Point 1 is that a black raven is not strong evidence against white ravens. (Said knows this, I think.)

A black raven is, indeed, not strong evidence against white ravens. But that’s not quite the right analogy. The more accurate analogy would go somewhat like this:

Alice: White ravens exist!
Bob: Yeah? For real? Where, can I see?
Alice (looking around and then pointing): Right… there! That one!
Bob (peering at the bird in question): But… that raven is actually black? Like, it’s definitely black and not white at all.

Now not only is Bob (once again, as he was at the start) in the position of having exactly zero examples of white ravens (Alice’s one purported example having been revealed to be not an example at all), but—and perhaps even more importantly!—Bob has reason to doubt not only Alice’s possession of any examples of her claim (of white ravens existing), but her very ability to correctly perceive what color any given raven is.

Now if Alice says “Well, I’ve seen a lot of white ravens, though”, Bob might quite reasonably reply: “Have you, though? Really? Because you just said that that raven was white, and it is definitely, totally black.” What’s more, not only Bob but also Alice herself ought rightly to significantly downgrade her confidence in her belief in white ravens (by a degree commensurate with how big a role her own supposed observations of white ravens have played in forming that belief).

Point 2 is that a behavior which displeases many authors can still be pleasant or valuable to some authors. (Said knows this, I think.)

Just so. But, once again, we must make our analysis more specific and more precise in order for it to be useful. There are two points to make in response to this.

First is what I said above: the point is not just that the commenting style/approach in question is valuable to some authors (although even that, by itself, is surely important!), but that it turns out to be valuable specifically to the author who served as an—indeed, as the—example of said commenting style/approach being bad. This calls into question not just the thesis that said approach is bad in general, but also the weight of any purported evidence of the approach’s badness, which comes from the same source as the now-controverted claim that it was bad for that specific author.

Second is that not all authors are equal.

Suppose, for example, that dozens of well-respected and highly valued authors all turned out to condemn my commenting style and my contributions, while those who showed up to defend me were all cranks, trolls, and troublemakers. It would still be true, then, to say that “my comments are valuable to some authors but displease others”, but of course the views of the “some” would be, in any reasonable weighting, vastly and overwhelmingly outweighed by the views of the “others”.

But that, of course, is clearly not what’s happening. And the fact that Benquo is certainly not some crank or troll or troublemaker, but a justly respected and valued contributor, is therefore quite relevant.

Point 3 is that benquo’s view on even that specific comment is not the only author-view that matters; benquo eventually being like “this critical feedback was great” does not mean that other authors watching the interaction at the time did not feel “ugh, I sure don’t want to write a post and have to deal with comments like this one.” (Said knows this, I think.)

First, for clarity, let me note that we are not talking (and Benquo was not talking) about a single specific comment, but many comments—indeed, an entire approach to commenting and forum participation. But that is a detail.

It’s true that Benquo’s own views on the matter aren’t the only relevant ones. But they surely are the most relevant. (Indeed, it’s hard to see how one could claim otherwise.)

And as far as “audience reactions” (so to speak) go, it seems to me that what’s good for the goose is good for the gander. Indeed, some authors (or potential authors) reading the interaction might have had the reaction you describe. But others could have had the opposite reaction. (And, judging by the comments in that discussion thread—as well as many other comments over the years—others in fact did have the opposite reaction, when reading that discussion and numerous others in which I’ve taken part.) What’s more, it is even possible (and, I think, not at all implausible) that some authors read Benquo’s months-later comment and thought “you know, he’s right”.

(Notably, benquo once publicly stated that he suspected a rough interaction would likely have gone much better under Duncan moderation norms specifically; if we’re updating on benquo’s endorsements then it comes out to “both sets of norms useful,” presumably for different things.)

Well, as I said in the grandparent comment, updating on Benquo’s endorsement is exactly what I was not suggesting that we do. (Not that I am suggesting the opposite—not updating on his endorsement—either. I am only saying that this was not my intended meaning.)

Still, I don’t think that what you say about “both sets of norms useful” is implausible. (I do not, after all, take exception to all of your preferred norms—quite the contrary! Most of them are good. And an argument can be made that even the ones to which I object have their place. Such an argument would have to actually be made, and convincingly, for me to believe it—but that it could be made, seems to me not to be entirely out of the question.)

I’d say it casts mild doubt on the thesis, at best, and that the most likely resolution is that Ray ends up feeling something like “yeah, fair, this did not turn out to be the best example,” not “oh snap, you’re right, turns out it was all a house of cards.”

Well, as I’ve written, to the extent that the convincingness of an argument for some claim rests on examples (especially if it’s just one example), the purported example(s) turning out to be no such thing does, indeed, undermine the whole argument. (Especially—as I note above—insofar as that outcome also casts doubt on whatever process resulted in us believing that raven to have been white in the first place.)