DJB would like to have a word with you ;) (I'm not sure if this article is not even a semi-troll ^^ - not that especially disagree with it though)
It is not a troll. There are pretty normie opinions across the board, except maybe for Filippo's Cobblestone construction.
I'm referring for example to the "classic mcellie is broken" even though my understanding is that the classic mcellie team is still contesting the results which are fairly new, his opinion that hybrid is a mistake for signature, the comment on the fact that P256 is equivalent or better that ed25519 (not in the article but a link to one of his email). I would not say this is controversial, but at least I feel this is debatable. Even the fact that PQ crypto is needed at all could be debated :) (Personally I'm convinced it's a good idea regardless of whether good enough quantum computers ever materialize, but it was the only risk, I'm not sure I would be so convinced)
These are all very normal opinions. Who's taking Classical McEliece seriously at this point? And there very definitely is a trend towards normalizing the P-curves; that's been going on since we got mainstream implementations of side-channel free scalar mult formulae for them. (Ed25519 is a signature construction, not a curve).