That alone is definitely a very skewed test of what makes someone an engineer. But the author writes: "The rest 199 do not understand why a bitcount of 1's com

That alone is definitely a very skewed test of what makes someone an engineer. But the author writes: "The rest 199 do not understand why a bitcount of 1's complement of bitwise XOR of two variables would give you the number of similar bits in corresponding positions in both variables, why one can not create a regular expression to match only strings with balanced parentheses, or how to find the shortest chain of connections between two friends in a tiny social network. Note that I have used 'or' as the conjunction and not 'and'." If I understood it correctly, anyone who knows at least one of those three problems passes your hypothetical test. This is not very skewed IMHO. If someone doesn't understand either one or two of those things, he may be a good engineer. But do you think any programmer worth his title would not know all three of those?