In the tables near the top of http://docs.python.org/library/itertools , some functions like cycle have an example that uses “seq”, which is interpreted as “sequence” but the functions do work with any iterable.

cycle() doesn't use seq, but p (p0, p1, ...). Others use seq (seq[0], seq[1], ...). How do you think these should be changed?

cycle was a bad example, but there are other: ifilter, islice, takewhile.  "it" (like what tee uses) instead of "seq" would work.

I would like to stick with "seq" because another other variant is less readable or understandable IMO.

Also note that the "seq[0], seq[1], ..." notation only makes sense for objects that are indexable.   It is very slightly misleading but does a good job of communicating what the itertools do.

The line before the second table says "sequence". That could be changed to "iterable".  Specifying "it" in parentheses just after "iterable" and use it consistently in the examples could make it clearer.
I agree that using [0] and [1] is OK even if not 100% correct though.

Per Raymond’s feedback, I will close this.  The table is useful as a quick reference, seq is a shortcut, “it” or “iter” are worse names, the first doc line says this is all about iterators, and the individual function docs do use “iterable”.

Thank you.