-----Original Message-----
From: Peter Saint-Andre [mailto:[email protected]] 
Sent: Wednesday, May 28, 2014 11:48 AM
To: Trevor Freeman; [email protected]
Subject: Re: [Uta] TLS BCP SSL v3 == TLS 1.0

On 5/28/14, 12:15 PM, Trevor Freeman wrote:
> I am curious why the draft treats SSL v3 (SHOULD NOT) differently to 
> TLS
> v1.0 (MAY)

We received feedback about that at the London meeting and haven't incorporated 
it yet.

>  From a security perspective, they are equivalent in that they are no 
> significant threats mitigated by TLS 1.0.
>
> TLS 1.0 is still vulnerable to the BEAST attack so it should be a 
> SHOULD NOT like SSL v3.

Ack.

> Given the vulnerability of these versions to the BEAST attack, we 
> should set a date to flip then from SHOUND NOT to MUST NOT to send a 
> stronger message to stop using these versions.

IMHO it's not the function of this document to be setting flag days.
[TF] The document is supposed to be defining best practice. That seems hard to 
reconcile with unbounded continued use of a protocol with a known 
vulnerability. If you don't want to set a date, why not flip the SHOULD to 
MUST? That may mean some are not compliant with best practice but is clarifies 
the position on use of protocols with vulnerabilities. 

Peter


_______________________________________________
Uta mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/uta

Reply via email to