One idea I've had is a content-type-restricted network that permits only text. That would allow utterly un-censorable communications: chat, planning revolutions, whatever, but wouldn't be useful for CP. (Unless you like ASCII-art CP.)
It could support ANSI. That would be neato. It would feel like the old BBS world. Wonder if anyone would still care if a name like ViSiON-X were stolen for it. :)
Maybe binary content propagates differently. Text that meets certain criteria is replicated indiscriminately, but binary content is only replicated when a user votes on it.
Edit: you could apply game theory to this problem. Model the network as a graph and write an agent-based modeling rule set for... say... CP-propagators and non-CP-propagators. Run iterative simulations of different propagation rule sets and weightings/parameters. Now introduce bad actors in the form of, say, government agents trying to suppress political discourse. The difference is that average-joe will cooperate in pushing out CP but will "defect" in a game with the other kind of bad-actor. You're looking for rule-sets and parameters where the CP gets pushed to the margins of the network or excluded but where the other kind of bad-actor is also excluded.
How about 256 words (or sets of words) representing each byte value? It's less dense by a factor of ~5, but it would work, be easy to decode, and be very difficult to identify, especially if you used sets of words. You could even cleverly generate in a way that is grammatically correct.
You're playing cat and mouse then, and you'll never win. If base64 was banned, I'd base26 encode it (i.e. letters of the alphabet), then use the NATO phonetic alphabet "alpha bravo charlie". The message size would be massive, but the message would get through.
Limitations of message size won't solve that because the messages will be broken into hundreds of pieces, just like they are on Usenet. In theory you could limit the total amount of content coming from an end point, but it's probably not feasible in an anonymous system.
How about a reddit-like ranking system as a portal do the darknet? It would easily categorize content in niches that would be curated socially. It could even help further ostracize the CP crowd to a marginalized role even in the darknet. It would enable anyone to flag out the abhorring niches and focus on the actual benefit of enjoying free exchange of information.
It doesn't get rid of them, but that's something that we wouldn't really be able to do (they all existed prior to the internet), but it would make the darknet usable.
If we are discussing potential solutions, maybe friend of a friend-network could somehow help thwart this issue. Even if CP etc was traded in the hypothetical foaf-network, it hopefully would remain in an isolated island of sorts. Anonymizing foaf-network could be bit of a challenge though.
Seems to me the only possible solution is a network of trust. Start small and build outward over time. This was the approach Facebook took to solve the "real identity" problem on the Internet at scale.
Yeah. You'd have to be more clever than that. There's a lot of interesting work on meaning extraction and text analysis. I wonder if you could have some kind of information density threshold.
It would be possible to encrypt text in such a way that things could be said about its information density but not about its meaning, too, though that would permit steganography to be used. But it would raise the technical bar for using the system for this purpose so high that it would probably drive away all the chickenboners.
There's also a dumb way: length limits. That would force binary data to be divided up into a huge number of posts, making it an annoying medium for file trading. Plotting the Iranian revolution would not require >1mb posts to a forum.
That's true. But, it's also completely possible to make a machine learning algorithms to separate "real text" (text meant for human to human communication) from text encoded images since they differ in significant ways. Granted, you could always try to make a text encoding format which resembles real text, but I'm fairly sure the machine learning algorithm could be constructed in such a way to make its usage unfeasible.
Yes, which is why you keep humans in the loop: you design a system in which the will of humans who want the system to be used only for text prevails and which is resistant to attack by humans with contradictory agendas.