[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Simple Multicast - building a case for a BOF or WG



I agree with that opinion. In fact, the wg could try to define some
services first, and look at bidirectional pim, Express and SM as
potential implementations of the service models. Then in a second step,
a new protocol could be defined.

Cheng-Yin Lee wrote:
> 
> Joel Halpern wrote:
> 
> > My conclusion is that rather than focussing on simple, express, or
> > any other solution, the focus, as with the prime purpose of the
> > "simple" work has to be on the problem.  We do not have a wide
> > understanding of what works, what does not work, and why.  We need
> > to have the public discussion (face to face) of this in order to get
> > the community to see past Van's argument at the Florida plenary.
> >
> > Yes, simple and express can be used as examples of other
> > approaches.  In particular certain of the properties of those
> > solutions (such as alternative group identification) need to be
> > strongly highlighted.  But the BoF / WG should not be about the
> > solutions.
> 
> I tend to agree with Joel's suggestion here.
> 
> We already have ISPs input on certain issues and requirement. It would
> be productive to discuss based on the requirement (with some due
> thought to long term needs, if possible) what is already available and
> what is still missing, and how to address the remaining problems. It's
> easier to be objective if we don't focus on the individual protocols
> itself.
> 
> What we might end up eventually is a list of 'to do' which can be
> realized in existing protocols in the way that Dave Oran suggested. If
> we can't decide on certain issues like bi-directional trees, perhaps
> we could let the market decide.
> 
> cheers,
> cyl
> 

-- 
=======================================================

christophe Diot         | tel: 1(650)375.4539
Sprint ATL              | fax: 1(650)375.4490
1 Adrian Court          | cdiot@sprintlabs.com
Burlingame CA 94010     | www.sprintlabs.com
USA                     |