Re: [Bloat] fcc input

[Matt Mathis <mattmathis@google.com>]

[-- Attachment #1: Type: text/plain, Size: 2676 bytes --]

I  encourage you to collectively think about a different strategy for
framing the problem.    Rather than talk about BB, and then forever looping
back and try to explain why it is important to non-network folks, develop a
tool (or a skin for an existing tool)  that estimates the suitability of a
link to carry video conferencing  and gamer traffic, using the
language from those communities (e.g. "lag").

No explanation is needed to tell a gamer why lag is bad.
No explanation is needed to tell a musician why lag is bad for a
distributed ensemble.
Very little explanation is needed to tell a VC user why lag is bad,
since these days everybody has experienced it.

Keep the up front the language at the application layer, but down inside
reveal that the largest culprit is often excess network queueing delay
under load, AKA buffer bloat.

Although a well framed tool could have high impact, you would get even more
bang for your buck to have a semi-standard jitter metric that could be
grafted into any real time application.  The goal would be to connect the
observed (and measured) application lag to the underlying network jitter,
queueing delay and bufferbloat.

Thanks,
--MM--
The best way to predict the future is to create it.  - Alan Kay

We must not tolerate intolerance;
       however our response must be carefully measured:
            too strong would be hypocritical and risks spiraling out of
control;
            too weak risks being mistaken for tacit approval.


On Sat, Mar 27, 2021 at 1:54 AM Jan Ceuleers <jan.ceuleers@gmail.com> wrote:

> On 26/03/2021 20:09, Dave Taht wrote:
> > I have often thought hard about trying to explain things better via
> > animations, and on more than one occasion tried to find an animator
> > who could turn this old 8m talk from "people as packets" into "unruly
> > animals as packets".
> >
> > https://www.youtube.com/watch?v=Rb-UnHDw02o&t=1657s
> >
> > But that requires both money and time, and I've usually had neither.
> > Grant money would be nice for a whole string of educational videos,
> > but not being part of the MICA complex has made that difficult. If I
> > ever get caught up on bills, and some other pressing problems, perhaps
> > we could do a kickstarter campaign to get some things animated.
>
> I have often admired Randall Munroe's ability to grasp and satirise
> sometimes quite complex concepts. Perhaps he could be prevailed upon to
> draw a bufferbloat-inspired cartoon? He publishes his work at xkcd.com
> under a Creative Commons license.
> _______________________________________________
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat
>

[-- Attachment #2: Type: text/html, Size: 3825 bytes --]