There's nothing inherently wrong with an opt-in coordination mechanism. It makes sense that if I trust someone to make a blacklist, I can have my node refer to that list. The problem is when an organization threatens you for "publishing" the wrong thing. That's a legal or social problem, not a technical one, and such organizations typically don't care in the slightest about how convenient it is for you to comply with their demands. If there's no blacklist mechanism, they're more than happy to demand that you shut down your node altogether.
I think of the two evils, the possobility of censorship or mistakes being eternal, I think I prefer the former.
For the same reason I find it comforting that an old fashioned bank transfer can be corrected if I transfer money and make a mistake writing the account number. Mutable history is a powerful feature.
> I think of the two evils, the possobility [sic] of censorship or mistakes being eternal, I think I prefer the former.
I get that. However, it's arguable that mistakes are all too often eternal against the most dangerous adversaries, such as authoritarian governments and other criminal organizations with state-level resources. And so it's arguably better to focus on resistance to censorship.
That’s technically correct but in practice it’s about as accurate as, say, assuming that you shouldn’t own anything expensive because it’s possible for anything to be stolen.
In real life, there are not billions of Mallorys watching your stuff constantly. Most people are decent and most of the others are deterred by laws, and the number of people who are willing to help abusers is relatively small.
Just using some real-life examples, think about doxxing or revenge porn. It’s technically true that this data cannot provably be removed from the internet but in practice most people didn’t save it and the ones who did became a lot more covert once the legal system caught up, which means that in practice far fewer people see it. The initial damage may have been done but that doesn’t mean we should give up and do nothing because there isn’t a theoretically-perfect option.
Sure. But if I e.g. accidentally uploaded something sensitive to GitHub (that can’t simply be changed to a new secret), I’d certainly delete it in a hurry, rather than shrug and say ”oh well It’s on the someone has already copied it so I’ll leave it”.
But in that case, are you saying you _wouldn't_ immediately change the credential you committed? Sure, the possibility of an adversary forking your repo after that commit but before your revision is small, but still exists.
Once a secret is exposed to the internet, it should be considered public and rotated. In this case mutability/immutability is moot though likely there are applications for other, non-credential secrets that are not so easily rotated (like your home address or something).
> an old fashioned bank transfer can be corrected if I transfer money and make a mistake writing the account number. Mutable history is a powerful feature.
That's not necessarily mutable history though. Such a correction will usually be made by an inverse transaction, not by wiping the original transaction from the record.
Humans. That’s the feature. That agreements between humans (contracts, transfers, ...) are often imprecise. Matters can be argued between humans (in companies, authorities, courts)
But in this case there is a conflict between the agreement between IA and the content-submitter, and the agreement between IA and some political power.
That's a funny take! At least the first two look like they need time machines, which would circumvent the solutions for the main issue in question here.
Downvote because I think the nationalistic "us vs them" is uncalled for. Let's not make it about country politics when the topic isn't already politics.
