Re: IESG WG Review: IP over DVB (ipdvb) - Synchronised delivery of IP packets

[Gorry Fairhurst <gorry@erg.abdn.ac.uk>]

I believe you're asing for more detai of the synchronised delivery 
service...

See in-line comments, below.


alain.ritoux@6wind.com wrote:

> > > IP over DVB (ipdvb)
>
>
> > >  ... Snip
> > > The current list of work items is:
> > >
> > > 1. Issue an Internet Draft specifying Requirements and Framework for
> > > supporting IP services via MPEG-2 transmission networks. Such 
> > > requirements
>
> So, its is more IP-over-MPEG2 rather tah IP-over-DVB. Am I correct ?

Yes, that always was the case.

> > > ... Snip
> > > 2. The working group will investigate and design an efficient 
> > > encapsulation
> > > method for IPv4/IPv6, and advance this via the IESG to a 
> > > standards-track
> > > RFC. The design needs to consider the need for MAC addresses, the 
> > > potential
> > > need for synchronisation between streams, support for IPv6 and 
> > > multicast
>   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> Is there not a risk of mixing different layers ? because this notion is
> more a DVB one ? 

To ensure we speak of the same thing the current requirements id (page 
12) says:

             "The standard PES Packet as defined in table 2-18 of [ISO-
              MPEG] can also be used as a container for data packets. The
              two values for the stream_id are "private_stream_1" and 
              "private_stream_2". The private_stream_2 permits the use of
              the short PES header with limited overhead. This makes it
              attractive for publicly accessible multicast distribution
              services. The private_stream_1 makes available the 
scrambling
              control and the timing and clock reference features of the
              PES layer. The PES_data_packet_header_length will be used in
              this context to insert stuffing bytes. This is an important
              aspect since the payload can be aligned to 32-bit word
              boundaries. "

I agree this is an MPEG-TS attrribute for PES-format streams.

Both DVB and the ATSC Data Broadcast Standard describe how this may be 
used (each with differing levels of detail).

At the moment the proposed charter says we should consider this need, 
and assess the impact on the design of the encapsulation procdure.

I have a question to implementors and potnetial users is do they 
envisage the need for this to be supported in the IETf-supplied 
encapsulation. if so please send potential use cases to the list!!!

> Are the systems over MPEG-2 such as DVB, ATSC, ISDB-T ... compatible in
> terms of stream synchronisation ?
>
> Anyway, what is the purpose of such sync between IP flows and other
> streams ? or between IP flows ? shouldn't it be more a pb to be solved
> at application and/or transport level (using RTP) rather than at a
> newtork level ? 

Not necessarily so, RTP can provide synchnosiation, but this isn't the same.

> Well, I maybe I missed some importants points, but if this feature is
> important/needed/whatever, why is it not in the requirement draft ?

Correct, we need text, are you offering text?

Does anyone else have text we can use to update the requirements draft 
with use-cases for this type of service?

> Regards.
> Alain