Hacker News Reader

Comment by brianwawok | Hacker News Reader

I would say it's not always as cut and dry for the benefits of Metric over Standard.

When doing math? Yes please, sign me up.

For measurements around the room? Inches is a pretty good unit. CM is too small, DM is too big. There is a reason 90% of tabletop wargames (even those made in the UK by Games Workshop) still use inches for movement. It doesn't matter in a wargame how many inches are in a foot, or how many feet in a mile. You stay 100% in inches, and it just works.

Another interesting place is in carpentry. You can divide 12 by 2, 3, 4, 6... it just works well. There is a reason the bootstrap CSS framework uses 12 divisions and not 10. Again, the number of feet in a mile don't matter when framing a house. Are the studs 18" on center, or 24"?

View on HN

> You can divide 12 by 2, 3, 4, 6...

I'm pretty sure that dividing 12 by 2, 3, 4, 6... works in metric countries, too. :-)

But seriously, take something like the spacing of things like studs, which is commonly 16" or 24" in the US.

24" would indeed be a pain to divide by 2, 3, 4, or 6 if we were instead working with the same spacing but in metric, where it would be 60.96 cm [1].

But in Norway, the first metric country I happened to find information for, they don't use 60.96 cm stud spacing. The use 60 cm spacing for studs, joists, and rafters (23.622"). Dividing by 2, 3, 4, or 6 for them is as easy as dividing by 2, 3, 4, or 6 in a place using 24" spacing--and they also get easy divide by 5.

If you've got some random physical thing to measure that was not build by humans, like say how tall a tree is, you are probably going to get some length that isn't exactly 2, 3, 4, or 6 times an integral multiple of the smallest division on your ruler, regardless of what system your ruler uses.

For things that are built for humans and for which we will commonly want to divide them by 2, 3, 4, or 6 (or other small integers that we might anticipate), we will tend to pick their sizes to facilitate that using the units we expect the users to be using.

[1] That one does actually come out as an integral number of mm when divided by 2, 3, 4, or 6 so it is not too much of pain. It's just having to deal with more digits and a decimal point as opposed to also having to deal with something that falls between the marks on your ruler.

> For measurements around the room? Inches is a pretty good unit. CM is too small, DM is too big

Can you perhaps elucidate further on which 'Measurements around the room' require multiples of 2.5cm units to describe, and that fail with 1cm units.

> There is a reason 90% of tabletop wargames ...

So:

a) tabletop wargames are not the way we evaluate the convenience of our measurement systems, and

b) you just said 90%.

90%

90 / 100

How wonderful is it to use a ratio that world+dog totally, immediately, nay intuitively understands -- a ratio of a given number to the fixed value of 100.

For people genuinely committed to factors of 8, 12, 16, 60 etc -- how can they be comfortable with percentages? It's a betrayal!

Surely fractions of, I don't know, let's say, a gross (144) would be more natural.

144 = 12 x 12 -- so you instantly have all the benefits of 12, multiplied by another 12. What's not to love?

Rather than saying 90%, you could more naturally say 130/144 (okay, that's only approximately 90% -- it's actually 90.27777...%) but as you say, base 10 numbers are difficult to properly convey your true intent, so you can come up with a better number there.

Tabletop wargames are a pretty important part of life. So I rate it pretty high.

For carpentry, 12-inches-to-a-foot isn't actually that useful, since wood is never the nominal size. For lumber, nominal sizes are measured assuming that sawing and planing remove no wood, but sawing and planing will remove about 1/4 to 1/2 an inch in each direction. That means that you can't stack two 2x4 to get an exact 4x4, and you can't use dimensional lumber to fill a nominal 1 foot gap without cutting.

Carpentry, both the house-building and furniture making kinds, have adapted by emphasizing tricks to measure lumber without rulers. That means starting with more wood than needed, and cutting down to make pieces fit.

Where do you use division in carpentry? My understanding is that it primarily deals with adding up or subtracting to some length, since you must nearly always account for thickness. In that sense, it doesn't really matter if you use inches or centimeters, since you never convert units and never do any calculation that is more complicated than plus or minus.

AFAIK, IKEA uses millimeters, for example.

Bootstrap CSS doesn't seem all that relevant to the metric vs standard debate either, since it's neither of them. Time and angles are also divisible by twelve, for example.

You use division when coming up with even intervals. For example, "nail this panel to that board at 3 evenly-spaced locations." This shows up in joinery, too.

That said, I do a lot of carpentry and woodworking, and have switched entirely to metric. I find it a lot easier to do addition and subtraction, as you say, but mostly it's more calculator-friendly. Adding ((273mm / 2) + 60.5mm + 546mm) is a lot easier to punch into a calculator than ((10 and 3/4 inches) / 2) + (2 and 7/8 inches) + (1 foot and 9 and 1/2 inches).

The HP 35s Scientific Calculator is a fantastic calculator and has fractional arithmetic support. I reach for it in the shop frequently.

Hey, that’s a great tip, thanks! I still have to use inches sometimes, so I’ve just ordered that calculator.

Nice! I think you'll like it. One thing, hope you noticed that it's an RPN calculator... takes a short amount of time to get adjusted to but is superior to algebraic entry.

Here is an expanded manual on working with fractions:

http://h20331.www2.hp.com/Hpsub/downloads/35_15_Working_with...

If you want to play with the HP family of calculators that are RPN check out Free42 (Android, Mac, and Windows emulator for a classic HP calculator. Note, it doesn't do fractions like the 35s)

https://thomasokken.com/free42/

Non-metric systems used industries were chosen through experience and optimized in the same way that genetic algorithms optimize.

The SI units are rarely, if ever, more convenient than whatever legacy standard they replaced, but the path dependent development of naturally defined measurement systems meant different results in every country. Shipbuilding, lumber, textiles, printing presses would also end up with varying degrees of compatibility or difficulty in working with materials sized in another domain.

As an aside, the same people extolling the value of the metric system's base ten nature have little problem explaining why the dimensions of A4 paper are not round numbers. The legacy systems in their countries didn't come from users being unaware of a base number system, they just used to prioritize other conveniences that a measurement system can provide, such as how A0 through A8 paper sizes can be simultaneously cut from one sheet of paper.

Your post seems well considered, but I apologise that I'm struggling to ascertain the point you're trying to make.

I can comment on this bit though:

> As an aside, the same people extolling the value of the metric system's base ten nature have little problem explaining why the dimensions of A4 paper are not round numbers.

Metric isn't about round numbers -- this seems to be a common misunderstanding.

Metric users are happy to dive into sub-unit numbers (6.35mm is a number we see an awful lot).

Metric, I think, is mostly about being aligned with most of the world. A common language, as it were.

Secondly, it's about some pleasant relationships between different units (length, mass, velocity, volume, energy, etc.

Probably equal second, it's about making calculations easier.

Round numbers are irrelevant for this, of course. But base-10 is a massive boon.

I think the metric system facilitates trade, especially in bulk quantities of raw commodities, between different industries and jurisdictions. And, I think the decisions made to arrive at a universal standard are a combination of top down and "perfect" that typify nineteenth century science.

Earth doesn't have a 40,000km circumference, but it doesn't matter because anyone who needs a precise measurement will use some sort of aid. Similarly the mass of water in 1 mm of rain over a hectare isn't actually 10,000kg, but you'll use some aid if you need a more precise number taking temperature and actual contents of rainwater into account. Of course using nautical miles or acre-feet with high precision required aids of some kind, too.

However, legacy systems of measurements were developed to minimize the number of instances where people like machinists on a factory floor need an aid such as pen and paper to do a calculation. Big problems like Fermi estimates are much easier with the metric system, but day to day problems within a specialized profession probably are not.

I do think the benefits of a more universal system vastly outweigh the costs to convenience for everyday users of measurements, but I am skeptical of many of the rationales used for advocacy. Perhaps if an international system of measurements was designed today it would be even better designed for binary computers as computational aids rather than decimals and paper, though that would be an even more difficult transition.

Interesting to note that decimal fractions (e.g. 0.3) are awkward to represent with floats in computers, whereas fractions based on 1/2, 1/4, 1/8, etc. that are often used in the non-metric system are a very natural fit for the float representation.

Nothing stops you for using fractions with metric.

Nothing stops you from using decimal in Imperial :)

True, but:

If you use decimal with metric, the units and prefixes play along with you very well (huge advantage). If you use other fractions with metric, you get no special bonus or burden (neutral).

If you use decimal with US customary ("imperial"), you get no special bonus or burden (netural). If you use other fractions with US customary, you get no special bonus or burden (neutral).

In summary, there is absolutely no rational reason to stick with US customary, besides legacy compatibility (technical debt).

But, conversely, converting 3 pounds to stones with floating point is not quite as natural. (-:

But literally no one uses stones. People in UK when measuring weight of humans? I have never seen it in any other case. You don't buy things in the US by Stone.

billti didn't say anything about who uses things, but you can enjoy how similarly unnatural binary floating point is for converting 8 inches into feet if you like, or 2 fluid ounces into U.K. pints.

Reply on news.ycombinator.com