It's rare for us to do anything like this. My guess is that as long as we tell people that we did it (and why; and as long as it was for a good reason), it's not something most of the community would object to. I don't think we need to build new bureaucratic mechanisms.

In this case, I could avoided it by posting a different comment with the same link and pinned that to the top instead, but then ChrisArchitect wouldn't be getting the karma for their contribution. That seemed unfair. Hence the above.

I do sometimes think about building an edit-and-diff mechanism for HN comments though! People could suggest edits to comments and they could be treated like pull requests. I can't tell if this is a great idea that would massively improve the threads, or an idiocy. (Edit: I was looking for a comment where I mentioned this not long ago: https://news.ycombinator.com/item?id=25459870, and found another one where I mentioned it 7 years ago: https://news.ycombinator.com/item?id=8214456. That's nuts!)

Yes, but as a matter of policy, all English language descriptions of what you did will have some wiggle room for interpretation. For something as serious as changing another person's words, you should have a special code path, like text in a special color that says "edited by moderator." Seeing that someone's post was edited shouldn't require expanding a reply and interpreting it, since editing someone's speech is far more delicate and prone to catastrophic outcomes than merely suppressing it.

I hear you, but the problem is that such systems also add cost, not just benefit. I don't only mean dev and maintenance costs (although those too), but the cost to the culture of making this place more bureaucratic. We've always avoided that. HN is a spirit-of-the-law place, not a letter-of-the-law one: https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que....

Edit: oh one other point: there will always be "wiggle room for interpretation". The idea of eliminating it is a fantasy, and trying to make systems to eliminate it is a wild goose chase.

I think some distinguishing attribute should be displayed to show that a comment was edited by someone other than the user. It's apparent to me that comments can be edited. How can I be certain that mods haven't silently edited a comment in the past?

This is the sort of thing I think would be a mistake. I posted a clear explanation of what I did and why. I believe that's enough to satisfy the community. It does leave a tiny minority unsatisfied who don't trust what we say, but they are not satisfiable. If we added the feature you're suggesting, how could you be certain that we weren't secretly overriding it sometimes? You would either have to take our word for it—in which case why not just believe us in the first place?—or you would need some other system to validate that one. This is how you get systems upon systems, with no actual solution, for a problem that doesn't even really exist. The same issue would recreate itself at every step.

Don't get me wrong - I'm not saying "just trust us". Trust is earned. But the way to earn it is to always be open to questions, answer them as well as we can, and when people point out mistakes, to admit them and fix them quickly. That's what we try to do, and it's informal and social. Formal/technical methods are neither necessary nor sufficient for community trust, and it would be a big mistake to think otherwise, because it would steal resources away from other things, like human-to-human communication and making the site itself better.

I suppose you are correct. I do trust you - you have earned that trust, and I don't think you have edited any comments yet.

More importantly - if you don’t trust him nothing short of pgp signatures or some form of a blockchain would allow you to be certain - as even if there was an obvious edit function it could be bypassed.

You’re thinking about the culture of your site, I am talking about the rights of your users. Your users create the value and deserve it (honestly it should be protected by law). If you edit the link and then someone takes over the domain and makes it a porn site, that can cause direct harm to your users. You need to accept the cost as part of the cost of doing business, and if you really edit posts very rarely than the bureaucratic impact should be minimal.

This strikes me as an archetypal example of the legalistic sort of thinking that is popular with a vocal minority of internet forum users who like to make litigious arguments, look for problems that don't actually exist in practice, then demand bureaucratic systems to "solve" the "problems". I don't believe that we're violating human rights by operating HN, and the idea of building software to address such cases as "someone taking over the domain and making it a porn site" is an example of what I mean by problems that don't exist in practice.

I don't mean to pick on you personally—it's a style of thinking that I've encountered on a regular basis. Being a moderator, and thus a sort of bureaucrat, attracts arguments from the bureaucratically minded. All I can tell you is that it's the opposite of how we think about HN. In my view it would be a big mistake to let it direct how we operate HN or what features to add. The way I look at it, HN users have a right to an interesting forum that doesn't suck (or at least relatively doesn't suck), that's a hard problem, and anything that detracts focus from it is a really bad idea. I get that you think differently and I don't mean to be disrespectful. But yes, I am thinking about the culture of HN.

I don't think what you did was awful, especially since you went to the effort of explaining exactly what was changed and why.

However, I also think there should be space for discourse about the mechanics of the action; however academic that discussion might end up being.

Personally, I think a suggestion to the OP encouraging them to edit the post themself might have been a less controversial method of achieving the same aim.

I disagree, as a user I think what dang did makes sense and was clearly explained to anyone viewing. Whether something does or doesn't have special treatment in the software is mostly orthogonal to the actual material impact of what happened, which was clearly minimal here.