〈 Berger | Dillon 〉

Looks great... buuuuut...sorry to be "that guy"... technically you're starting from the assumption that the series converges at all, which you do not prove with this. Similar to the "1+2+3+4+...= -1/12" argument, right?

Suppose S is finite in step 1 & there we have it the difference between this heuristic and the -1/12 one is that 1-1+1.. diverges always This converges for |r|<=1, and so while nits can be picked, at the end of the day I didn’t want to code up the condition

Yes. But you have to make that assumption, which was not visible in what you wrote or animated. iirc, the -1/12 problem goes quite a bit deeper than 1-1+1... not converging in the "for all epsilon..."sense. Indeed, I am being nit-picky, but that's math, right?