[SoftTalker]

Imperial measurements do have the benefit of more even divisors than metric.

Pretty common to talk about measurements of 1/2, 1/4, 1/8, 1/16 of an inch and find those graduated on a ruler. Or 1/2, 1/3, 1/4, 1/8 of a cup for liquid measures, etc.

But then machinists generally work in thousandth or ten-thousandths of an inch.

[crote]

The problem is that fractions suck to do math with. Rather than doing basic addition you have to do a divisor conversion and then do an addition. Heck, the third-pounder burger failed because a significant fraction of the American consumers believed it was smaller than a quarter-pounder!

And most of the kitchen stuff is a chicken-and-egg problem: the US used 1/2 cup because that's what it is used to, the rest of the world has recipes calling for 50g of flour. If the US was used to weight-based measurements everyone would have a kitchen scale with 0.1g precision lying around, instead of a bunch of measuring cups.

[tpmoney]

The only evidence we have for the “Americans were too stupid to understand that 1/3 was bigger than 1/4” story is from the memoirs of the A&W CEO at the time referencing some unnamed market research performed by some unnamed firm. A person who has a vested interest in explaining why their failure to turn around a sinking company was due to external factors beyond their control. It also implies that in a culture where every kitchen has at least one set of stackable measuring cups in 1/4, 1/3 and 1/2 sizes, and a culture where you would order meat from the deli counter at your grocery store in 1/4 or 1/2 lb increments was suddenly and completely incapable of remembering their basic fraction sizes when comparing the cost of a hamburger. Personally while it makes for a cute story, I really just don’t believe it.

[dylan604]

You sound like an 8th grader asking the teacher when will we ever use this in real life type question. Fractions aren't that bad when like anything in life you are practiced in using them. Let's see what's faster in a kitchen, scooping something by measuring cup or trying to get a specific weight by 0.1g precision? I'm home after working all day trying to feed myself and mine. I'm not trying to win a start for my restaurant by the Michelin man.

These incessant arguments of "why is someone doing something different than what I'm used to so stupid" are funny if not tiring. Why not ask why are there so many different spoken languages in the world instead of just speaking like me? We could go into if Rust is better than Go, or why Romulans are better than Klingons. The problem is that nobody wants to understand the differences and just want to rag on the person opposing their views. Yawn

[digitalPhonix]

> scooping something by measuring cup or trying to get a specific weight by 0.1g precision

Why are you conflating speed and precision? A cup-ish measure would be 200g and no one is measuring that to .05% accuracy.

On the flip side, good luck getting .1g precision out of a measuring cup if you need it.

[dylan604]

I'm pretty sure my reasoning was in my comment. Nobody making dinner wants to weigh things out. That's why there are measurements like 'add a pinch', 'just a skosh', or similar. Pretty much the only people I know that measure would be bakers.

The point you're not willing to accept is that there's no dying need to be that precise and using "cup-ish" measures is good enough and works just fine.

[digitalPhonix]

“About 5 grams”, “approximately 250ml”.

Precision is orthogonal to the unit.

[SoftTalker]

Fractions are often easier. What's half of 3/8 inch? Easy, 3/16. And you'll find both those graduations on most rulers or tape measures.

[analog31]

I'm a fairly experienced machinist, though it's not my occupation. At the present time, machinists have the least problem with metric. The unit conversions are built into all of the machines and measuring tools. You press the inch / mm button. The ruler has inches on one side and mm on the other.

Everything is standardized on IEEE floating point. ;-)

It's a headache to maintain collections of parts and tools such as taps and dies for both standards.

The biggest shift is simply the obsolescence of old stuff, and emergence of new stuff. And industries have adopted the practice of reducing the overall variety of parts needed. I work in the development of industrial measurement equipment, and where a design might once have had 30 different sizes of fasteners, now it's 5, all metric. Designs rarely need nuts and spacers any more. Washers are integrated into the screws. No more "philips" or flat head screws. And so forth.

[SoftTalker]

> Designs rarely need nuts and spacers any more.

What is the modern approach where nuts/bolts and spacers would have been used?

> No more "philips" or flat head screws.

Torx? Rivets? Something else?

[analog31]

Threaded holes and rivnuts. Now a rivnut is a part, but the fabricator keeps it in stock instead of us. More shaped stampings and castings so that spacers aren't needed. Reducing parts also reduces handling of parts. Probably CAD makes it easier to design these things in.

At where I work, we use hex head (trying not to say Allen). I see lots more Torx in products as well.