Greetings, all! Following the seminal work on MUTE and ANts, both clients for anonymous peer-to-peer filesharing, I was intrigued by the idea, but both clients lacked functionality that the other had but that they both should of had (end-to-end encryption, point-to-point encryption, etc.).
Here's some links for these projects... MUTE http://mute-net.sourceforge.net/ ANts http://www.myjavaserver.com/~gwren/home.jsp?page=custom&xmlName=ants So, I decided that I wanted to do something similar, but a bit different, and certainly more robust. Here's the project summary; To develop a single, modular daemon which will establish connections with other anonymous peers, and provide the following features 1) Goal #1 is to provide a way for peers to connect, and relay information, without each peer knowing who specifically sent a packet, or what that packet might contain (this method has already been established in MUTE and ANts). 2) AES "Rijndael" encryption in the standard public key/private key exchange method, with digital signatures, in both an end-to-end method (from sending peer to recipient peer), and a point-to-point method (encrypting the packets sent from one relay peer to another). 3) Routing which is governed by self-organization of the network, without compromising the identity (or timing values) of the peers on the network. 4) Filesharing, including multiple, deep-propagation searches, and encrypted transfers. 5) An HTTP-type browser and daemon, which uses a client ID (but not his IP) for delivery or request of files. Instead of "www.puryear.com" you would have a query packet specifically addressed to user ID "puryear" and the page filename as the search keyword in the packet. This way, you could also do web-spider-less searches across the network. 6) "Endlessly Cyclingl E-mail", which is to say, propagation of e-mail packets throughout the network, with termination of propagation at the recipient (this will only work with clients who stay attached to the network all the time). 7) Basic Instant Messaging, (*Not detailed as of yet*) I am going to be testing this protocol this coming weekend. I am interested in hearing your thoughts and ideas, and definitely welcome anyone who wants to help me test it. David Jackson
