Re: [MMUSIC] Updated MSID draft

[Harald Alvestrand <[e-mail]>]

On 10/30/2013 03:38 PM, Paul Kyzivat wrote:
> A few comments about this version. Actually the comments aren't
> specific to this revision, but they relate to things I didn't notice
> earlier.
>
> Msid-Semantic syntax:
>
>      attribute =/ msid-semantic-attr
>      msid-semantic-attr = "msid-semantic:" token (" " identifier)*
>      token = <as defined in RFC 4566>
>
> The usage of "*" is backward. (Following regex syntax rather than abnf
> syntax). This ought to be:
>
>      attribute =/ msid-semantic-attr
>      msid-semantic-attr = "msid-semantic:" token *(" " identifier)
>      token = <as defined in RFC 4566>

Thanks!


>
> Just following that:
>
>    The semantic field may hold values from the IANA registries
>    "Semantics for the "ssrc-group" SDP Attribute" and "Semantics for the
>    "group" SDP Attribute".
>
> I have a couple of issues with this:
>
> Section 5 defines a new registry "Semantics for the msid-semantic SDP
> attribute", and defines a new semantic "WMS" in it. The text above
> doesn't even reference this registry.

Yep, that section needs to be re-thought. The SSRC-group registry is now
irrelevant, since we don't do things on SSRC any more.
>
> Because the two mentioned registries are independent, AFAIK there is
> no guarantee that they couldn't someday contain different definitions
> for the same "semantic" name. Then, if that name were used with msid
> the proper definition to use would be ambiguous. (I realize this is
> implausible, but nothing prevents it.) IMO the *right* answer would be
> to merge all these grouping semantics into a single table. But that is
> a hassle right now. Perhaps it would be sufficient to say that only
> semantics that are uniquely defined across all of these tables may be
> used with msid. E.g.,
>
>    The semantic field may hold any value that is uniquely defined
>    in one and only one of the IANA registries "Semantics for the
>    msid-semantic SDP attribute", "Semantics for the "ssrc-group"
>    SDP Attribute" and "Semantics for the "group" SDP Attribute".

That would work - it should then ask IANA to update the instructions for
those registries to disallow the registration of new values with
conflicts between them.

This effectively turns them into one registry...

>
>     Thanks,
>     Paul
>
> On 10/30/13 3:13 AM, Harald Alvestrand wrote:
>> Roni and Flemming pointed out that I hadn't managed to submit the fixes
>> I made to -msid after the reviews of -01 before the deadline, and had
>> even managed to forget to send it to the list.
>>
>> I had missed a number of occurences where "SSRC" needed to be changed to
>> "m-line" in the text (with some more adjustments to language around
>> that).
>>
>> Here it is. I'll submit it on Monday, when the I-D repository reopens.
>>
>>
>>
>> _______________________________________________
>> mmusic mailing list
>> [e-mail]
>> https://www.ietf.org/mailman/listinfo/mmusic
>>
>
> _______________________________________________
> mmusic mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/mmusic


-- 
Surveillance is pervasive. Go Dark.