My favorite interview question is asking presenters to predict a flip of the coin.

You would be amazed that half the candidates cannot accomplish this task. I'm talking about people with doctorates from Harvard, open source projects a mile long, they still can't do it.

The remaining half are the ones who will truly shine on their job.

That is a pretty nice response, even if it is a bit snarky :-)

I don't think you, Daniel, would do well there. When I was interviewing people at Google I would always try to remind the candidate that whether or not the interview proceeded to the offer stage said more about the people who interviewed them than it did about the candidate's capabilities.

I like the refer.ly question because it allows the candidate to talk about something they are passionate about, or have thought deeply about. I'm not sure that it informs on their organization as much as the interviewer would like.

Be aware that you incorrectly interpret results of the question: all candidates guess correctly, but only half of them do it in your Universe.

Well, let's be fair... the method the article describes is a bit better than random. It's just asking for a work sample, except the work you are asking for is efficient communication.

I doubt the "inventor" of this method really thought it was the one true question, we can attribute that tone to the author of the article.

It's refreshing to see that some people have an ounce of sense when it comes to interviewing :)

Previous discussion here http://news.ycombinator.com/item?id=4613413

There is a potential problem with topics chosen by the candidates. The passion they have for their favorite topic might never appear in their daily work.

Sometimes it's nice to at least know they are passionate about something, anything at all.

I'd be very interested to know the percentage of people asked this who responded by explaining a project that directly related to the job they are applying for, and those that explain something from their everyday life.

I wonder which would be more effective. Passion in the topic of choice is without a doubt key.

This sounds like a trick question. If you want to measure passion for the job, ask about the job and listen to them. Don't ask them about something else and wait for them to bring it back up.

I'm not sure if I understand you correctly, but the question is a perfectly valid way to measure the ability of a candidate to explain complex topics to people with varying levels of pre-existing knowledge.

Here it is presented as a question used by Referly. But for historical accuracy; weren't early Google interviews famous for using this question years ago?

I heard about this questions while working at Microsoft in 1998, before Google existed. I am sure someone was asking it even before Microsoft existed.

Per the discussion I linked, Google used it; it was also used before that by Microsoft and Novell.