Re: [TLS] Prohibiting RC4 Cipher Suites

Hi Andrei,

Of course current Web browsers all support better ciphersuites. But 
there is huge diversity here: people may have very old browsers and are 
not allowed or cannot install newer versions. Or people may be using old 
kiosks that are totally out of their control. Or admins may have set up 
a silly group policy and users are stuck with it.

And in general, many people (most people?) will do what comes easier 
rather than what makes sense from a security point of view. We should 
maximize the security of users, even if they're acting stupidly.

Thanks,
	Yaron

On 2013-08-23 02:24, Andrei Popov wrote:
> Hi Yaron,
>
> Thanks for the feedback.
>
> Are there any major web browsers that only support RC4? I am not aware of any.
>
> Arguably, the common fallback for web users is to install a different browser, rather than try to find a non-TLS service (which is not always an available option).
>
> Cheers,
>
> Andrei
>
> -----Original Message-----
> From: tls-bounces@ietf.org [mailto:tls-bounces@ietf.org] On Behalf Of Yaron Sheffer
> Sent: Thursday, August 22, 2013 12:36 AM
> To: tls@ietf.org
> Subject: [TLS] Prohibiting RC4 Cipher Suites
>
> Hi Andrei,
>
> Thank you for the new draft. While I agree with the motivation and with the first two recommendations (do not offer RC4, do not accept RC4 - if
> possible!) I disagree that it is better to completely reject a client that offers only RC4, because the intuitive fallback for Web users is simply, don't do TLS. In a world of pervasive passive surveillance (see https://www.ietf.org/mailman/listinfo/perpass), we would prefer sessions to be encrypted even if it means that an active attacker, working hard, can break into them. And yes, the is the age old "false sense of security" discussion, yet again.
>
> Thanks,
>       Yaron
> <https://www.ietf.org/mailman/listinfo/perpass>
> _______________________________________________
> TLS mailing list
> TLS@ietf.org
> https://www.ietf.org/mailman/listinfo/tls
>