[cap-talk] Covert channels (was Re: Capabilities for immutable data)

[Bill Frantz]

On 2/20/11 at 6:43 PM, daw at cs.berkeley.edu (David Wagner) wrote:

>With confidentiality and covert channels, I can articulate a security
>claim, but without trusting all code that has access to the confidential
>data, I don't know how to be sure that I've achieved the security claim
>or not.  My ability to know whether I've achieved the security claim is
>limited by inherent unsoundness in the patterns of reasoning we use.

I don't even know how I can gain trust in code that has access 
to confidential data.

Covert channels and side channel analysis are really the same 
animal. We assume a willing sender with a covert channel and an 
inadvertent sender with a side channel. But we really don't have 
any idea of how many modulation schemes there are that we 
haven't thought of. The RSA implementations that leak keys 
weren't written to try to leak the key.

Cheers - Bill



-----------------------------------------------------------------------
Bill Frantz        | Snow shuts down federal      | Periwinkle
(408)356-8506      | government, life goes on.    | 16345 
Englewood Ave
www.pwpconsult.com |  - Associated press 2/9/2010 | Los Gatos, 
CA 95032