imogold

Certainly, but if the examiners would be accused of asking a too difficult question, as the article alleged, this would be defended by showing a more elementary approach. Or both of them, wouldn't hurt to have more ways of solving it. Unless the Soviet science was banned from using reductio ad absurdum proofs in some math version of Lysenkoism.

Out of curiosity, I've searched it and this article was posted to HN at least four times. Opinions aside, turns out it was merely a footnote in the exam pipeline https://news.ycombinator.com/item?id=4761846

Let's put this in the context: the article is about problems which are supposedly killer problems because they require some hard-to-get hint to solve. Therefore, a hint is low quality, if another hint exists that is simpler, more elementary, or the problem is solved straightforward anyway.

In this case, a simpler hint is: "Do not use derivatives or anything fancy really. Only solutions are constant functions. Proof by contradiction. Use binary search".

One may only wonder what level were the problems given to regular candidates.

Unfortunately, these "ideas" for the problems are low quality. Problem 2: "use derivatives". How the heck that helps if F does not have to be continuous, let alone differentiable.