Is it possible to build classical hidden services?

I wonder if Tribler can forward any normal UDP packets to a specified destination anonymously, just like the classical hidden service or the Oneswarm service-sharing?
If possible (it seems likely because downloading also uses a socks proxy), will it be implemented in the official Tribler release at any time?
I think it can be a faster and more efficient replacement of Tor hidden services when I need less anonymity.

Well, the functionality needed for that is more or less there. When you do a fully tunneled download, it will look for rendez-vous points on the DHT by using a modified version of the torrent’s infohash. So something could be build to use that to tunnel other sorts of UDP traffic through the tunnel community.

We can’t invest time on that right now, but we might someday find a master student that wants to this as a final project or maybe someone from the community could jump in and give it a shot (we would love to have more people contributing to the project)

We are working on automated bandwidth investment now, so peers acting as hops in the tunnel community get digitally signed certs for the bandwidth they provided so they can spend the credits to get privileged bandwidth priority for their traffic in the future. It will take a while to have this implemented but we already have the basics of the automated torrent picking and seeding part running.

Interesting and exciting :slight_smile: I am super excited to see that you are working on incentives that could speed up anonymous sharing!

If you have time could explain in more detail how this would work?

Well, we have 3 parts there:

  • The investment part, which picks up the most potentially profitable torrent swarms from an arbitrary set of sources such as Tribler channels, RSS feeds, etc. (we are planning on a full-auto popular channel discovery too) This way you can choose which kind of torrents you want to invest too (depending on what’s legal on your country, if you want to boost some particular channels you like, etc…). This will then select the most rare pieces on that swarm and seed them for as long as possible. It will also have a persistent piece pool, so if for any reason a swarm becomes profitable again after a while, you won’t have to download already downloaded pieces.

  • Bandwidth “certificates”: Every X megs uploaded to a given peer, you will get a double-signed certificate crediting you for the amount of data you have provided to that specific node. Which can be used to “give” to other peers to get first on the line for the same amount of traffic. This way you will be able to leave your machine acting as a tunnel hop or seeding torrents when you don’t need the bandwith and then use the collected certificates to get a better service when you actually want something to be transferred as fast as possible.

  • Certificate propagation: There will be a community that will take care of propagating the certificates so people can check if the certs they are receiving match, etc.

The first part is mostly there, the second we have a mostly working prototype.

What prevents you from counterfeiting certificates?

What prevents the spending of certs from deanonymizing the downloading of initial issuer? Doesn’t the fact that I have a cert from Bob indicate that I was part of the anonymous route Bob took to download content? If I then give that cert to someone else, what prevents them from deanonymizing Bob?

I don’t know how this part will work, sorry.

For the first prototypes I guess we will probably use separate keys for your identity and for the bandwith certs, but something more advanced will be needed to check how much you trust a given cert (otherwise it would be trivial to simply generate the certs yourself with random keys).

Maybe @root has a better idea of the plan.

Correct, an incentive system leaks information.
Having 1 bad hop in between your path should be OK, if you set the path length sufficiently long.

Tor is always the better option if you have the time and expertise to use it correctly. We aim for HD streaming in the future, using our incentives. Different use-case. A detailed scientific answer is here: “Paying the Guard: an Entry-Guard-based Payment System for Tor”. http://fc15.ifca.ai/preproceedings/paper_112.pdf -j

I do. Look at the Beta … There testing a Emercoin Plugin.
I looked over your note and Code its very clear you going to setup your own Blockchain or piggy back on anther…
Oh some one could hack it but at that point there mining and it be cheaper to share it. And Tracing? Nope all you have is the amount of coin a Random Placebo IP Generated… All very Elegant. Makes me wish I kept up with my coding.

Coin for bandwidth We will have Bandwidth coin Miners… Very Nice considering are current net is so DNS dependent and vulnerable.