>>>>> "Tom" == Tom Petch <[EMAIL PROTECTED]> writes:
Tom> The logical conclusion of your approach is that what we need
Tom> is encryption, encryption and encryption, and oh, we could
Tom> throw in a little MAC here and there. I think this makes it
Tom> too complex, too costly with the result that the security
Tom> that is needed, and could be provided more simply, will not
Tom> happen.
I will say that encryption and macs are very easy to do and I think
you'd need to show a strong argument that they will not perform well
before performance concerns can be taken seriously.
The question I really need answers to is:
* Does the WG believe that Authentication of the origin of the message is a
requirement for the mandatory to implement approach?
* Does the WG believe that integrity protection independent of transport is a
requirement for the mandatory to implement?
I want these questions answered independent of particular choices
about implementation complexity.
Ultimately this comes down to Chris judging the consensus of the WG
based on the discussion here.
--Sam
_______________________________________________
Syslog mailing list
[email protected]
https://www1.ietf.org/mailman/listinfo/syslog
