On Thu, Nov 20, 2003 at 11:58:45AM -0600, Edgar Friendly wrote:Toad <[EMAIL PROTECTED]> writes:I do now. However, it is a way that a node could hurt the network without being proportionately punished for it in the estimators - if I reinstate it, then nodes that QR with HTL 0, killing the request, will be treated the same as nodes that QR at any other HTL. This seems very bad.
It's bad, but only very mildly. Evil nodes can kill request at will; just DNF.
On the other hand, QRs are in the current thinking pretty bad, and are punished quite heavily anyway (backoff).
can we all agree that this punishment is entirely voluntary ? Not that we must address it immediately, or that we ever even can.. but someone can always code an abusive version in the future. Okay, let's not head down that road again! And the heaviness is due to *excessive* backing off...
ken
_______________________________________________ Devl mailing list [EMAIL PROTECTED] http://dodo.freenetproject.org/cgi-bin/mailman/listinfo/devl
