Re: [Ace] Adoption of Low Latency Group Communication Security Work in ACE

[Michael StJohns <[e-mail]>]

On 7/21/2016 5:26 AM, Carsten Bormann wrote:
> Michael Richardson wrote:
>> Why will ACE succeed when DICE failed?
> Because DICE tried to hack something into TLS.  That had no support.

Actually, that's not the complete story.  It was one of the things that 
finally killed this off (e.g. DICE was supposed to make a profile of 
DTLS for constrained devices, BUT DTLS didn't already support multicast, 
so its difficult to profile it in...; we have to come up with message 
formats for a DTLS extension)

It wasn't the only thing.  Again, there's a very long record of why this 
was a bad idea in DICE.  It's trivially easy to map each and every one 
of those arguments to why the equivalent thing in ACE is bad.

>
>> Does ACE now have some knowledge or mechanism that DICE couldn't have created
>> because it was out of scope?
> ACE has COSE.

*sigh* If this had any application to the stated lighting problem, then 
sending a COSE message with a public key signed payload to trigger state 
changes would be the solution, not a symmetric group multicast key.

E.g. use section 4 of the 
https://datatracker.ietf.org/doc/draft-ietf-cose-msg/ document.  Do NOT 
use any of the symmetric key sections.

  I've said similar things before, but there continues to be this belief 
from certain folk that its too expensive to do public key cryptography 
for lightbulbs.

So to be clear - yes COSE is useful.  No, it does not actually do 
anything to fix the problem of symmetric key group communications UNLESS 
you stick to the public key sections.

Later, Mike


>
> Grüße, Carsten
>
> _______________________________________________
> Ace mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/ace