Re: [lamps] [EXTERNAL] Re: I-D Action: draft-ietf-lamps-im-keyusage-00.txt

[Rohan Mahy <[e-mail]>]

Hi,
IMO, anyone who is issuing TLS server certificates would blissfully ignore
IM extended key usage and carry on doing what they are doing.
Matrix/Element clients would happily verify certificates from their TLS
certificates not expecting this extension in that context. They would
continue to require a key usage of serverAuth from the server certificates.

Matrix clients could also enroll user @alice on the server
matrix.example.com with a certificate that uses this extended key usage.
When Bob's client receives Alice's certificate (ex: in MLS, in the
signaling that carries DoubleRatchet encryption, via some key transparency
protocol, etc), Bob can validate Alice's certificate.

Thanks,
-rohan


On Wed, Apr 17, 2024 at 11:22 AM Stephen Farrell <[e-mail]>
wrote:

>
> Hiya,
>
> On 17/04/2024 18:37, Rohan Mahy wrote:
> > We clearly don't want someone reusing keys for multiple purposes.
>
> In general, we don't.
>
> Some IM servers today can operate web front ends (e.g. matrix
> synapse), with the same server identity being seen by both
> web browsers and installed clients (that may well just be a
> shim on a browser window).
>
> I'm not clear if what you're proposing would end up making
> that harder? Would it? If so, what's the justification for
> that?
>
> If such setups are unaffected then pardon the intrusion:-)
>
> Ta,
> S.
>
>