NODE 2fdb491aRemailing
frc%bwnmr4@harvard.harvard.edu (Fred Cooper)Tue, 7 Sep 93 16:16:45 PDT
-----BEGIN PGP SIGNED MESSAGE-----
Would it be possible to incorporate a ping into the remailer? Say give it
a list of known remailers for the chain and have it pick one at random,
then ping it. If the remailer responds then send out the mail, otherwise
pick a new remailer, and repeat....
Or this there a flaw in this that I'm just too ignorant to see?
FRC
-----BEGIN PGP SIGNATURE-----
Version: 2.3a
iQCVAgUBLI0UtbbAlE4AqlTZAQEtTQP/SZlTdkSn+fwiO13SybCHGArjWPpl+xg8
CnSxgw7coYUWdMUU/kwnCQWuLKVsZigcLQmlm1SUvvf6MADmkENF1A0V1Q7kej7Q
CCHEjvKA78oHAFBjw3b/BuBAGIibtuGubrSDp/65fe3jF2obDwK4mJqSlD9dcWLC
SzU58NCnl9M=
=zmFE
-----END PGP SIGNATURE-----
NODE 8082b912Re: Remailing
Marc Horowitz <marc@Athena.MIT.EDU>Tue, 7 Sep 93 16:52:41 PDT
>> Or this there a flaw in this that I'm just too ignorant to see?
It's fine, if you don't mind making traffic analysis completely
trivial.
Marc
NODE 56c83efcRe: Remailing
frc%bwnmr4@harvard.harvard.edu (Fred Cooper)Tue, 7 Sep 93 17:11:46 PDT
-----BEGIN PGP SIGNED MESSAGE-----
How does this make TA trivial?
I am assuming that batch mailings or staggered mailings are occuring.
The remailers are known locations so any traffic between mailers is
assumed to be observed. The key is making it impossible to correlate
arrival and departure time of any particular message. Am I missing
something?
-----BEGIN PGP SIGNATURE-----
Version: 2.3a
iQCVAgUBLI0h2bbAlE4AqlTZAQExOAQAs6lLhVMjm5hi9swLgkLFJUSz8SazhlfW
5RDTBxNJs2oxt0oit1oEvKqBR26zKKayrwQf2O1DIxQD/f08qfIRS5dbLiz8c4VE
5XiP5j+HBr9j/mt5EiN8uCukpi1eP4pCq/cl82UDqkA8kkosvNDfSY26ubmf97FH
3uNbJ6vkWv0=
=qryI
-----END PGP SIGNATURE-----
NODE 233635f2Re: Remailing
Marc Horowitz <marc@Athena.MIT.EDU>Tue, 7 Sep 93 20:56:47 PDT
>> How does this make TA trivial?
>> I am assuming that batch mailings or staggered mailings are occuring.
That is not currently the case.
And even if it were, sending messages to yourself is going to add a
significant "known quantity" to the remailer traffic. The more an
attacker knows, the less he needs to figure out.
Marc
NODE c6cbe5deRe: Remailing
frc%bwnmr4@harvard.harvard.edu (Fred Cooper)Wed, 8 Sep 93 19:08:04 PDT
-----BEGIN PGP SIGNED MESSAGE-----
> And even if it were, sending messages to yourself is going to add a
> significant "known quantity" to the remailer traffic. The more an
> attacker knows, the less he needs to figure out.
How is pinging a known address considered sending messages to
yourself?
FRC
-----BEGIN PGP SIGNATURE-----
Version: 2.3a
iQCVAgUBLI6O3bbAlE4AqlTZAQE2qAP7ByAj4hf0IMYBqBhTQDuLY0mSIDGrYSdn
ahpDILrnvdPDcz6Lh72lAdT5QIXa8e+2hi/4PLXXPtQpmuW52uSSZjhN4w2FaLUl
/Y6ZccEWbbau4Okyw+RUUjJYZbWT4EUjv2vLiN06NjJaPztJJXwFKKWeLnJNixci
2xjZOa3iMKw=
=oUif
-----END PGP SIGNATURE-----