I had an idea to block the RIAA spoofed files from the network.

The RIAA hold these files from a private network with high bandwith... They have a lot of ultrapeer computers, and a lot of broken files. That makes their presence on the network considerable. And the effects of it are also considerable.

What can we do about it?
certainly add their computer to a ban list in the gnutella client.
but their is a problem... we just hide behind our blankets, and the RIAA still uses the network's resources (slow down the queries, using bandwith, etc.)

What we can do is having a shared ban list. But there are huge problems with it. First: it has to be hosted on a server (we *could* live with that - for a certain time. Second: it would need to be modifiable (clients could add unwanted hosts to the list). There's the huge problem with that: it would give all the tools to the RIAA to shut down the network. ARGh!

Wait a second, Gnutella is peer to peer... let's think p-2-p a little bit! Imagine if every client has its own ban list (downloaded along with the application), the servers on which the requests will be passed will filter the results (using their version of the ban list) and only return the results of the hosts that were able to pass through.

That means that if a particular server is banned only by one computer, the information will still pass through, by flowing around the server responsible for the block. It would be like a rock in a river... blocking the flow for some...

When more computers share the same ban list, it will contribute to clean part of the network, removing unwanted requests in certain parts. The part of the RIAA on the network will be reduced considerably, limiting their action and their "control".

Let me know what you think... it can't do any harm to the project to implement that kind of stuff!