And some things you don't have to prove, because you've seen them with your own eyes. Have you ever seen a programmer that is crazily more productive than another? If yes, then you know what we're talking about.
It might not be 10x or 20x --which is hard to measure anyway--, but we've all seen programmers that are far far better than others.
What's crazy to me is why would anyone try to argue otherwise. Isn't the same true for EVERY other field of activity?
There are painters (e.g Picasso) that are crazily more prolific than other painters. There are inventors that are crazily more prolific than other inventors (cough Tesla cough). There are athletes that are crazily better than other athletes. Writers. Composers. And so on. On some of those fields (like sports) we even have objective ways to measure how better they are. But in all fields, we know there are geniuses, crazy good guys and average ho-hum dabblers.
So, why exactly, would it be any different in programming?
>Thomas Kinkade was, I believe, far more prolific than Picasso. But I do not want Kinkade-grade developers.
I meant "prolific" in the sense of producing lots of GOOD work, not merely churning stuff.
Which for Picasso, it was. If anything, the critics agree (which is as far a metric as you get in the fine arts).
>There needs to be some other metric for software development. Figuring out what that should be, and how to accurately measure it, is non-trivial.
Sure, but the "good programmers are X times more productive than bad" (as I see it) is not about producing more code, but about producing more of what a project needs. I.t they are qualitatively more productive, not quantitatively.
Come to think of it, it sounds more of an obvious statement, almost a tautology or truism: "good programmers produce MORE good code than average programmers".
