Re: [MBONED] call for adoption of draft-acg-mboned-multicast-models

Inline:

On Mon, Dec 4, 2017 at 12:18 PM, Manfredi, Albert E <
albert.e.manfredi@boeing.com> wrote:

> I remain somewhat disturbed by this underlying thinking (Section 7.6):
>
>    Thus, in theory, it should be possible to port ASM-only applications
>    to be able to run using SSM, if an appropriate out-of-band mechanism
>    can be chosen to convey the participant source addresses.
>
> Most of the document is fine by me, as is its title, "Deprecating ASM for
> Interdomain Multicast." The problem is when the wording strays from that
> interdomain scenario. Specifically, in the quote above, "if an appropriate
> out-of-band mechanism" is a big "if." Almost like saying, it's not an IETF
> problem, so we'll pretend it's not a problem. Use SSM, create whatever new
> and previously unnecessary out-of-band mechanism you might need now, and
> trust us, this will be better.
>

How did the application learn about the group? The port? The codec? No new
mechanism required, just an additional field to your mechanism of choice.

> There are intra-domain use cases in which ASM is by far the simplest
> approach, and there are no downsides to it. I'm not sure why the IETF
> should be making "strong" recommendations against such specific
> applications of ASM. How about at least remove the word "strong," and
> remove the "deprecate" in intra-domain cases?
>

I agree with softening the intra-domain wording.

> I'll go back to my previous comment. SSM is very much like IP multicast
> over ATM networks. The model is based on having to create those separate
> pt-mpt connections, which at the time, were an artifact caused by the ATM
> way of doing things. At the time, having to accommodate ATM in this way was
> more or less of a nuisance, especially in intra-domain cases. Yes, it
> required this extra step of knowing all of the possible sources, in a model
> that was happily purely leaf-initiated. Now the IETF is expanding that
> intra-domain nuisance even for Ethernet networks, where it's not needed.
>

Well, that hurt. ;-) SSM was about simplification when the source was
well-known. Any similarity to ATM is a painful and unintended coincidence.
:)

Greg

> Bert
>
> -----Original Message-----
> From: MBONED [mailto:mboned-bounces@ietf.org] On Behalf Of Leonard
> Giuliano
> Sent: Monday, December 04, 2017 14:46
> To: MBONED WG <mboned@ietf.org>
> Subject: [MBONED] call for adoption of draft-acg-mboned-multicast-models
>
>
> We would like to initiate a call for adoption of the "Deprecating ASM for
> Interdomain Multicast" draft in MBONED:
> https://datatracker.ietf.org/doc/draft-acg-mboned-multicast-models/
>
> Please respond by Dec 18 if you do/do not support adoption of this draft.
>
> If you are listed as a document author, please respond to this email
> whether or
> not you are aware of any relevant IPR. If you are not listed as an author
> and
> are aware of any relevant IPR, please respond as well.
>
>
> -Lenny and Greg
>
> _______________________________________________
> MBONED mailing list
> MBONED@ietf.org
> https://www.ietf.org/mailman/listinfo/mboned
>
>
> _______________________________________________
> MBONED mailing list
> MBONED@ietf.org
> https://www.ietf.org/mailman/listinfo/mboned
>