Thanks for the feedback. I was trying to make that clear with "grep reference", but you're right, it's still ambiguous. I'll add a note to clear it up. Both "wc" and "grep" don't solve the problem, they're just comparisons to see how fast other programs can read the input (and in wc's case, tokenize into words). For grep, I'm using "grep foobar".
[1] https://github.com/benhoyt/countwords/blob/eb2a8adf21c895907...