On Wed, 13 Mar 2013 07:43:12 -0400 Greg Troxel <[email protected]> wrote:
> > "\"Daniel \".koolfy\" Faucon\"" <[email protected]> writes: > > > It's not up to the OTR protocol to define logging policies, so we > > must make sure OTR implementations behave responsibly. > > I'm not sure I agree with this. I think it's entirely reasonable for > the protocol to say that clients MUST NOT log, and if so MUST send a > "i-am-logging" notification to the other parties. Of course, there's > no guarantee, but the protocol spec might as well define the expected > behavior. I know it can be tempting at times, but I think it's actually a bad idea. The thing is: it's too easy to cheat with that. Even with such a notification, you can't prevent a copy/paste or a screenshot. You could argue that against a malicious interlocutor, we're screwed anyway, and that thus mechanism would be aimed at distracted or clumsy people, and I almost agree. My biggest problem is that the protocol would try to provide an information on something it has no real control of. And by doing so, it will create a false sense of security between the parties. The bottom line is: whatever happens, both parties engaging in OTR conversations should at all times evaluate the risk of being logged without consent. What a clean log management across OTR implementation does, is minimize the frequency at which this assumption is true. Without altering the user's perception of the "logging state" of the other interlocutor. The last thing we want is out encryption protocol being deceptive about the state of the exchange. This is when everything starts to fall apart :( -- Daniel ".koolfy" Faucon Tel: France : (+33)(0)658/993.700 PGP Fingerprint : 485E 7C63 8D29 F737 FEA2 8CD3 EA05 30E6 15BE 9FA5
signature.asc
Description: PGP signature
_______________________________________________ OTR-dev mailing list [email protected] http://lists.cypherpunks.ca/mailman/listinfo/otr-dev
