Re: [AVT] FW: New Version Notificationfor draft-begen-avt-rtp-cnames-01

> > RFC 3550 says that one of the usage of RTCP is for monitoring and "To
> > facilitate third-party monitoring, the CNAME SHOULD be
> > suitable for either a program or a person to locate the source."
> 
> 
> Two points about that:
> 
>   o  This requirement is at odds with the need, in some RTP
>      applicatoins, to thwart traffic analysis.

So, our suggestion is to have multiple methods to come up with a CNAME, and some of them will be sticky across sessions and will not be opaque whereas others won't be sticky but will be opaque. Application should be able to choose whatever satisfies their needs. Each will have a con vs pro.

>   o  I find it VERY confusing to need to read both RFC3550's
>      section on CNAMEs and draft-begen-avt-rtp-cnames and reach
>      some create my own Vulcan mind meld.  This is unnecessarily
>      confusing.  It would be far cleaner to have one canonical
>      place to read about CNAMEs.
> 
>      So, can we pull all of the requirements and justification
>      text for those requirements into draft-begen-avt-rtp-cnames,
>      and have that supercede RFC3550's suggestions and
>      requirements about CNAMEs?

I agree that this will be a clean solution. We can add text around this.

> > Other comments
> >
> > 1. For the UUID I assume the proposal is to use the string
> > representation of the UUID maybe as a URN.
> 
> The CNAME doesn't need the 9 characters of "urn:uuid:", though.

Correct.

> > 2. As for the structure of CNAME, section 6.5.1 of RFC 3550
> > has the cname
> > composed of user and domain name. it also says that
> > "The user name SHOULD be in a form that a program such as
> > "finger" or "talk"
> > could use, i.e., it typically is the login name rather than
> > the personal
> > name.  The host name is not necessarily identical to the one in the
> > participant's electronic mail address. "
> > Maybe this document should recommend how to create a unique
> > host part of the
> > CNAME and not the whole CNAME.
> 
> I don't know how to do that, but maybe someone else does.

So, the methods we will have will be used to create the "host" part of the CNAME, and when combined with the "user" part, things will work out fine.

-acbegen

> -d
> 
> 
> > Regards
> > Roni Even
> >
> >
> >
> > _______________________________________________
> > Audio/Video Transport Working Group
> > avt@ietf.org
> > https://www.ietf.org/mailman/listinfo/avt