Neat use of web torrent. Now if only we could do it with live data instead of persistent data, have built in encryption, and anonymize the peers we'd have the perfect decentralized internet :-)
Where is the anonymity? It is unreasonable to use TOR in my opinion because despite being only a few exit nodes, it's not truly P2P then (and therefore can be slow). Tribler tried something similar. Someone needs to build an onion router of peers without defined relays or exit nodes (think of it as everyone-is-a-relay TOR, it's one of my future ideas[0]). This will speed things up substantially and can definitely work when you don't have the limitations of TOR like working w/ all TCP connections (e.g. HTTP). Maybe they could make every ZeroNet node a hidden service, I dunno.
Also, I have researched a lot about plausible deniability of publicly available yet distributed data. I have never seen a system that stores public data without them (clarity edit: node runners) being able to find out whether they are storing a piece of something they don't like. I started a thread about it on the Maidsafe forum recently [1]
From what you've described, it seems like i2p would satisfy the requirements of your first paragraph and Freenet would satisfy the requirements of the second. Both of these networks have been around for over a decade, so they're not as in vogue as newer projects.
I have only slightly looked at i2p, but sadly appears the only impl requires a JVM. I have researched freenet and it does not appear to provide plausible deniability for public data (sorry on mobile and can't link easily).
It does provide plausible deniability, since you have no control over what files go to your node (files are spread as people request them) and they are encrypted, hence you can't easily know what they contain.
Both [0] and [1] seem to suggest that if you make the data "public" then someone can know they have a piece.
"It is hard, but not impossible, to determine which files that are stored in your local Freenet Datastore"
"Of course, the decryption keys, which are contained in links to the files, may be publically posted on some other site - they have to be if the site creator wants people to visit their site. But if you've never had knowledge of that link, which is very plausible if there are thousands of Freenet sites, you can't be expected to know what is contained in the encrypted files in your Freenet node."
If you feel brave, you could try the C++ implementation of the i2p router, purplei2p[1] (aka i2pd[2]). Last time I tried, there were a few rough edges, but it is now over two years old, so it has probably improved (or you may even improve it yourself!).
Feross is awesome, I've met him several times and WebTorrent is driving the future forward. I have a complementary Open Source project ( https://github.com/amark/gun ) for live data sync - it is already decentralized and anonymous by default (other than a session-ID).
Adding encryption is pretty easy, now with WebCrypto! The future is looking exciting, between WebTorrent, IPFS, and other projects!
None in particular since you still have to connect to a traditional HTTP or WebSocket (although WebRTC is coming soon) servers. So if you connect to a malicious peer they can get your IP can get leaked. But the messaging algorithm is very ad-hoc mesh-networky and UDP-ish. At its core, messages only contain a message ID and a body, these are then daisy chained throughout the servers and clients (they're all peers). Meaning that just because a peer sent you something does not mean they are the originator. However, nothing stops peers from broadcasting their session ID or IP or X-Forwarded-For header. But the counter is also true, the messaging system still works even without that information - which if there are enough peers preserves anonymity. I of course should say the usual disclaimers that small peer groups can be attacked, and larger networks if you have enough intelligently placed peers you can probably calculate triangulation and stuff like that. When you get down to the actual "physics" of stuff, there are lots of tricks/hacks to break any network based on timing and patterns alone.
