It's a bit annoying to me mainly because this sort of thing is really common among heuristics developers (even in papers submitted to conferences/journals). They develop a heuristic for a problem, and then show that it beats the exact, guaranteed optimal algorithmic solution. But that's only the right comparison if it's the first-ever heuristic for a problem! Otherwise it's a bit of a straw man, because its real competitors are other existing heuristics and randomized or approximate algorithms for that problem, and/or general, domain-independent heuristics like randomized hill-climbing.
Apart from the claim that it beats competitors, it's a really interesting post, though; light-weight heuristics based on clever connections are interesting in their own right. (So I sort of agree on the tone.)
This is extremely common in compression algorithms "research" as well. At least in video compression, the vast majority of papers are complete tripe, comparing only to algorithms known to be bad (i.e. an exhaustive search). You can easily spot the few good papers simply by looking for the ones that compare themselves to good algorithms.
Apart from the claim that it beats competitors, it's a really interesting post, though; light-weight heuristics based on clever connections are interesting in their own right. (So I sort of agree on the tone.)