“log” is “whatever base makes most sense in context”. In a pure mathematics context, sure, “log” is base e, but in some places it’s base ten, in computer contexts it’s almost always base 2 and elsewhere it could be anything.
“ln” is unambiguous in all contexts. logarithmus naturalis is always base e. And so, since I’m a cross-discipline amateur, I’ll use “ln” every time.
Consider WolframAlpha that likes to give results in terms of “log”, meaning base e. If you feed its output back into it, it will give the option to change which base your log is supposed to be because it can’t be sure. It’s like it can’t read its own handwriting. Use “ln” and it won’t do that.
You’re right, I just wish it was unambiguously meant to be base e :(. There are many notations in math that are context dependant, for instance the volatility or variance in Gaussian distributions or the scale or location parameter in Poisson and exponential ones, or the integration symbol and “dx” order.
For me it’s the other way around. Log should always be the natural logarithm, other bases can be made explicit with an underscore.
“log” is “whatever base makes most sense in context”. In a pure mathematics context, sure, “log” is base e, but in some places it’s base ten, in computer contexts it’s almost always base 2 and elsewhere it could be anything.
“ln” is unambiguous in all contexts. logarithmus naturalis is always base e. And so, since I’m a cross-discipline amateur, I’ll use “ln” every time.
Consider WolframAlpha that likes to give results in terms of “log”, meaning base e. If you feed its output back into it, it will give the option to change which base your log is supposed to be because it can’t be sure. It’s like it can’t read its own handwriting. Use “ln” and it won’t do that.
This is perfect, I understand! My partner approves of this so I’m marking this as the accepted answer.
You’re right, I just wish it was unambiguously meant to be base e :(. There are many notations in math that are context dependant, for instance the volatility or variance in Gaussian distributions or the scale or location parameter in Poisson and exponential ones, or the integration symbol and “dx” order.
Agreed. I can’t tell if the joke is we should write the operation as ‘ln’ or if the joke is that we prefer base e.
I’m pretty sure it’s the second, but to me log and ln both have base e.
Other bases can be made explicit by division 😹