Of course I have read the rules, I've reviewed them since you asked me to and I'm familiar with them. And I like the rules a lot and I respect you a lot as the moderator. But I'm genuinely not sure which of the rules I've violated, especially to the extent that it's something you usually ban for. I can try to construct a case for some thing or other, but they all seem a bit tortuous to me and I'm guessing you have a clearer answer.

I tried pasting the thread word for word into a few LLMs after anonymizing the conversation and trying to faithfully reconstruct the deleted comments. My prompt was just the chat then "what do you think of this exchange?"

I got several takes, including

- one from Gemma saying I (person B in the anonymized chat) was the voice of reason and trying to calm down a heated situation.

- an independent run from Gemma saying I was right in my response to person A (the user talking about Sarkozy) but that maybe my response was "too measured" and suggested a "stronger condemnation...would be more appropriate".

- one from ChatGPT 5 saying "Person B – Principled and fact-focused. They maintain the moral high ground, challenge the call for cruelty directly, and back up their concern about rising political violence with data. This is the adult in the room—ethical, rational, and de-escalating."

So based on the raw chat input, these LLMs are interpreting my responses somewhat opposite to you -- as a sort of anti-flame style.

In my original comment I was addressing their claim that others shouldn't act like they have the moral high ground just because they're advocating a violation human rights. I'm pointing out that if you advocate for violating human rights then you're putting yourself below the globally understood minimal ethical commitments. You can't both advocate for human rights violations and also say that other's don't get to be morally superior. The user explicitly acknowledged that they don't believe in going high when others go low. So they've deliberately occupied the low ground and I pointed out that if they do that they can't also condemn everyone else for being above them. I meant it as a literal statement: they logically have to pick one or they other.

Similarly with my other comment. I didn't understand what Trump or Sarkozy quotes had to do with the discussion the other user and I were having about the rising trend of people advocating for political violence on social media. I expressed that I didn't see any relevant connection, I asked why they would be relevant. I spent some time considering what good faith motivations there would be. I confessed that I couldn't think of any and it seemed like the only reason to want that would be to derail the conversation from talking about the trend of political violence toward the question of whether the anger and thirst for political violence is justified. I've also seen this pretty often on HN, as I'm sure you have. Their response made it clear in my view that their goal was indeed to make a pitch for not feeling sorry for these people, giving them a taste of their own medicine etc.

Anyway, I'm not disagreeing with you that I've violated any rules. And I don't think LLMs are the source of truth in any sort of moderation discussion. They were just the tool I had at hand today to get other points of view. But I did want to convey that as far as I'm aware I've been operating in good faith and that I've made an effort to understand what in this conversation is so problematic.

For example in a followup question I asked which participants would likely see moderation comments on HN and ChatGPT says of me "Safe. Civil tone, evidence-based, de-escalatory. Fits HN’s preference for reasoned, data-backed argument."

My comments are sometimes terse and blunt, but they're rarely snarky or sarcastic. Sometimes they're slightly tongue-in-cheek, but in a playful way not a sarcastic way. And at any rate the comments in this thread weren't tongue in cheek. I tried to stick say what I was thinking quite literally and assume good faith.