Exactly. If you just stare and be like "I dunno" you may be a bad candidate. If you try to resolve it, even though you are way off with the answer, it means you may be a good candidate.
Well, putting my natural instinct to disagree with most of what Joel says aside, I've never liked that approach. Those are the kind of questions that make me think I probably don't want to work at wherever I'm interviewing. If I really needed to solve that problem, I would probably look up the answer. I certainly wouldn't make a whole bunch of estimates that all have ridiculous margins of error; that shows a willingness to problem-solve coupled with an inability to recognize your own ignorance, which isn't a super combination
You're right on the fact it's a stupid question you have to look up, and brings barely any value to the interview (I think in a follow up version he removes the question).
Regarding not recognizing your own ignorance, I think you definitely have to mention in your answer that really you are making up all the numbers and just tackling it in the first logical way you came up with.
