Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

As a scientist, use scientific notation off the base units; you eliminate opportunities for confusion. The prefixes don't really add anything.

The Sun is 1.5 x 10^11 meters from the Earth.

Calling it 1.5 x 10^8 kilometers doesn't help you visualize the distance better, nor calling it 150 gigameters. Now you have three numbers that mean the same thing floating around, and if you accidentally write 1.5 x 10^8 m somewhere instead of km, or read 1.5 x 10^11 km as m, you've just introduced a thousand-fold error.

Common units (eg, Angstrom) make sense when you do not need to convert between them; it's convenient if they are easily convertible to your standard unit, but having mg and ug and g and kg floating around is just an unnecessary headache, you inevitably accidentally interpret a microgram dose as a milligram dose from time to time and poison a patient or eight.



Personally, I find the prefixes helpful when memorizing approximate values, eg for back-of-the-envelope estimates (proton mass is 1GeV, electron mass is .5MeV, room temperature is 25meV, h_bar c is 197 MeV fm, ...)


A valid defense of the metric prefixes (I mean, I still think they're on the balance a mistake, but you do have a point), but ironically, electron-volts are not an SI unit.

Why 1 GeV instead of 160 picojoules?


Unless you need to make a dial on a physical piece of equipment. In that case it's helpful not to have to append x10^N to all of your numbers.


Or if you need to write it down anywhere at all; that's why programmers use eN or e+N. Although we really need to get around to standardizing a hexadecimal equivalent.


If you don't mind I'll keep using milligrams and micrograms, micromolar, nanomolar etc. instead of writing all that out.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: