You are viewing a single comment's thread.

view the rest of the comments →

[–] Merchant_Menace 0 points 0 points (+0|-0) ago 

For precision work we use "thou" which are thousandths of an inch and "tenths" which are a tenth of that. The fractional lengths you're talking about with other bases are more used for standardized sizes like screws, wrenches, etc. Incidentally, you use our 1/4", 1/2", and 3/8" standard for your socket wrenches. You know, on account of standards being useful and convenient to just keep in place.

[–] prairie 0 points 0 points (+0|-0) ago 

Incidentally, you use our 1/4", 1/2", and 3/8" standard for your socket wrenches. You know, on account of standards being useful and convenient to just keep in place.

Thinking about all this there are different uses of measurements. One, like with sockets, where you need the exact size, and the units don't really matter. Here the values are essentially just unique names for each. In other uses, like cooking, you just need to be able to add and divide the units, so imperial versus metric isn't an issue (and imperial might win due to its fractional heritage). And then there are more general uses, where ease of representation as a decimal value and moving measurements between different domains are important for calculations. The point is that these can coexist without much of a downside as compared to using the same units for everything. Where it matters, one can use metric throughout, as my example of using metric for measurement.

[–] Merchant_Menace 0 points 1 points (+1|-0) ago 

Yeah I mean it's not hard to do unit conversions. It depends on what you're doing. It's fairly reliable to divide something into two equal parts by eye. Not so easy to do thirds. Even harder to do fifths. Almost impossible to do tenths by eye. If you'll notice our fractional system is all based on dividing by halves then adding halves. For example, 1/2, 1/4, 1/8, 1/16, 1/32 Then 3/8ths, 5/16ths etc. It's really useful for work where "close enough"-type work will suffice. If you're measuring across too many orders of magnitude then you just switch to a decimal representation and it's fine.