On Wed, 2003-10-22 at 14:31, Dan Merillat wrote: > On Wed, 15 Oct 2003, Edward J. Huff wrote: > > > Freenet uses a lot of bandwidth. Some of it might be avoidable. > > Somewhere I saw an argument that content must pass through each node > > along the chain so that they can all verify that the content matches the > > hash. But there are other ways of verifying this. > > Initial assumption wrong, rest of message invalidated. > > Freenet routes backwards through the search path to preserve anonymity, > not to verify content integrity.
Well, not to preserve anonymity, but to preserve plausible deniability in the face of mole nodes. I am interested in stronger methods of preserving actual anonymity, and point out that if a stronger method is used, you don't need to route backward through the search path. > Besides, since it's a (fairly) static connection mesh, popular content > gets pcached near where it's desired. The bandwidth waste comes from > transferring the same data to the same node multiple times when half the > requests are no-longer needed. Using the same data to satisfy multiple > requests would be a fix, but will require protocol changes. (Probably > a part of multiplexing anyway) > Yes, that is a bandwidth waste, but there is also another: suppose that several nodes on the request path are low bandwidth, but the requester and the supplier are high bandwidth. Then it would be better if the data did not need to pass through the low bandwidth nodes. -- Ed Huff -- Edward J. Huff <[EMAIL PROTECTED]>
signature.asc
Description: This is a digitally signed message part
_______________________________________________ Tech mailing list [EMAIL PROTECTED] http://hawk.freenetproject.org:8080/cgi-bin/mailman/listinfo/tech
