The juxtaposition of these two statements, coming from
the same source on the same day, is interesting.
What is the difference between a "network trace"
and a "wiretap"? An application architecture that
stores trace data does not require modification (hooking)
of the protocol. And a modification to the protocol
to transmit digitally signed data changes neither
the confidentiality properties of the protocol nor
the existence of application architectures that
store trace data.
Note that:
"the crap with the digital signatures is entirely
unnecessary for technical analysis."
is entirely correct. But Danvers is a non-sequitur.
Ignoring lawyers, non-repudiation, and protocol
analysis, the question remains: is there any benefit
to providing data origin authentication at the TLS
layer? In other words, assume the existence of an
evaluated high assurance "TLS Appliance" fronting an
application that is much more complex and dynamic
and difficult to characterize from a security
perspective. Is there any benefit to having the
TLS box sign data instead of having the application
sign data itself? I'm not convinced there is,
but regardless, it would be better to focus the
discussion on the relying party's ability to
validate signatures for its own benefit rather
than discuss taking those signatures to court.
Dave
-----Original Message-----
From: Martin Rex [mailto:martin.rex@xxxxxxx]
Sent: Friday, January 05, 2007 4:10 PM
If one needs cleartext network traces of what goes in TLS and
what comes out TLS, that can easily be done today (in fact,
an application architecture which can not do that today is
seriously broken-- it is impossible to provide REAL support for
larger mission-critical application software without capabilities
to trace at various protocol levels and API/component level),
and the crap with the digital signatures is entirely unnecessary
for technical analysis.
-----Original Message-----
From: Martin Rex [mailto:martin.rex@xxxxxxx]
Sent: Friday, January 05, 2007 9:08 PM
Mark Brown wrote:
> App doesn't need to know. App can ignore all Evidence- messages
> so long as some component somewhere is storing them.
> ... maybe a wiretap on the TLS session is the preferred
> method of persisting the TLS Evidence.
It's time to shut down the discussion in this forum for
procedural violations of the IETF Danvers Doctrine.
The IETF will NOT put wire-tapping hooks into its protocols.
_______________________________________________
TLS mailing list
TLS@xxxxxxxxxxxxxx
https://www1.ietf.org/mailman/listinfo/tls