On Friday 29 August 2003 08:18 pm, Tracy R Reed wrote: > On Thu, Aug 28, 2003 at 11:45:19PM -0400, Andrew Rodland spake thusly: > > What this is, is an idea for decreasing the number of nodes that data > > passes through on freenet, and so also freenet's bandwidth usage. It > > would not be a compatible change, but it's just an idea. :) > > There is a very good reason for passing through so many nodes. It makes > the requester and the source of the data harder to track down. > Shortcutting as you propose it would reveal the identify of the requesting > node which totally defeats what freenet is all about.
No, not if it's probabilistically reset along the way. We *might* be sending it back to the requester, but we also might be sending it back to someone who is going to forward it further back any number of hops. Just like now. The higher numbers are less likely, but still entirely possible. As long as freenet hides its traffic decently well, it would be difficult even for a global observer to be sure. > Plus returning the > data back along the same path helps propagate good routing info and lets > the nodes along the path know that the data was found so they can route > similar requests back along this path. Ideally we want the data to pass > through as many nodes as possible with the user still getting his result > in a reasonable amount of time. True, I had forgotten that routing info only gets set when we actually see the data come back. And for good reason, too. Unless the node sees the data and gets a chance to verify it, then routing could be spoofed. Didn't catch that one. Seems sort of sad, though. The rest was nice. --hobbs
pgp00000.pgp
Description: signature
_______________________________________________ Devl mailing list [EMAIL PROTECTED] http://dodo.freenetproject.org/cgi-bin/mailman/listinfo/devl
