplicity is a 1 m long string with a lead weight on the end, which makes a full swing in two seconds, or half-swing in one. That is a very accessible definition and still very accurate.

      1. After all, when you construct a grandfather clock and tune the escapement to advance one tooth per second, you’ll find the center of mass of the pendulum is swinging in a 1 meter radius arc, give or take couple millimeters to account for latitude and local gravity.

        1. The seconds pendulum was proposed by several people but never adopted by France as part of the metric system. At 45° latitude, per Wikipedia, the length is 994 mm, not all that close to 1 m, in fact a larger error than the polar quarter meredian, about 10 002 km vs intended 10 000 km. Further it depends on local gravity which depends on latitude, altitude above sea level, and local mass disribution errors in the earth. A poor basis for a standard in a (modern) era were the expectation is 8 or more significant figures.

          1. I’d say an error of 0.6% due to differences in latitude is pretty close for a standard proposed in 1668. For practical trade with items such as cloth, grain, wood, wine etc. it would have been perfectly adequate as a universal standard.

            In any case, that’s where the metre comes from – it aligns nicely with clockmaking. The meridian definition that was offically adopted was a political choice – the error in the polar quarter meridian comes from the fact that the French neglected the fact that the earth isn’t a perfect sphere, measured the distance wrong and the prototype metre bar became 200 micrometers short – which was then grandfathered in to every subsequent definition of the metre.

          2. It may have been a political decision, but it is also a defensible accuracy decision. Once you measure the earth correctly (I agree on the mistakes made then) it is more precise and repeatable than lugging a pendulum around and trying to measure local gravity with a length unit yet because you haven’t corrected your pendulum for local gravity. I’m not sure they had a good gravity model back then

            I would guess you are hung up that it started as a French invention, but so what? It became international in 1875 when the Treaty of the Meter was signed.

            By the way, the French chose the meridian method and then the bar in the 1790’s. Too lazy to Google the year. But it was long over and the French were giving copies to anyone who was thinking of going metric long before 1870. The US was given one around 1820, used for coastal surveys until we got a better once following the treaty of the meter.

          3. Or more to the point, if the definition of the metre had indeed been standardized on the pendulum, 994 mm would be exactly 1 meter today. The actual error due to latitude is just 3mm which can easily be taken into account – likewise the local gravity, altitude etc. measured and compensated for.

            For everyday use, just about anyone could have produced a good enough metre. For precision use, further refinement and cross-checking would have been needed but the system would have been independent of a prototype bar sitting in Paris.

            Hence why the meridian definition was chosen in 1870 instead of the pendulum definition.

          4. >” but it is also a defensible accuracy decision. ”

            It’s more defensible for surveying reasons than accuracy and repeatability, because it was and still is practically very cumbersome to measure an earth quadrant. For example, finding true north is tricky because of variations in the local magnetic field, and the north pole isn’t the geographical north pole anyways, and the length of the day is variable, and the night sky doesn’t hold perfectly still… and the uneven mass distribution of the earth means that the length of the quadrant varies depending on where you measure it – the earth is not only a bit squashed, it’s also lumpy, so you have to calculate some sort of idealized curve rather than the real deal and that throws your measurement off.

            And, the most accurate way of measuring the direction of true north is is… guess what… observing the precession of a pendulum or a gyroscope.

            The surveying argument comes from the fact that the French were using a system of angles where a full circle is 400 units, which made the math of adding up angles very simple. For that reason, they attempted a definition that would put a factor of ten conversion between angles and kilometers, and why they ignored the off-spherical shape of the earth – they wanted the kilometer to be the same north-south and east-west.

          5. It’s really immaterial. They made a bar and it was the meter until 1960, when they redefined by a certain red wavelength of an excited cadmium atom and later the speed of light. I agree the quadrant measurement was neither convenient nor accurate (but better than the seconds pendulum). Once there was a bar, no point in flogging the meridian measurement to death. I’m actually surprised they were as accurate as they were and suspect a bit of luck.Had they measured the meridian a number of times, I suspect the standard deviation would have been worse that the one-off result.

            Once they made a bar, it was no different than measuring the king and making a bar. What was clever was using the bar to define a volume, then using a standard (water) to set the mass standard. They apparently understood air buoyancy and correcting for it too, as the error is only about 28 ppm, not bad for the 1790s. The British independently picked the pound as 7000 grains of something, then defined the gallon as a ten pound bag of water.

      2. Well if you dwell far enough to the past all metric units were based on prototypes. One should probably talk about Imperial vs. SI units as the metric system was more or less French giving the finger to the Royal Society. The problem in engineering is that we have different numerical systems for time and angle. My former boss once said we should use 60 as the base radix for everything and the classical physics would have easier time with the associated math.

        1. The French developed the original metric system as a uniform mercantile system. It only covered length, area, volume, and mass. Scientists later added the concepts and derived units that made it useful for science and engineering, and members of the Royal Society were major players in that development, and those derived units are named for those principle players. So the notion of giving the finger to the Royal Society is complete and utter bullshit. I should also add that British firms manufactured most of the physical standards used in the early history of the metric system.

        2. Measurement of *anything* is based on some arbitrary relationship to something, and the only one that seems universally accepted is time… the Earths orbit divided by some esoteric numbers (which made good sense when it was done). The whole metric/imperial argument as presented here is farcical. The real argument has NOTHING to do with the UNITS, it is about how they relate to each other.

          Metric is always based on multiples of 10, whereas imperial is based on a random number sequence. THAT is what the argument should be about, not trying to sidetrack it as the author has done.

          I put this sort of “journalism” into the same category as HaD “journalists” arguing against the 3.5mm (is that a metric measurement…) phone jack.

          I am *VERY* disappointed in the quality of many recent articles.

  1. Exactly! I mean, the smart people at NASA who make orbiters have to convert between units all the time and they would never make a basic conversion error. I mean, I’m not going to Google to check that I’m correct or anything, but I’m sure they just deal with it!

      1. [synthetic voice] Nasa? Look, seems to me that Mars is way too near right now, and growing rapidly. Can you check that?
        [Nasa mission control] … It’s okay Orbi. You’re supposed to be at 12000 miles away from surface. More than enough room to dance.
        [Orbi] Ok then. I was wondering….WAIT you said MILES, not KILOMETERS…
        …. boom!

          1. Except that people who use metric on a regular basis don’t really see n, mu, m, k,M,G,… as units but as numeric prefixes, integer powers of 1000, i.e. we don’t rapidly confuse km and m as they instantly convey a scale…

          2. Exactly. I had an issue at work this week because Texas Instruments specified the delay on a regulator IC as being 200ms, i.e. milliseconds.

            Turns out it was microseconds, and metric doesn’t prevent unit conversion errors.

          3. @Steven Gann: The metric system prevents also those errors, as micro should only be abbreviated with a small u or μ. Doesn’t prevent against TI having typos in their datasheets, though.

          4. @Steven Gann: I can understand that if you are a hobbist (witch I am), you can err by a thousand, but, you say you were at work, so apart from the error from TI (main fault, of course), you should have know better what’s the typical response time of a kind of IC that you work in, you can even look for others IC in the family or similar. But apart from that, I totally agree with Dan, I don’t care about the imperial units, I try to keep it as far a I can get.

          5. Exactly. I had an issue at work this week because Texas Instruments specified the delay on a regulator IC as being 200ms, i.e. milliseconds.

            Turns out it was microseconds, and metric doesn’t prevent unit conversion errors.

            Except 200 microseconds should be written as 200µs, or 200us for the keyboard-challenged. Whomever at TI thought “200ms” was the correct way to express that unit of measurement should be taken out and given a lesson on SI prefixes.

          6. Let me ask you: What is the typical current that will flow through a mosfet when it’s supposed to be blocking current?

            The actual number in practice can easily be a factor of 1000 off from the value in the datasheet. It’s all fine: The number in the datasheet is a max value, and the actual measured value is below that. It’s just that it’s a value that engineers don’t care about. low enough is low enough. Real devices often have have a value that’s 1000 times better, but if things degrade by a factor of 1000, too bad, things will still work.

    1. Awww. You beat me to it!

      To the author: Generally, I don’t think it’s the scientists that have an issue. It’s the engineers who have to convert the scientists’ arcane doodles (and archaic units) into reality. The rest of the world (for the purposes of our economy, close enough) uses metric, and there should be an assumption that anything that is going to travel abroad should be designed in the appropriate system. Making engineers switch between systems is always annoying, frequently expensive, and sometimes spectacularly catastrophic. But whatever – the imperial system works for you when you drive home and buy groceries.

      The only argument I can think of for sticking with Imperial is that . . . we currently use imperial. Skipping lots of the detailed arguments, If the Imperial system functions so well, I’d expect to see more countries still using it. Instead, the US finds itself in the refined company of (drum-roll please) Burma and Liberia – engineering powerhouses, to be sure. Let’s hang out with the Germans for a while and see how it works, ja?

      1. No, there was another argument, and you missed it.

        Imperial is a very HUMAN system. A foot is about a foot. An inch is an easily divisible chunk of a foot. For intrinsically human things, imperial is easier.

        The divisibility is another thing that didn’t get mentioned, but it’s another very human thing. Consider: You’re making a recipe, but you’re cutting it in half.

        375 g sugar.

        How much is a half recipe?

        2 cups sugar.

        How much is a half recipe? And that works reasonably well in most imperial things, because it’s NOT decimal. Again, a foot is divisible into inches by 2,3,4, and 6. A meter has to be divisible by 2 or 5. Or you can use metric, and use a half dozen significant digits. That sounds easy, ja?

        1. 2 cups of sugar is 400 grams, or a shade under 500 ml. Pretty sure most people can handle halving those numbers just as well as they can 2 cups. But if you want to artificially construct your argument, you can pick numbers to make either system look easier or harder.

          To make a more generally robust argument I’d say that Imperial units are fine for eyeball and rule of thumb measurement, but metric is easier if you’re concerned about precision. That’s my experience anyway, being comfortable in both systems. (Except weighing people in kg, and measuring fuel economy in L/100km. Screw those two ideas.)

          1. @John Grimoldy:

            And points of reference a huge part of what this debate is about, aren’t they? Which highlights how futile it all is, because none of that is going to change for anyone.

          2. Why? I prefer the l/100km way of expressing fuel consumption.

            But if you want to convert between mpg and l/100km, that’s easy. Take the number 235, divide by what you have (mpg or l/100km) and you get the other number.

            (Ok, it’s not quite 235 but 234.something but it’s close enough for daily work)

        2. You are picking numbers to make metric seem complicated and imperial easy.

          How about half of 1.6 kilogram to grams vs half of 1.6 pound to onces?

          A meter can be divided into anything down to sub-millimetre if needed, always in base 10. That’s what cm and mm are for. They’re also easy to add up because you just move the decimal point. But with imperial you have to deal with fractions such as adding 9/64″ with 15/32″ and 7/16″.

          1. Fractional inches are useful until they’re really, really not.

            Machinists usually solve this with mils, I think it’s woodworkers who tend to really get stuck with that crap.

          2. The inch can be divided into anything down to the microinch if needed. It can be in base 10 just as well as base 2. The base 10 divisions are easy to add up because you just move the decimal point. The base 2 can be derived on the fly; if I have an exact inch it is easy to scribe it to as fine a resolution as I need without creating a separate item divided into 10 parts.

            And that would be (9+30+28)/64 = 67/64. I guess it’s a challenge for those lacking multiplication skills. I will say that 35 years as an engineer I never came across the need to add fractional inches.

          3. I picked that because it was something I’ve actually seen and used. It was not the kind of recipe where they could adjust to 400, and it was 375. Some things are not easily adjustable up 1/15 to make the math easier.

            If the recipe was formulated in imperial, conversions are a problem. And until you can find me a metric egg, just adjusting the recipe is not an answer.

        3. You are really missing the point. The base units don’t matter at all. Using meters or inches is irrelevant. What makes the metric system so widely accepted is its scalability. When I place 10000 20nm transistors side by side it is trivial to figure the total width is 10 * 20 um or .2mm. Using imperial units it is much more difficult to scale and hence airplanes fly at 30000 feet instead of 6 miles.

        4. My nine year old daughter would like to talk to you about foot size.

          She’s also not so hot on your idea that 3/16 of something is a reasonable way to describe something, that might be compared to objects that are 11/64,5/32, or any of a bushel of other annoying conversions – but she’s nine, and she’ll figure out the details soon enough. “why can’t they just use the same fraction [sic]

          I’m sure that if I reminded her of how easy recipe halving was while she was making pie crust, which calls for 1-1/3 cup of flour, she’d have words with me after she spent a while looking for the 1/6 cup measuring cup. On the other hand, she’s perfectly capable of working out the long division of 375/2, and it won’t be long before she can do that in her head. But by then, though, she’ll be a deft enough hand at making cookies that she’ll know that the difference between 375 and 380 grams per batch isn’t enough to mess them up.

          Also, one of the best pastry chefs I know is from Germany, and he was flabbergasted that most US recipes call for cups and ounces, so I expect he’s probably got the metric system working for him too . . .

          I’ll grant that dividing a whole thing in half is easier than dividing an arbitrary three digit number, especially when you have a measuring tool that that comes in “halves”.

          1. > My nine year old daughter would like to talk to you about foot size.

            A valid point that will get you nowhere with most of the Imperial-wielding adult males in this sort of conversation.

            Things like “a foot is roughly the length of your foot” are just post-hoc rationalizations. I could say “10cm is roughly the width of my hand” but that doesn’t make it a sensible justification for the metric system. People internalize units that make sense to them at different scales, and then that measurement system just makes sense to them.

            I grew up with both systems and for certain circumstances I find one or the other feels more natural, so I use it. If I had to pick just one I’d go with metric, because the math is just so much easier. Other people would obviously choose otherwise. Some people who get into these discussions really need to take a chill pill.

          2. I don’t understand the US/UK/AUS insistence on using grams for cooking in metric.

            It seems there’s some sort of fetish or misunderstanding about being precise because metric is supposed to be that way, but in reality it’s just complicating things and nobody else does that.

            Metric cooking is done by the deciliter (1/10th of a liter), and your measuring cups come in fractions of a dL which is easily convertible between mL and cL. Teaspoons and tablespoons are used, but those standardize to 5 mL and 15 mL. Just about the only thing that’s in grams is butter, and the butter stick has a scale on the wrapper to tell you where to cut for what amount.

          3. Dax, the reason for using grams is that measuring by weight is a *lot* more precise than by volume. Even better, if you simply place your mixing bowl on the scale and zero it every time you need to add something else, you can add ingredients straight from the container. (…as long as you are careful about it.)

          4. >”the reason for using grams is that measuring by weight is a *lot* more precise than by volume. ”

            But such precision isn’t needed, and it makes the receipes tedious to scale up and down.

            And you can’t actually use such precision in real cooking, because things like flour vary in grade and quality. If you substitute one brand with another, you need to adjust the amount of liquid etc. so the receipes are very approximate – you start with the amounts specified and then adjust according to how it goes.

          5. >”if you simply place your mixing bowl on the scale and zero it every time you need to add something else”

            That’s assuming you can simply throw everything together without mixing, kneading, whisking… how do you gradually add 100 g of flour to a bowl on a scale when your fist is in the dough?

        5. You make good points. I particularly like the recipe example.

          Born in the U.S., I’m comfortable and familiar with the Imperial system — right up to the point of needing a wrench or socket. If a 5 is too small, grab a 5.5 or a 6. If 7/16 is too small, what do you reach for? I know the answer is 1/2, but I think you get the idea. Imperial can be annoying at times.

          Also, there is an undeniable brilliance of 1 cubic cm of water = 1ml = 1 gram. Water freezing at 0 and boiling at 100 (at sea level).

          I hate buying gas in Canada because I need to convert to Imperial and also convert the currency in my head. I simply am not comfortable with how much a liter is.

          I’m not interested in leaving Imperial, but lets admit that its cousin, Metric, is beautiful in many ways.

          1. Fractions are a stupid way to measure length. Full stop. Try measuring a stray allen wrench or drill bit with a micrometer sometime. However, I see nothing brilliant about basing the range of 0..100 in a temperature system on the life of water (Celsius) rather than the life of humans (Fahrenheit), since an awful lot of temperature measurements concern humans (think weather). 1ºC is too coarse of a measurement for a thermometer or thermostat, whereas 1ºF is fine.

            I do remember reading some instructions that said that if something was too hot it should adjusted by -5ºC (or 23ºF). Sigh.

            Speaking of units, if you have not found Frink (https://frinklang.org/) go get it and start using it! For doing physical calculations there is nothing like it. Not only does it know units (umm, all of them?), it also know physics (how units combine to make other units). If a square 50 miles on edge gets 1/4 inch of rain, how many gallons is that? (50 mi)^2 .25 in -> gal 1.0861714285714285714e+10. About 10 billion gallons.

        6. The argument that imperial is a more “human” system is just completely and utterly bogus. That’s pretty much on par with arguing that your own mother tongue is so much easier than anyone else’s.

          As someone who has spent about half of my life in Europe and half in the US, I can assure you that once you get used to it metric measurements feel just as intuitive and natural as imperial.

          I can easily guesstimate the length of a room in 1m paces, much faster and easier than pacing it out in feet (and about as accurate, considering that my feet aren’t 1ft anyway). A cm is about the width of my finger. I can just tell the difference between 19 and 20 Celsius, but I would be really hard pressed to tell the difference between 67 and 68 Fahrenheit. Grams and kg are easy – I know what a 20kg sack of potatoes feels like, and what 100g of sliced salami or 200g of flour looks like. Plus you can get a very easy size/volume estimate for many materials by keeping in mind that 1 liter of water weighs one kg.

          By the way – fractional inches, and having half a dozen length units with weird-ass conversions between them, are just a huge pain in the ass. I remember one news story during the Iraq war about Saddam Hussain ordering trenches full of crude oil being set ablaze. They had this many miles of trenches, with each trench being so many feet wide, filled with so many inches of oil. I was just waiting for them to turn it into a highschool math problem and ask how many gallons or barrels of oil that would take. In metric, this would be dead easy: simply multiply kilometers, meters and centimeters together, make sure you keep your orders of magnitude straight, and you can covert everything straight to liters…

      2. Then there’s the UK, which uses units like “stone”. I’m not sure what system that is, but it sure isn’t the metric we use in the USA.

        Which, for the record, every American engineer, scientist, and physicist worth their salt uses metric. Imperial seems to only be used for commerce, cooking, and car mechanics.

          1. In laying out PCBs, I usually have to deal with a random assortment of parts created on an imperial grid, parts created on a metric grid, and parts created on no discernible grid at all. I’m just as comfortable with millimeters as mils, but I’d be happy to get rid of one or the other. I was glad when Ford decided to have the U.S. join the world by signing the the Metric Conversion Act of 1975, and not very happy when Reagan put an end to it.

        1. I wouldn’t jump to this conclusion. I tend to be pretty proud of much of my work and am most comfortable in imperial. I also deal with miner’s inches, both western US versions. Cubic feet per second is just my default. My pet peeve is using units different than those developed for the situation or mixing. I remember another group’s project where rebar was spec’d in imperial converted to metric, causing some bars to be increased, converted back to imperial increasing other bars thereby increasing cost until the mistake was caught.

        2. The stone is part of the Imperial system, which the UK uses. The US uses Customary and does not use the stone. Some units are common between Imperial and Customary, and some aren’t. Basically some units predate the Imperial Act of 1824 and were used here by King George III’s redcoats before they were sent home in 1783. Those units changed or introduced by the Imperial Act of 1824 are not used in the US because “MURICA!”

          However, I am unable to explain how continuing to use the pre-revolutionary units of our former colonial masters represents “freedom units.” That one just stumps me,

          The primary differences between Imperial and Customary are units of liquid and dry volume (gallons and bushels), and larger units of mass (stone, hundredweight and ton)

        3. We only use ‘stones’ for people’s weight oddly enough, and I’ve never seen anything other than people weighed in stones. The only explanation I can think of is you’ll usually weigh a number between five and twenty stone, and one stone is a large enough quantity that small fluctuations (eg if you’ve just eaten something) won’t change your weight much. In metric your weight can change by a kilo over the course of an hour or so (eg if you drank a litre of water).
          I think there’s a very slow move to using metric instead of stones, for ease of maths etc.

          1. small error though, the temperature in my sauna is sometimes over hundred degrees of Celcius and I’m not dead. As far as Fahrenheits and Celsius are concerned it is more about the exposure time although I think we have cryogenic treatments where the room temperature drops close to 100K so this I guess falls into the same category. 0K I agree, one would be dead … oh wait did I hear someone to say suspended animation?

      3. Even Burma & Liberia have now switched to Metric. The USA is literally the last country on earth that still uses Imperial measurements. Personally, I think that if America is stupid enough to shoot itself in the economic foot by demanding archaic & arcane screw threads, tools, etc, we other 7+ billion people will just shrug & continue to benefit from the economies of scale that come of going with the flow.

  2. The issue of needing to remember odd conversions within the Imperial/American system is another red-herring. Those working in inches tend to stay in inches, as do those working in linear feet, or yards, or miles. The fact is in practice one is rarely called upon to convert or even use mixed values.

          1. It only became a problem because the airline brushed off a safety rule about not flying with all the fuel gauges inoperative – and the flight crew did not request that the tanks be ‘dipped’ (poke a stick into the fill hole) to ensure they were full the last time they got fuel.

            Just a lack of dumb simple common sense. If you’re violating a safety rule relating to being able to tell how much fuel is in the tanks *always* request the fuel crew put in however much it takes to fill them.

          2. There were no safety rule violations in this incident. The MEL allowed operation without the gauges, and the checklist neglected to assign —clearly and specifically— the responsibility for calculating the fuel load in an abnormal situation. It was not a lack “dumb simple common sense” and no airline fills tanks to the max at every fuel stop.

        1. As a Canadian, I have to say that your “perfectly functional system” is a nightmare. We’ve had to live with both metric and imperial every day of our lives for decades now. You deal with something made in Europe, Japan or China, it’s metric. Something made in USA, it’s imperial. Made in Canada, could be either one. We’re probably one of the few countries on the planet where you actually NEED your rulers to have both measurements on it. And the only reason we use imperial too is because of the construction and manufacturing industry getting a lot of supplies from the USA.

          So my reason for dumping imperial would be to have a single measurement system for the whole planet and stop wasting time and making mistakes for no good reason.

          1. Do you understand you cannot ‘dump imperial’ like it was a girlfriend you’re tired of, legacy issues mean we are married to this system in several key sectors and nobody will be happy to pay what a full divorce would cost.

          2. As DV82XL says, it’s not easy, not that we’re in a good place. There will be imperial bolt sizes and threads (and plumbing and on and on) for a long time even if we “drop” one system right now. As a maker with a machine shop, having two is the main evil – I’m fine with either, and when I do science or any subset (say, chemistry) it’s often in metric. As soon as I go down to the shop…well, it’s another story. I can eyeball all the standard thread and wrench sizes easily – for imperial. I get fairly close on metric. Which would be a heck of a lot easier if I knew going in which I was working with. Having to own roughly twice the tools is a pain. Although, if you’ve got a half inch bolt, sometimes either a 12 or 13 mm will do well enough…or 15 for 9/16. But when threading say, a gun barrel, it’s even worse. not all threads are…60 degrees (see: Whitworth). having two systems is a pain, and having to stock two sets of tools and parts is expensive and time wasting, and telling which you have just now is also a holdup sometimes. For a number of years, American cars had both systems, depending on where the part was sourced. Talk about a nightmare. Just be glad we didn’t fall for it with volts, ohms, amps, farads and henries…but we do have radians and Hz (which were CPS when I started).

            I’d rack up that Ti error to simple incompetence, perhaps along with the lack of the correct unicode u on the keyboard. Read forums on the net these days. Look up say, Mark Dices “how many books have you read” on youtube. It seems most of the population can’t tell loose from lose, to-too-two, and let’s just not even get into affect and effect. And that’s all one language…supposedly.

            Any long time engineer or maker knows darn well to check data sheets for reasonable content. What semiconductor is that slow? Duh. It’s always been caveat emptor. If an opamp is fast and quiet, it’s a power hog , or has hella bias current(and they leave that out of the front page). It used to be that a 12 bit a/d had 1/2 lsb accuracy. Um, nope, not even close anymore. And that crap, mentioned in another thread here about kiB vs kB – oh please $deity spare us from marketers. It’s not the system as much as what we’ve let them get away with!

      1. The orbiter bit the dust BECAUSE of the metric system. NASA wanted to go metric and insisted on foisting this nonsense on its vendors.
        NASA still insists on metric for a few of its programs. All that means is vendors design in INCHES and put dual dimensions with metric as primary and inch as secondary. You know because CAD software has this new fangled dual dimension capability.

        1. It’s got to be at least somewhat annoying that you then have to fabricate parts that wind up being something akin to 92.23 mm in primary size, rather than a “nice” round number though, right?

          1. Then a shop takes your 2 digit truncated metric number, converts it to inch and totally looses the intent of the design. The resulting nonconformance ends up costing the taxpayers $10,000 to resolve for a use as is disposition. All becaue metric is so much better…

        2. NASA ought to shift to metric – c’mon, it’s not rocket science! ESA does everything in metric, Russia does. Jaxa do everything in metric. The Chinese space agency do, the India space agency does.

          What’s so hard about counting to 10? If they can’t do that, they don’t deserve to be launching rockets.

          1. My understanding (not verified) is that NASA and scientists in general use Metric almost exclusively in its work, but switch to Imperial when speaking with the public. Which is incredibly annoying when one these persons is on an interview here and they speak about miles, miles per hour, Fahrenheit … which means just about nothing to most of us.

          2. @Leonard ” Which is incredibly annoying when one these persons is on an interview here and they speak about miles, miles per hour, Fahrenheit … which means just about nothing to most of us.”

            Well.. then I am guessing this probably means that “most of us” aren’t paying the taxes that paid those scientist’s salaries are you?

        3. This is silly in the extreme. The U.S. Space Transport System bit the dust because it was antiquated and cost more to launch than expendable rockets. No shop declines to bid on a part because it’s specified in metric. They are all happy to do the conversions for the business.

          And by the way, if you decide to go with an ESA or SpaceX launch, be assured that payloads are specified in metric units.

        4. And all vendors but one followed the terms and conditions of the purchase order and went metric. One violated the terms and supplied thruster data in lbf·s instead of N·s so that all course corrections were magnified course deviations. The orbiter bit the dust because one vendor hid the fact that he refused to use metric in spite of the specification and terms of the purchase order. But in your view, it is the fault of ALL the OTHER vendors for obeying the spec? WTF?

    1. I’m probably one of the few who have to do conversions almost daily when producing articles or videos online and have to cover a worldwide audience or answering the questions that result. But converting is easy, just google “23/16 inches to mm”. (They think of everything, don’t they!)

    2. I don’t know if it’s just Britain (I imagine it might be similar in the ex-colonies), but we tend to use both metric and imperial interchangeably. Sometimes I’ve asked my dad to pass me (eg) a bit of wood that’s two inches square and forty cm long.
      What units you use often depends on who you’re talking to, I might tell one person I’m 5′ 10″ and someone else that I’m 177cm.

      1. Sure, that’s the way it is here in Canada too, and it’s not going to go away anytime soon because of our proximity to the States. Then again I don’t see many people getting that upset about it.

      2. I can’t speak for all post-colonial countries (India for instance), but in South Africa, Australia and New Zealand, Imperial is dying -, apart from usage by the older generations or in niche areas. In most instances, if you speak with a teenager in Imperial you might as well be speaking in a foreign language.

      3. This is also the case in the U.S. We pretty much use whatever units we’re given – nobody ever buys a 16.9 oz coke; they buy a half-liter. Even waay back when we played LPs on turntables, the tracking force was measured in grams, not ounces, even though grams wasn’t an appropriate unit. All of the measurements used in sports that we participate in internationally use the same measurements as the rest of the world. Even our tires are measured metrically, although the wheels they go on aren’t. Just like most of the rest of the world. The whole world uses electronic components with dimensions specified in both inches and mm, and the U.S. is no different. There is a lack of standardization in electronic parts footprints that goes far beyond U.S. influence, forcing every one of us to physically check the parts that we get before committing a PCB design to fab. Even Chinese breadboards have their contacts on 0.100″ — oh sorry, 2.54 mm centers. Nobody has adequately answered my question about speedometers – where in the world are speedometers calibrated in FULLY SI units? And don’t tell me that “hours” are a derived SI unit, because they’re not, any more than an inch is a derived SI unit. Seconds are in fact a poorly thought out unit, since they don’t easily relate to the most important clock in our world – the motion of the earth. 86,400 seconds/day? That sounds like something only the British could come up with, and yet NO country was willing to give that one up, nor its base-60 foundation. We use what we’re given. I would guess that MANY countries that have adopted SI still use their old systems of measure to some extent, making the claim that “The U.S., Burma, and Liberia are the only countries not on the metric system” somewhat disingenuous.

        1. Heh, tires and wheels are really screwed up, you get a 205R15/70 tire that’s 205 mm wide on a 15 inch wheel, that has a 5 on 4.5″ bolt pattern for which you use M14 threaded lugnuts that might have a 3/4″ cap on. Then inflate to 240 kiloPascals and torque to 90 ft.lb…

        2. Nobody has adequately answered my question about speedometers – where in the world are speedometers calibrated in FULLY SI units? And don’t tell me that “hours” are a derived SI unit,

          1) Nowhere. All metric countries use km/h for speedometers and speed limits.

          2) Hours are not a derived unit. They are a “non-SI unit accepted for use with the SI” like the liter, tonne, hectare, degree, minute, and second of arc. They are covered in Table 6 of the SI Brochure. They may be used alone or in compound units like km/h or kW·h. Obviously, in calculations, you have to substitute the definition 3600 s to maintain coherence.

          No one really cares how many meters their car goes in a second, because few journeys are only a few seconds. People want to know when will I get there, will I be arrested? I don’t feel this makes a metric country non-metric, but I suppose opinions can vary. Certainly some metric fanboys think m/s should be mandatory on speedos and speed limits.

          Are you surprised that liters are not “official” SI units? Since they are “accepted,” it is not that big a deal. All nations use them in commerce, and Table 6 basically admits there is no hope of eliminating these units, they will be used indefinitiely and the SI accepts that.

          1. And the SI Brochure is a free pdf download. Everybody in this debate should have a copy.

            NIST publishes an Americanized version (American spellings of meter, liter, metric ton, deka, preference for L as the symbol for liter, etc) as NIST SP 330, also a free pdf download. Finally, NIST has an excellent metric style guide binding on NIST employees, NIST SP 811. Honestly, it is a bit picky on issues where I am not quite so picky, but it is free and therefore better than paying for ANSI SI10, the main industrial spec for metrication and using the metric system in the US.

            Note that Congress has delegated its Constitutional power to “fix the system of weights and measures” to the Secretary of Commerce, who has internally delegated it to NIST. NIST SP 330 is the official guide to US usage of the SI; nothing like learning straight from the horse’s mouth.

        3. That’s a very interesting set of observations. It may be that Imperial (US customary units?) are not quite as pervasive as I have assumed (based on the comments in HaD). As for decimal time … that was an interesting discussion. I would be very happy to go with decimal time – calculated by the third most intelligent species on an utterly insignificant plant orbiting a small, unregarded yellow star.

          I would take the view that the use of Imperial for a small fraction of niche/legacy applications does not prevent a country claiming to be Metric.

          1. Yes you can. It’s easy.

            You take a piece of paper and fold it in half multiple times until you get more than 10 divisions. Then you place your ruler on the paper at an angle so that there are exactly 10 creases crossing the side of your ruler over the distance you desired. In this way you can actually divide a line into arbitrary equal divisions:

          2. Another way of doing it is to construct a wooden frame with hinges at the corners, divide the two opposing sides into equidistant divisions, then pull thin strings or metal wires across the frame. As you skew the frame, the distance between the wires decreases while maintaining equal division.

          3. Dax, that isn’t precise like folding tin or gold foil. And you need to already have a way to make a series of parallel lines hat are more accurate that the folded ruler. The error bars for the plane geometry method is roughly width of the lines times sin of the angle of intersection. A fine point scribed in a metal with no parallax might do OK, but it assumes a lot that you don’t need to do the folding.

    1. I am sure you could at least use a “woodworking calculator” that would make these calculations easier though. Not that they shouldn’t be more exact to begin with but at least it’s an easy workaround.

      1. “Shouldn’t be more exact?”

        Wood, see, it’s not always the most precise thing in the first place. There’s no real point in going out to sub-one-thou tolerances when you’ve got something that can grow or shrink by 20 or 30 thou just depending on the humidity in the air.

    2. So… you get a stick with 1/32″ markings on it, and shave off 5 of those. Why is that insane?

      More importantly, how is that different than taking something at 1.805 and shaving off 0.005? You still go get a stick with 0.001 markings on it, and shave off 5 of those.

      1. Sure, except that generally you’ll calculate what thickness to shave down to and set your equipment accordingly. It doesn’t matter that your shaving is an exact thickness, you’re going to be more concerned about the thickness of the finished part.

        Now let me see, 1.805 and I need to reduce it by 0.005, whatever could that final thickness be…

        I’d say the real problem here isn’t really Imperial units so much as doing something in fractional inches that really needs the precision of mils. Those woodworking instructions are whack.

          1. I know what mils are, but you don’t work with mils in the form n/1000″, you deal with them as decimals. With the exception of a few fields like woodworking, using fractional inches is pretty rare.

            If you wanted to you could do all your math with roman numerals too, but it’s pretty uncontroversial that doing so is harder than using Arabic numerals. Fractions are harder to work with than decimals in the same way.

            Oh wait, are you thinking I’m making a metric v Imperial argument? I’m pretty ambivalent on that part. I lean metric, but use whatever you’re comfortable with as your base units, that’s fine by me.

        1. even that wouldn’t be the problem, if people stuck with 1/32s. But since they’re fractions, and we’ve been taught to reduce fractions whenever possible, so we never see anything specified as 16/32; it’s just 1/2. So they end up mixing 1/2s, 1/4s, 1/8ths, 1/16ths, and 1/32nds.

          And let’s not even get started on why 1/16 is a “sixteenth”, while 1/32 is a “thirty-second”.

  3. I think you’re missing the point by a kilometer.

    Conversion between metric units is very easy:

    10 millimeters (mm) = 1 centimeter (cm)

    10 centimeters = 1 decimeter (dm) = 100 millimeters

    100 centimeter = 1 meter (m) = 1,000 millimeters

    1000 meters = 1 kilometer (km)

    Also:

    1 liter = 0.001 cubic meters

    1. What I’m trying to say is, it’s easier to learn and makes more sense.
      The fact that you as a scientist are able to juggle multiple standards does not mean it wouldn’t be benificial for most people to stick to the more logical system.

      1. “What I’m trying to say is, it’s easier to learn and makes more sense.”

        It’s easier to learn?? Really? If I give you the volume of the Earth in cubic zeptometers, it’d be easy for someone to tell me how many zetaliters that is? And it’s easy to figure out how many hectometers you travel in a week when you’re moving at 1 micrometer per nanosecond?

        Just because you as a metric acolyte are able to juggle multiple prefixes does not mean it wouldn’t be beneficial for most people to stick to units convenient to whatever they’re measuring.

          1. …? Aren’t you making my point for me? It’s OK for computer systems to use convenient units which sound exactly like metric units, but which are close enough, but it’s not okay to use imperial units?

          2. “You can deal with TERABytes, ZETABytes, YOTABytes”

            To be fair, only one of those is actually used. More importantly, those prefix have nothing to do with SI prefixes, unless a kilometer has suddenly become 1,024 meters.

            …which it really should.

          3. Actually, tera, zeta and yota are SI prefixes and therefore base 10, so not really suited for quantities of digital information. There are derivations of metric prefixes established by the IEC for base 2 units. So, tebi, zebi, yobi.

        1. 1 cubic metre (PLEASE spell it properly, Americans!) = 1000 litres. So the answer is 1000 x however many zeptometres. Pretty easy, really, you just say “thousand” at the right place in the sentence, or add 3 zeroes.

          The prefixes are standard SI ones, and the same ones apply to each unit. And most of all THEY’RE ALL BASED ON TEN! If I gave you the volume of the Earth in cubic feet, could you convert that to gallons? And what do you measure the Earth in, zeptoyards? Yottagallons?

          The popularity of scientific notation proves the metric system is easy, it’s the same system effectively. Moving decimal points left and right is utterly easy. Powers of ten are a piece of piss.

          1. “1 cubic metre (PLEASE spell it properly, Americans!) = 1000 litres. So the answer is 1000 x however many zeptometres. Pretty easy, really, you just say “thousand” at the right place in the sentence, or add 3 zeroes.”

            I love it when someone proves my point. And, by the way, just because your spelling has been corrupted by the French doesn’t mean ours has to be, thank you very much.

            Volume is distance cubed, so the conversion from cubic meters to liters has nothing to do with the conversion from zeptometers (10^-21 meters) and zettaliters (10^21 liters). You can go from zettaliters to cubic meters, but then to go from cubic meters to cubic zeptometers it’s 10^21, cubed, or 10^63.

            And boom, there goes the metric Mars rocket…

          2. This is the argument that I hear most often when trying to explain why metric is easier, because moving a decimal around on a still arbitrary scale makes more sense. People don’t typically talk in Dekaliters or femtograms so that “convenience” is largely irrelevant. Also, just because there are 100,000 cm in a km doesn’t mean there is any greater concept of understanding what that length is by learning the metric system.

            What’s the SI unit you would use if you, say, had 1000 kilograms? Oh, there isn’t one? Lets just call that a tonne. Do you ask for a pint at the pub or 0.6L? Have you ever used the word stone to describe your weight? How many BTUs are in a Joule? Don’t act like because a country is on the metric system that’s the only units that are used commonly.

            You know what other SI units are popular in science? moles, parsecs, candelas. Those aren’t clean base-10 values. You would get just as weird and unintuitive a value for these units if you used metric or imperial.

            I work as an engineer in injection molding so machining is done with decimal inch, part volumes are calculated in cubic inch, and material barrel volumes on our presses are in ounces. Doing calculations for space missions isn’t a concern of mine

          3. Fine, oh foolish old me. I’m not actually au fait with the more far-out extremes of SI prefixes, just off the top of my head, and I therefore fell into the trap of assuming Pat would ask a sensible question. But nope, one was big and one was small! Well set trap, Pat.

            And it doesn’t matter. It’s still scientific notation, (Whatever x10 ^ Whatever). Cubic metres to litres isn’t even a conversion, they’re both units of volume.

            Do it with gallons and yards or whatever. Show me the working that’s easier than just adjusting the exponent.

            Just out of interest, how would you specify [an absolute shitload] of yards? Or Zetainches or whatever? The SI prefixes go with the Metric system because both are base ten. But they’re not a necessary part, just a very handy and easy part. The Metric method, you just change the exponent. Show me the old-fashioned analogue method. Use any Imperial units you like.

            As far as the French go, since they actually invented half the fucking language, let’s use the proper spelling. I’d be embarassed if Daniel Webster had vandalised my language, because he thought it was a bit too difficult for my countrymen to be able to manage.

          4. >”Do you ask for a pint at the pub or 0.6L?”

            Neither. It’s “A beer please”, and you trust that they know what sizes they come in and how much to charge.

            If you need to specify further, it’s a “A small beer please”, or “A large beer please”. If you’re being anal about it, then it’s “oh four/five/six” for the exact sizes of 0.4, 0.5 and 0.6 liter pints, which you will find marked on the side of the glass. In any case, it doesn’t matter exactly how much beer you get as long as you’re paying the right amount for the volume, because you can always ask for more.

        2. “It’s easier to learn?? Really? If I give you the volume of the Earth in cubic zeptometers, it’d be easy for someone to tell me how many zetaliters that is?”

          psss… zettaliters = cubic zeptometers*10^-45

          1. If I ever see someone using yottas and yoctos I guess I’d care, but all I ever see for extreme scales is scientific notation applied to base units like metres or kg. It’s not like US units scale well into those ranges either, so I don’t really get your point.

          2. You can ask Google!?!? I can just say “thousand”. Or if someone’s asking me a silly question of a vast amount in a tiny unit, I suppose I’d do the mantissa / exponent notation. You don’t need Google to multiply by 1000, or 10, or 100, or whatever else.

            Shall we start teaching schoolkids to just ask Google? Cos I don’t fancy their chances of managing more than a couple of different Imperial units in one equation, and even that needs a calculator. And thereby risks losing precision.

        3. Yes, absolutely, it’s so much easier to convert between different kinds of units: Gram to ml to individual dimensions. Total doddle! It’s even easier to convert from miles to metres (x1608) and then do the maths on that basis.

          Metric: it’s just so easy!!! Stop fighting it, let the power of 10 work its measuring magic :-)

        4. I think you are missing the point of prefixes. The prefix should be chosen so the numerical portion is a sensible sized number.. The diameter of the earth is around 12.8 Mm (megameters, if you are not used to it) so it would be incredibly stupid to express it in zeptometers. The earth is actually an ellipsoid with the polar diameter is slightly smaller than the equatorial, so the volume is about 1083 Mm³. Many metric practice guides (in particular SAE TSB003) discourage prefixes larger than one with the liter, so I’ll just stick with cubic volume. But if you really insist, it is about 1083 ZL, or better, 1.083 YL.

          Problems like “how many hectometers you travel in a week when you’re moving at 1 micrometer per nanosecond?” are a way for teachers to teach scientific notation and teach children to hate the metric system (the American way). Clearly you are moving at 1 km/s, a week is 7 x 86400 s, so 604.8 Mm/week, and it would be stupid to express in hectometers. (6 048 000 hm)

          1. I think you’re missing the point, too – why bother having prefixes when you could just say that the diameter of the Earth is 12.8E6 meters? Now you just have to learn scientific notation, which is useful to learn anyway, rather than having to learn all of the prefixes.

          2. The “E” is widely used in high level computer languages for this purpose, but it may not be understood by people other than programmers. It is also used to represent 14 in hexadecimal. The more usual scientific notation replaces “E” by “times ten to the power of ____”. But scientific notation is available as an option to prefixes, and might be worth doing for the extreme prefixes that most people have little experience with. I’ve thoroughly memorized and regularly use prefixes from pico to tera. Beyond that, I usually cheat and check. YMMV.

            If prefixes are actually the hurdle in you using the metric system, by all means using scientific notation. Just metricate. But the truth is that you will find most other people prefer and use prefixes, at least the more common ones. The SI does allow some choice, but choose sensibly.

    2. No you are missing the point: who cares? Just how often do you need to do this calculation? I worked as a chemist and a metallurgist in aerospace for forty years, I used both systems daily and it was rare that I needed to convert between systems or within a system. Values were expressed in one way throughout any process, and no one was all that inconvenienced.

      Finally, on those occasions where one had to convert EVERYONE used a calculator, every single time, because no one did calculations of any sort by hand anymore. We could not spare the time, nor the possibility of error.

    3. Okay, but who cares? How often do you really do this? I honestly cannot recall the last time I had to convert miles to feet or feet to inches, or gallons to cubic feet. “But what about scientific uses?” On the one hand, most science is done in metric, and has been for ages, and even when it’s not, conversion is still a fairly unmotivating argument.

      1. I think “who cares” is an easy position to take for those who have end to end control of the results. Where I’ve seen “caring” was required is when one didn’t have that control, and had to deal with what was given.

        1. The thing is I was that person for most of my career and it really wasn’t that big a deal. Real-world numbers are seldomly nice and round anyway, and I don’t have to take off the other glove to count the times the significant figures were a concern in cases where there was a conversion. On top of which, I always used a calculator, more often than not, stapling the tape to the document. Was it an extra step? Yes. Could I have done without it? Sure. Was a full company-wide conversion worth the hot steaming turds that were going to drop on my lap making that happen? No.

      1. Yes. Making a ruler by folding in half and then again and again is a terrible idea. Using rational numbers, and rational arithmetic, which is exact, is a bad idea. The dials on an oscilloscope going 1, 2, 5, 10, 20, 50, 100 instead of 1, 10, 100, 1000 is a bad idea. Something smaller than a factor of 10 is a bad idea. Etc. Give it some thought.

        1. In all fairness, I will say that using powers of two in a base 10 counting systems worked acceptably well until we got our brains screwed on straight and figured out something better.

          It also helps that with commonly available tools, any idiot can measure something with reasonably accuracy and precision now. When “close enough” was have the width of your thumb, there was probably less need for a decimal system to describe distances (volumes, intervals, etc) that were less than a whole unit of measure. We’ve since moved on to needing a more robust system.

          1. For me, there are two problems with SI. One is cosmetic and is the nomenclature. The French System arose from elitist (it’s just the way it was) eggheads who wanted to communicate scientific ideas that covered huge ranges of scale and allowed easier manipulation of very large and very small numbers. They also new Latin and Greek, so the use of Greek for prefixes less than 1 and Latin for greater than one was very clever and handy – for them. But the names are very unsuitable for a work environment or communication where there is noise, meaning anything from sound to poor handwriting. The Imperial/American systems use single syllable words that never similar sounding, and it evolved this way because it works better. The SI use of words easily mistaken for each other, and in fact more confusing because they all end in ‘meter’, lets one can think they heard it right. Terrible human factors choice.

            The other is the scale. If it had all been defined so that a liter were about the same as an 8 ounce drinking glass, the whole thing would fit human use much better. A yard is the typical step of an adult and the average distance around the waist. Centimeter is a bit small for human scale activity and a meter is a bit large. A liter is awkward as well, being too big to be convenient for cups and glasses and too small for for typical water and fuel tanks.

            And scales that change by factors of ten look good in charts and tables but are terrible for actually using them. Plus, you can’t use rational arithmetic.

          2. A question: If Imperial works so well, why do only three countries still make it their official system of measurement?

            If you are communicating critical details by voice in a shop, there may be other issues. I’d suggest a highlighted set of prints, perhaps? That’s what we’ve used in every MFG facility I’ve ever worked in. It doesn’t seem like the Imperial units are any more immune to bad handwriting than metric, so I’m going to ignore that one.

            The knuckle on my thumb is almost exactly 20 mm wide. My outstretched hand spans 22 cm. My stride is almost exactly a meter. A yard is usually closer to the distance from the chin to an outstretched index finger (for American adult males) or half their height – whatever. A meter is 10 percent longer – whatever. As for the rest, I have a hard time believing that everyone in the US should stick to Imperial because you’ve managed to calibrate it to your various body measurements and such. When I go to Europe, no one gets so depressed by 0.5 L glasses that they feel like they need to fill it with beer just to drown their sorrows.

            That’s not sufficient for me. When you say imperial is better because it is a “human sized”, system what I hear is: “I don’t want to deal with the change.” You’re going to coming up with reasons not to switch as long as I indicate that I’m willing to listen.

            The thing that I hear lots of people missing though, is this: Most things that are made now, anre not made at a human scale anymore. If you are making ANYTHING, the error in the system is important. Today, many, if not most, of the things we build have tolerances that are smaller than any reasonable “human-sized” measurement. When I have parts manufactured (at Imperial-system suppliers), I expect machined parts to be accurate to 0.005 inches (I can, and have send parts back if they come in at 0.006″ from the nominal.). In this case, 1/128 inches (0.0078″) is not sufficient. If I’m working on more critical things, my tolerances may be tighter, 0.002 inches, or even 0.0005 inches. (Then there’s semiconductors, where we start counting in nm.) There’s nothing particularly “human” about these things and I need to be able to describe them accurately

            But here’s the deal: I actually don’t care what units you use or talk about in your daily life. If you want to use cubits, fingers, stone, or mouthfuls for measuring, that’s fine. It won’t work for me, but if it works for you do it. But, having the US use an entirely different system for measurement than every other country that my company’s products go to is annoying as all get-out. Screws are a perfect example: If I need a specialty screw, I may actually have to import it, or have it manufactured for me because most screw manufacturers in the US carry their full line only in imperial sizes. If I’m lucky and they do have the m6 screw of length L and steel grade X, I’m going to pay more for it here than a similar Imperial-threaded screw. If I use the imperial screw, I get push-back from our customer service department, because foreign customers don’t like to deal with them. You may say that Imerial units are comfortable, and you like them, and they match your body bits. I say that they are annoying and cost my company money.

        2. The 1-2-5 sequence on oscilloscopes is based on the fact that scaling something by 10 or by 1/10 is usually sub-optimal for measuring amplitude or period, because something would go instantly from too-small-to-measure to off-the-screen, so we split the decades into three rough parts. Similarly, bandwidth selection on spectrum analyzers is usually in a 1-3 sequence because that’s usually good enough and it would cost a lot more for the extra filters to do that in 1-2-5 sequence. Note that both 1-2-5 and 1-3 sequences were extensively used in analog instrumentation, all around the world, regardless of the units. An advantage of the 1-2-5 sequence on analog scales, is that when you split each major division into five minor divisions, it still makes sense, because for any of the “1” ranges, each minor division is 0.2, on the “2” ranges it’s 0.4, and on the “5” ranges it’s 1.0 units. The 0.4 is kind of awkward, which is why analog meters are often in 1-2.5-5 sequences, so for the “2.5” ranges the minor division is 0.5. I’m not sure why this wasn’t done on oscilloscopes.

          While we’re on weird divisions, think about the old ASA (later ISO) system for film sensitivity, which was in increments of 2^(1/3), rounded to nice numbers, which is why we had ASA settings on our cameras of 25, 50, 64, 100, 125, 160, 200, and so on. Note the discrepancy at 125, which to be consistent should have been 128, but the system was fudged to make it easier to use. I can’t think of anything else that ever used 2^(1/3), but there it was. And the DIN system was also based on this increment, but was logarithmic in the base of 2^(1/3), so each increase of 1 on the DIN scale is a sensitivity multiple of 2^(1/3), or about 1.25. Most digital cameras, while they still kind of use this system for their “ISO” sensitivity settings, eliminated all but the whole powers of 2, times 100. So that’s an ISO standard very definitely not based on SI or metric.

          And THEN we have music, where pitch in a chromatic scale is measured in intervals of 2^(1/12), known as semitones. America isn’t the lone holdout on this arcane scale – it’s still used everywhere, and if you’ve ever heard a scale based on ten notes to the octave, you’ll probably agree that the very weird chromatic scale has its benefits.

          1. The actual increment(ASA) is 10^0.1, approximately equal to 2^(1/3), but repeats from decade to decade. It is part of a system of preferred number or Renard series, equally spaced logarithmically. The same sequence is used in frequency analyzers for octave and one-third octave band filters.

            American Wire Gauge uses the 39th root of 92 (the 40th root of 100 would have been SO much better).

          2. No, the standard ASA film speeds do NOT repeat every decade. The series is (within the limits I show, I assume the scale goes further in both directions) 25, 50, 64, 100, 125, 160, 200, 250, 320, 400, 500, 640, 800, … Note that each value is double the third value below it until you get below 100. That’s 2^(1/3), which is approximately 10^(1/10), not the other way around.

          3. You may wish to check Wikipedia article. There were additional (intermediate) film speeds beyond your listing.
            https://en.wikipedia.org/wiki/Film_speed
            Also note that DIN, which is now “related” with ASA uses exactly 10 steps per decade.

            The R10 Renard series is 1.0, 1.25, 1.6, 2.0, 2.5, 3.15, 4, 5. 6.3, 8, (10) and repeats in decades. Admittedly, ASA rounds 3.15 to 3.2, and 6.3 to 6.4, otherwise, these values repeat in decades. to VERY high film speeds. Also looks like 1.25 rounded to 1.2, but not in higher decades.

          4. I did leave some values off of the ASA scale, but the ones I stated are identical to those shown in the Wikipedia article. This goes way beyond the decade repeating rate that you claim. Note that the “rounding” of 3.15 to 3.2 and 6.3 to 6.4 (how is that rounding?) agree more closely to the 2^(1/3) values than 10^(1/10).

            I see that I was wrong about the DIN scale, though; thanks for the correction on that.

          5. I think the weirdness in films speeds is related to F-stops or Water-stops. The holes used to stop down lenses needed to relate to focal length, and to make it “easy”, the diameter of the hole. They wanted each stop to be twice or half as much light, but halving light means halving the area of the hole. So, start with diam = focal length, start halving the area and find diameters and you get the ratios of diameters and they are the numbers you get in F-Stops. Simple :-)

          6. No, not at all. F-stops are related to the lens opening, and doubling the diameter of the opening allows four times the light through. An “F-stop” was defined as a change of square-root-of-2 in diameter, which corresponded to a doubling or halving of the amount of light passing through. ASA film speeds were different, as it took three steps of ASA speed change to be equivalent to one F-stop. The ASA speeds were linear, so it took half as much light to expose ASA 200 film as it took for ASA 100 film, for the same result. As JohnS points out, the three-steps-per-doubling aspect of ASA speeds was more about there being roughly ten different film speeds over a range of 10:1, so for example it took ten steps of the film speed dial to get from ASA 100 to ASA 1000, in logarithmic steps. Thinking about it again, I now realize that my observation that it was three steps for each doubling was just a happy coincidence that happened to make the numbers work better than they did in base 10 logarithms.

          7. ASA is linear in the same way F-stops are linear. A doubling step in ASA is one “stop” in a lens. They each represent a doubling or halving of the amount of light.

          8. Not quite, depending on what you mean by “linear”. Doubling the NUMBER of the aperture results in a QUARTER the amount of light, which is two F-stops, whereas doubling the ASA number doubles the sensitivity, i.e., one F-stop.

    4. “Conversion between metric units is very easy”

      *Those aren’t different units.* They’re different prefixes. None of them are based on a different *thing* than “distance traveled by light in 1/299792458 of a second.”

      SI has exactly 7 base units. Meter, kilogram, second, ampere, kelvin, mole, and candela. That’s it. SI also has a system of “prefixes” which people tout as being a great thing. They’re not a great thing. They’re stupid. Why should you remember that 1 centimeter is 1E-2 meters? Why not just write 1E-2 meters? Why not call it an neg2-meter, so you don’t have to remember the damn prefix? What the hell is a yottameter anyway?

      There are also a *huge* number of units *derived* from SI that are *conveniently near* things that are easy to remember or measure. A nautical mile is the distance on the Earth of about 1 arcminute of latitude. The speed of light is about 1 foot per nanosecond, a foot is about a third of meter, and a foot per second is damn close to a kilometer per hour. An astronomical unit is about the radius of Earth’s orbit. *All of those* are metrically-defined units, and the conversions between them are not “easy.” None of them are exact, either, but they’re close enough.

      Heck, the “metric system” (not SI) still has subsets of bizarre unit groupings between them, with units derived from centimeter-gram-second and units derived from meter-kilogram-second. How many dynes in a newton? How many gauss in a tesla? Or, of course, my favorite, how many statcoulombs in a coulomb? (Heck, you can’t even answer that last one!)

      The whole “imperial versus metric” obsession that people have misses the point. Scientists and engineers use whatever unit is convenient to use as the base unit *all the time*. They have to remember bizarre units *all the time*. Adding in a couple more (feet, mile) is no big deal.

      The entire advantage of SI wasn’t that its units or converions made sense. The advantage of SI is that you can convert between convenience units *exactly* – everything is derived from *one base standard*. And those convenience units *include* the old imperial units.

    5. “the meter is based on phenomena that are completely inaccessible to the people who will use is, and unrelated to their daily lives.”

      An water boils at 100°C, freezes at 0°C at one atmosphere with correspond to an altitude of 0 meters. water having a density of 1 so 1kg of water weights one liter. Very complicated, fortunately your got a calculator in the pocket.

      Imagine one second doing dealing with imperial system in electronics… and conversion tables for Ampere, Watt, Ohm, Hz and all the like…. do you really think science in general would have developed that fast/far with a system that takes i roots in pre-Gutenberg era ? Metric system didn´t pop out of unicorn farts. It´s an evolution of science, which developed because it was needed.

      Although it is, “just a way to measure the world” it lacks scientific logic behind. The only merit it has it that one can´t get learn to use imperial system without acquiring one of the most fundamental mathematics concept: fractions.

      But it seems the Mormon that sleeps in each of the Imperial Subjects still bites, denying logical facts because it feels attacked in its identity.
      From Wikipedia: “The metric system has been officially sanctioned for use in the United States since 1866,[2] but the US remains the only industrialized country that has not adopted the metric system as its official system of measurement: although In 1988, Congress passed the Omnibus Trade and Competitiveness Act, which designates “the metric system of measurement as the preferred system of weights and measures for United States trade and commerce”.

      But it´s not so bad, keep it, feel free to use it for the hundred years to come! It´s not the worst archaism of U.S. , anyway … which unit(s) does one use to measure the size of a… wall, for example ?

      1. You know what’s hilarious about your comment? That’s not how the temperature unit is defined.

        1 Kelvin is actually defined as 1  ⁄ 273.16 of the thermodynamic temperature of the triple point of water. That is, it’s 1/273.16th of the thermodynamic temperature (as defined by energy content – so really this is defining the Boltzmann constant) of water in a state such that it can be liquid, gas, or solid simultaneously.

          1. The kelvin scale is defined by that point (as zero) and by the triple point of water 273.16 K. In the future, it is expected to be defined by fixing the Boltzman constant.

    6. This. And for all those doubting, quick now, in your head: You have a cubic tank, one yard on a side, full of water, hanging perpendicularly from a lever that’s 20 inches long. How much torque does it exert? Roughly. Give or take ten percent. Need a calculator? That’s what’s wrong with imperial units.

      The equivalent problem in metric is easy. A tank one metre on a side, full of water, hanging perpendicularly from a lever 20cm long; the tank weights 1000 kg, which hanging from a lever 0.2m long exerts 200Nm of torque.

      Examples abound. Your one short tonne rocket has 50 lb of force applied from a standing start for a thousand seconds. What is the change in its speed? How much energy has been expended? If you’re burning diesel at 10% efficiency, how much fuel has been consumed? All trivial to do in your head in SI units, so long as you know the energy density of diesel is roughly 10 kWh per L. Even the non-SI kWh is easy to convert to joules; 10 kWh is 36MJ. Done in my head. If our rocket weighs one ton and has 50N of force applied, it will accelerate at 0.05 m/s/s. After 1000s, the change in speed is 50 m/s. It will have traveled 25,000 m and 1.25MJ will have been expended. A bit over 0.3L of diesel.

      I’m not saying any of this terribly complicated, or even impossible to do in your head. But it’s not *easy*, and it’s not easy to see when you’ve got it wrong.

          1. “A tank one metre on a side, full of water” followed by “the tank weights 1000 kg.” That easy part only works because the density of water is an easy number to remember.

            Kindof like how the speed of light in feet per nanosecond is easy to remember.

          2. “Metric is awesome because a liter is a cube 10 cm to a side” Gee, and how often does that come up in a conversation you’re having over beers with your mates?

        1. “So metric is awesome because water has a nice density?”
          No metric is awesome because it’s properties are well defined.

          It is a system that has been designed with ease of use in mind, rather than one that has been bodged together out of generally random measures some of which were chosen with no more care than to feed the ego of a now long dead king… and I say this as someone who actually enjoys the fact that I understand inches,feet,yards and miles just as clearly as I understand mm,cm,m and km –

          If I am going to design something new.. I have to admit I’ll plan it in metric units .
          If I am repairing something old, I full expect to be measuring in inches and feet.
          In summary SI for the future, Imperial for the past.

          Neither the future nor the past will vanish any time soon, assuming we can keep our current crop of fruit loop politicians, and power crazed kings’ egos under control of course. .

        2. Metric is awesome because it’s easy to compare similar things with good resolution. (Although it is also good for comparing things that are quite different with good resolution, and a few other reasons too…)

          Here’s a practical example:

          You just spilled a complete set of drill bits on the floor. Some unknown quantity of bits may have been lost under nearby heavy equipment. Fortunately for you, about half of the drill bits are actually marked, but the rest are not (too small for text). Here’s a set of the shop electronic calipers. Placing so much as a single bit in the wrong holder will result in your carelessness being publicly discussed by all who use the shop.

          Now, the question, are you hoping that this is
          – a metric drill set (Mcmaster PN: 3085A11), which has 118 bits.
          – an imperial drill set (Mcmaster PN: 8834A12), which has 115 bits.

          Which set is going to be faster to sort back into their holders?

          1. I’m going to go with the imperial set. After all, electronic calipers were specified. I’ve yet to see those without the ability to change modes, and with 3 less bits to pick up and sort, I’ll got with the imperial. Now, given dial calipers or a micrometer, it’s dependent on which set matches the units on those (I’ve used decimal and fractional versions of both metric and imperial. And yes, fractional metric is an abomination).

      1. You lost me there with your French system.

        Torque is r x F (r cross F). Force is mg. In your example, this is 1000kg * 9.8m/s/s = 9800kgm/s. r is 0.2m and torque will be (assuming right angle) 1980Nm. Isn’t that right? Did SI lead you astray?

      2. Since you feel like being such a high-and-mighty scientist/engineer/know-nothing, I’m going to skip over mocking you directly for how your arrogance has led you astray, and, instead, simply point out that you didn’t provide enough information to come to an accurate conclusion, and that your conclusion was only correct within the confines of the information which you provided, which is to say… Wrong.

        You see, you made the mistakes that a lot of newbies make when they come into my shop: Forgetting that not only does everything have three full and significant dimensions (unless you’re building it out of single sheets of paper, at which point one dimension becomes negligible, but I’d like to see you store 1685.56lbs of water in a box made completely of paper and glue), but also that everything has weight (and we’ll say for the sake of argument that fasteners have negligible weight). So, buddy, what’s the box made out of? Is it welded-together sheet steel? What’s the lever made out of, and what are its other two dimensions? Is there a lid on the box?

        Sheet steel is horribly heavy at approx. 490lbs/cu. ft, and thereby adding 22050lbs lidless, or 26460lbs lidded. Using the lidded version for the sake of simple engineering (in particular, it gives us an easy place to attach the lever), that makes the box alone a bit over 15 times heavier than the water itself. With a lever of 1″ solid steel rod at 20 inches, that adds another 9800lbs, bringing the total weight of the tank, the lever, and the water to right about 37946lbs, or, for the sake of this being an argument and not an engineering test, 38000lbs.

        20in is equal to 0.508m. 38000lbs is equal to 17236.51kg, which is equal to 169032.42N. Ultimately, this gives us a final result of 85868.47Nm, with a difference of 42900% between my engineered number and Tom’s ass-pulled one.

        And yes, I know that using something as heavy as steel is an extreme, but it serves to make my point: Every aspect of a system has weight, and Tom, for all his higher thinking and metric superiority, didn’t even think to include a way of containing the water, just sort of assuming it would stay connected to a lever. And before anyone brings up ice, it has a significantly different density than water (0.92g/cu. cm, as opposed to water’s 1.0g/cu. cm, or 920kg per cubic meter), and, besides, he specified using a tank.

        By the way, you just got told off by a theatre major.

          1. A 17200 kg object should weigh around 169 000 newtons, in a normal-ish Earth gravity of about 9.1 m/s^2. Apart from sloppily saying they are “equal”, aren’t the numbers pretty much ok?

          2. The 169,000 N number (you meant 9.8, not 9.1, but you did the math right) is correct. What’s not correct was the 200 Nm number from the previous calculation, which was wrong because the person multiplied 1000 kg times 0.2 meters, and got 200, forgetting to multiply by gravity.

            This actually gets at the reason why certain convenience-units stick around – because they end up being harder to screw up in useful situations. We measure energy consumption in kW-h, because, well, people run things for hours, not seconds, so you do less math. Foot-pounds can be easier to teach because something that weighs a pound, at a foot, generates a foot-pound of torque.

            Imperial often gets dinged for using a force measure as a mass measure, but there’s no free lunch here – if you predominantly use mass measures, you can’t forget gravity when you need force, torque, energy, etc. Newton-meters would be fine… except that it’s easy to forget that a 200 kg object doesn’t weigh 200 N. Lots of people would say this is a good thing, and in lots of situations it is, but there are plenty of cases where it wouldn’t be, too.

        1. You’re absolutely correct about taking all the components into account. But a difference of 42900 %, or 429 times sounds rather large.

          490 lb/cubic feet (or about 7800 kg/m^3) for steel matches the numbers I can find. That’s only about 8 times the density of water and IIRC even solid gold has about 19 times the density of water. That won’t explain it all.

          Let’s look at the other numbers: A box with weight 26460 lb would have a volume of (26460 lb) / (490 lb/cubic foot) = 54 cubic feet, meaning it would be e.g. a solid steel box with the a size of 3 ft x 3 ft x 6 ft. Two times the original cube which was said to have a side of one yard. I think you calculated the area of one side of the cube (3 ft x 3 ft), and then multiplied that by 6 to get all six sides. But that only gives the total surface area of the box, not the volume of the (hollow) container. We’d need to take the thickness of the steel into account to get the total volume and total mass of the steel. For the steel thickness, I can only guess, but probably not more than some millimeters?

          As for the lever, 9800 lb / (490 lb/cubic foot) = 20 cubic feet, which would be a solid steel box of 1 ft x 1 ft x 20 ft. I think you meant a rod of 1 inch x 1 inch x 20 feet, or 0,083 feet x 0,083 feet x 20 feet = 0,139 cubic feet in volume.

          I don’t think that still makes up the full difference, but the metric and imperial examples of the original post weren’t equivalent to begin with. Especially there’s a big difference between 20 inches and 20 cm.

        2. You’re absolutely correct about taking all the components into account. But a difference of 42900 %, or 429 times sounds rather large.

          490 lb/cubic feet (or about 7800 kg/m^3) for steel matches the numbers I can find. That’s only about 8 times the density of water.

          Let’s look at the other numbers: A box with weight 26460 lb would have a volume of (26460 lb) / (490 lb/cubic foot) = 54 cubic feet, meaning it would be e.g. a solid steel box with the a size of 3 ft x 3 ft x 6 ft. Two times the original cube which was said to have a side of one yard. I think you calculated the area of one side of the cube (3 ft x 3 ft), and then multiplied that by 6 to get all six sides. But that only gives the total surface area of the box, not the volume of the (hollow) container. We’d need to take the thickness of the steel into account to get the total volume and total mass of the steel. For the steel thickness, I can only guess, but probably not more than some millimeters?

          As for the lever, 9800 lb / (490 lb/cubic foot) = 20 cubic feet, which would be a solid steel box of 1 ft x 1 ft x 20 ft. I think you meant a rod of 1 inch x 1 inch x 20 feet, or 0.083 feet x 0.083 feet x 20 feet = 0.139 cubic feet in volume. (If I didn’t screw up the numbers again, that is.)

          Also, the original imperial and metric examples weren’t exactly the same either (mostly, the lever was 20 cm in the other, and 20 inches in the other). Of course as others have noted, the original calculation was off by a factor of ten, too.

          1. TheRegnirps, if you’re referring to me, yes, you’re right, tomkcook’s original metric example has a 1000 kg mass, and your correction about the end result being off by ~10x is quite right. However, the imperial example had different numbers, and would work out to a different mass and different torque. si_jblomberg’s answer, which I tried to reply to, had numbers which were quite different still.

        3. How thick are the walls of your box. You’ve used 45 ft³ of steel to make a box that surrounds 27 ft ³ of water. I’m guessing those walls are a lot thicker than required.

          Ignoring the weight of the lever, the center of mass of the box and water is not at a lever arm of 20″, but around 20″ + 18″ (half the dimension of the box assuming the walls are thin but adequate)

    7. Even scientists weren’t happy with the metric system. Nor was anybody else. Nobody ever based their measurements on meters, grams, and seconds, because these don’t scale well together. We have the MKS system – meters, kilograms, seconds – and CGS – centimeters, grams, seconds. And centimeters aren’t consistent with the rest of the commonly-used units, because “centi” is the only commonly-used prefix that isn’t a power of 10^3. So scientists SHOULD be using millimeters, grams, and seconds. But they don’t because they’re just as f’d up as everybody else.

      1. No, they didn’t use MGS because there wouldn’t be any way to line up with the common electrical units (volts/amps). That is, there would be no way to make the simple derived unit of energy (mass times length squared over time squared) line up with a volt-amp. You needed a system that boosted up from the cgs system such that if you scaled length (cm) by N magnitudes, and mass (g) by M magnitudes, 2*n+m = 7, which was the scale from ergs to joules. Meters and grams wouldn’t do it.

          1. Surely in this metric vs imperial discussion you mean 80 chains, see that’s almost metric.
            The only problem with the imperial system is, stupid people can’t use it, so something simple was invented.

          1. The acre is one furlong by one chain. Along a road it was marked and called a King’s Acre and was how taxes were paid. The King got the produce from the King’s acres.

  4. I’m comfortable with both SI and Imperial units, but I find Imperial units to be easier to deal with on a daily basis. I think it’s because of the way SI units are defined, without regard to the “humanity” of the end user. Whereas Imperial units are much more people centric, being originally based on measurements people could do with their own bodies. One isn’t better, overall, than the other, it’s just that they are each better suited to different problem domains.

    1. Be serious! I use both system too and I don’t care on what are based those units. When using a measuring tape I just measure. Nobody care where the foot or the meter come from. When buying a measuring tape I buy one with both system on it and I use the system according to the plan I have on hand.

          1. The problem there is the ‘one third.’ Because 3 is a prime number it doesn’t play ball with base 10 counting schemes (which has factors 2 and 5). So in metric you get awkward decimals, and in Imperial you get awkward fractions (or decimals, if you swing that way).

          2. You can use rational arithmetic in the Imperial Metric. As you say, 10 is only commensurate with 2 and 5. 16 is commensurate with 2 4 and 8, which is better. Base 12 is best of bases 2 to 16, what with 12 being commensurate with 2, 3, 4, and 6.

        1. The reason that this question is aggravating is sadly that you like most people trained after 1950 ask the wrong question and do the hard math. The solution to this is to use the right tool and geometry. To divide the space you do the following:
          1) Drop a perpendicular from the 23 14/16″ edge.
          2) Use a long ruler to make a right triangle between the perpendicular and the edge to be divided.
          3) Adjust the length of the hypotenuse so that it is easily divisible by the number of sections. In this case 36″ would be good.
          4) Divide and mark the hypotenuse at even intervals. For 36″ and 3 sections, mark every 12″.
          5) Use the marks at 12″ and 24″ as the centers for the cuts

          Most practical problems are easy with the use of the appropriate geometric relationship which translate problems into whole numbers and the correct tools like a string or gauge which eliminate the use of math altogether.

      1. Human factors is a serious business. Imperial evolved over a long time to fit human use, like once syllable names that do not sound like each other, and sizes that fit normal activities. I wonder what a metric GUI would be like.

    2. You should adopt metric system immediately: you´ll get 2.5x increase, immediately, without pills or pump. It does not cost anything but may be painful for another part of your body: your brain

        1. A, uh, “friend” has confirmed that a relatively normal sized human can hold more than half a fluid ounce in their mouths. More like 2 oz. With some variation for how hard they are trying not to laugh.

          1. Yah, fill your cheeks like a hamster you’ll get 3oz in there, but it’s more like the amount possible to hold with the front lip over the teeth with the jaw level and open…. in common parlance now, it’s the tablespoon.

          2. Our highly scientific test show that two oz was easily sucked into the mouth out of a soda cup with a straw.

            RW, I’ll take your word on that, but I’m suddenly feeling an urge to start verifying that people are primarily using an actual tablespoon to make my food. (And is that with the tongue up or down?)

          3. Just tried it on a platinum iridium BSA standard human, tipping a teaspoon into that space, and it didn’t seem to make much difference whether tongue was pulled back, tip down, or lifted high, because space would be deeper when tongue pulled back, but floor of mouth lifted when tongue lifted.

        2. If a gill is 5 fl oz, then it is a Imperial gill and an Imperial pint is 1.25 lb. If it is an American half fluid ounce, then 8 in a gill (4 fl oz). 16 US fl oz in a US pint and it weighs 1.04 lb. So you are mixed up on whether you are using Imperial or Customary, which is just one more reason not to use english measure.

          Of course, I’m not sure where your 0.5 fl oz comes from. If I take a drink from a glass, each sip is 2-3 oz. What exactly is your “friend” doing?

    3. I hear from imperial fanboys the argument about imperial being mode “people-friendly”, but after 15 years of dealing but inches and pounds, I’m still hoping to find where they this friendliness is hiding. Whatever imperial units can do, metric system can do in much easier way. Using power of 2 fractions for dividing materials might have been advantageous 800 years ago when accurate measurement tools costed a fortune and were among the most valuable possessions of traveling tradesmen, but today it’s completely lost its meaning and became obsolete. Millimeter is much easier to use than dividing an inch into fractions. The same applies to other units.
      This wouldn’t be such a big problem if all the relations were based on powers of 2. Try to ask on a store a random person how many ounces are in a pound, and watch the terror on their face as they try to recall the number. Have fun! :)

      1. By definition of IQ half of everyone is below average intelligence, then among those above, some may be in error, or have been absent that day in school, so there is a greater than evens chance of that test proving you right.

      2. I see many arguments that the advantage of imperial is thit it is more people friendly, and that the advantage of metric is that it is easier. I flat out reject both arguments.

        Firstly fractions are hard. Doing them in base ten hides some of this but does not actually reduce the difficulty, it is still there waiting to bite you. Learning fractions is good for you. I used to work at a paint shop where the interview test was “you have some two part epoxy paint. The mix ratio is two parts of A to one part of B. you have a gallon of each. How much mixed epoxy can you mix?” In all my time there I was the only applicant to not have to think about it. Average time to answer was seven minutes (usually five to ten minutes), and most got it wrong. If you learn to do the hard things, you don’t get stuck when things get hard. The advantage of imperial is that it is hard.

        The advantage of metric is that it is standard. A gram is a gram and a liter is a liter whether you are in France or New Zealand. (Although tones are another matter.) There is no issue with which ounce is this and be glad we are no longer using stones (which used not only different standards town to town but also trade to trade and sometimes multiple standards within a trade such as butchers who had one standard for live animals and another for cut meat and the difference was what they earned)

        ps. my favorite unit of measurement is the cubit which the best we can tell is about eighteen and a half inches.

        1. Standardization may have been the original motivation for developing the SI, but today, inches, feet, miles, gallons – all of these are just as precisely defined and standardized as the SI units. So there goes your argument for the advantage of metric. As for “the advantage of imperial is that it is hard”, you could as easily say that the English language is hard (due to the large number of exceptions to the large number of rules), and therefore should be made the world standard, because it forces children to learn complex things just to get along in the world.

          1. Try to learn Polish – the only rule that actually works for entire language is that every other rule has a metric tonne of exceptions, and each of them has it’s own exceptions…

          2. “gallons – all of these are just as precisely defined and standardized as the SI units.”
            LOL, no. There are multiple definitions of gallons & ounces. And don’t even get me started on cooking measures, like cups.

          3. Not at all. Like the rest of the world, our STANDARD for volume measurment is the liter. Conversions are used to get to the binary units (ounces, cups, pints, quarts, gallons). There is no U.S. Standard Gallon.

            Jeez. Everybody just stop whining.

          4. Your comment about the gallon is below the level at which replies are possible.

            The US gallon is defined as 231 cubic inches (Dept. of Treasury, 1832). It is the Queen Anne wine gallon defined by British Parliament circa 1700 and left behind when King George III’s redcoats went home in 1783, Of course, the inch is defined in metric, so a quick multiplication by (0.254 dm/in)³ quickly converts it to liters, with a great many decimal digits. (3.785 411 784 L)

            About 30n years ago, the British redefined the Imperial gallon (about 20% larger) from a 10 lbm bag of water under specified conditions to 4.546 09 L.

          5. As you say, the U.S. gallon is DEFINED as 231 cubic inches, and the U.S. inch is DEFINED as 2.54 cm. This means that the U.S. gallon is based on the METER. I really don’t give a rat’s ass how the British Parliament defined the gallon. You can take that up with them. So what’s your point?

          6. No, much added confusion. The US uses both the Survey foot (1200/3937 m) and the International foot (0.3048 m), this then goes on to create two yards, miles, etc. The US liquid gallon, the US dry gallon, and the Imperial gallon all differ from each other, as do long and short hundredweights and tons.

            Each is precisely defined in SI, but the qualifiers are usually omitted leading to ambiguity. In the SI, there may be spelling differences (metre/meter) but everybody agrees on the size; there is no national meter, International meter nonsense.

            However, you nailed the inch. There can be only one (25.4 mm) because surveyors use decimal feet and no inches.

          7. [JohnS], you misunderstand me. I do NOT defend the U.S. insistence on keeping these antiquated units alive. I was quite dismayed when that fucking idiot Reagan abolished metrication in the U.S. But whenever I see somebody arguing that one system is more “human scaled” than the other, or when nuthead claims that one system is more precisely defined, or when somebody either condemns the U.S. system for using fractions or even worse, claims that fractional units are “easier”, I will jump in every time to correct their ignorance. Just can’t help it.

          8. [BBJ} Quite a few of the posters here are British and talking Imperial while the Americans are talking Customary. They are talking past each other without even realizing (for the most part) that they are talking about two standard systems not one. Things like the fact that there is an Imperial gallon, a US liquid gallon and a US dry gallon (1/8 bushel) are part of the problem. People drop the qualifier, just talk gallon (or pint) and there is no way to determine what volume they speak of as there are three choices. An American in Canada hated buying in liters and had to convert to Imperial; I’m sure he really wanted to know the answer in US gallons. Did he he use the right conversion? My point is the same unit word (without the proper qualifier) having multiple meanings is a major problem with standard measure that is entirely avoided in metric.

    4. Agreed. This is most evident in temperature, Celsius vs. Fahrenheit. At work I use Celsius, because the engineers that wrote the operating manual for my aircraft wrote the charts and tables using Celsius, and I work very easily in that. However, when I get dressed in the morning I still check the weather in Fahrenheit. Fahrenheit was designed to cover the vast majority of most weather experienced in the populated portions of Earth’s surface (or at least Europe) between 0 and 100F.

      Of course if you grow up using C, then you can pretty easily get dressed properly using C. I still can’t do that. You just deal with frequent negative numbers in the winter.

          1. Almost all physical proprieties of elements and matter are expressed with Celsius/Kelvin. Celsius is actually a very understandable unit for common peoples. As human you constantly deal with water (how many in your body ?). Observing water boiling or freezing is very easy and give directly usable information, for cooking, for walking or driving… I’ts probably the best unit ever created in regards of usability. Even length, mass, time units (regardless of metric or imperial) are more obfuscated in there definition.

          2. I’m pretty sure that as a human, I don’t shove my hand into boiling and freezing water everyday. Yet I do go outside and experience 0 F – 100 F degree weather everyday.

          3. This is the problem with arguing that SI is superior. It has plenty of aspects that are arguably NOT superior. Celsius is one of them, not because it’s a derived unit, but because it has an offset zero. So Kelvins are used as the base unit. But Kelvins are impractical in daily life because there’s a very narrow range in Kelvins that is useful in everyday life, and after all, it’s humans you have to sell it to.

            Which is why my point all along has been, there is no “superior” system. The U.S. AND EVERYBODY ELSE should convert to SI units. Why? Because SI won a long time ago, and everything else is just reactionary stubbornness. A am American, but not “murican”.

          4. For commerce and daily use, I think that is correct.

            It looks like you do engineering calculations too. For those calculations, the fact that the SI is a rational coherent system is a HUGE advantage, but that doesn’t mean much to people who don’t do those kinds of calculations. It leaves STEM people making a “no sale” argument to about 80% of the audience.

          5. 1) I do plenty of engineering calculations.
            2) Be careful how you use the word “rational”. Inasmuch as it means “Mathematics
            (of a number, quantity, or expression) expressible, or containing quantities that are expressible, as a ratio of whole numbers. When expressed as a decimal, a rational number has a finite or recurring expansion”, U.S. customary units are quite rational. Clarification: they are rational but not sensible.
            3) I argue that the ONLY argument that matters is conformance with what’s accepted all over the world. This should sell just as well to non-technical people as it does to engineers and scientists.

          6. The BIPM uses “rationalized” in a slightly different way to relate the development of the derived units (particularly electrical) to physical laws, and coherent to explain the units are chosen to correspond to physical laws without arbitrary scaling constants, power in watts is force in newtons times velocity in meters per second as an example (utterly unlike horsepower).

    5. A tip: one can measure metric stuff using ones body too, with the same kind of precision! The length of an adult thumb is ~2.5 cm, the length of an adult foot is ~30 cm. That’s directly comparable with imperial units (well, the conversion factor for inch (=length of thumb) is 2.54 cm).

      For a bit of better precision I recommend to actually measure ones own body parts. The width of of ones hand to a certain point, the width and diameter of a finger, the length between the tip of the index finger to the base of the thumb etc. Just keep the pants on, the nether regions aren’t suited to measurement use…

    6. I find many parts of PCB design are a bit easier to express in mils vs. mm, too. 5, 6 or 7 mil spacing between conductors is easier for me to wrap my head around than 0.127, 0.152 and .178.

      And like you say, there are just certain measurements in the imperial system where the difference between 2 whole units makes a lot of sense in the context. The different between 1 inch of height, for example, is perfect when talking about two people. 1cm is a bit too small to care about and 1m is obviously too large. I also like pounds for dealing with any large object that I might hold.

      That being said, I prefer in my work life to use C for temp but I use F when talking with friends. Maybe we’re just a bit weird.

      1. I do remember that Soviet and US DIP chips had different contact spacing. Soviet ones have their legs slightly closer, due to the fact that the spacing was in metric and was exactly 2.5 mm, while the US ones had spacing in mils and the step between contacts was 2.54 mm. You could easily find a replacement, but chips foreign to the device got stuck way easier than the original ones.

      2. I’ve spent many years designing PCBs, & while I agree that it’s a pain to do so in Metric units, that’s only because the US popularised them using a 1/10″ base. If they’d originated in a Metric nation, of course they’d be based on millimetres.

  5. Be careful with terminology here. Referring to the system that the US uses as “Imperial” might cause someone confusion later. There are differences between the Imperial system and the “US customary” measurements. If people find your article and then look up conversions to Imperial units, they are going to get the wrong answers.

    For specifics: https://en.wikipedia.org/wiki/Comparison_of_the_imperial_and_US_customary_measurement_systems

      1. Well, the SI asserts that all of metric that didn’t carry over to the SI is obsolete and deprecated.

        Whereas Imperial and Customary are systems used by the UK (and a few lingering uses in the Commonwealth) and in the US. And the same word literally has two meanings, like gallon (well bad example, three actually, the US has a dry gallon too). The long ton is not used in the US and I assume the short ton is not used in the UK, they differ by more than 10%, so confusion rips somebody off.

        There is no comparable example between metric and SI

          1. In the US, the preferred spelling is dekameter to avoid that. However, since one length is 1/100 of the other, I think that would be obvious.

            On the other hand 5 Imp gallons = 6 US gallons roughly, close enough to fool me, far enough apart to rip me off.

          2. It’s obvious when it arrives on your doorstep.

            I don’t know quite know how it works, but there’s this thing called interweb commas, where people correct your spelling, but you can order stuff like a sears catalog.

          3. When I order that way, they usually show me a picture, tell me the shipping, and I know the difference between 10 dm and 10 dam. Nice troll.

            If I order a gallon of something from Canada, I literally would not know whether it was Imperial or Customary as they dabble in both. Fortunately, it would also be marked in metric and I could use that. So whether it is Imperial or Customary, the “standard” is unnecessary. Can’t say it plainer than that.

    1. Indeed. I understand (not verified) that it’s still common in France to see building dimensions stated as metres and decimals of centimetres, so 1.523 m (or 1523 mm) would look something like 1.52.3 (1 m, 52 cm, 3 mm). Very Metric, but highly confusing to those accustomed to using m and mm.

      1. Maybe “metric,” but absolutely not allowed in the SI, it must be either meters or millimeters, no compound mix of units like feet and inches.
        However, I have seen French document (I hope historical) that used it.

    1. Feeding the troll status report;

      Chucked out of the third sequential “All you can eat” buffet and now trying Joey’s Diner infamous “if you can finish it it’s free” 70oz steak with 3lb of fries and trimmings…..

  6. Do whatever you want. I’ll continue chuckling and wearily shaking my head whenever I see someone mucking about in fractions of inches when they have to calculate something. Or find the drill size for a #4 thread tap or something equally silly that works much easier in metric (Or even just converting to a single standard and working in decimal instead of fractions)

      1. Do you have any idea how many industries retool on a yearly basis? The biggest problem is resistance from the people using the tooling. The investment in terms of conversion costs spread over a reasonable time wouldn’t be that high

        1. Far better than you do since I worked in these places and the fact that you confuse the term ‘retooling’ with ‘changing standards’ indicates to me that you don’t really know what you are talking about. To shift an whole industry is next to impossible as you would be demanding that everyone up and down the supply chain, and every third-party support firm change as well. Just not going to happen because there is very little to gain for the magnitude of the expence.

          1. And yet, that’s exactly what the rest of the world did. Never mind that at the time they did it, there WERE no standards, or worse, many conflicting standards, so they were switching from utter chaos where nothing was interchangeable, to a well-defined system. The problem with trying to switch the U.S. exclusively to SI today (or even 30 years ago) is that we already HAVE well-defined standards and interchangeable parts, so there’s no overriding advantage to make it worth the cost.

            Instead, we have slowly adapted our systems so that inch-based and mm-based parts can coexist. Machine tools have sufficient resolution in their scales that parts specified in either mm or inches can be made on a machine that itself is based on either inches or mm. And because the inch was re-defined at some point to be exactly 2.54 cm, even gear-ratio-gased thread cutting machines can create both metric and inch threads. PCB design tools at one time had to operate on a grid, which had to be specified in mm or inches, but since so many parts today are specified both ways, the tools have adapted to be essentially gridless. Food products, which must have their quantities specified in both imperial and metric units, often have round numbers of ml OR round numbers of fl.oz.. Or to put it another way, we ALREADY operate in SI, we just haven’t given up our traditional units yet.

            So very likely, we will continue to have two systems of measurement (and two systems of fasteners) for the foreseeable future, and discussions like this amount to just so much hot air.

          2. Your grasp of how the Industrial Revolution played out on the Continent in comparison to how it did in Britain and Her empire is flawed if you think that this is the way it happened.

          3. The short answer is that on the Continent, the state of measurement standards was grim depending on local values – a foot in Paris was different from a foot in Lyon, a pound here different from a pound there and so on. Between the need to impose taxes and tariffs and the need to industrialise, European governments realised they needed a single system and between political and economic pressures turned to metric. The British, and by extension the rest of the Anglosphere, had already gotten control of weights and measures to the extent that they needed to for commerce because they had a head start in the Industrial Revolution which had forced the issue earlier, and while even at the time there were those that could see the value of metrification, by the time the process started to be taken seriously in Europe, it was too late for the British and Americans because the cost would have been prohibitive in comparison to the gains – they were after all at this point already on top.

            Yes I’ve glossed a few things over for the sake of brevity here, but it’s the general drift of the story.

          4. The difference is that those countries that metrified early, didn’t have an installed industrial base to worry about. That is also true of most of the Third World when they adopted metric – there was very little cost involved for these countries as there was little industry. Thus the assertion that other countries when through the process so the States could too (which seened to me to be the crux of your argument) is not really valid because the situation was very different.

          5. “To shift an whole industry is next to impossible as you would be demanding that everyone up and down the supply chain”
            Oh bullshit. We in Australia did it back in the 70s, as have many other nations over the years. It wasn’t that big a deal. These days, it’s even easier, because the entire rest of the world uses Metric, so the parts/tools/machines are cheaper than your weird-ass single-country specialised crap.

          6. Yes for countries with the GDP of Australia perhaps, for ones like the US the cost would be prohibitive. The fact is when it made the shift, the manufacturing sector in Oz was relatively small and the trading zone it was in already metric. It was not the same situation at all.

          7. Nonsense. GDP doesn’t matter one whit. The people who have to deal with the expense of conversion do it within their own domains, and a 1% increase is a 1% increase anywhere. Very weak argument.

          8. Sure it does because it determines what end of the market calls the shots. Oz HAD to convert, because their potential business relationships demanded it and their production sectors were not big enough to say no. The Americans just didn’t have to because they are big enough not to if they don’t want to. Look I’m Canadian, and for the forty years I worked in industry I had to deal with both systems probably more, and more often than those working in other countries ever do, and it was just not big a deal. Certainly far less than the sturm und drang I’m seeing from people that obviously don’t have to in this thread would suggest. It pisses you off that the Yanks can do what they want in this regard? So what? If they want to pay the costs, it’s their affair and they are still a larger economy than the sun total of the old British Empire would be if it were reconstituted today. Legacy measurments are going to be with us for awhile, and even in long metrified countries there are corners that still use traditional units and will continue to do so for the foreseeable future because it’s just not worth it to convert and the people that use them don’t care. God knows there is enough to be upset about in this world, why this topic winds everyone up is beyond me.

          9. I hope this appears in the right place; thread is becoming very difficult to follow. “It pisses you off that the Yanks can do what they want in this regard? So what? If they want to pay the costs, it’s their affair and they are still a larger economy than the sun total of the old British Empire would be if it were reconstituted today”. “God knows there is enough to be upset about in this world, why this topic winds everyone up is beyond me”. A wise politician once made the observation “Decisions made in Washington have more influence on our country than decisions made in this Parliament”. Many people object to the ‘big kid on the block’ making decisions in their interest that clash with the objectives of everyone else. I believe that the politics of power (i.e. who gets to decide what we do in our own country) is part of why this topic winds everyone up. Being honest; it is a factor in my responses. Yes, I acknowledge being somewhat small-minded in this respect.

          10. As I wrote elsewhere in this thread, the Chinese still use traditional weights and measures in daily commerce known there as the “market system” despite the fact that they are regarded as fully metric. The trend that I see in the States is to something similar as industry slowly shifts to metric – what happens on the retail side is nobody’s business but their own. But the biggest point is that the debate is sterile, in that the decision will have to be made by them in the end and I don’t see that happening anytime soon. Yes the UK, Australia, and Canada made the switch, but the fact is in all those cases the population was smaller, and the economic pressures higher.

          11. It winds people up because MOST of the discussion is such rubbish, it just pisses people off to read it. There is ONE and only ONE reason to update ALL measurement systems to SI, and that’s because for a one-time expense, we can get everybody using the same units. SI is used by a sizable majority, no matter whether you’re counting by population or GDP. SI WINS. Just effing DO IT.

          12. It has a lot to do with political attitudes I think. Australians, Kiwis, etc are a lot more amenable to government-driven directives than US people, and capital exerts more influence in the US. I imagine that it would be difficult to impossible to impose a once-off change in the US unless prompted by some form of national crisis.

          13. We can pat ourselves on the back post-facto as much as we want, but as someone that lived through the shift here in Canada, I can tell you the resistance was stiff and vocal. But it was made very clear that it wasn’t a choice for a small economy like ours, if we wanted to trade with the rest of the world and while many sulked about it was clear that no political party was going to stop the process from going down.

        1. Do you know how much it costs?

          Nothing. The conversions are exact. You don’t need to compare the size of someone’s foot to the distance travelled by light in some portion of a second. You just multiply by a number.

          And don’t say “it costs millions, look at that NASA rocket that exploded due to unit conversions!” That didn’t happen because of someone screwing up unit *conversions*. That happened because of someone screwing up *units*, period. It would still have happened if they were given meters/second instead of kilometers/hour or something similar. The mistake there wasn’t the conversion. It was not paying attention to the units.

          1. The conversion is exact in theory, yes, but industry plays with the numbers to short-change people so it definitely is not ‘nothing’. The whole point of this discussion is that a single system is ideal for trade. One system isn’t necessarily better than the other, but one system is better than two. Since most of the world uses metric, that is the logical choice as it would be cheaper to switch the States over to metric than it would be to switch everybody else over to freedom-units as the muricans are so fond of calling it. Alternately, let’s all sit down in a room and agree on a single new system.

          2. [Smashley]: You’re just trolling, right? Which means I shouldn’t bother with you. You bring up “freedom units” and “murican” just to light people’s fuses, right?

            Look: when I go to a hardware store to pick up parts for a project, even though I’ve been in electronics and aerospace engineering for decades, and am probably every bit as fluent in SI units as you are, and PREFER to use metric units, do you know what I’m going to buy? WHAT THEY HAVE. And if I’m designing a product to be built in the U.S., what size screws am I going to specify? THE ONES THAT COST 1/4 AS MUCH. We are not neanderthals, we are people doing the best with what we have available. If you’ve grown up in a place where everything is sold in metric units, well la-de-fucking-la, that just proves you to be intellectually superior, doesn’t it? So take your hate words and shove them up your fucking ass.

        2. That is the one valid argument in this whole domain. Yes it would be better if the world was on one system when it come to trade, but it really wouldn’t matter which one. So spare me the bullshit about SI being more logical or easier to work with, because neither are true, and stick to this one point.

          After all it was to make taxes and tariffs easier to impose that the system was developed by the French in the first place.

          1. Mathematics is base ten. Metric is base ten. Imperial is base 12 / 14 / 16 / 200 and that’s just the ones I can remember. Metric is ten, ten, ten, ten, hundreds and thousands, all day long.

            Oh wait, Imperials use 1/32 of an inch instead of millimetres. Makes sense.

            There’s a little cheat for multiplying by ten, you just put a zero on the end! Sssssh! Don’t tell the French we know their secret!

          2. Mathematics isn’t base ten, counting is base ten, mathematics can and is done in several bases, when it needs to deal with bases at all. Fine measurement in industries still using the American system use decimal fractions so that argument is mostly without foundation.

          3. Unless you’re trying to break things in half multiple times.

            Here’s a hint. Put meters/centimeters/millimeters on one side. Put metric feet/metric inches/fractions of a metric inch on the other side. Why don’t people do this??? I would be happy to switch to metricized feet/inches.

          4. Okay, but just about all mathematics, or strictly speaking arithmetic, that deals with measurement, is in base ten. You don’t use base 12 (0123456789AB ?) to calculate how many square inches of carpet to cut. You have to do it in base 10. Most people can’t do maths in any other base. I can do it in binary but that’s because I’m a geek, and I can’t multiply or divide. Well I could, (shift, test, add, repeat), but I’d try not to.

          5. Real world measurement, almost never requires a shift of scale – if one starts in inches (or centimetres) one stays in that scale. If one is working in yards one stays in yards and so on. This whole rationale for metric revolving around the fact that it is base 10 is just not something that those that deal in measurements every day at their jobs are all that concerned about. What they are concerned about is convenience and cost and wholesale changes to embedded established systems requiring massive investments in tooling and equipment for very little gain.

          6. I’ve been doing measurements and calculations in both imperial and metric units long enough to realize that only in the most trivial cases using simplistic examples (like “1M x 1M x 1 M box”) does the powers-of-ten aspect of metric make a difference. With any realistic dimensions, you actually have to do the arithmetic. Typically, once I’m in a unit (like cm), I stay in that unit and use engineering notation (like scientific notation but only using powers of 1000 for the exponent), until the very end, at which time I will convert units if necessary/useful.

          7. Every British tape measure I’ve ever seen has inches down one side and Metric on the other. Same for rulers. We try not to use inches, but we can if we need to.

          8. Try Products Engineering Corp (they also go by PEC). They make a variety of rules with every conceivable scale. I have one 300 mm, by mm on one edge, by 0.5 mm on the other edge. The scales repeat on the opposite side, Numbers are 10, 20, . . .290 mm (there isn’t room to print the numbers for 0 or 300, just the lines). Mine is reverse color, white lines and numbers on a blackened background. Love it! It is my favorite rule by far.

            They have SO many choices, you have to hunt for what you want.

          1. Ben: what it means is that if you buy a 3.5 mm machine screw in the U.S. or anywhere else, it will work with a 3.5 mm nut bought anywhere, and if you buy a #6-32 in the U.S. or anywhere else, it will work with a #6-32 nut bought anywhere.

            What would happen if American Standard Thread fasteners were banned after a five-year phase-out? It would mean that repairing old equipment using those would require either making replacement fasteners, or drilling and re-threading the machine to use metric fasteners, or scrounging through junkyards for them. Or more likely, it would mean that those threads would be re-defined in metric terms, bearing some weird numbers, but being fully interchangeable with the “legacy” fasteners. Which is how it’s gone in the electronics industry: it used to be that integrated circuits in DIP packages had their leads spaced at 0.100″ on centers. But not any more. Now those packages are only available in 2.54 mm spacing, at least from some manufacturers. I see a LOT of products neither designed nor manufactured in the U.S. that contain chips with 2.54 mm or 1.27 mm spacing. And they can call it metric all they want – it’s still 0.100 inch.

      2. You know the domestic auto industry did exactly this in the 70’s to save money (admittedly, we’re multinationals and one benefit was better coordination with our foreign operations which were already metric). Those suppliers who changed with us (at their cost) stayed in business. We had to support older vehicles, but we simply designated a model year for which the design would be metric.

        It is probably true that if you don’t have the leverage to force your supply chain to change, metrication will stumble and you will live in a land of conversion hell. Example: NASA

      1. Why do people keep bringing up woodwork? Dividing by two is exactly the same, whether you’re making something out of wood or titanium. It’s just as easy to divide 1.625 inches in half as it is 41.3 mm. It’s only when people pretend that standard stock is in exact units, like a 2×4 being exactly 2″ x 4″, that it appears to be easier. And even then, you have to consider the saw kerf. So give me an example from a real thing, where dividing a dimension of a piece of wood by two is easier using inches than mm.

      2. Or maybe you’re talking about the difference between using binary fractions (1/2, 1/4, etc) vs. decimal fractions (1.1, 1.2, etc). Okay, so quickly: what’s half of 3-1/2″? (And by “3-1/2”, I mean “three and one-half”, not “three minus one-half.) Maybe this is easy for YOU, but for me, I’d have to say, “okay, that’d be half of 3 is 1-1/2 plus half of 1/2 is 1/4, which together is 1-3/4″, but then I would be somewhat unsure that I did it right, whereas if I want half of 3.5″, that’s 1.75”, done easily in my head.

        Okay, I did the bad thing – I used “easy” numbers. You’re right. So let’s try it with hard numbers: take half of 13-13/16. Now, again, maybe you’re better than me at this, but I wouldn’t even TRY that without a calculator, and since my calculator doesn’t happen to have fractions, I’m going to enter 13 / 16 = (answer 0.8125) + 13 = (answer 13.8125) / 2 = (answer 6.90625). Is there a shortcut for this in fractions that’s easier?

      3. All but one of my woodworking machines are metric, the imperial machine being a Canadian sawmill but like me it’s not the machine that’s imperial or metric its only the scale, remember it’s only the scale.
        Americans, you should push for a conversation to metric because for reasons that are beyond me it floods the market cheap milling machines and lathes.

      1. If they’re using a meter stick, I highly doubt they need submillimeter accuracy. 33cm, 3mm, and ballpark the 1/3 mm. Next cut is at 66cm, 6mm, and ballpark the 2/3 mm. Remainder is the third piece.

        Or, understand that your “one meter” probably isn’t that exact anyway, so you’re probably safer cutting things a little large and grinding/sanding/filing to fit.

        Just don’t tell your math teacher.

        1. “If they’re using a meter stick, I highly doubt they need submillimeter accuracy. 33cm, 3mm, and ballpark the 1/3 mm. Next cut is at 66cm, 6mm, and ballpark the 2/3 mm. Remainder is the third piece.”

          So… instead of just defining a unit as 1/3rd of a meter, and having a mark on it, you have to mark a distance between 2 points that are 1 mm apart? And you’ll be able to do this regardless of how good your eyes are, right?

          Also: “if you’re using a meter stick”? What the heck else do you use to measure other than a stick or a tape measure?

          The metric system is fine as a base set of units. The absolute dumbest thing that was ever done in measurement was not defining a “metric foot” as 1/3rd of a meter and a “metric inch” as 1/12th of a foot.

          1. Calipers and micrometers (the tool, not the unit) are frequently used to make fine measurements of relatively short distances. Meter sticks and tape measures are used to make coarse measurements of relatively large distances.

            Fine measurements of large distances are tricky. Laser interferometry is one way of doing it, but that only measures changes in position, not absolute location. LIDAR is another technique, but I’ve never used it personally, so I can’t comment on the pros and cons.

            If your eyes aren’t good enough to make a mark between two lines separated by 1mm, then you’re going to have a difficult time marking a point corresponding exactly to a 1/3 meter marking anyway. To say it another way, if the 1mm lines all blur together for you, your 4-inch pencil marking isn’t guaranteed to be within half a millimeter of 4-inches.

          2. You do realize the ridiculousness of bringing up laser ranging measuring methods when talking about *cutting something to split your base unit in thirds*, right? This is not something that should require precision tools! This is not something that should require precision *anything*.

            “If your eyes aren’t good enough to make a mark between two lines separated by 1mm, then you’re going to have a difficult time marking a point corresponding exactly to a 1/3 meter marking anyway.”

            Diffraction disagrees with you.

            Locating the middle of a peak (the middle of the dark line) is easy if the contrast is large – in other words, if there are only markings every ~inch or so. Locating the middle of a dark line with lines separated by 1 mm is much, much harder.

          3. I agree that it shouldn’t require precision anything. It’s all about tolerances. Over such distances, 33cm is probably close enough, but you can measure out another 3mm if you’re so inclined. Mentioning laser-assisted measurements was just for completeness for the rare case of needing a precise measurement of a large object.

            Locating the middle of a peak is easy, sure. Locating the middle of the peak within half a millimeter is not as easy. If locating the middle of a peak is so easy though, approximating 1/3 of the distance between two peaks shouldn’t be too difficult.

            If you have a meter stick without millimeters marked, cutting at 33cm and approximating 1/3 of the way to 34cm would probably be good enough, but I’ll concede that having a bare meter stick with markings at 1/3 and 2/3 may yield a finer measurement for someone with bad eyes who’s taking their time.

            I’m not defending the metric system. I’m saying that it’s not generally difficult to make a unit system work, even if the system is a poor match to the measurement at hand. I was always taught to measure every digit that the tool has marked and then to estimate one more digit. If that’s not enough precision in practice, then I’m using the wrong tool. With bad eyes, that final estimation may be more difficult, but that’s a problem of a tool being built for people with better eyes. I’d say it’s not an inherent problem of the unit system.

          4. Heck, I’m not attacking the metric system either. I’m attacking the morons who thought it was a good idea to only divide something to measure things in 10 and 100.

            I’m also attacking people who think that the great advance of SI was that it’s base 10, rather than defining a single standard for *everything else to be based on.* The power-of-10 prefix thing is just dumb.

          5. Use a strip of paper or string. Fold it in thirds (Z shape flattened out). Layout and mark the work piece. Or use some geometry and lay out a hypotenuse of a length easily divided.

      2. Curious, what do you do when you need something that is 1/3 of an inch?

        Or a third of a pint?

        As for me, I don’t always divide 30 cm objects into thirds, but when I do, I divide it into 10 cm pieces.

      3. You can divide a foot into thirds handily. But in US residential construction, stud spacing is commonly 16″. Could you please illustrate the simplicity of dividing that spacing into thirds with a Customary ruler? Things which are not multiples of three are not easily divided into thirds.

        Since 2 x4’s are only 1.5″ by 3.5″, the open space between studs is 14.5″. That isn’t so easily divided into thirds either. If you cut an 8′ stud into three equal lengths, you will lose two saw kerfs (each about 1/8″) and your equal lengths need to be about 31.916 666 666 666 666 …”. You are just cherry picking easy examples.

      4. All I can do is tell you what I do as a matter of routine if I need to divide a dimension into equal parts – three parts of a metre in this example. Measure 333 (or 333.5 if I’m being silly), then 666 (ho boy – now I’m in trouble) or 666.5, then 1000. Accuracy (assuming my eye and pencil are tracking decently) is within about 0.3 mm. Good enough?

  7. The most common measurements are comfortable whether you have 12 or 10 fingers, but when you start doing real engineering it gets truly awful in Imperial units: Pounds of force versus pounds of mass (cf. Newtons and Kg), slugs, kips poundals and all the rest. The SI system neatly dissects all this and allows people (particularly students) to keep it all straight, especially in fluid mechanics and dynamics.

    I tell my students that if they’re faced with an Imperial units problem, to convert the whole thing to SI units, and then convert the result back at the end and they’ll find (as I did many years ago) that the unit conversion vortex won’t cause them to mix up things because of a poorly written subscript.

    1. if you keep track of units (and you should because it will tell you if you’ve made an error), it is not an issue at all. 12″/1ft = 1, for example. you can divide out units just like they are numbers. if you tell your students to convert to metric and then back, they are already keeping track of units so you just added one or possibly 2 additional unnecesary steps.

      1. It’s not just units. The formulas have to change (some random constant) or made-up units have to be introduced. The SI is based on F = ma. Imperial is based on F ≠ ma, because the pound is both a unit of mass (lbm) and force (lbf). To fix it you either have to make up units, like slugs (ever try to buy a slug of potatoes at the store) or poundals (ever measure your tire pressure in poundals/in²) or otherwise introduce the factor of 9.80665 m/s² divided by 0.3048 m/ft to fix the problem and get “correct” answers. And this is the simplest formula involving the issue. I must admit I once completely failed to recognize Bernoulli’s equation in Customary and said “WTF is this.”

        Aviation may enjoy air density in slugs per cubic foot, I much prefer the metric formulation of the US standard atmosphere.

        1. In all fairness physics contains great deal of mystical constants derived from models approximating real world phenomena. Couple I could quickly think of are PI, G, Euler’s, Boltzman’s and electron volt. There are myriad of others too. Problem with all is they do not have precise value in radix 10 arithmetic but when approximated and placed into the formula as a single glyph they make our lives much easier. I do not know however if these have their counterparts in Imperial units.

          1. And there are physical reasons for those constants. F ≠ ma stems from incoherent units while the SI defines coherent units for basic concepts of acceleration, energy, power.

            Pi usually comes about because the only coherent definition of angle is in radians whether you need the arc length (energy or power from torque) or the series for a trig function (solving a triangle) and pi radians is 180°. Degrees, minutes, and seconds are allowed for use with the SI but must be changed to radians for coherent computations.

      1. For someone who is more familiar with Metric calculations of this type, it would seem to me that the simpler conversions at the inputs and outputs is a good strategy, even if it does introduce addition steps that represent their own opportunity for error. Thanks for the interesting observation.

      1. CGS certainly has its problems (like abominable electrical units), but at least it has separate units for mass and force, the gram and dyne, just like its big brother, the SI, and unlike Imperial/Customary.

        I sort of understand physicists, they measure little tiny things. I don’t understand astronomers liking CGS; they measure huge distances and things with little bitty units.

        1. Yeah. At least it has that. That way people using SI units can screw up torque calculations by factors of 10 by forgetting to multiply by gravity. Or factors of 1000 in cgs.

          Astronomers don’t measure huge distances with centimeters. They measure them with parsecs. Astronomers end up using cgs because it’s older than MKS, and when the numbers are huge anyway there isn’t a bit of difference between a meter and a centimeter, and when old papers have measurements in cgs units, you might as well keep using them.

          It’s not just astronomers, though. I don’t think there’s a single branch of science that actually uses SI base units (and their ‘simple’ derived units – e.g. newtons, volts, ohms, pascals, etc.) exclusively. Particle physics uses eV for energy measures, barns for area, and grams per square centimeter for column density. Astronomers use Jannskys, ergs, parsecs, and a ton of others. Atomic physics has bohrs and hartrees.

      1. Metric already asks people to convert things. It also asks them to memorize a bunch of arbitrarily-named prefixes. If they’re capable of doing that, what’s wrong with having a few other useful convenience units?

        1. That’s not what I’m saying at all.

          I’m saying if the problem is giving you one unit system and asking for an answer in the same unit system, converting to a 2nd system because you can’t do the math in the first one is an easy way to introduce error.

          1. Yes. But the only solution to that is for everyone to agree, in the beginning, what *units* they’re going to use, period. And that doesn’t mean you can use ‘cm’ and ‘m’ interchangeably. It means you specify ‘millimeters for everything’. You can screw up mm vs cm just as much as you can inches vs mm.

        2. They are not arbitrary to a scholar of the time. Prefixes for less than 1 are Greek. For greater than 1 they are Latin. It helped keep the riffraff out of the sciencey club.

        3. Not really. The prefixes are just a convenient shorthand for scientific notation. You are welcome to use unprefixed units and scientific notation, but in verbal communications, “times ten to the power of ___” gets to be a drag. If you do elect to use them, they are reusable with any unit and a power of ten. For me, at least, that is much easier than all the arcane ratios in Imperial/Customary.

          1. Using prefixes is not mandatory (although I do because they are convenient. You can use the SI entirely in scientific notation if you wish (I choose not to).

            If you use prefixes, you should usually choose the prefix so that the numeric portion is in the range 1 – 1000.(however, there are valid reasons to make an exception in tables or when quantities are compared). Hence you would not express the diameter of the earth in zeptometers, but in megameters.

          2. “You can use the SI entirely in scientific notation if you wish (I choose not to).” I cannot recall one place I worked that permitted such a choice, nor would I be happy if they did. Furthermore, the use of these arcane prefixes is more likely to make you look like a fool than impress anyone. Scientific notation is used almost universally for a reason.

          3. Finally! Something that DV82XL and I agree on completely. PS: It may not seem that way, but I find your comments to be insightful and thought-provoking, to the extent that I look for them.

  8. Watching all those youtube videos about wood working where those poor wood making bastards are trying to divide their material correctly measured in inches is really a pain to watch. And when you write them a comment that it is really stupid what they are doing, their best reply is: “I was taught that way.”
    Whatever. Eat your inches.

    1. Watch Norm Avram of the New Yankee Workshop (and this old house). He has been known to take plans in inches and build with baltic birch (which is metric) and do all the conversions to compensate for a different thickness of wood.

    2. This is not a problem with inches vs. mm. This is a problem with decimal math vs. fractions, which is completely independent of the base units. I have a metal scale that’s marked in both 64ths and 100ths of inches. It makes this kind of stuff easy, because you can stick with inches (if you want) and NOT get messed up by the fractions. I also have a scale in mm, so don’t judge.

  9. The author wrote: “Why should I be forced to remember that there are 5,280 feet in a statute mile? … My snappy retort to that is, again, “So what?”

    So what? Tell that to NASA and the guys who crashed the Mars Climate Orbiter because they used non-SI units. This is a fricken stupid article just use metric and drop Imperial so we can standardize with the rest of the planet.

    1. It’s not because they used non-SI units, its because the other team wasn’t aware each were using different units. That’s a very big difference.
      If you stick to one unit system there’s no reason you can’t build a nuclear reactor or land on mars. Heck, even using 6 unit systems, you just need to be aware of it.

    2. I thought it was because one team did not use rocket industry standards for force. Poundals IIRC. Someone in navigation assumed it was something else. It is more about the dangers of assumptions, than systems of units. The same would happen if something was in kg mass and assumed kg force (kilopond or kgf). They would be off by a factor of 9.8.

      1. “The same would happen if something was in kg mass and assumed kg force”

        This is a corruption of SI. There is no such thing as a kg force. There is “the amount of force exerted on a 1 kg mass in a 9.81 m/s^2 gravitational field”, but it’s just plain laziness to reduce that to “kg-f”.

        This is also an example of how SI isn’t always easier. In imperial, we measure mass and force in the same units, with a tacit assumption is that the mass is measured as a force in a 1 G field, so this kind of awkwardness just doesn’t happen. And yes, I am aware that there is a traditional unit for mass, but I’ve never seen a scale calibrated in slugs.

        1. In Imperial you should be using slugs for mass. All the text books used to. And the Poundal is a unit of force in the Imperial system, in units of foot-pound-second. If you are using weight for mass, you are doing it wrong.

          1. We should be – if we were doing work anywhere where gravitational acceleration was significantly different from 9.81 m/s^2 (sorry, I don’t know what that is in ft/s^2 and don’t want to calculate.) The point is that although using a force measurement as a substitute for mass in the “customary” units meant that you usually got something close to the right answer, but in metric, you can’t just substitute mass for force because the next guy is going to say, “hey, that’s a force, I have to convert it back to mass”.

            In other words, pretty much everybody ignored the distinction between mass and weight in the English system, whereas metric forces you to either fudge the numbers and pretend mass is weight, or do the actual conversion.

            I know this sounds hypocritical, but kgf was invented because engineers working on earth-bound things decided they didn’t want to always be converting from mass to force and back again when it was just wasted arithmetic. But it is most definitely NOT an SI unit, and a possible source for error.

          2. No, there are two (actually three) gravitation english systems. The US mostly converts mass to slugs and uses the pound-force for force. The UK system mostly uses the poundal for force and the pound-mass for mass. (the third is use lbf and lbf, measuring acceleration in multiples of standard gravity not ft/s². They all introduce a constant correction factor of 32.174 to fix F ≠ ma in Imperial/Customary. But you must only use one of the three artifacts.

            The better solution is to get the customer requirements, figure out where they mean lbm, and where lbf, converting to kilograms and newtons.

          3. There’s no point in bringing the Brits back into this, since they use SI now. I suspect you’re doing that just to make it look like there are more standards than there are. In my whole life in several engineering fields, I have seen pounds and pounds-force specified many times, but never slugs. Trying to apply that sort of simplification to metric measurements is just a corruption of SI.

      1. As I’m sure you know, in SI units the only time unit is the second. So really, people who calibrate speedometers in “km/hr” are using another corruption of SI. This is where the “metric is just as human-scale as imperial” argument breaks down, because nobody cares how many cm their car goes in a second, or even in 1000 seconds, so they revert to a customary unit that they understand – the hour.

        I would claim that the fact that people who grew up with SI break down and corrupt the system to make the units understandable, indicates that the basic units used in SI are not ideal for human use.

        At the risk of beating a very dead horse, I would suggest that the second was a bad choice. It was taken as 1/86,400 of a day, which was not consistent with the rest of the system. If the damned French were putting THEIR money where their mouths were, they would have redefined the second as 1/100,000 of a day, the minute as 100 seconds, and the hour as 100 minutes, leaving us with 10-hour days. The second would still be roughly the same as a heartbeat, the minute would be about 1.4 “imperial” minutes, and the hour about 144 “imperial” minutes. Sure it would be tough to get used to, but for people born to it, km/(metric)hour would make intuitive sense because a metric hour would be closer to the scale people expect when expressing travel time.

        1. They did do that, but in doing so they made the week 10 days long and that gave people fewer holidays than the 7 day week, so the people who came up with the idea got guillotined and the old system went back to use.

      2. For coherent calculations, you just convert all time units to seconds.. Not the problem Imperialists attempt to make of it. The minute, hour, day, week are not SI units but accepted for use with the SI with the caveat they must be converted to coherent units for calculations.(or the coherence of the SI is lost).

    1. The SI base unit of mass isn’t gram. It’s kilogram. Gram is a derived unit. Technically a gram should be a millikilogram, but of course everyone thought that was stupid, so “mass” is magically treated differently.

      Why? Who knows!

      1. Mass isn’t treated differently. Gram and kilograms are mass not weight. SI unit for weight is newtons. People use electronic scales and get a reading in grams based on some approximation of gravity but since it does not know the elevation in most cases or the true gravity value, it is just that, an approximation. That’s why you use a balance when dealing with mass (grams/kilograms) instead of a scale to cancel that approximation out.

        Also the gram is not a derivation of kilogram. It’s the opposite. I mean, kilogram literally means 1000 grams.. how could the gram be the derivation when kilogram is defined based on grams. The real answer is that before SI, there were two flavors of metric (OMG it wasn’t always standardized???) which were MKS and cgs, and one used the gram as the unit of mass and the other the kilogram. When they finally standardized under SI, they chose the kilogram over the gram. They also chose the meter of the centimeter, which the two groups also differed.

        1. You’re wrong.

          The base unit for mass in the SI system is “kilogram.” There is no international prototype gram. There is an international prototype *kilogram*. A gram is defined to be 1/1000 of that. For every other SI unit, putting a prefix in front of the base unit changes it by that exponent, however, for mass, because the ‘kilo’ prefix is already there, you adjust everything else by that.

          “how could the gram be the derivation when kilogram is defined based on grams.”

          Because France!

          1. Yeah, but cgs was a special kind of screwed up, since its practitioners typically used electrostatic units, making comparisons totally impossible, and confusing the hell out of a generation of astronomers.

            And cgs is still screwy, just with length. No centicentimeters, sadly.

          2. [Pat]: holy crap. I didn’t know that. Apparently this got worked out, else we’d have two sets of electrical units as well. Feels like a dodged bullet to me.

  10. What really bothers me: why do I have to have two sets of certain tools like wrenches or hexagon socket wrenches? One would be enough if this one single country would have opted in for metric and would save us and the planet a lot of tools produced and taking up space.

    1. The French did not invite the US to the 1798 meeting on the French Metric. The US Congress formally allowed the French Metric for any purpose (made it legal for court document and all transactions) in 1866 – there were concerns that after Napoleon was gone it would all revert in Europe. US weights and measure were based on the kilogram and meter from 1893. It was sort of mandated in the 1970’s. Attempts to convert school students to metric enthusiast have gone on since the 1880s, and it just won’t stick.

      1. The attempts were destined to fail, because they only mandated labeling, not any kind of change in standard quantities. How is somebody going to be swayed toward the metric system, when their can of coke is labeled as 12 fl.oz./354 ml? 12 is a lot “easier” than 354. Would it have made a difference if they had somehow mandated the standard quantities? I mean, Coca-Cola probably would have been happy to switch to a 1/2 liter size because it’s clearly bigger than a 12 oz. can, to be labeled as 500 ml/16.9 fl.oz.

        In fact, this happened without a mandate, because multinational corporations changed many of their packaging sizes over time to standardize their own manufacturing. Most people now know without thinking how big a 2 liter bottle is, or even 1/2 liter, and have no idea how many fluid ounces are in them. And people DO know that a 2-l bottle takes up a little more space in the refrigerator than a 1/2-gal milk carton, so yeah, a half gallon must be slightly less than two liters.

        1. Coke makes 12 fl oz cans and 500 mL bottles (both labeled in dual units as legally required). 500 mL is also a popular personal size for bottled water. Almost all bottled water except distilled water has round metric fills.

          1. That was my point. Before the metrication in the 1970s-80s, the standard soda size was the 12 oz. can, or 16 oz. bottle. Since the U.S. gave up on metrication, most of the standard packages shifted to plastic bottles in sizes that are round numbers in SI, so the conversion is happening in spite of reactionary efforts against it in the U.S.

      2. The Metric Act of 1866 was passed so we could legally manufacture metric goods to export to Europe.

        The French did invite us and others to the 1875 Treaty of the Meter (or Metre Convention). We and others signed. At that point it ceased to be exclusively French and became international, with all the signatories having a vote on further improvements, which include MKS replacing CGS, the incorporation of sensible elctrical units, making it MKSA, the development of the SI in 1960, and continued refinement of the definitions of the base units.

  11. I’ve got two–fairly–complete sets of hand tools in four tool boxes: half are metric and half are SAE. Together they keep my house anchored to the ground. My only complaint about the system is that there is no rhyme nor reason as to whether the fastener–same car!–will need a metric or an SAE tool.

    Things could have been much worse, folks: I still have a memento from the good ol’ days in my tool box. When I was riding motorcycles, I had to buy Whitworth wrenches (oops; spanners) in order to work on my BSA banger. I kept one spanner to prove that they really used to exist.

        1. Unless it’s old and British, in which case it’s a 1/4whitworth, and scaffolding fittings are 7/16 whitworth.
          so all you dual toolset guys can stop moaning, some of us need 3 tool sets especially with old brit motorcycles

  12. There are a couple of sensible reasons to stick to imperial. In my opinion there are more reasons to convert to metric. But in the end none of those reasons do really count, people just want to stick to what they are accustomed to.
    But I wish all the Imperialists would be forced to change to the metric system. Then all these discussions would have an end. Just think about the amount of productive time that could be liberated by freeing our minds from such a stupid topic.

    1. You know, long before there was many others units in uses. Each worked very well for the peoples that use it. Peoples realize that there lost time and effort when there exchange with others using a different unit, so there agree to use a common unit. USA is just the exception that confirm the rule…

    2. If your country’s industry and infrastructure was destroyed in WWII, it was much easier to go all SI. I don’t see any need for folks in the US to apologize for not getting blowed up. It is also the reason we have antiquated phone and power systems and all those other things that are strung on poles or buried underground.

          1. Nothing related at all to the age of the infrastructure. The historical fact is that most Western European countries developed the metric system long before the WWII, please read https://en.wikipedia.org/wiki/History_of_the_metric_system#Adoption_of_the_metric_weights_and_measures

            You can observes that at that time the arguments against the metric system was the same used today by some USA peoples. Yet all countries in this world but 3 successfully adopted the metric system because the advantage of a good common system is high.

  13. “Doing the conversions between imperial units and SI units is tedious and error prone, they say. Really? Perhaps I’d buy that argument a hundred years ago, or even fifty. But with pervasive technology that can handle millions of mathematical operations a second, there’s not much meat on that bone.”

    How short your memory is. Did forget that not so many years ago a NASA probe crashed on Mars. The investigation related the crash to >>>unit conversion error<<<.

    https://en.wikipedia.org/wiki/Mars_Climate_Orbiter#Cause_of_failure

    This king of costly error could be avoided if everyone was using SI.

    1. Did you even read what you posted?
      It had nothing to do with an error in unit conversion, it was the fact that no conversion was done. They were given unitless numbers and assumed the numbers were in the units they were used to.

        1. Metric isn’t a single standard. It actually legitimizes *tons* of different ones. Things aren’t always measured in meters – they can be measured in 20 different standards, 19 of which are different from “m” by a single letter. Except for deca. Because it’s stupid, and no one likes deca.

          In order to avoid any problems, two groups have to agree on what the units are that they’re specified in. In other words – exactly the same problem as the NASA guys had.

          1. *Everything* at this point is defined on the SI standard. Inches, miles, etc. It’s all the same.

            The problem is that with metric prefixes, you now can have measurements specified in micrometers, millimeters, nanometers, centimeters, etc. Yes, you can convert between them easily, but why the hell are there all these random new measurement types?

          2. This is not the problem you imagine it to be. Engineering drawings normally have a general note regarding the unit of unitless numbers, and require marking the units on any dimensions which deviate from that. The normal convention is millimeters unless noted, which may be used up to 100 m or even more. I suppose IC chips might use µm.

            As for NASA, the purchase order SPECIFIED that the data be in newton seconds, and the supplier violated and provided pound-force seconds. It was not a “mix up.” It was a contract violation. In accepting the contract, Lockheed tacitly accepted the terms and conditions of the contract, then violated them. NASA made the mistake of believing a supplier would honor the T&C of the contract.

          3. This is at least the third time you’ve made this claim. Do you have any evidence that this was a WILLFUL act? Because that’s what you’re saying.

          4. Is this a reply to the wrong thread? I willfully put notes on drawings that dimensions are millimeters unless noted (well actually my drafter took care of the general notes, but I always checked for the note before I signed.) It saves thousands of repetitions of “mm”.

          5. Apparently so. My reply was to your latest statement to the effect that Lockheed-Martin willfully delivered software that issued values in customary units where the specification staed SI units. My question is where you get that this was a willful act. Because barring that, your statements amount to libel.

          6. Reckon BrightBlue is referring to the contract violation, which could be wilful or non-wilful. I would assume that it was non-wilful (ie an error).

            As for drawing dimensions – yes, “dimensions in mm” stated once and not a “mm” anywhere else.

          7. >This is at least the third time you’ve made this claim.

            I found what you meant and thought I replied correctly last night, but I can’t find it. I’ll paraphrase what I typed and lost.

            I never sold anything to NASA but I have sold stuff to the military (when with my first employer, non-automotive). Our contract adherence officer was up my ass over every phrase in the contract wording and whether I had complied. The process was a right PITA and the company officer was worse than the military contracts guy. Maybe Lockheed and NASA are different. but based on my experience, it is just inconceivable that this could be missed (at Lockheed). No proof it was deliberate, but it was EXTREME, EXTREME negligence, both by the technical group and the contract adherence group at Lockheed, that would require numerous employees to “fall on their sword.” I trust it resulted in a massive overhaul in how Lockheed verifies contract compliance.

            However, NASA kept fairly silent about that and took most of the blame on themselves for failing to catch it in software review. I disagree with that approach. In automotive, I understand engineering mistakes; we make them and our suppliers make them. However, we expect suppliers to big boys, understand the contract and comply. I can’t pass it off as an “ordinary” or “understandable” mistake and would demand a MUCH deeper investigation of how it was made, how it got shipped, and how the system was changed to prevent recurrence of any similar mistake (part of “the eight disciplines of problem solving”). Certainly, Lockheed would owe me a new rocket and payload. But, apparently, we are a “harsher” company in supplier negotiations.

          8. Thank you for clarifying this.

            I once worked for a short time for a company that produced power supplies and other subassemblies to military customers, and I have to say that I have seen examples that extend past the boundaries of negligence (e.g., parts that passed tests but were known to have defects that would show up after a short period of normal use), so I do understand that sometimes companies will ship things they KNOW aren’t right, for whatever twisted reasons they may have. I just wanted your confirmation of what you appeared to be accusing Lockheed-Martin of.

      1. The error was in a function in a computer program. Internally, virtually all quantities in embedded computers are unitless. This is the sort of thing that software specifications are supposed to ensure, and software reviews are supposed to check. At the software spec level, you KNOW the units had to be specified. If not, that’s a huge design error right there, NOT a conversion error.

          1. And the software review missed it. Or are you implying that Lockheed Martin Astronautics – a company whose business is space flight – sabotaged the mission by deliberately delivering results in the wrong units without telling anybody?

          2. That or illiteracy. As failing to meet the contract specifications usually results in not getting paid or getting sued or both, most suppliers read them quite carefully and have a group that reviews contract adherence.

            At least in my world (automotive) this is not an acceptable “oops” and would result in the contractor being liable for the cost of the mission.

          3. Still, there’s a clear difference between negligence, which is what I read into the error, and malice, which your use of the word “disobeyed” implies, don’t you think? It makes your assertion a bit provocative is what I’m saying.

          4. @BrightBlueJim:
            Can we compromise on “violated.” They failed to comply whether it was intentional or accidental. I violate speed limits all the time. When the cop catches me, he doesn’t judge, he just writes me a ticket.

          5. But the judge DOES judge. In the U.S., the state does not need to prove intent in penalizing people for infractions, but in criminal law this is a very important consideration. Your continued insistence on using terms that imply intent without evidence are irresponsible. Yes, I would agree that Lockheed-Martin “violated” the contract. Disobeyed carries a completely different meaning.

  14. It’s amazing how all of the people that can’t be bothered to learn how to do unit conversions get so angry at anyone who actually has the nerve to use imperial units. Poor babies. I guess they didn’t have as thorough an education as those of us who were taught both systems and how to easily use and switch between both.

    Grow up. Just because you don’t like something, don’t expect someone else to change.

    PS: You call the imperial system “random”, so tell me the definition of a meter again? How about a kilogram? And temperature? You do realize that water never actually changes state at precisely the same temperature?
    Get a grip, people. All measurement systems are random, and it just doesn’t matter.

    Great opinion piece, [Dan]! Keep ’em coming!

    1. It´s the same when one finds the highway encumbered with horse carriages: use the horn. And when some explode on the way like a NASA space shuttle, it´s a mess. US would have not adopted the metric system at all, would not be an industrial power today.

        1. Yes they were, the unit of idiocy used to describe NASA management too large, like a farad was said to be, therefore 200 picostupids didn’t seem to be enough to cause a problem, even when the average city planning committee ran at a few nanostupids.

          1. Yes, the Farad was unfortunate, with probably the greatest consequence being that it is fairly easy to misread or miswrite a pF as a 𝜇F or vice-versa. But to be pedantic, “stupids” is not an SI unit.

            Keep in mind that the development of the Space Transport System spanned the period when metrication was mandated by Congress, so it wasn’t NASA being elitist, but being proactive.

          2. One metric stupid is defined as the amount of mental aberration induced by the introduction of a cubic meter of LSD into the worlds fresh water supply. I think you can see how it ended up rather cumbersome.

          3. Which is the same kind of non-human-scale standards that mess everything up. I think it SHOULD have been based on the mean stupidity of members of the Polar Bear club while diving into a lake at the triple point of water.

    2. ” so tell me the definition of a meter again?”

      It’s the distance light travels in 1/299792458 of a second. Even with a reference to the speed of light, metric is much nicer than imperial, since the speed of light approximates to 300000Km/s. Easy!

      1. “Or 180,000miles/sec”, also easy. Not as easy, obviously – because multiplication and division by 3 is easier and you get 99.93% accuracy; whereas 180,000miles/s isn’t even accurate to 2 significant figures – you should have 190,000, which is 2% out. Division and multiplication is harder with 19 right?

        Of course, metric isn’t ‘more’ arbitrary than Imperial, especially when the US and other imperial users have different standards for the ‘same’ units!

        1. No, we (the U.S.) don’t. We use exactly the same standards for our metric units, and we apply conversion factors to get imperial units. The U.S. doesn’t have a primary standard for the inch. We have primary standards for the meter (based on the speed of light), and all inch measurements are based on those. Same as in the rest of the world. Similar for other standards. The last I heard, the U.S. standard for mass is the same as the international one. There is no “standard pound”, and hasn’t been for decades.

          We are not country bumpkins, and the people who are really into precision measurement are fully committed to SI. Imperial measurements are used in the U.S. only for convenience.

    3. The nice thing with the metric system, it’s streamlined, not random:
      1m= 1000mm, 1l= 1000ml, 1000l= 1e3 liter = 1 m3.
      or 1 l is 0.1mx 0.1mx 0.1m = 0.001 m3
      1l of water = 1Kg = 1e3 g, 1ml= 1g of water
      Water freezing 0 deg.C , boiling 100 deg.C
      Perhaps not that precise but it works.
      Green light is 520nm or 0.52 micrometer or 0,000 52mm or 520 x 1e-9 (nano) meter.

      The biggest problem of the imperial system are the different divisors for different size within one category.
      ex: the short ton is 2000lb (1:2), 1 lb is 16 ounces (1:16). or the yard, feet, inch.
      You could make your life as difficult as you want.

      1. No. Ask anybody who calculates things using the cgs variant of SI: they do all length measurements in cm, like 3.6×10^9 cm. Which is itself ridiculous since a cm is 10^-2 meters, but the point is, if you use what’s become known as “engineering” notation, the multiplier is in powers of 1000, so there’s none of that “deka, hecta, deci” crap to “convert” between.

        Machinists using inches measure everything in a single unit – inches. You’ll never see a machinist calling something “one foot, three inches”. They’ll just call it 15 inches. And rarely will you find one who measures in binary fractions of an inch, either – it’s all decimal.

        You can make your life as easy as you want. Units of measure have nothing to do with it.

        1. The hecto, deca,… aren’t used as much anymore, they slowly get phased out of the common language, even the use of the common cm is diminishing. Where the cm is perceived, more as an approximation for a value. The other common thing is the deciliter 0.1 l because it’s handy, but otherwise the use of hecto/a, deca/i is more perceived as hyping/advertising/cheating old fashion style.
          It’s reminds me, that in the past, there were the same troubles with money… and at least, we evolved.

          1. hPa is a standard unit of pressure. Deca isn’t even a prefix – it’s deka, not deca. Deciliters gets used in recipes.

            The fact that those prefixes are being phased out isn’t a good thing. It just shows it was dumb to ever have them in the first place, and now you’re just going to be left with a bunch of archaic prefixes that get used in a few minor spots and people have to remember what they were.

          2. It is deca- in British or International English, deka- in American English. See the SI Brochure and NIST SP 330.
            I think it is deka in American English because that has two letters different from deci, but it could be we just like being different than the British, Daniel Webster, etc. We spelled it deka in the Metric Act of 1866.

  15. The issue i have with the imperial system is when it gets smaller than an inch (i.e 3d printing, milling or turning) then the whole system just gets crappy…

    “I need to make this 7/16ths of an inch part larger by 25 thousands of an inch.”

    AAARGH

  16. This article is very silly … you started it by explaining why it’s necessary to have a global way of measuring things (and I totally agree with that !) and then you finish it by saying : ” So we can continue to use our own unit system, we don’t care ! “

    1. You’re confusing “global standard” versus “global method of measuring.” You’re never going to get everyone to agree on ways to measure everything, in all disciplines. Even in metric, that doesn’t happen. SI’s good part was telling everyone “yeah, your convenience units are fine, but the conversion between them is going to be *defined* by these numbers.”

      It’s hard for us, now, to understand that prior to global standards like this, you only had reference objects. And to convert from, say, pounds in one country to pounds in another, you needed a place with both reference objects so you could compare the two.

      This is *still* the case right now with the kilogram. You’ve got *one* international kilogram, and then everyone else has copies that are *close to*, but not exactly the same as, that other kilogram. Then any other units get calibrated to that, and obviously, eventually, you get error that continues to creep in. But you *know* those errors, because you know all the other errors along the way.

  17. In aviation you measure height in feet, distance in nautical miles and visibility in kilometers so there’s that. It was a while ago now, but unit conversion have caused incidents like the Gimli Glider (https://en.wikipedia.org/wiki/Gimli_Glider).

    Having one common system would be awesome, but I don’t expect to experience that in my lifetime since I think it’s more of a political issue than anything else.

    1. Maybe those invoking the Gimili incident should actually read the reasons this happened. It was NOT a simple conversion mistake that caused this problem, but rather a string of errors, equipment failures, and procedural issues that mitigated together as proximate causes. Furthermore, to the extent this was an issue of confounding measurement systems, the fact is that it was a consequence of a company trying to shift from one system to another in its daily operations that was at the root of this. In other words, this incident is just as much an argument to not force a change.

      1. Several people had to screw this up for it to happen. Everybody KNEW that the company was switching from pounds to kilograms in fuel measurements, but nobody said, “Now, that’s in pounds, right? Right?” And then, neither of the pilots seemed at all concerned that the aircraft rotated early, which should have at least made them ask, “did we miscalculate our take-off weight?”.

        If anything, the Gimli Glider incident could be used to argue AGAINST changing standards.

        1. There was also the fact that in Canada there were at the time many people that were still irritated at metrification and the press emphasized that aspect a bit more than was warranted. Oh, and the flight crew didn’t have all that much time on the equipment which might explain not picking up on the rotation clue.

  18. Now can we have another article written by someone who, say, moved to the United States from metric-land and explains why it’s retarded that not just kids, but grown ups don’t know how many feet are in a mile, how many ounces in a cup, how many cups in a gallon? We do these conversions every day even if you don’t think so.

    I am such a person, and I’m glad that at least in hospitals there’s an oasis of sanity where scientists understand how important it is that mistakes in simple conversions such as between cubic centimeters and liters can cost lives. True, that may not be the case in everyday situations and a ruined meal because of a misread recipe may not be that severe. But the point is that a world where calculations are easier, mistakes are less likely.

    The imperial system is already based on the metric system: one inch is 2.54 centimeters by definition and one degree Fahrenheit is 9/5 degree Celsius, and I agree it doesn’t really matter which unit of measure you use to measure something. Actually in some cases imperial units are easier, for example on most freeways you can get from one place to another in about 1 mile per minute. But I think you American knuckleheads are sticking with the Imperials system just because, well… you’re American.

    ===Jac

    1. “I’m glad that at least in hospitals there’s an oasis of sanity where scientists understand how important it is that mistakes in simple conversions such as between cubic centimeters and liters can cost lives.”

      Yes, that exactly my point. There is a time and a place for metric, and life safety is certainly one of them. Baking a cake or tiling a floor? Not so much. Actually sounds more knucklehead to mess with that than to stick with what works.

      And just as an aside, doctors do a pretty good job of being as unscientific as possible when they use SI. In the lab, we generally measure concentrations in terms of molarity – moles per liter. So we end up talking about millimolar solutions or micromolar concentrations. Doctors think it’s keen to express concentrations in “milligrams/deciliter” concentrations. My diabetic daughter’s blood glucose is at 166 mg/dL right now. Does that make any sense? Wouldn’t molarity make more sense? Or even go with mg/mL – we do that in the lab sometimes too. At least a mg/mL is the same as a g/L or a ug/ul.

      Point is, even SI units can be screwed up by a properly motivated party. Figures it would take doctors to do it, too – they pronounce “centimeters” as “sonometers”.

      1. I agree that SI units can be screwed up by a properly (or improperly) motivated party. Other examples are how Americans use “calories” instead of “kilo-calories”, an upper case “L” as abbreviation for liters, or “MFD” as abbreviation for microfarads. I know why that happens and I don’t have a big problem with that, just like I don’t have a problem with tachometers in cars that have a “x100” scale even though they only use numbers for which a “x1000” scale would be sufficient, or “IIII” on clocks with roman numerals on their faces: it looks better and/or is less confusing and/or is more convenient (I don’t have a Greek letter mu on my keyboard and can’t be bothered to look up the Alt-code right now either).

        But you have to admit that a conversion between, say, mg/dl and g/l is much easier than a conversion between, say, cups and ounces, especially with fractions, and especially for someone who’s used to SI prefixes everyday (which everyone in the world is, except Americans). And don’t get me started about fluid ounces versus ounces by volume. Not to mention the weird habit of specifying things like amounts of flour as volume, when a weight measurement makes more sense.

        That’s the last I’m going to say about it. The article reads mostly like a troll feeding invitation anyway. So I did. I hope you’re 11 pounds and 0.370 ounces of happy in a 5kg bag :-)

        1. >an upper case “L” as abbreviation for liters

          The liter is the only unit with two symbols, upper and lower case L. This was specifically authorized as Resolution 6 of the CPGM in 1979 and is published in the SI Brochure. The American version, NIST SP 330, specifically defines “L” as the preferred symbol for liter in American usage.

      2. [Dan Maloney]: sonometers – that’s those damned French again.

        You’re right about the deciliters nonsense, but that’s exactly the kind of thing you don’t want to fix, because if you try to change it to liters, then in many cases a factor of 10 error may not be obviously wrong, but lethal. Yes, milliliters would be better, but that would still screw up everybody’s sense of what the nominal values should be, for a while.

        And yeah, molarity would make more sense, but the Mole is not an SI unit. Well, it’s not a unit at all, since it’s unitless.

        1. >but the Mole is not an SI unit

          Well, damn the idiots at the BIPM who write the SI Brochure (the official definition of the SI). They assert it is an SI unit (amount of substance) in section 2.1.1.6 of the SI Brochure. It is also listed as the sixth of the seven SI base units in section 4.1.

          However, molarity and its symbol M are considered obsolete and deprecated. It should instead be expressed as moles per liter (mol/L) with a suitable prefix if the concentration warrants it.

        2. Where did you get the idea that the mole isn’t a unit? Of course it’s a measurement unit. It’s the amount of a substance that contains as many individual units as 1/1000 of a kilogram of carbon-12. (Gets reported a lot as ‘1 gram of carbon-12’, but a gram isn’t a base unit, so technically it’s 1/1000 of a kilogram).

          It converts number of individual units (e.g. atoms) to mass, just like density converts mass to volume. If binding energy/isotopes didn’t exist, it wouldn’t really be necessary because everything would just be multiples (e.g. the molar mass of hydrogen would be 1). But since they do, you have to measure the molar mass of things, which means you need a definition of a mole.

  19. I live in Panama, which is a highly “USA-biased” in many ways…yet very international in all others.

    Here you will find yourself on any Friday night buying a “half-gallon of Rum and 2 liters of Coke”.

    We measure large distances in Kilometers, but small distances in feet. “I am a 6 foot tall man living in a city 10km wide”.

    Sodas are measured in Liters while gas and Diesel in Gallons.
    We recently standardized gas pricing to liters, the country collapsed because no one had any idea of how many Liters their car used…mine takes 17 gallons.

    Its a fun country.

      1. Sometimes they do.
        Ever buy a 1/5 of liquor? That’s 1/5 of a gallon. Beer kegs are measured in gallons (15.5G) and pints are well, pints. Lots of licensed ‘moonshine’ is sold in quart jars, as well as the real stuff sold in quarts or gallon jugs.

        1. The fifth changed to 750 mL around 1980. Certain standard metric sizes are specified for wine and spirits and those are the only legal volumes in which they may be sold. As part of our duplicity, beer, on the other hand, must be labeled and sold in Customary quantities based on the (British) wine gallon, which we use for all liquids.

  20. i think its a misconception that the usa is strictly an imperial outfit. i encounter both systems on a daily basis almost, its why i own 2 sets of tools for metric and imperial. even doing some construction i found a few building materials were imported from canada and came in metric sizes. and obviously you minimize cuts so you can keep as many factory edges as possible. everything is also labeled in both systems, from the speedometer in your car to food packaging at the store. i still cook in imperial units, mostly because thats the units all my american food recipes are in. every now and again i get some italian or asian food recipe in metric, but its not really that hard to convert if need be.

    1. The difficult thing in recipes is converting North american volume measures to Euro weights… 1/4 cup of flour and 1/3 cup of oatmeal, might make you figure you need to use half as much again in flour as oatmeal in grams, but might be more like double the weight.

      1. That’s not imperial vs. metric. That’s volume versus weight. You’d have the same problem if you tried to do that in metric. It’s just that volumetric cooking is more common in the US than in other areas of the world, apparently.

  21. With suitable education, yes, we can handle everything. Forget science, forget engineering, forget all the highly-trained disciplines… we scholars can measure lengths in Niagara Falls days per Utahs if we needed to. What about the people who are not so educated? It’s easy to forget that unit systems are used outside of engineering, science, construction, and manufacturing.

    This recipe calls for 3/4 cup of skim milk. How many batches can I make with the unopened quart I have in the fridge? Oh, and I’ve already thoroughly washed my hands, so I don’t want to touch my disgusting phone. Let’s see, I know there’s 128 fluid ounces in a gallon. I think I remember that a cup is three… somethings… maybe fluid ounces? Three tablespoons seems too small…

    For someone who hasn’t gone to college (maybe who didn’t even finish secondary school) and who doesn’t use such mathematics every day, that problem seems impossible.

    A quick Google suggests that I was mixing up my conversions… it’s three teaspoons in a tablespoon, not three tablespoons in a cup. If you don’t use it frequently, well… who remembers this stuff?

    In my opinion, a “good” unit system is accessible to the (relatively) uneducated. Anyone can count a length in feet, but that’s only one unit, not a unit system. The conversions are important.

    1. “it’s three teaspoons in a tablespoon, not three tablespoons in a cup. If you don’t use it frequently, well… who remembers this stuff?”

      Here’s the point: almost everyone in the US could figure this out super-fast. You know why? Because they *have* a teaspoon/tablespoon measuring set, and a set of measuring cups.

      I actually have no idea how people in metric-land measure stuff out. I mean, I know that scales are super-common elsewhere, but measuring cups with gradations on them take *way* longer than just levelling off a cup. The “I have no idea” part was not meant to mean that you can’t do it – it really was meant as a question. I’m guessing it’s just a very different method of cooking.

      1. Around here it’s teaspoon (5 ml), tablespoon (15 ml), and dl. At least in the recipes I use. The only time I weigh something is when making pancakes and I’m unsure how much milk I have left. Just put the carton on a scale and if it’s at least 4 hg, or so, I go for it (the recipe calls for 6 dl, but I substitute the missing amount with water).

        1. Those are the base units, though. What about halves/thirds of those? In the US you’d normally have 1/8th, 1/4, 1/3, and 1/2 tsp, 1 tsp, 1 tbsp, and then for cups, you’d have 1/4, 1/3, 1/2 and 1 cup. 1 dl is about the size of a graduated measuring cup in the US.

          “Just put the carton on a scale”

          Lots of US kitchens don’t have a scale – it’s all volume measurements.

          1. I checked my cookbook and 1/4, 1/2 and 3/4 are used frequently, didn’t see 1/3 so I guess that’s uncommon. My 1 dl cup also has markings for the former. For smaller measurements we also have a spoon, literally called seasoning spoon, which is 1 ml. I assume these four “bases” of measuring are enough, cooking is hardly an exact science… Regarding the scale, that was for my benefit to see how much is left using the relationship between dl and hg.

          2. “I checked my cookbook and 1/4, 1/2 and 3/4 are used frequently, didn’t see 1/3 so I guess that’s uncommon.”

            Brain… exploding. I can’t even grasp the idea of not having a 1/3 measure. Sooo many things are in 3:1 (or 5:3, which is (1 2/3):1) relationships in cooking that it’d be a pain in the neck to have to do everything in quarter increments. I mean, of course you can do it, but it’s just really surprising.

    2. The measuring cup markings will show you that 3/4 cup = 6 fl oz, but you’d have to figure out that a (US) quart is 32 fl oz, and the answer is 5 and a bit.

      Actually, the nutrition label on my gallon of milk tells me there are 16 one cup servings in a gallon, so I assume your quart will tell you it is 4 cups, the 4/0.75 gives 5 and a bit.

    1. The Fahrenheit temperature scale is a bit odd, but actually logical to a degree, and more precise than Celcius without having to resort to decimals. I’m unsure why he based 0deg on a brine solution, but the scale puts plain water’s freezing and boiling points exactly 180 degrees apart.

      The Kelvin scale uses Celcius degree intervals and starts at absolute zero. Likewise, the Rankine scale also starts at absolute zero, but uses Fahrenheit degree intervals.

      1. It’s not even defined that way, really. It’s defined as a fraction of the triple point of water, with water being defined as a specific isotopic mix.

        Why did it have to be defined that way? Because the original definition (freezing point of water = 0, boiling point of water = 100) sucked.

        1. See, this is just fundamentally misunderstanding how temperature scales came about and why they’re useful.

          Celsius, as temperatures you can estimate *without any reference*, is stupid – you can come across ice pretty easily, but you’re not going to stick your hand in boiling water for the 100 C reference. And room temperatures certainly vary a heck of lot more than, say, your body. (Plus you usually *want to measure* the temperature in the room). So you end up having to remember “37 C” for your body. Which is obviously just as stupid as remembering “32=ice”.

          Fahrenheit isn’t a ton better, because the 0 reference is dumb. The 100 reference would be better if human body temperature were 100, but at least in practical experience, the 0-100 Fahrenheit scale is “it’s effing cold out” is 0 and “it’s effing hot out” is 100. 0 = ice and 100 = body temperature would be smart, but that scale doesn’t exist, damnit.

          So why did everyone fall in love with the fact that Celsius has 0 = ice and 100 = boiling? Because *scientists could calibrate thermometers better with it*. But, in fact, those calibrations *weren’t good enough*. Eventually, temperature became defined by thermodynamic energy of water’s triple point, because that’s how you get the most accurate measurements. Having “0=ice” and “100=boiling” wasn’t so it would be easier to remember, and it certainly wasn’t because it was more useful.

    2. Who cares about the definition? It’s how useful it is that matters. Most livable portions of Earth range between 0 and 100 F. Almost all of the outdoor temperatures you encounter are bounded between 0 and 100.

      If you think the definition is stupid, find a better one that’s close, and use that. That’s what metric’s been doing with its units for a hundred years.

        1. Yeah, probably, but really, people are fine below 0 for a good while. In fact, amazingly, 0 F is actually really useful – on a typical day, which in the US has wind speeds around 10 mph, below 0 F you’re at real danger for rapid frostbite. In other words: 0 F is pretty much a “you shouldn’t be going outside” point.

          If you just treat Fahrenheit as “don’t go outside below 0 or above 100” it’s a damn useful scale.

  22. The metric system has more benefits than converting miles to inches. That alone is plenty annoying- 63,360 inches in a mile vs. 1,000,000 millimeters per kilometer.

    Gasoline is 75% as dense as water. How much does a 15 gallon gas tank weigh? No earthly idea. How much does a 40 liter tank weigh? 40*.75=30 kilograms. That simplicity isn’t just convenient, it bleeds over into every aspect of design and engineering. By knowing the density of something you can automatically know the weight just based on size or volume. I can guess the weight of a 350 milliliter can of soda within 1%. I would bet that less than 1% of people would have a half decent guess on how much a 12 ounce can weighed. You can automatically have a mental idea of how much every part weighs and how big it is and how much of x you need to make it. It has to be done consciously in imperial.

    1. Hell its useful for everyday life, not just engineering. How much of a workout are you getting by lifting two gallons of milk? How many pounds is that? 8 liters of milk is about 8 kg.

      Can you put a 2 liter on a shelf that can support 5 kg? Yep. Can you put a gallon on a shelf that can support 5 lbs? uhhhh…

    2. 15 gallons of gasoline: 1 gallon of water weighs 8 pounds (“a pint’s a pound the world around”, times 8 pints). 15 gallons of water weighs 15*8=120 pounds. Gasoline is 3/4 (75% or .75) the density of water, which generally means 3/4 the weight by volume. 3/4 of 8 = 6, therefore a gallon of gasoline weighs 6 pounds and 15 gallons is 90 pounds.

      1. Exactly- three unit conversions means you have to do it consciously rather than just knowing it, as you do in metric. Not to mention that I’ve never heard that saying in my life, had no idea it was true, and a pint isn’t the same volume in every country.

        1. Why would I need to convert a gallon to cubic inches or cubic feet, when all I need to know is how much weight 15 gallons of gasoline adds to my car? I’m not building a rocket within certain physical size constraints, and having to determine how much fuel I can carry, I’m filling the gas tank in my car. Maybe the car manufacturer has to determine what space is available to carry fuel, but they do that calculation once and then they build the thing. All I care about is how far I can drive on a full tank, and how much it’ll cost me to fill it.

    3. So metric is awesome because the density of water is convenient?

      So you’re saying it makes sense to use a measurement system when a common measure is convenient? Which is like… the entire basis of the imperial system.

      1. No, metric is awesome because volume, size and weight are related as simply as possible. The fact that all the units can be scaled up and down sensibly is just a bonus.

        This “Imperial is convenient” argument is an insane piece of cognitive dissonance. It’s convenient because feet are a useful measurement, but at the same time putting every sub-inch measurement in fractions is just not a big deal. I’m 6′ 1″, 1.85 m. That’s literally the only time you can ever argue feet>meters. For anything more than 10 feet, meters are more convenient. For anything less than 3 feet, centimeters are more convenient. Even in that narrow range feet really aren’t any better, because you still have to say 5′ 10″- three digits, same as centimeters, but 254% more vague.

        People who say meters are unintuitive just haven’t wondered if they are only unintuitive because they didn’t learn them.

        1. “That’s literally the only time you can ever argue feet>meters. For anything more than 10 feet, meters are more convenient.”

          So you’re saying that feet are more convenient than a certain other measurement system? Great! So use feet/inches for that, and use meters for other ones where meters are useful.

          Don’t like having to convert between things? You already do, all the time – you had to memorize which prefixes corresponded to which exponent. You had to memorize how many seconds in a minute, how many minutes in an hour, how many degrees in a circle. God, could you imagine metric degrees? Instead of a 30-60-90 triangle, you’ve got a 8.33-16.67-25 triangle. Equilateral triangles aren’t 60 degrees each, they’re 16.67. Yeah, that’s totally better.

          Hate the random 2.54 cm = 1 inch? Fine, get rid of it. Define a metric foot as 1/3rd of a meter, and a metric inch as 1/12th of that. Then you’re just about 5′ 7″.

          Metric already has bizarro things about it: kilogram is the base unit of mass, but it’s a “kilo.” You’ve got prefixes for 10X, 100X, and 0.1X that practically no one uses unless they think it sounds cool. Why not just add a modifier for 1/3rd, and 1/36th? Do that, and mark those on a meterstick, and I’d be thrilled.

  23. Its ironic to see the USA clinging to the detritus of a defunct Empire they fought hard to escape from.

    You might forgive the British, since they invented most of this nonsense and imposed it on their conquests, but even here we are as likely to work in kilograms and litres as lbs and gallons. I buy my fuel in litres, but measure my cars performance in mpg and read road signs that are in mph. We at least have the excuse of the delusions of our supposed former grandeur for clinging to this arcane system. The USA arguably does not, but clings even harder than us deluded Brits, Ho hum… now where did I put my noggin of ale and my three fingers of Scotch?

    1. Well the British system came out of the Roman system which was widespread in Europe, and then French and German systems had their livres and pfunds? So euro colonies established before Napoleon would have had their nearly compatible versions, wavering a percent or two on definition. Then Napoleon got the metric bee up his behind and based it off not quite accurate measures of the earth and so on, and since he was a tyrant and got beat, the British felt no obligation to adopt his faulty measurements or drive on the left.

    2. “read road signs that are in mph”

      Aaaah, but have you noticed that the distance signs on motorways and major A roads are given in Km? I’m talking about the little, blue ones you see every 20 seconds, they provide the distance from the beginning of the road in Km, usually jumping 0.5 Km at a time. Thus there are probably far more distance signs in Km in the UK than ones in miles ;-)

  24. To everyone telling us how it would be impossible to switch from Imperial to Metric, BULLSH*T!

    Lots of other countrys have already done it, just learn from them, set a date, and be done with it!

      1. And like the U.S. Believe it or not, we measure a LOT of stuff in SI units. The only thing we’re really stuck on is nuts and bolts. Everything else could change overnight and not really hurt anything.

        1. And really, even if nuts and bolts changed, you’d still need tooling sets for both for… pretty much forever. SI is well structured and standardized now, but metric was really a fair mess for a while. You had CGS units, MKS units, MTS units, and even the quadrant-eleventhgram-second system for electrical engineers. (Yes, ‘eleventh-gram’, or 10^-11 grams). I don’t terribly blame the US for not switching – metric probably didn’t look like a big win in the early 20th century.

          And SI is only a post-WWII thing, so you’ve got this huge amount of early-20th century mechanisms that need to be supported.

          1. >quadrant-eleventhgram-second system for electrical engineers.

            Never heard of it.

            The International or “practical” system of electrical units developed by the IEC was used from the end of the 19th century. It is VERY similar to the MKSΩ system proposed by Giorgi or the MKSA system approved by the CPGM (the values are very slightly tweaked)
            https://en.wikipedia.org/wiki/International_System_of_Electrical_and_Magnetic_Units

            I had some exposure to these old units, but for the most part, the minor difference is just ignored. My university taught MKSA, although the SI had been released by then, and these older units were just treated as equal to modern units.

          2. The practical electromagnetic units (ohm, volt, watt, etc.) along with the definition of the permeability of free space as 1 is the QES system.

            Giorgi’s proposal tossed the idea that the permeability of free space should be 1, which pushed it into tiny-land: 4*pi E-7 since people decided they really wanted Maxwell’s equations to look clean. This brought the length/mass measures back up into reasonable levels (meter/kilogram), and formed a complete set of units.

            And all of that was happening in the early 1900s. The point here is that you actually had two disjoint sets of units at the time – you had electrical units (the ‘important ones’ were seconds, joules, volt, ohms, farads, henries, since things were already measured in those units), and you had practical units (centimeters, grams, etc.). Joining them was not something you could do arbitrarily – only certain combinations of units would do.

    1. Australia switched over when I was a kid, & it really wasn’t that big a deal. We used both for years, & gradually phased out Imperial. Anyone born here after the 70s has used Metric all their life. Americans are just putting off the inevitable & making it harder for themselves.

  25. Okay, either trolling / clickbait here, or somebody had a deadline and a case of writer’s block.

    The imperial / metric wars crop up here now and then naturally. There’s no need to seed them. Nothing is ever achieved, we all just repeat the same stuff everyone’s said about the subject for decades.

    Can’t you Google “technology” or something, if you’re hard-up for an article?

        1. And again. It’s always going to get people arguing over it, and many of us just can’t resist a chance to put right those idiots in the other camp. But it’s still a worthless trolling click-bait article. What’s been achieved in these comments? What’s learned? What of value is said?

          Pretty much nothing. And it’ll be the same the next time somebody accidentally triggers the Metric vs Imperial button. Always and forever. But pressing the button on purpose is unforgivable. Craps up the site. I’d accuse it of wasting time, but of course my time is worthless!

          1. I’ve learned quite a bit myself, mostly where I was previously wrong. That makes this a good thing. I’m hoping a few of some other folks’ misconceptions have also been corrected as well. If you don’t like this discussion, why are you still here?

          2. I agree – I’ve also learned stuff. Some factual, some misconceptions. It takes time to dog through the dross to find the comments that add value to the discussion, but I would say that it’s well worth the effort.

          3. Here’s what I learned this week:
            1) People think that “U.S. Customary Units” equates to using binary fractions (1/2, 1/4, etc), even though that’s a separate thing.
            2) Many countries BESIDES the U.S. still use their old measurement systems in many areas. I was particularly surprised that Great Britain apparently still measures their roads in miles, whatever kind of miles those may be.
            3) Many people regard Americans as idiots. This is kind of hard to dispute when considering our election results, but I do try to make the point sometimes that not ALL Americans are idiots.
            4) That there are STILL multiple standards in the U.S. for pints and gallons, and that these differ from what is still used in other countries that use pints and gallons.
            5) That the liter is not an SI unit.
            6) That the electrical units in the cgs variant of the metric system are different from what I thought were the ONLY electrical units.
            7) That some Canadians consider themselves bound to U.S. customs due to their proximity. This is baffling.
            8) That Russia had a traditional system of measurements. I thought they just used whatever units were used in the technology they were stealing. (But seriously, I do get tired of people claiming that the U.S. stole every bit of technology we have. Let’s face it, advances are made by individual minds, and it’s just been luck that many important discoveries have been made by Americans. And by Russians.)
            9 – 12) Other things I can’t remember right now.

            12 new things from one article isn’t bad, though 12 new things in 800 comments does seem inefficient. I never know what I don’t know until I learn it. How do YOU know what you don’t know?

          4. “That some Canadians consider themselves bound to U.S. customs due to their proximity. This is baffling.” Nothing baffling about it: when I drive my car across the border, I’d better be able to function in miles, miles-per-hour, gallons, and Fahrenheit at a minimum, do a bit of food shopping and the other weights and measures come into play. Most of us, even the young, like my kids, who were born after we went metric make the shift almost unconsciously.

          5. > 6) That the electrical units in the cgs variant of the metric system are different from what I thought were the ONLY electrical units.

            Yes, but as an electrical engineer, I have no idea who used them. I know I never learned them or used them, nor seen any evidence of their use. In addition to the bizarre cgs electrical units, there was a “practical” set of units, that evolved and were used from the 1870s.. The practical electrical units were VERY close to modern, a small fraction of 1%. I don’t know the story but I suspect the SI ampere was chosen to correspond with some minor rounding to improve the definition.

            The idea that electrical units could be integrated with mechanical units is attributed to an Italian named Giorgi around 1903, but it took until 1946 for the CGPM to agree on exactly how, define the ampere and replace MKS with MKSA. The (degree) kelvin, mole, and candle already existed as units in certain fields, so that 1946 system is essentially what became the SI in 1960.

          6. “Yes, but as an electrical engineer, I have no idea who used them. ”

            Astronomers, chemists, atomic physicists. You don’t usually see statcoulombs (same reason you rarely see coulombs), but you do see Debyes all the time in dipole moments, which are 1E-18 statcoulomb-centimeters, or 1 attostatcoulomb-centimeter (which clearly just rolls off the tongue, so who knows why they didn’t use that).

            Why they didn’t just define some arbitrarily tiny MKS-based dipole moment, I have no idea. Although hilariously, Wikipedia notes that there is no prefix you could use to get from the base SI unit (coulomb-meter) to atomic scale.

            There are soooo many quirky bizarre units lying around in physics everywhere that I have no idea why people think scientists like SI units. We abandon them as soon as we find something else which gives us numbers close to 1.

  26. I use both and am perfectly comfortable in both. What usually sticks in my craw is the perceived belief by people not from the US (usually Europeans) is that we in the US are neanderthals because we use the imperial system. This is often the source of many memes on the internet and a running joke (along with the whole kinder egg thing). The truth is that in the US, we are taught both systems. In addition, we are taught how to convert back and forth between them (although granted, it’s not always nice conversions and sometimes require a calculator). I concede that the metric system is far superior in engineering and sciences. Due to this reason, everything I do in these fields are done in the metric system. Where it really gets annoying is the belief that all aspects of my life should be converted to the metric system. Sorry, but what difference does it make if a trip will take me 500 miles or 805 km? Especially when the signs are all in miles. The kicker is the ridicule about using the imperial system comes from non-scientific individuals. Nothing like someone supporting their use of the metric system as being better for science and they never do anything in sciences or engineering.