Re: [Moq] Small MoQ charter questions/suggestions

[Ted Hardie <[e-mail]>]

On Mon, Jul 25, 2022 at 7:25 PM Ian Swett <ianswett=
[e-mail]> wrote:

> The first paragraph of the proposed charter
> <https://github.com/moq-wg/moq-charter/blob/main/charter.md> is currently:
>
> "Media over QUIC (moq) will develop a *simple* low-latency media delivery
> solution
> for ingest and distribution.  This solution *may* address use cases
> including live
> streaming, gaming, and media conferencing and *will scale efficiently*.
> The
> solution will be implementable in both browser and non-browser endpoints."
>
>
> What does *simple* mean in this context?  Is it low overhead in terms of
> bytes on the wire or CPU or simple to implement, which seems implementation
> dependent?
>

Generally speaking, "simple" is used in charters as a signal that the
proponents know that there are a lot of potential bells and whistles that
might be added, but that the focus is on getting a workable system for this
era of the proposed WG's life.  The reality is that many protocols that
start out with simple in their charter (or title) end up with all the bells
and whistles over time, but this is a hint that the folks writing the
charter want iteration one to be able to focus on the core system at this
stage.*


> Both of these seem valid, and maybe the current text is sufficient, but if
> there are key goals besides low-latency, it'd be great to ensure they're
> enumerated as clearly as possible.
>
> I could suggest changing "The solution *may* address..." to "The solution
> *will* address..." unless we're unsure if the MoQ WG will address all of
> these use cases?
>
>
I think that "may address use cases" is there because the later phrase
starts with "including" and there may also be some use cases that are not
eventually in scope (e.g. gaming over store-and-forward systems based on
BUNDLE, to enable terrestrial and non-terrestrial gamers to play together).


> Does *will scale** efficiently* imply cacheability, or is there something
> else to ensure efficient scale?
>
>
I think the references to relays, caches, and replication points,
referenced later in the charter, areTh the more concrete aspects the WG
knows it wants to work on, but imagine it wouldn't turn down other ideas
that are presented as the work progresses.

The charter looks very good to me overall, so these are nits, but I wanted
> to bring them up.
>
> Thanks for the close reading,

Ted

*The meaning of "core system" is left as an exercise for the reader.




> Thanks, Ian
> --
> Moq mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/moq
>