The idea that being a "dumb" programmer helps with debugging reminds me of a tactic I see people use on StackOverflow. Some programmer will post a problem they're having, and lots of ultimately unhelpful solutions will be suggested until someone else asks "what are you trying to accomplish?" and then it's suddenly made clear that the programmer's problem was being approached in the wrong way.

I've seen this happen so many times on SO, with good results, that it's become the way I want to think about other problems in life.

This is common enough on SO to have a name: the XY Problem.

http://meta.stackoverflow.com/questions/66377/what-is-the-xy...

In that regard, see the three classic virtues of a programmer - laziness, impatience and hubris. http://www.cranked.me/2008/07/laziness-impatience-hubris-thr...

Efficient != lazy Inquisitive != dumb

I agree with the overall gist of the article, but the title is obvious link-bait and his valid points have been twisted just to fit it.

"a good programmer must be dumb. Why? Because if he’s smart, and he knows he is smart, he will:

a) stop learning b) stop being critical towards his own work"

No. Just, no.

Exactly. That is pretty much the opposite of the Dunning–Kruger effect.

http://en.wikipedia.org/wiki/Dunning–Kruger_effect

So being smart equates to not being able to put oneself into question ? Not the way I'm used to think about intelligence. Though, programming cultivates the habit of rethinking your premises, I think we can all testify.

Your post adds far less value. Please leave the snark at home.