Hi, let me explain. I'm developing a new P2P system called Rucio. My idea with this software is basically to combine the best features of BitTorrent with those of the ed2k network, so that it feels like a natural evolution of the ed2k network.
Right now, the software is very stable and working very well. It functions both as a client on its own Rucio network (which I’ll describe later) and as a client on the ed2k network in KAD, since the goal has always been for it to be 100% decentralized (which is why it doesn’t support ed2k servers).
I've documented the network itself in detail here, but basically it's a network based on libp2p (a fairly standard library for P2P projects) that uses Kademlia DHT and a BLAKE3 bao-tree for the hashes of the files and chunks. This way, you have something similar to Torrent (in that it uses DHT) but adapted to a system where you share thousands of files. The search then uses the Gossipsub protocol, which is ideal for this situation.
If you start a node locally using the container, you can simply navigate to http://localhost:3003/api/docs, where you'll find all the documentation for its API (which is fully RESTful), which should be all you need to complete the integration.
Anyway, if you like the idea and need a hand with anything, just let me know and I'll be happy to help.
Let me know what you think.
Hi, let me explain. I'm developing a new P2P system called Rucio. My idea with this software is basically to combine the best features of BitTorrent with those of the ed2k network, so that it feels like a natural evolution of the ed2k network.
Right now, the software is very stable and working very well. It functions both as a client on its own Rucio network (which I’ll describe later) and as a client on the ed2k network in KAD, since the goal has always been for it to be 100% decentralized (which is why it doesn’t support ed2k servers).
I've documented the network itself in detail here, but basically it's a network based on libp2p (a fairly standard library for P2P projects) that uses Kademlia DHT and a BLAKE3 bao-tree for the hashes of the files and chunks. This way, you have something similar to Torrent (in that it uses DHT) but adapted to a system where you share thousands of files. The search then uses the Gossipsub protocol, which is ideal for this situation.
If you start a node locally using the container, you can simply navigate to
http://localhost:3003/api/docs, where you'll find all the documentation for its API (which is fully RESTful), which should be all you need to complete the integration.Anyway, if you like the idea and need a hand with anything, just let me know and I'll be happy to help.
Let me know what you think.