Re: Pub request for draft-ietf-sip-dtls-srtp-framework-03

Let us try it another way:

In the U.S., we have a strong regime of lawful intercept, that  
includes the courts for wiretap orders and multiple, independent keys  
required to do the wiretap.

With all of this protocol and legal protections...

Most of the carriers executed unlawful intercept. Our answer was to  
give them blanket immunity.

Do not be surprised if soon it will be illegal to use cryptographic  
means to secure communications.

Thus...
Unless you have a written directive to ensure insecure communications,  
this model is hopelessly broken.

(signed with my cryptographic signature, if only so you can detect  
integrity degradation)

On Sep 25, 2008, at 1:51 PM, Schneider, Peter (NSN - DE/Munich) wrote:

> See inline
>
> Peter
>
>> -----Ursprüngliche Nachricht-----
>> Von: ext Dean Willis [mailto:[email protected]]
>> Gesendet: Donnerstag, 25. September 2008 21:32
>> An: Schneider, Peter (NSN - DE/Munich)
>> Cc: [email protected]
>> Betreff: Re: [Sip] Pub request for
>> draft-ietf-sip-dtls-srtp-framework-03
>>
>>
>> On Sep 25, 2008, at 2:50 AM, Schneider, Peter (NSN - DE/Munich)  
>> wrote:
>>>
>>> Dean, I assume that you refer to the proposal to use SDES that is
>>> discussed in 3GPP. However, 3GPP does not focus on that approach.
>>> Other signaling path solutions are discussed in 3GPP that exclude
>>> all intermediaries from access to the key. Clearly,
>> allowing lawful
>>> interception is a requirement for 3GPP, as is preventing "unlawful
>>> interception".
>>
>> The two forms of interception cannot be technically differentiated.
>> Anything that allows one allows the other.
>>
> Allowing lawful interception does not imply allowing anyone to  
> intercept the communication. You must have access control for the  
> facilities that allow lawful interception. Compare this with the  
> authentication service described in RFC4474 (SIP identity). Who  
> controls that service, can mount a man in the middle attack that  
> cannot be detected by the means provided by DTLS-SRTP.
>
>> The USA has recently shown that it is possible to convert unlawful
>> intercept into lawful intercept retroactively, despite a
>> constitutional bar on ex post facto laws. If you're going to
>> break one
>> law, you might as well break 'em all, right?
>>
>>> The middlebox issue is NOT a pretense for allowing only weak
>>> solutions for 3GPP.
>>
>> I believe you.
>>
>>> My proposals concerning the framework draft wouldn't make
>> DTLS-SRTP
>>> any weaker, right?
>>
>> I'm not sure. Security is such a frail thing, and it is easy for
>> unintended consequences to occur. You may well be right, but I'm a
>> little slow to commit.  Further study is required.
>>
>>> And making DTLS-SRTP more adequate for 3GPP/TISPAN scenarios would
>>> be a good thing, wouldn't it?
>>
>> It might, it might not. It depends on what your definition of
>> "adequate" entails and which scenarios you are talking about. The
>> requirements rejected by the IETF during the Media Security
>> Requirements drafting arguably provide a proof case against this
>> assertion.
>>
> Well, making DTLS-SRTP more adequate for 3GPP/TISPAN scenarios  
> (excluding lawful interception) without making it weaker would be a  
> good thing - better now?
>
>> --
>> Dean
>>
> _______________________________________________
> Sip mailing list  https://www.ietf.org/mailman/listinfo/sip
> This list is for NEW development of the core SIP Protocol
> Use [email protected] for questions on current sip
> Use [email protected] for new developments on the application of sip

_______________________________________________
Sip mailing list  https://www.ietf.org/mailman/listinfo/sip
This list is for NEW development of the core SIP Protocol
Use [email protected] for questions on current sip
Use [email protected] for new developments on the application of sip