4 ms·

I needed this in the past but didn't know of it! It seems convenient, because when you write km and kWh in some places and then just write 'k' behind another nu

by lucb1e 5y ago

I needed this in the past but didn't know of it! It seems convenient, because when you write km and kWh in some places and then just write 'k' behind another number that is simply a unit, it looks like you're missing something, or might have written Kelvin in lowercase, but really you mean a thousand units. Writing kU is then unambiguous. Not sure why `response to the proposal of the uno "had been almost entirely negative"` according to Wikipedia. Does anyone know more about that? It's not as if I need this daily, most times it's clear from context, but still I remember wishing for such a unit in the past.

dragonwriter 5y ago
> Not sure why `response to the proposal of the uno "had been almost entirely negative"` according to Wikipedia. SI already has at least two different unit names for dimensionless 1 (rad and sr), so a third identical unit was probably seen as superfluous.
- HPsquared 5y ago
  Radian and steradian are specific to measurement of angles. You wouldn't provide a Reynolds number in radians, for example.
reikonomusha 5y ago
As someone who works in science, it’s because it mostly feels like a solution looking for a problem. The “unit” U gives the same vibes to scientists as some classes/abstractions thought up by creative programmers feel to software engineers. Imagine your colleague saying you should use a function called def doc(val, docstring): if __debug__: print("using value described as %s" % docstring) return val like in x = doc(5, "number of fingers on one hand") because “this leads to incredibly readable code, and the compiler won’t compile it if you don’t include the string.” Could it be valuable in principle? Sure. In practice? Just a nuisance. A comment will do. I don’t think I, or my colleagues, have ever run into an issue with ambiguity arising from unitless quantities.
- jhardy54 5y ago
  I hate to ask, but is the example you suggested A Thing? I agree that it’s terrible for a handful of reasons, but it also seems kinda nice that you could annotate any variable with a description that’s passed around through its lifecycle.
  - reikonomusha 5y ago
    I don’t think it’s a Thing (TM), but it’s the kind of “abstraction” that someone might feel justified in proposing. Similar things get proposed all the time on programming projects, especially around configuration management, documentation, code generation, resource management, etc.

by lucb1e 5y ago

dragonwriter 5y ago

> Not sure why `response to the proposal of the uno "had been almost entirely negative"` according to Wikipedia. SI already has at least two different unit names for dimensionless 1 (rad and sr), so a third identical unit was probably seen as superfluous.

HPsquared 5y ago
Radian and steradian are specific to measurement of angles. You wouldn't provide a Reynolds number in radians, for example.

reikonomusha 5y ago

As someone who works in science, it’s because it mostly feels like a solution looking for a problem. The “unit” U gives the same vibes to scientists as some classes/abstractions thought up by creative programmers feel to software engineers. Imagine your colleague saying you should use a function called def doc(val, docstring): if __debug__: print("using value described as %s" % docstring) return val like in x = doc(5, "number of fingers on one hand") because “this leads to incredibly readable code, and the compiler won’t compile it if you don’t include the string.” Could it be valuable in principle? Sure. In practice? Just a nuisance. A comment will do. I don’t think I, or my colleagues, have ever run into an issue with ambiguity arising from unitless quantities.

jhardy54 5y ago
I hate to ask, but is the example you suggested A Thing? I agree that it’s terrible for a handful of reasons, but it also seems kinda nice that you could annotate any variable with a description that’s passed around through its lifecycle.
- reikonomusha 5y ago
  I don’t think it’s a Thing (TM), but it’s the kind of “abstraction” that someone might feel justified in proposing. Similar things get proposed all the time on programming projects, especially around configuration management, documentation, code generation, resource management, etc.