[SSL Observatory] SSL CA compromise in the wild

ArkanoiD ark at eltex.net
Wed Mar 23 14:39:30 PDT 2011


I do not see how HSTS helps. It does nothing besides enforcing TLS, trust model is still the same.

On Wed, Mar 23, 2011 at 12:48:51PM -0400, Steve Schultze wrote:
> SSH is just TOFU.  So, the current CA model with HSTS is at least as good as SSH, and probably better because at least you first have to compromise a CA.  Mozilla and Chrome implement HSTS.
> 
> http://en.wikipedia.org/wiki/User:Dotdotike/Trust_Upon_First_Use
> http://en.wikipedia.org/wiki/HTTP_Strict_Transport_Security
> http://blog.sidstamm.com/2010/08/http-strict-transport-security-has.html
> 
> On Mar 23, 2011, at 12:28 PM, Chris Palmer wrote:
> > On 03/23/2011 08:52 AM, Matt McCutchen wrote:
> > 
> >> The SSH "model" is a cop-out.
> > 
> > Its problems are significantly less bad than the status quo. Right now,
> > anyone who controls any one of the 1,400+ signing certificates (~650
> > organizations, many of them proven not to practice even basic sanity
> > checking on what they sign, and now at least one affirmatively pwned)
> > can own the entire internet.
> > 
> > Obviously, The True Internet Authentication Scheme will be better than
> > SSH. But we could do a lot worse than adopting the SSH model in the
> > medium term.
> > 
> > 
> > -- 
> > Chris Palmer
> > Technology Director, Electronic Frontier Foundation
> > https://www.eff.org/code
> 
> 
> email protected and scanned by AdvascanTM - keeping email useful - www.advascan.com 
> 
> 




More information about the Observatory mailing list