That one's really damning. It's implying that they're using some sort of syntactic equivalence to compare solutions, which is a big no-no. It's not that difficult to generate a bunch of tests and run both solutions through an interpreter, and in this case, it's not that hard to compare them both on every possible input by running them through an abstract interpreter.
