You are too polite. Where I come from, we call this "a pile of horse shit."

You said it, not me :-)

Maybe the point was about specifically "trivial" questions. The article's author may have been attempting to draw an analogy between the questions we ask in interviews, and asking a doctor to state the normal range for human body temperature (e.g.), which probably only happens once during early-stages of training.

I think that is a possible interpretation of this sentence, but I'm not sure that I agree with it in the larger context of the piece.

In the preceding paragraph, he states that applicants are expected to know "the latest RB tree research." In the following paragraph, he states that applicants are asked "technical minutiae." These examples to me don't seem to imply he thinks that "trivial" questions are being asked in interviews. The entire piece seems to argue that questions which are too difficult are being asked in order to unfairly weed out engineers who cannot answer them, which seems to me to be the opposite of trivial.

To contrast, he explicitly used physicians as an example where occasional questions (which are 'trivial') have to be answered early in training in order to be licensed, a statement which I submit is false. There doesn't seem to be a corollary either stated or implied that physicians are otherwise asked nontrivial questions when they are not asked trivial questions.

I was a little unhappy to read this line as well. I thought this blog post was insightful and addresses an important topic, and I do think that the difference between the interview process in medicine and programming is worth investigating.

For starters, yes, you're absolutely right. Doctors must pass very difficult exams, and the statement is too inaccurate to go unchallenged.

But please consider the next statement as well: "These fields have the benefit of well-defined bodies of knowledge and monopolies over their skill, so in that regards I guess they have a huge organizational advantage to prevent incompetents from walking through the door."

I think there's so much of importance to consider here that I hope we don't stop after rebutting an exaggeration that I will fully acknowledge is very inaccurate.

Of course, it's always going to be difficult for a mid-career programmer to comment intelligently on interviews for mid-career physicians and vice versa. But my impression is that while physicians do take rigorous exams (and must continue to do so throughout their lives), the career and study path is much better defined, at least in part because Physicians are in far greater control of their profession than programmers are.

For instance, when a physician interviews, is the candidate possibly facing back-of-textbook questions about organic chemistry? Do they need to keep that book around and study it for 40 hours (or more) when they change jobs? If there are additional training requirements as the field evolves, how well defined is the study and exam path?

The impact of this particular blog post was weakened a bit by the inaccurate depiction of what it takes to be a physician. But let's still consider the issues. In other HN discussions about this topic (interviews), someone wrote that they'd happily take a brutal "Bar" exam for software if it meant that they could finally be free, for once, from the highly unpredictable exams they must take every time they interview.

For final thoughts - I was a math major (with some CS coursework) and I have an MS in Industrial Engineering. I interviewed for a company that does software in this area. During my interview, I was asked (from memory here)

Prove the dual of the primal is the primal of the dual; Now, with matrix notation; Various binary tree operations; Print all permutations of a string; Find the long term state probabilities in a Markov chain; Write a query testing to see if I know outer joins; Swap two integers without creating a third integer (asked over lunch); Refactor code with lots of if-then (to demonstrate that I know what a factory and strategy patterns are); Detect a cycle in a linked list; …

That wasn't all, because I interviewed at three companies over the course of a week and a half, but I think I'll stop there. At one point or other, in my life, I've taken exams on all this. I wasn't sure of what would and wouldn't be asked, and of course, I don't walk around with all this loaded into exam-ready memory. My guess is that a physician wouldn't have equivalent coursework loaded into short term memory either, and wouldn't be grilled on it in interviews, but what do I know?

I don't know what the solution (if any) will be, but I think it's a good start to recognize that 1) it's horrible, and 2) other professions - even ones that require massive amounts of education - don't apply the process so arbitrarily and capriciously (for instance, I had no idea I'd be asked about strategy patterns or asked about optimization math proofs. The next week, I was asked about database transactions and locking).