[stan_rogers]

The body temperature the Fahrenheit scale was made to work at was axillary (armpit) temperature, at 96°. That's a multiple of 32°, and both are evenly divisible by 16, 8, 4, and 2, making marking of the scale easy. Both are common temperatures that a person would be likely to experience (whereas 100°C is something you'd hope not to experience, but rather to observe). That leaves zero as "frikkin' cold" by most European standards. It's a perfectly reasonable scale if you're not obsessed with base-10 values.

[int_19h]

It's hard to not be obsessed with base-10 values, when your numeral system is base-10.

I am sympathetic to the argument that 12 (but not 16) would make a better base. But that argument only works if it's uniformly applied bottom up - base-12 everywhere, and then we define base-12 metric prefixes, use 144 degrees between defining points for temperature etc.

As it is, the incoherent mix of base-16 and base-12 that is common for American customary measures, and base-10 used to actually write them down, is a mess.

[wayn3]

if you think that they're "easily divisible by 2", why not tell me what 832 is in terms of exponents of 2.

Thought so. Most people have enough difficulty understanding multiples of 10. No need to make it more difficult for the mentally challenged. For us scientists, factors of 10 prove really helpful.