Hello Bungee!It depends on how you look at it from what I can see. I think something like this is a good idea... for regular people. I mean this form of a social network would appear to offer a lot to some people interested in privacy, but we want something much stronger than privacy, we want complete anonymity, those are two different, but related animals.Issues (partly to do with their idea, partly just in general with these things): - You're expected to connect via clearnet, thus your real IP address goes somewhere. So ok, you might not, but that's the default assumption.- You have to trust that the admins aren't simply lying through their teeth. In practice few people can analyze whether such a things is working as advertised.- Anonymous payment isn't trivial for most people, even with bitcoins. Bluntly, people are lazy and don't employ 2 Factor B$ anonymity, even on SR, let alone on the wider net. They want a 1 click solution. I don't know what can be done about that, other than upgrading cryptocurrency to superior anonymity implementations (which is ongoing, eventually it'll get there) or plain old natural selection.- let's say everything else is fine. Admins can't access your data. Everything is encrypted, secure, locked up tight. But there is still traffic analysis. With a persistent identity, this can become a problem, which is why many people continually swap alts.In all honesty, all solutions like this tend to be of the Hushmail variety. That is: You are super duper secure, unless you piss off somebody powerful. Actually, the people who run Hushmail were lucky not to get killed, from what I've read I'm not sure they comprehended that was even a possibility, which I find slightly scary. The DEA just snapped up small fry, if they bumped into somebody serious, their office building would have been converted into a morgue. But anybody serious was highly unlikely to use Hushmail, so there's a sliver lining. In many ways, many LE agents have never come across a real adversary in their entire careers, but I'm getting off topic.Point is, if you take anonymity seriously, then coercion cannot be possible from without or within. That is very important. People using your system must not be able to manipulate you. People outside your system must not be able to manipulate you. This is a tall order for any centralized entity.In practice, only decentralized applications will be capable of fulfilling those needs. It doesn't matter how smart you are, or how clever your system designers or programmers are. A central point is a point of critical failure because of rubberhose cryptanalysis (nice way of saying the government or some other entity beating the crap out of you until you give them what they want).Unless your system can resist an enemy capable of murdering everybody within informational reach, it is not genuinely secure, not really anonymous. Otherwise you are merely depending on somebody else's morality, when you look at it like that, it's a dubious proposition.Building decentralized systems that work is much harder than building centralized ones, where all elements are under your control and everything's nice and tidy. Because a decentralized system is closer to an eco-system than a normal piece of software, it evolves and adapts on its own, or at least it should if it's working properly.Bitcoin, BitTorrent and Tor are good examples of such systems. If you are to build a anonymous social network, you'll have to going along those lines. It's tough, but obviously it is possible. The key thing is to build in the assumption right from the beginning that if the developers got hit by cars, that the network would continue and others will take up the baton.For example, if the Tor project people were all kidnapped by UFOs tomorrow, then the Tor Project would continue. The software is readily available, anybody can download and set up their own personal Tor network, there are actually a few versions of Tor outside of the Tor Project software, which have either been built from the start, or branched off.The Tor Project main guy has implemented a RFC system similar to how the Request for Comments implementation of TCP/IP worked online. This way anybody can follow in his lead and build on what went before him. So that is smart, it's a kind of 'meta' backup if you like. You don't need everybody to grasp how it works, this way you just need one or two computer geeks with enough knowhow to rebuild the project if/when disaster strikes (and the Tor Project has > 3000 people involved indirectly, so you bet your ass there's enough of the right people out there!)Slightly Distracted Thought: I don't know why there isn't a copy of Tor built into most copies of Linux. It's fairly lightweight, I think it should be a default app like the text editor or calculator.Btw: most of what they have to say on piddler is good, but their advice for setting up a passphrase is a bit silly. Every password attacking software knows to interchange common letter substitutions. Because, you know, if it's any kind of standard alteration many people will think of, then it makes its way into password crackers because hackers simply aren't completely retarded.tldr; my advice to you, if you're serious about building such an anonymous network of any kind, is to study existing decentralized networks first as case studies.