Well if you can get Raymond to agree on that too I suppose you can go ahead. Personally I'm -0 but I don't really write this kind of algorithmic code enough to know what's useful.
Actually, you do - but you don't _think_ of problems in these terms. Neither do I. For those who do: consider any program that has state and responds to inputs. When you get a new input, the new state is a function of the existing state and the input. That's `accumulate`! It generates a sequence of new states as the sequence of incrementally updated states derived from a sequence of inputs according to the function passed to accumulate.
In that view of the world, specifying a starting value is merely specifying the program's initial state - and from that view of the world, not allowing to specify a starting value is bizarre.
While Will Ness's wheel sieve program, and my Collatz glide-record-finder, don't _derive_ from that view of the world, it turns out that specifying (and returning) an initial value is also useful for them, and despite that they're just doing integer running sums.
A simple example from the broader view of the world is generating all the prefixes of a list:
from itertools import * list(accumulate(chain([], [8, 4, "k"]), lambda x, y: x + [y]))
[, , [8, 4], [8, 4, 'k']]
That's obviously easier to follow if written, e.g.,
list(accumulate([8, 4, "k"], lambda x, y: x + [y], first_result=))
I do think that the new parameter name ["first_result"] is ugly. But maybe that's the point.
I noted later that the `accumulate` in the Python itertoolz package names its optional argument "initial". That's not ugly, and works just as well for me.