No, multiplying by 7.48 isn’t hard. The hard part is knowing that 7.48 is the number you need to use. And that’s just one of dozens of different random-looking numbers you have to keep handy, and remember which one is which.
It’s curious that people here are referring to the American units as “English”, since the English system of measurement is actually significantly different; the units have the same names but some have different values. For example, an American pint is 473 mL, but an English pint is 568 mL. Also different are fluid ounces, cups, quarts, gallons, and tons. And some customary English measurements, such as the stone, are used rarely or not at all in America.
Although this invariably the example that is cited whenever the issue of SI versus US Customary units, the reality is that this was a problem of requirement verification, not an innate problem with the units system. Spacecraft, even those built natively using SI units, have a wide array of different unit conversions and coordinate transformations that have to be performed correctly. Engineers designing structures will use different units or scales than those working on guidance, navigation & control, which will used different units than those in software. Factors and conversion functions will be applied for different conditions and responses, and if any of these are wrong, things don’t work. In this case, in an effort to be “faster, better, cheaper” they eliminated verification testing to make up schedule, and thus didn’t catch the error.
This is because the auto industry sources so many components internationally and often even has systems or whole vehicles assembled in nations where US Customary units are not used (thus, making tools and tooling more difficult to acquire). The auto industry started going through this conversion in the early 'Eighties, and it was by all accounts a painful process, though more straightforward than for, say, construction or aerospace insofar that the modern auto industry is large enough to order mill runs and full lots to spec of materials and fasteners, and completely redesigns product lines on a roughly five to eight year cycle, meaning that they only need to maintain tools, tooling, and spares for a period of twelve to fifteen years. A manufacturer that doesn’t have that kind of purchasing capability and tooling turnover has to bear a much higher cost of obtaining SI-compatible tools and materials, which can be considerable indeed.
Relying on a calculator to perform unit conversions is lazy and prone to generating unintended errors. Wherever conversions are used in engineering or scientific calculations the conversion should be an explicit step with the appropriate factors for verification. In the case of using decimal-based units, it is a fairly trivial matter to confirm that the exponents are correct.
But there is an additional reason to prefer decimal-based units, and that is for the ease in applying and using logarithmic scales, Although analog calculators such as slide rules are no longer in common use, the base-10 log scale is still very useful for power and field factors that function on a power law, reliability and hazard calculations that fit a lognormal distribution, dealing with floating point calculations, and linearizing scaling factors in astronomy, climatology, biology, et cetera, that nicely fit to factors that are multiples of ten. Technically speaking, factors of 10 are arbitrary; it could as easily be an octal, hexidecimal, or base-13 scale, but being able to scale to a coefficient of an exponent is the basis for many methods of reduction is of inestimable value, even with digital computers and calculators that can perform unit conversions automagically.
Actually, both Fahrenheit and Celcius are problematic, as they are units based on relative scales that only work if you are performing calculations on a differential factor (difference or rate of change). Temperatures should properly be measured and stated on their absolute scales.
Stranger
That’s what the internet is for. It’s not like I don’t already have to keep track of hundreds or thousands of numbers anyway, even with SI, when doing any remotely serious calculations–for a fish tank, that might mean everything from the density of seawater to the mean packing factor of gravel. Of course, if all you want is to know how many jugs of distilled water to buy at the supermarket, then “7 or 8 per ft^3” is perfectly adequate. For that margin of error, it’s not even hard to guess based strictly on a mental image.
Of course, none of this is to argue that SI isn’t superior for many or most purposes. I just think many of the claimed benefits are overblown (especially for the “man on the street”), or have nothing to do with SI.
Also, quick: how many kilowatt-hours is 100 megajoules?
Units aren’t decimal (what would that even mean?). Prefixes are decimal-based, but within a calculation you shouldn’t be mixing prefixes anyway: a calculation that contains joules, kilojoules, and megajoules as units is inviting diaster. Use scientific notation (which has nothing to do with the metric system) instead and stick with one unit.
The only thing remotely decimal about metric units is the difference between MKS and CGS, which of course differ only by factors of 10, but you really shouldn’t be switching between those systems.
My point is that we almost never need to convert at all. If there is a large difference in the same measurement, we’ll use exponents. The only conversion I have to do is from grams to moles, where I have to look up the molecular weight anyway. There are advantages to metric units, but they are overstated.
the ability to scale quantities more easily makes the metric system valuable.
Chemist here…almost everything is in SI units, but some literature still uses calories (or kcal/mol) rather than joules (or kJ/mol). Yes, it’s simple enough to convert, but after a while you have an intuitive feel for what a reasonable value for, say, a bond strength is in kJ/mol, and it’s hard to apply that to values in other units - it’s a bit like someone from the US knowing that 72 F is comfortable, but not having much of a feel for 18 C.
Theoretical chemists (and many theoretical physicists) use atomic units.
They are another internally consistent set of units which has the property that many of the common constants (mass of electron, speed of light, hbar=Planck’s constant/2pi, 1/(4pi permittivity of vacuum)) are equal to one. It also avoids using extremely small numbers. For example, hbarhbar shows up a lot. In SI units, it’s approx. 10^-68 Js. Years ago, some computer systems turned numbers this small into zero… The atomic units for length, mass, energy and even time are all different from their SI counterparts.
Funnily enough, that one’s easy for me, because my work involves lots of converting between kWh, MJ and therms.
Hmm… I think the last few calculations I did (in the U.S.) used mg/l, inches and acres and convert back to pounds. Which perhaps reflects the mix of scientific, civil engineering, and legal definitions I deal with.
And the reason you almost never need to convert is that you’re using sane units. If you were using customary units, you would have to convert a lot.
SI units are incremented on a scale that is consistent with the decimal (i.e. base 10) counting system that is used in virtually all hardware engineering. Similarly, units that are used with discrete math are (generally) consistent with the number bases that are used in native arithmetic systems on digital computers (binary:base 2 or hexidecimal:base 16).
Again, in large scale systems where different technical disciplines use different tools, measuring systems, and modeling approaches, conversions are inevitable. You wouldn’t build a spacecraft using kilometers as a standard unit, or define a trajectory in millimeters, just because of the problem of dealing with tolerencing on each scale. But at some point you’ll have to relate one scale to the other, and there you will have to perform a conversion, which is much easier to sanity check if the units are in a consistent number base.
Stranger
Your absolutely right. I don’t mean to imply there aren’t advantages to metric. I used to be a hard core metric supporter. My contrarian personality eventually made me look at it from another angle though, and I think that the advantages are sometimes a bit overstated. I have never in my life had to deal with pressure in Pascals. I deal with Torr at low pressure and PSI at high pressure. I’ve never had to convert the two, because the two systems just never mix.
I think the most insane unit I have to deal with is the MIL. Thicker coatings are measured in MILS, and thinner coatings are measured in microns. At the scale I deal with, it’s 50/50 what unit you will use. It’s particularly confusing because MIL sounds like it should be metric. A MIL is 1/1000 of an inch. So, it’s sort of metric.
I still have no idea what that means. The SI prefixes (kilo, mega, giga) are decimal-based, but the units are not. The only confusion that the kilogram is considered a base unit in SI, even though it appears to be a compound unit.
Perhaps you are implying that, for instance, the kilometer is a unit in and of itself. It’s not; the meter is the unit, while the *kilo- *modifies the unit.
Since the prefixes are the only place where these factors of 10 appear, and not the units themselves, I claim that it makes no sense to say that the units are decimal-based.
All of this stuff is easy if you do it all the time; I didn’t even need a calculator to come up with “about 28 kWh”. But it’s still a conversion that you have to check and double-check, and in this case it’s not one where it’s orders of magnitude different and it’s obvious what the unit is.
Then I think you grasp SOAT’s point, and can address it, rather than an extended discussion on semantics of “decimal-based units” versus “power-of-10 prefixes.” Move along, Counselor.
The pound is a slug*ft/s^2
The pound-foot is… a pound-foot.
Hz may be SI, but it ain’t uniquely SI.
Of course, all of these quantities may come in other units: pound(mass), BTU, rev/min, etc. The same is true of metric units. Hell: how many German or French automobiles come with tachometers labeled in Hz? None, you say?
I think it’s an important distinction, though. Most of the complaints about non-metric units involve the non-power of 10 conversions. 12 inches per foot? 5280 feet per mile? Nonsense!
The thing is, even with American standard units, there’s nothing stopping me from simply converting to one unit at the very beginning and sticking with it, retaining all the advantages of decimal. In fact, I do this all the time: I do machining as a hobby, where just about everything is based on thousandths of an inch. Yes, it’s kinda stupid that my drill bits are labeled 17/64" or whatever; no matter, once I convert to decimal I don’t have to deal with that nonsense. Sometimes I’ll use millimeters instead and it’s exactly as easy.
On the other hand, the metric prefixes are overrated. There’s a reason that essentially nobody uses megameters or centiseconds or hectojoules: it’s because they don’t correspond to quantities that people care about. Instead, people pick a unit that works well for their purposes (nanometers f you’re a computer chip designer, femtoseconds if you do serious laser research,etc.) and pretty much stick with that. The powers of 10 again don’t really enter into it.
Even aside from that, I’m not sure SOAT’s point makes sense, although I’m willing to be corrected. You should essentially never find yourself taking a logarithm of a number with a unit, because you can’t actually take a logarithm of a unit. Instead, you take logarithms of ratios, so that the units drop out. And the ratio kilometers/kilometer works just as well as feet/foot or even rods/rod.
I hear people working in megameters all the time. It’s a very practical unit for dealing with sizes of features on the Sun. Why, what’s the customary equivalent, and is it used more often?
I think it’s fair to say that “people who regularly deal with features on the sun” qualify as “essentially nobody” (with all due respect to them).
Of course, now they have to use a new constant if their formulas are to work properly. If they are trying to calculate the magnetic field of a coronal flare, they either need to convert to meters first (trivial, but nevertheless an extra step), or remember that their new permeability of free space is 1.256x10^-12 T*megameters/A.
If they wanted, they could simply use feet or miles or whatever and get used to having scientific notation everywhere. Personally, I would stick with MKS and use E notation (1.23E9) for conciseness, but I’m not them and think they should use whatever they find convenient.
Anyway, I am by no means *advocating *customary units, particularly in the sciences. Quite the opposite; I think it’s great to have multiple systems, and metric happens to have some nice properties. The fact that there are multiple metric systems (generally but not always related by factors of 10) means there is even more flexibility in choosing a convenient system. Just like languages (both human and computer), some systems are better suited to a particular application than others, and everyone benefits by knowing more than one.
Of course, standardization is useful. Metric isn’t inherently better for, say, the auto industry, but it is useful if manufacturers don’t have to make two lines of nuts, bolts, sheet metal, etc. Due to the vagaries of history, this means it’s easier for the US to switch to metric than the converse.
This hasn’t happened universally, though, and so (for instance) for aviation the knot is preferred over km/h or m/s. In fact, the knot is “accepted for use” within SI due to its common use.