Hacker News Reader

Comment by scrooched_moose | Hacker News Reader

The biggest conflict I see is in many fields were you are dealing with dimensionless numbers, you are dealing with a lot of them once.

Having to specify "With a Reynolds number of 1300 U and a Prandtl number of 1 mU" is a lot more cumbersome than common convention of "With Re=1300 and Pr=1e-3".

View on HN

There's also the problem that 1U×3U=3U² and the fact that you wouldn't be able to add 3U² to 2U since the'd be different units.

Aye, but 1^2 is 1,so it's a bit of a special wash. Like 0!.

The problem is that U isn't 1, it's meant to be a unit, but then it doesn't behave like a unit in which case, why bother with it at all?

This isn't a problem in practice though, you just get "strange" constants with units equal to some negative power of U in your equations. The same thing happens when you move from electrostatic units to SI: suddenly Coulomb's constant appears. Many people advocate ESU for precisely that reason: no Coulomb's constant.

That's not how you'd use it. It's a "dimension" for dimensionless numbers. You don't append measures of length with "length" or volume with "volume" but with the actual units of length or volume.

I developed code to do unit conversions and I do include a "dimension type" for dimensionless numbers, which is effectively a 1.

how is 1 mU more cumbersome than 1e-3? The whole point of the prefixes is to avoid those cumbersome exponents, isn't it?

Reply on news.ycombinator.com