Re: [rtcweb] More Comments on draft-ietf-rtcweb-use-cases-and-requirements-07

Hi,

My suggestion would be to remove the statements from each use case regarding eavesdropping and replace it with bullets in section 4.1 for considerations which apply to all use cases. These should state that all media streams and data channels must be integrity and confidentiality protected.

Regards
Andy

> -----Original Message-----
> From: rtcweb-bounces@ietf.org [mailto:rtcweb-bounces@ietf.org] On
> Behalf Of Harald Alvestrand
> Sent: 10 June 2012 23:51
> To: rtcweb@ietf.org
> Subject: Re: [rtcweb] More Comments on draft-ietf-rtcweb-use-cases-and-
> requirements-07
> 
> On 06/10/2012 05:33 PM, Hutton, Andrew wrote:
> > Hi,
> >
> > Some comments on draft-ietf-rtcweb-use-cases-and-requirements-07.
> >
> > Section 4.1.1.1 (Simple Video Communication Service) contains the
> statement "Each user can also pause sending of media (audio, video, or
> both) and mute incoming media".  I think this and the associated
> requirements need some clarification for example whether the
> requirement is to control the direction of the media and therefore
> requires the generation of a new SDP offer with for example a direction
> of "recvonly/inactive".  The only associated requirement seems to be A8
> but this does not help clarify what is meant by "pause" in the use
> case. It needs to be clarified whether the pause/resume/mute are
> actions within the browser or are actions which result in notifications
> to the remote user.
> >
> > Maybe we could add a use case which involves controlling the
> direction of multiple streams to emulate switching between different
> users in a multiparty call scenario.
> >
> >
> > The statement "It is essential that the communication cannot be
> eavesdropped" appears in all use cases so could be moved to section 4.1
> but I think it needs to be reworded in terms of the media always being
> encrypted as I don't like the term "eavesdropped" which is not defined.
> If we want to define the term, we might want to use the term
> "wiretapping", and refer to RFC 2804 for the definition.
> 
> (having been involved in that discussion, I've got some personal
> fondness for that definition...)
> 
> 
> >
> > Regards
> > Andy
> >
> >
> >
> > _______________________________________________
> > rtcweb mailing list
> > rtcweb@ietf.org
> > https://www.ietf.org/mailman/listinfo/rtcweb
> 
> _______________________________________________
> rtcweb mailing list
> rtcweb@ietf.org
> https://www.ietf.org/mailman/listinfo/rtcweb