> The only thing the nodes know is the hash. So if A > seems better to C it is > only because A is faster overall for C biased on the > requests that C gets. So > if C routes lots of requests to A, A's > specialization will slowly become > closer to C's. However C will never route to A for a > request if it thinks B > is a better match. If C does not know B very well > and so does not know that > it is a better match, and it routes a request to A, > yes that is non optimal, > but the only solution is for C to learn more about > B. Yeah, as real life person can only know what he knows based on his perceptions of the world. Nobody can claim to live optimally.
A node itself knows better than its neihbors how well it routes, but we can't trust it, at least not by itself. I think I remember reading way back in the big fat HTML paper, that freenet was designed for popular data used by people topographically close to each other. It maybe that Freenet acutally puts people close to each other that use the same content. We might be able to use see if this "subjectivity" pops up more or less with different routing algorithms, if it's correlated to any problems we're having, and if it gets worse as the number of nodes grows. All we'd have to do is compare what a node thinks is its probablity/time to resolve key X verses what this neighbors think it is. The routing does have much to do with it. Imagin if all nodes did was query thier "best" neighbor regardless of the key. That kind of a network would probably quickly turn into a bunch of link minded clusters, especially if they had populations of users with no overlapping tastes. The entire network might even subdivide into unconnected networks! Ok, we'll never do something that dumb, but it illustrates the routing problem that could theoretically crop up, maybe even with some of the more elegant systems. Bottom Line: It sure would be nice to have some way of testing 1000+ node systems and begining able to study things like this. What's the closest we have to this? Woops sorry that got long. __________________________________________________________________ Gesendet von Yahoo! Mail - http://mail.yahoo.de Logos und Klingelt�ne f�rs Handy bei http://sms.yahoo.de _______________________________________________ devl mailing list [EMAIL PROTECTED] http://hawk.freenetproject.org:8080/cgi-bin/mailman/listinfo/devl
