Agreed. I was trying to point out how refusing to be friendly, even from a cynical point of view, is counterproductive. 

First: no, it absolutely is a “major weakness in people” that they prefer to avoid engaging with relevant criticism merely on the basis of the “tone”, “valence”, etc., of the critics’ words. It is, in fact, a huge weakness. Overcoming this particular bias is one of the single biggest personal advances in epistemic rationality that one can make.

Or put another way, "Your strength as a rationalist is the extent to which it takes more Charisma to persuade you of false things, and less Charisma to persuade you of true things"
 

I do think many people could be served by trying to find the truth in harsh criticisms, to wonder if part of the sting is the recognition the critic was right. You're example of ArsDigita was quite helpful in getting a concrete demonstration of the value of that kind of critique. 

The thing is, Greenspun failed.

People are not empty-machines of perfect reasoning. There's an elephant in our brains. If the critique is to land, if it is to change someone's mind or behavior, it has to get through to the elephant. 

Second: you imply a false dichotomy between the “improv session” sort of faux-criticism I describe, and “exuding disgust and contempt”. Those are not the only options! It is entirely possible to criticize someone’s ideas, very harshly, while exhibiting (and experiencing) no significant emotionally-valenced judgment of the person themselves.

Indeed. It is also possible (I claim) to give pointed criticism while remaining friendly. The elephant doesn't like it when words look like they come from an enemy. If you fail to factor in the elephant, and your critique doesn't land, that is your own mistake. Just as they have failed to see the value of the critique, you have failed to see the weight of the elephant. 

The executives and other board members of ArsDigita failed, but if Greenspun could have kept their ear by being friendlier, and thereby increased the chances of changing their minds or behavior, Greenspun also failed at rationality. 

If it is rational to seek the truth of criticism even when it hurts, then it is also rational to deliver your criticism in a friendly way that will actually land. Or put another way, your strength as a rationalist is the extent to which it takes less Wisdom to notice your plans will fail.

So you want a culture of competing with each other while pushing each other up, instead of competing with each other while pushing each other down. Is that a fair (high-level, abstract) summary?

I see the disagreement react, so now I'm thinking maybe LessWrong is trying to be a place where both competitive and collaborative dynamics can coexist, and giving authors the ability to ban users from commenting is part of what makes the collaborators space possible?

Yes, shockingly, people have preferences about how people interact with them that go beyond obvious unambigious norm violations, what a shocker!

This seems to be just another way to describe what I wrote in the grandparent, except that your description has the connotation of something fine and reasonable and unproblematic, whereas mine obviously does not.

This seems to me to be the crux of the issue. 

There's a thing that happens in sports and related disciplines wherein the club separates into two different sections, where there is a competition team and there's everybody else trying to do the sport and have a good time. There are very sharp differences in mindset between the teams.

In the competition team every little weakness or mistake is brutally hammered out of you, and the people on the team like this. It's making them stronger and better, they signed up for it. But if a beginner tried to join them, the beginner would just get crushed. They wouldn't get better, and they would probably leave and say their competitive-minded teammates are being jerks.

Without any beginners though, there is no competition team. The competitors all used to be beginners, and would have gotten crushed in the hyperbaric training chamber of their current team culture. 

I think you are trying to push for a competition team, and Habryka is not. 

Competition teams are cool! I really like them in their time and place. I think the AI Alignment forum is a little bit like this with their invite-only setup (which is a notable feature of many competition teams).

You need the beginner space though. A place where little babbling half-formed sprouting ideas can grow without being immediately stomped down for being insufficiently rigorous. 

Another angle on the same phenomenon: If you notice someone has a faulty foundation in their house of understanding they are building, there are two fundamentally different approaches one could take. You could either:

1. Be a Fellow Builder, where you point out the mistake in a friendly way (trying not to offend, because you want more houses of understanding built)
2. Be a Rival Builder, where you crush the house, thereby demonstrating the faulty foundation decisively. (where you only want the best possible houses to even be built at all, so whether that other builder comes back is irrelevant)

I think Habryka is building LessWrong for Fellows, not Rivals.

From the New User's Guide:

Is LessWrong for you?

LessWrong is a good place for someone who:

...

• wants to work collaboratively with others to figure out what's true

My impression is that you want LessWrong to be a place of competitive truth-seeking, and Habryka is guiding LessWrong towards collaborative truth-seeking. 

I think it's fine to want a space with competitive dynamics. That's just not what LessWrong is trying to be.

There are so many critical posts just here on LessWrong that I feel like we are living in different worlds. The second most upvoted post on the entire site is a critique, and there's dozens more about everything from AI alignment to discussion norms.

If the critique “lives in the larger LW archipelago”, then:

1. It won’t be highly upvoted, because…
2. Almost nobody will read it; and therefore…
3. It won’t be posted in the first place.

Both regimes share the property wherein someone can disagree and write a lengthy critique as a top-level post. This empirically does happen, and they are sometimes highly upvoted and widely read. Hence existence proof. The regimes are not different in this regard. 

Doesn't the existence proof of long, well-written and highly upvoted critiques disprove your point?

There's plenty of comments that are critiques, and the author of the post doesn't ban them because the critique wasn't cruel. Even if an author starts constantly banning anyone who disagrees with them, they'll get a reputation pretty fast. 

It feels to me like you are vigorously defending against a failure mode that is already handled by existing reputational effects like karma, and people being free to write their own post in response. 

Agreed on wanting LW to be better than the rest of the Internet. 

My model is something like: The site dies without writers -> writers only write if they enjoy writing on the platform -> writers don't enjoy writing without the ability to ban commenters that they find personally aversive -> Give authors the ability to ban commenters on their posts.

I'm cognizant of the failure where good critiques get banned. Empirically, however, I don't think that's a problem here on LW. Long, well-written critiques are some of the most upvoted posts on the site. I think it's fine if the critique lives in the larger LW archipelago, and not in the island of one author. The rigor lives in the broader site, not just in an individual posts comment section. 

So like, do you distrust writers using substack? Because substack writers can just ban people from commenting. Or more concretely, do you distrust Scott to garden his own space on ACX?

Giving authors the ability to ban people they don't want commenting is so common that it feels like a Chesterton's Fence to me.