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.
You seem to be using a very narrow definition of "crypto".. I'm not sure whether you're just being pedantic about definitions, in which case you may be correct, or if you're actually disputing the substance of what I'm saying. To answer your question, I'm not a cryptographer, but I have a CS degree and am quite capable of reading and understanding crypto papers (though not of retaining the knowledge for long)... it's been several years since I read the relevant papers, so I might be getting some of the details wrong in how I'm explaining it, but ...
New proposed censorship policy:
Any post or comment which advocates or 'asks about' violence against sufficiently identifiable real people or groups (as opposed to aliens or hypothetical people on trolley tracks) may be deleted, along with replies that also contain the info necessary to visualize violence against real people.
Reason: Talking about such violence makes that violence more probable, and makes LW look bad; and numerous message boards across the Earth censor discussion of various subtypes of proposed criminal activity without anything bad happening to them.
More generally: Posts or comments advocating or 'asking about' violation of laws that are actually enforced against middle-class people (e.g., kidnapping, not anti-marijuana laws) may at the admins' option be censored on the grounds that it makes LW look bad and that anyone talking about a proposed crime on the Internet fails forever as a criminal (i.e., even if a proposed conspiratorial crime were in fact good, there would still be net negative expected utility from talking about it on the Internet; if it's a bad idea, promoting it conceptually by discussing it is also a bad idea; therefore and in full generality this is a low-value form of discussion).
This is not a poll, but 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, and possibly bet on with us if there's a good way to settle the bet?'
Yes, a post of this type was just recently made. I will not link to it, since this censorship policy implies that it will shortly be deleted, and reproducing the info necessary to say who was hypothetically targeted and why would be against the policy.