If Walled Garden communities upholding strong ideals are great, then what's going on here? This sounds like someone who is trying, deeply, to subvert a walled garden. The community they're in has decided to have some type of bias, and instead of rejecting that community (because it had, perhaps inevitably, grown to encompass so much of life), they seek to undermine the rule.

I don't know they're wrong to do this, because being factually wrong about things really can lead to arbitrarily huge paradoxes. But having feelings about things is how we remember their general merit. If you have a prejudice about "communism", that's probably healthy – even if you don't know why, it seems to routinely lead to excruciating suffering, starvation, deprivation and lack of dignity. Is it even important to know why? You could be completely factually wrong, and your feelings would still be doing you massive favors. So trying to correct facts in a way that violates social norms really is an attack on the foundation of any reasonably efficient society. Unless everyone is an excellent scholar, they won't be able to handle it the way that post implies people ought to.

But I'm more curious as to how we notice mistakes in our walled gardens, and how we try to fix them. Persuasion, obviously – but people being persuaded by things isn't always a perfect reflection of what's good and bad. If it was, we wouldn't ever have any disagreements, except from wholly original thinkers.

I'm noticing that the post is concerned with advocating a principle – in this case, scholastic honesty, even when it is unpopular (and it always has been). Maybe we need to abandon particulars more often, and defend principles in the abstract?

Of course, this is essentially the opposite of what I present here. At the end of the day, it might be cowardly to defend a principle outside of the specific instance you need it. The person posting on Less Wrong is likely concerned with some strange consensus over nonsense they found (very likely in left-leaning politics, if the footnotes are to be believed). They should defend an object-level disagreement, as well as the principle that the object-level disagreement they make is important. It's an example of when the current rules of the Walled Garden are wrong. And I think we need more discussions like that. We need to prevent making things an orthodoxy when they're wrong, and I see plenty of that happening (perhaps it is my own strange outlook, but the serious politicians not getting challenged on claims like "all of humanity will end in two generations if we don't do something about climate change" is one of the more absurd examples – no experts are coming even remotely close to making that claim).