Respectfully, I think you're wrong on both counts. The first is an attempt at drawing a correlation. "When you observe fact-pattern X, you can expect that pro

Respectfully, I think you're wrong on both counts. The first is an attempt at drawing a correlation. "When you observe fact-pattern X, you can expect that property Y obtains." It might be derisive, but it's not AdHom. He's not using an unrelated character assassination to discredit an argument. I concede that it smells a bit ad-hom-y, but only because he has a lot bad to say about the article (all of it supported, imo), and so it kind of builds up momentum. That's the article's fault, though, for being so wrong about so much (although the article is right that WP is a complex (social) system, not perfectly documented). (A related mistake, which I don't think you're necessarily making, is that of thinking that anything that attacks credibility is AdHom. If you say "Bob is always wrong about biochemistry, so he's probably wrong about this claim about haemoglobin", that's not AdHom either.) As for the second, your inference might have been reasonable (if uncharitable), except that he later explains exactly why he brought it up: the AfD process has to operate at scale, it's not expected to be infallible. Combined with his other points about how notability has to work, it's build-up for his paragraph that starts with "Anyone who can snark... betraying a deep ignorance ...". And I think it's a pretty decent comment, overall, because it has a virtue I prize, of looking at a wrong-but-seductive view, and giving enough additional context to help readers form a better picture. Though of course the magnitude of the upvoting depends on name-recognition.