This article is very subjective. From the point of view of the author, he might be right.

But take this example: Web3 is less censorable. Once something is committed, it cannot be removed, unless a hard-fork event occurs, and even then there can be a portion of the network that decides that the old chain is the correct one (see ETH vs ETC). This still validates the decentralization aspect overall, so the comparison in the article, in my view, lacks of some additional perspective cases.