[barrkel]

The rate of increase must have been declining (i.e. rate of rate of increase negative) in order for it to have peaked. This was probably detectable.

[lifthrasiir]

That makes me think: is there any instance that the rate of increase of packages or similar metrics ever went below zero (presumably due to removed packages)? I'm not sure if it is generally impossible in typical package manager models though.

[quotemstr]

If one is paying attention, one can spot early signs of decline in lots of things that seem very strong.

[benj111]

"and rate of rate of increase negative"

Are you a politician? Because that's a very round about way of putting it. :P

If the rate of increase is negative, there is an actual absolute decline. Is that want you were meaning to say?

[barrkel]

I said rate of rate of increase. The second derivative is negative at a local maximum.

https://en.wikipedia.org/wiki/Derivative_test#Second_derivat...

[caf]

...and the sign of the second derivative flips at an inflection point.

[benj111]

My mistake, I misread.

[antisemiotic]

"rate of rate of increase" = "second derivative", I presume. Negative second derivative doesn't necessarily mean a decline, it can be just growth slowing down.