On Thu, Oct 30, 2003 at 06:57:29PM +0000, Toad wrote: > On Thu, Oct 30, 2003 at 11:07:45AM +0000, Ian Clarke wrote: > > Is it fair that when 3 nodes might have sent me a vast number of > > requests, thus overloading my node, that a 4th node which has never > > asked anything of my node before is refused? Probably not, but how can > > this be addressed? > > > > The obvious option is to bias QRs towards nodes in proportion to the > > degree to which they are responsible for the QR being sent. Now > > clearly, this is an imperfect solution as it relies on negative trust, > > and there really can't be any reliable negative trust on the Internet. > > Having said this, while this mechanism might not be impossible to > > circumvent, it could be enough to encourage developers of high-traffic > > applications such as Frost to make their users try to be better Freenet > > citizens. > > I'm not sure it would. It might be more likely to encourage their users > to run several Freenet nodes - even with Fred's bloat, RAM prices > continue to fall, and Fred will not always be as bloated as it is now > (in RAM/CPU usage), if I have anything to do with it. > > However, another possibility: have some positive trust, not by WoT, but > by a much simpler mechanism of favouring requests from nodes in the > routing table when we are overloaded... or from nodes who are not only > in the RT but have been answering requests well for some time.
The difficulty in the last sentance is that nodes can insert themselves into our RT if it is to their advantage, by announcing at HTL 1... > > > > Thoughts? > > > > Ian. -- Matthew J Toseland - [EMAIL PROTECTED] Freenet Project Official Codemonkey - http://freenetproject.org/ ICTHUS - Nothing is impossible. Our Boss says so.
signature.asc
Description: Digital signature
_______________________________________________ Devl mailing list [EMAIL PROTECTED] http://dodo.freenetproject.org/cgi-bin/mailman/listinfo/devl
