Which category is it if you decide based on multiple factors, ONE of which is PR? And why is this a bad thing, if that's what you believe?

You're flat-out wrong about #3. Encryption is just a mathematical algorithm, it doesn't care who uses it, only that you have the key.

In short, encryption is just a very complex function, so you feed Key + Message in, and you get an Output. f(K,M) = O

I already have access to Key and Message, so I can share both of those. The only thing you can possibly secure is f().

If you have a cryptographic program, like OTR, I can just decompile it and get f(), and then post a modified version that lets the user manually configure their key (I think this is actually trivial in OTR, but it's been years since I poked at it)

If it's a website where I login and it auto-decrypts things for me, then I can just send someone the URL and the key I use.

Point 2 seems to reply on Point 3, and as far as I'm aware the only formally mathematically provably deniable method WHEN THE KEY IS COMPROMISED is a one-time pad.

I'm not sure how much crypto experience you have, but "and no one else knows the key" is a foundation of every algorithm I have ever worked on, and I'm reasonably confident that it's a mathematical requirement. I simply cannot imagine how you could possibly write a crypto algorithm that is secure EVEN with a compromised key.

EDIT: If you still think I'm wrong, can you please give me a sense of your crypto experience? For reference: I've met with the people who wrote OTR and hang out in a number of crypto circles, but only do fairly basic stuff in my actual work. I do still have a hobby interest in it, and follow it, but the last time I did any serious code breaking was about a decade ago.

This seems to me like a deliberate misunderstanding. But taking it at face value, a story in which violence is committed against targets not analogous to any present-day identifiable people, or which is not committed for any reasons obviously analogous to present-day motives, is fine. The Sword of Good is not advocating for killing wizards who kill orcs, although Dolf does get his head cut off. Betrayed-spouse murder mysteries are not advocating killing adulterers - though it would be different if you named the victim after a specific celebrity and depicted the killer in a sympathetic light. As much as people who don't like this policy, might wish that it were impossible for anyone to tell the difference so that they could thereby argue against the policy, it's not actually very hard to tell the difference.

I am asking in advance if anyone has non-obvious consequences they want to point out or policy considerations they would like to raise. In other words, the form of this discussion is not 'Do you like this?' - you probably have a different cost function from people who are held responsible for how LW looks as a whole - but rather, 'Are there any predictable consequences we didn't think of that you would like to point out

Eliezer, at this point I think it's fair to ask: has anything anyone has said so far caused you to update? If not, why not?

I realize some of my replies to you in this thread have been rather harsh, so perhaps I should take this opportunity to clarify: I consider myself a big fan of yours. I think you're a brilliant guy, and I agree with you on just about everything regarding FAI, x-risk, SIAI's mission.... I think you're probably mankind's best bet if we want to successfully navigate the singularity. But at the same time, I also think you can demonstrate some remarkably poor judgement from time to time... hey, we're all running on corrupted hardware after all. It's the combination of these two facts that really bothers me.

I don't know of any way to say this that isn't going to come off sounding horribly condescending, so I'm just going to say it, and hope you evaluate it in the context of the fact that I'm a big fan of your work, and in the grand scheme of things, we're on the same side.

I think what's going on here is that your feelings have gotten hurt by various people misattributing various positions to you that you don't actually hold. That's totally understandable. But I think you're confusing the extent to which your feelings have been hurt with the extent to which actual harm has been done to SIAI's mission, and are overreacting as a result. I'm not a psychologist - this is just armchair speculation.... I'm just telling you how it looks from the outside.

Again, we're all running on corrupted hardware, so it's entirely natural for even the best amongst us to make these kinds of mistakes... I don't expect you to be an emotionless Straw Vulcan (and indeed, I wouldn't trust you if you were)... but your apparent unwillingness to update in response to other's input when it comes to certain emotionally charged issues is very troubling to me.

So to answer your question "Are there any predictable consequences we didn't think of that you would like to point out"... well I've pointed out many already, but the most concise, and most important predictable consequence of this policy which I believe you're failing to take into account, is this: IT LOOKS HORRIBLE... like, really really bad. Way worse than the things it's intended to combat.

Implement a full on OTR style system providing full deniability through crypto. Rather than stopping content from being copied, just make sure you can claim any copy is a forgery, and nobody can prove you wrong. A MAJOR engineering effort of course, but totally possible, and 100% effective.

I can't help but see two major flaws:

1) If I link to a major, encrypted offshoot of LessWrong, people will AUTOMATICALLY be suspicious and it will damage PR.

2) Why would it be any easier to cry "it's a forgery" in this situation versus me posting a screenshot of an unencrypted forum? o.o Especially given #1...

3) I can share my password / decryption key / etc..

Taboo "driving" and "informing" and explain the difference between those two to me?

Or we can save ourselves some time if this resolves your objection: Eliezer is saying that he is adding the OPTION to censor things if they are a PR problem OR because the person is needlessly incriminating themselves. I'm not sure how that's a bad OPTION to have, given that he's explicitly stated he will not mindlessly enforce it, and in fact has currently enforced it zero (0) times to my knowledge (the post that prompted this was voluntarily withdrawn by it's author)

"it might make us look bad" is a horrible argument

You can argue that LessWrong shouldn't care about PR, or that censorship is going to be bad PR, or that censorship is unnecessary, but you can't argue that PR is a fundamentally horrible idea without some very strong evidence (which you did not provide).

-

It's almost tautological that if a group cares about PR, it HAS to care about what makes them look bad:

If Obama went on record saying that we should kill everyone on Less Wrong, and made it clear he was serious, I'd hope to high hell that there would be an impeachment trial.

If Greenpeace said we should kill all the oil CEOs, people would consider them insane terrorists.

If the oil CEOs suggested that there might be... incentives... should Greenpeace members be killed...

Well, are you?

No. To prove this, I shall shortly delete the post advocating it.

True, but you have said things that seem to imply it. Seriously, you can't go around saying "X" and "X->Y" and then object when people start attributing position "Y" to you.

Point one: We never said X->Y. We said X, and a bunch of people too stupid to understand the fallacy of appeal to consequences said 'X->violence, look what those bad people advocate' as an attempted counterargument. Since no actual good can possibly come of discussing this on any set of assumptions, it would be nice to have the counter-counterargument, "Unlike this bad person here, we have a policy of deleting posts which claim Q->specific-violence even if the post claims not to believe in Q because the identifiable target would have a reasonable complaint of being threatened".

This must be why the media companies haven't given up on DRM yet. They think if they can just unmask and arrest the ringleaders of the "organized conspiracy out there" then copy protection will start working, when in reality any random person can become a "conspiracy" member with nothing more than a little technical knowledge, a little free time, and a moral code that encourages copying.

To be fair, the "vetting" and "full deniability" options don't really apply to the ??AA. The best pre-existing example for those kinds of policies might be the Freemasons or the Mormons? In neither case would I be confident that the bad PR they've avoided by hiding embarrassing things hasn't been worse than the bad PR they've abetted by obviously dissembling and/or by increasing the suspicion that they're hiding even worse things.

I wouldn't have posted the following except that I share Esar's concerns about representativeness:

I think this is a good idea. I think using the word "censorship" primes a large segment of the LW population in an unproductive direction. I think various people are interpreting "may be deleted" to mean "must be deleted." I think various people are blithely ignoring this part of the OP (emphasis added):

In other words, the form of this discussion is not 'Do you like this?' - you probably have a different cost function from people who are held responsible for how LW looks as a whole

In particular, I think people are underestimating how important it is for LW not to look too bad, and also underestimating how bad LW could be made to look by discussions of the type under consideration.

Finally, I strongly agree that

anyone talking about a proposed crime on the Internet fails forever as a criminal[.]