People who are on non permanent connections are not going to connect from the same ip next time anyways, so ignoring requests from that ip is hardly going to be very helpful. For that sort of thing to work you need to have some sort of centralized user lease system, which we obviously cannot have.
Also, I think you misunderstand the system. You will get a DataReply from a node only if you sent it a DataRequest. So if you are getting more DataRequests from a node then you are getting DataReplies, that is simply because you are sending less DataRequests too it. It might be possible to have a system where a node starts at zero trust, and has to build trust by helping supply data before it can request anything (sort of like ratio warez servers), but this would be a major stick in the wheel of anybody wanting use Freenet. I could see such a feature as a part of the sort of underground web of trust based Freenet that we would have to build should it prove impossible to run it as a public service because of floods. On Tue, 02 May 2000, degs wrote: > Hi, > > I think without an incentive to contribute, that most people will access > Freenet via a client rather than a node. I think Human nature is pretty > certain on this one, especially given that there is still a fair premium on > bandwidth and a fixed IP address (at least in the UK where I am). So I guess > this begs the question - what ratio of clients to nodes will cause Freenet > to fail? Can Freenet nodes allow themselves to be leeched off freely? My > scheme was an attempt (maybe misguided) to allow nodes to act together to > ensure they are not exploited to the point where Freenet breaks down. I'd be > interested if anyone has made educated guesses about how many nodes would be > required to make Freenet viable and how Freenet's performance would change > if a large proportion of these nodes chose not to participate in data > storage. (Waaay beyond my maths...) > > Degs > > -----Original Message----- > From: Oskar Sandberg <md98-osa at nada.kth.se> > To: freenet-dev at lists.sourceforge.net <freenet-dev at > lists.sourceforge.net> > Date: 02 May 2000 12:40 > Subject: Re: [Freenet-dev] I had a mad idea > > > > > >I don't see the point of this at all. This is just a going to make the > entire > >network a downward spiral in performance. The prisoners dilemma doesn't > really > >relate to this situation at all. > > > >In the current network, a node that does not reply to requests will rather > >rapidly by forgotten completely by the rest of the network. I am convinced > this > >works well enough (possibly even to well) at filtering out such nodes. > > > > > > _______________________________________________ > Freenet-dev mailing list > Freenet-dev at lists.sourceforge.net > http://lists.sourceforge.net/mailman/listinfo/freenet-dev -- Oskar Sandberg md98-osa at nada.kth.se #!/bin/perl -sp0777i<X+d*lMLa^*lN%0]dsXx++lMlN/dsM0<j]dsj $/=unpack('H*',$_);$_=`echo 16dio\U$k"SK$/SM$n\EsN0p[lN*1 lK[d2%Sa2/d0$^Ixp"|dc`;s/\W//g;$_=pack('H*',/((..)*)$/) _______________________________________________ Freenet-dev mailing list Freenet-dev at lists.sourceforge.net http://lists.sourceforge.net/mailman/listinfo/freenet-dev
