On Thursday 31 May 2007 16:16, Andrew McIntyre wrote: > How can we be confident that it was you who wrote the document. In this > case its irrelevant but this is not the case wrt medical communication! > > Andrew McIntyre (perhaps)
Look at the sender of this email reply. Who does it say it was sent from? If I have malicious intent, I can always spoof the header. But why stop there? In most cases where there is malicious intent in this context, there is physical proximity, and possibly access to the dongle of the other person too. Somebody else using your dongle would actually work in favour of the malicious attacker, because it increases your burden of proof. I am reasonably happy to accept burden of proof if I can be confident that I am the only one ever having had access to my private key. I can impossibly have that confidence if somebody else generated that private key and gave it to me. Likewise, how can a third party be confident of the sender if it was signed with what they *KNOW*is a potentially already compromised key - we are in fact no better off than just taking the email at face value and deriving from content + common sense that the sender *probably* was who the email stated it was. So, this "probably" is good enough for practical purposes to me if the alternative would be unilaterally unreasonably increasing burden of proof towards the sender's side. Horst _______________________________________________ Gpcg_talk mailing list [email protected] http://ozdocit.org/cgi-bin/mailman/listinfo/gpcg_talk
