►
From YouTube: MOPS WG Interim Meeting, 2020-10-30
Description
MOPS WG Interim Meeting, 2020-10-30
A
B
A
F
B
G
D
B
D
H
I
I
did
I
finally
figured
out
how
to
make
webex.
Mostly
let
me
change
my
name,
I'm
quite
pleased
with
that
yeah.
How
do
you
do
that?
Because
I
need
to
do
that,
so
what
I
did
is
I
went
through
my
browser
and
I
removed
all
cookies
associated
with
webex
and
now,
when
I
start
webex
before
it
before
I
joined
the
meeting
on
the
top
left
of
the
corner,
there
is
a
like
it
has
your
name
with
a
little
pencil
icon
and
you
can
change
it
before
I
deleted
my
cookies.
H
I
H
H
D
I
know
I
learned
that
for
the
for
the
ietf
presenter
account
to
log
in
in
an
in
in
an
incognito
window,
because
otherwise,
like
I
will,
I
am
ietf
mops
forever
and
I
haven't
been
able
to
figure
out
how
to
how
to
stop
that.
From
being
true,
you
know
erased
cookies.
Everything
right,
you
know,
deleted
all
the
local
browser
store
and
nonetheless,
every
meeting
I
joined.
I
was
ietf
mops.
I
I
They've
obviously
never
joined
the
pre-iesg
tiller
chat
thing,
which
is
I
mean
this
is
hugely
intelligent.
Compared
to
that.
That's,
like
five
minutes
of
dribble,
about
making
coffee
or
like
a
bird
I
saw
outside
my
window
or
something
or
about
infected
mushroom,
hey
that
that's
reasonable
and
intelligent
conversation.
H
D
C
D
D
H
D
C
D
D
J
J
B
B
Great,
you
still
hear
me:
okay,
good,
okay,
so
the
there's
two
issues
that
have
been
that
I've
gotten
some
updates.
I
I
plan
to
spend
the
well.
Maybe
it
won't
take
very
long,
but
one
of
them
had
some
scope,
questions
raised
and
so
we'll
go
over.
That
I'll,
give
an
update
on
the
solicitations,
we're
seeing
or
not
so
much
and
go
over
the
kind
of
list
of
open
issues
and
seek
some
comments
so
yeah
next
slide.
B
The
short
answer
on
our
milestone
status
is
that
we're
behind.
We
were
supposed
to
be
aiming
to
go
into
last
call
this
november.
I
I
gather
at
least
from
my
side.
This
has
been
severely
impacted
by
the
sort
of
crazy
world
we
live
in
this
year
and
the
the
demands,
the
unusual
level
of
demands
coming
out
of
that
you
know
I
I
do
think,
there's
still
interest
in
this
topic.
I
think
we've
seen
you
know,
certainly
on
my
side
and
and
from
at
least
some
of
my
co-authors.
B
I
I
hope,
I
hope
all
of
us
and
the
so
I
I
think,
there's
still
the
will
to
to
finish
this
up,
but
we're
gonna
have
to
move
the
milestone,
so
I'm
thinking
march
or
july
would
be
reasonable.
I
know
for
myself
I'm
still
planning
to
have
an
unusual
level
of
demands
on
my
time
through
about
november
through
about
february,
at
least
and
and
perhaps
into
march,
but
I
would
nonetheless
aim
to
carve
out
some
time
to
work
on
this.
B
C
D
J
B
B
B
The
the
opinion
of
several
of
us
was
that
there
are
things
we
could
say
that
would
not
be
overreaching
the
scope
of
this
working
group,
so
we
put
together
a
pull
request
with
some
proposed
text,
got
some
good
comments
on
it.
Thank
you
to
all
those
that
made
these
comments,
and
I
don't
know
if
we
want
to
actually
pull
it
up
and
look
at
this
in
the
context
of
this
discussion.
D
B
Sure,
but
the
context
here
is
really
about.
Like
you
know,
I
think
what
I
tried
to
do
is
is
just
sort
of
say
that
these
are
appropriate
technologies
given
different
domains
of
of
latency
targets
and
not
really
give
any
recommendations
on
how
to
use
it
or
much
in
the
way
of
it.
Much
beyond
just
pointers
to
the
technology,
just
to
sort
of
bring
it
up
as
something
that's
useful
for
someone
who's
trying
to
operate
this
kind
of
a
media
service
or
or
interact
with
it
in
some
cases.
So
I
think
it's
reasonable.
B
F
While
the
issue
come
out,
maybe
as
one
of
the
person
that
was
feeling
strongly
about
this,
I
wanted
to
say
that
the
new
text
is
going
in
a
great
direction
and
I,
like
it
very
much,
I
think
this
thing
the
use
case
and
which
protocol
could
apply
very
in
a
neutral
way,
is
the
direction
to
go
great.
Thank
you.
B
D
D
B
B
I
think
the
the
sort
of
one
that
matches
most
closely
with
the
questions
about
scope
is
the
one
that's
right
here.
This
would
be.
This
would
be
the
the
kind
of
ultra
low
latency
portion
of
the
latency
considerations.
So
here
we're
talking
about
yeah.
For
me,
this
is
still
a
bit
tight
to
read
on
the
screen,
but.
D
B
B
I'll
be
touching
that
up
the
bit
that
that
kind
of
so
what
I'm
trying
to
do
here
is
sort
of
describe
context
for
the
kinds
of
use
cases
where
this
we're
trying
to
have
sub-second
glass-to-glass
at
media
latency
is,
is
commonly
encountered
and
and
comes
up.
This
touches
on
things
that
are
not
necessarily
about
the
kind
of
media
broadcast
stuff
that
I
understand
to
be
more
the
focus
of
this
group,
but
also
gets
into.
B
B
I
think
I
did
have
a
mention
of
having
to
in
somewhere
of
like
talking
about
the
d
jitterbug
buffer
with
a
reference
to
an
appropriate
rfc
for
it,
but
did
not
like
dive
into
details
about
doing
that
or
make
any
recommendations
for
it
just
sort
of
flag
it
as
when
you're
trying
to
do
this.
Probably
you
should
take
a
look
at
this
kind
of
stuff.
B
H
Well,
so
the
one
input
I'll
make
is,
I
know
if
people
have
been
following
there's
been
a
discussion
over
on
dispatch
involving
a
proposal
there
for
doing
a
media
ingest
or
effectively
webrtc,
which
is
you
know,
the
real
use
cases.
I
understand
it
is
really
around
this
ultra
low
latency
ingest
mechanisms
so
that
they
can
do
live
streaming
of
things
like
games
through
webrtc,
and
so
we
might
wanna
cross
reference
here
or
link
up
somehow
this
discussion
and
and
make
note
of
that.
F
As
a
as
one
of
the
co-author
of
the
of
the
proposal,
I
can
maybe
give
a
few
details.
This
is
just
a
signaling
and
control
protocol.
So
what
about
this
here?
Rtc
web
at
ietf
had
only
the
the
media
stack
and
the
encryption,
but
he
didn't
have
the
signaling,
so
people
could
not
use
it
like
rtmp
or
srt
or
wrist.
You
need
to
have
a
proprietary
control
or
signaling
protocol
that
comes
with
it.
F
You
can
see
it
as
the
same
thing
as
rtsp
and
rtp
rtp,
taking
care
of
the
media
and
rtsp
taking
care
of
the
control.
So
when
we
check
the
charter
of
mobs,
the
control
of
the
media
transport
protocol
was
out
of
scope.
So
that's
why
we
put
it
at
this
path
and
it
looks
like
in
concept
it's
the
same,
an
rtsp
that
was
designed
within
the
m
music
group,
so
we
suggested
to
the
to
dispatch
that
that
might
be
a
good
working
group
to
land
further
discussion
and
further
work
on
it,
but
yeah.
F
B
Right,
some
of
the
comments
noted
that
there
was
one
comment
about
like
this:
doesn't
reflect
the
sort
of
ongoing
changes
to
what's
going
here.
So
I
think,
there's
going
to
be
an
update
that
tries
to
fold
some
of
that
in,
and
there
also
is
a
good
thread.
I
think
it
might
be
this
one
further
down
where
maybe
raised
the
idea
that
we
need
another
issues:
yeah
that
was
the
yeah
yeah.
B
If
you
try
to
use
an
ip
network
to
do
the
transport
for
these
between
these
different
sort
of
working
stages-
and
you
know
again
which
which
technologies
are
commonly
used.
What
are
the
pros
and
cons
that
you
find?
What
are
the
kind
of
failure
modes
that
you
hit?
So
I
think
that
was
a
really
good
idea.
I
got
a
plus
one
from
spencer
on
it,
I
think
so
far
and
and
so
we'll
probably
be
raising
an
issue
to
take
that
further
as
well.
B
I
think
we're
still
in
the
stage
that
we're
generating
issues
faster
than
we're
closing
them,
but
yeah.
It's
that's
just
the
way.
The
way
writing
works,
I
guess
so
yeah.
I
guess
I'm
looking
really
for
like.
Does
anybody
think
that
this
does
overreach
our
charter?
And
if
so,
could
you
give
an
overview
of
why?
I
guess.
C
I
mean
my
personal
point
without
any
hat
right
now
we
can
say
first
as
an
author,
you
can
say
whatever
you
want
now,
it's
a
working
group
document.
So
it's
up
to
the
working
group
to
to
share
this
point.
But
it's
observation
right
as
long
as
you
don't
try
to
define
new
protocols,
we
are
fine.
If
we
think
there
is
a
gap
somewhere
and
there
is
a
new
protocol
needed,
then
we
can
do
something
like
this
patch.
Basically
and
then
this
part
should
be
the
real
working
group
to
do
it.
C
D
I
mean
this
this.
This
seems
to
be
more
like
more
like
defining
a
problem
statement.
Right
I
mean
it's
not.
I
don't
think
it's
stepping
on
anyone's
anyone's
toes,
basically
saying
hey,
there's
this
problem
we
need
to.
We
need
to
solve
it,
here's
the
way
that
we're
formulating
the
problem
and
that
can
be
used
as,
as
you
know,
input
to
a
work
to
a
technical
working
group
that
is
actually
going
to
start
driving
a
solution
to
it.
J
J
Do
a
quick
mental
revisit
of
all
the
discussions
through
the
chartering
phase
of
this
working
group
and
and
while
I
certainly
embrace
eric's
perspective,
I
would
like
to
make
sure
that
we
don't
wind
up.
Looking
like
we
are
trying
to
do
artwork
in
the
in
the
in
the
ops
area,
and
that
was
part
I
think,
of
the
concern
about
whether
we
were
going
to
get
into
real-time
communications
or
not.
J
C
That's
basically
a
kind
of
a
flashing
light,
a
reminder
don't
have
a
step,
but
we
are
far
away
from
the
stepping,
and
I
see
that
barry
the
rtd
is
in
the
cold.
So
if
we
were
completely
overstepping
we're
stepping,
I'm
pretty
sure
that
barry
would
have
said
something
all
right.
You
want
to
say
something.
J
F
If
need
be,
what
have
been
seen
done
in
different
groups
was
to
make
a
separate
chapter
that
was
explicitly
listed
as
non-normative,
and
so
that
would
be
completely
explicit
and
separated
from
the
rest
of
an
otherwise
normative
document.
But
I'm
not
sure
I
think
it's
clear
in
the
current
wording
and
that
might
not
be
necessary.
D
J
B
E
Sorry,
if
I'm
overstepping
here,
I
don't
know
the
usual
flow,
but
I
did
want
to
just
kind
of
clarify.
The
document
that
we're
working
on
now
seems
to
be
oriented
largely
around
current
industry
practice.
Is
that
what
we're
kind
of
attempting
to
do
is
kind
of
destroy
here's
the
status
quo?
Here's
what's
actually
happening
and
here's
you
know
the
tensions
that
we
see
you
know
going
forward
and
maybe,
if
we
specifically
around
protocols,
say
here's
the
protocols
that
are
in
use
today.
B
Yeah,
it's
it's
a
it's
a
good
question.
I
think
there's
a
lot
of
things
about
the
future
that
we
just
can't
know.
So
this
is
necessarily
gonna,
be
a
snapshot
of
some
sort,
but-
and
maybe
we
should-
we
should
make
that
more
clear.
The
goal
with
making
rfc,
as
I
understand
it,
an
informational
rfc,
is
to
make
something
that
would
have
some
archival
values.
So
you
don't
want
anything,
that's
going
to
go
immediately
out
of
date.
B
B
B
There
was
so
we
you
know.
I
I
gave
that
a
quick
review.
It
looked
like
it
covered
some
good
stuff.
I
thought.
Maybe
you
know
with
regard
to
the
interest
in
video
that
certainly
in
a
lot
of
the
kind
of
dash
related
presentations
and
some
of
the
recent
hls
work,
there's
been
some.
You
know
some
discussion
of
cache
footprint
with
regard
to
using
fragmented
mp4
as
opposed
to
ts
segments
and
how
much
it
helps
for
this
kind
of
thing.
B
B
B
B
B
So
I'm
looking
to
get
some
at
least
a
couple
of
people
active
in
the
industry
to
add
some
comments
with
their
knowledge
in
it
and
to
get
some
review
on
this
just
through
the
back
channel
connections
that
I've
got.
I
hope
a
couple
of
the
co-authors
would
do
the
same,
but
I'm
not
sure
if
we
need
to
reach
out
to
others
through
some
other
venues
and
sort
of
highlight
this.
B
I
don't
know
maybe
get
a
five-minute
slot
on
a
couple
of
other
conferences
and
and
tell
people
that
we're
trying
to
write
this
thing.
We're
soliciting
feedback
from
people
who've
been
operating.
These
things
and
we'd
love
to
hear
your
stories
of
of
the
issues
you've
encountered
that
had
to
do
with
networking
so
that
we
can
document
them
and
get
them
out
to.
B
J
So
I've
mentioned
it
a
couple
of
times.
I've
mentioned
the
mops
working
group
a
couple
of
times
on
the
streaming
video
alliance
network
working
group
who
one
of
whose
chairs
is
actually
conveniently
on
this
call
as
well,
and
I
think
that
if
we
think
we're
at
a
point
where
we
have
specific
questions
where
we
we
could
point
people
out
the
document
and
say
you
know,
could
you
provide
a
review,
particularly
from
the
lens
of
x?
I
think
that
that's.
J
H
Well,
yeah,
I
mean
we
certainly
can
include
that
leslie
and
get
the
word
out
there.
I
I
think
that
strong
linkage
between
the
two
groups
is
something
we
need
to
promote
and
that
it's
a
good
example
of
a
way
to
do
it.
We
also
may
want
to
poke
at
a
few
different
places,
so
I
I
mean
ali,
maybe
there's
a
way
we
can
throw
something
up
on
the
upcoming
mile
high
video,
maybe
there's
like
a
break
or
something
you
can
throw
up,
and
just
let
people
know
that.
K
E
J
B
B
Yeah
so
anyway,
the
the
submissions
we've
gotten
filling
out
the
template.
Maybe
the
template
is
not
right
like
that,
might
be
off-putting
to
people.
It
asks
maybe
too
many
things,
I'm
not
quite
sure,
but
for
whatever
reason
we
haven't
gotten
a
lot
of
it.
So
you
know
if
we
started
getting
a
trickle
of
those.
That
would
be
great
anybody
who
can
make
them
happen.
Please
do.
B
All
right
and
let's
go
to
the
last
slide,
then,
which
is
the
an
overview
of
the
open
issues,
so
I
put
green.
We
think
we're
about
to
close
these
we've
got
some
proposed
text
submissions
I
haven't,
put
the
the
other
new
one.
We're
gonna
be
trying
to
add,
based
on
the
feedback
for
the
what
was
it
having
the
the
stages
of
the
workflow
and
their
their
networking
considerations,
but
I
think
that'll
be
useful.
These
are
the
other
open
issues.
I
think
there's
a
one
more.
B
I
also
didn't
mention
it's
more
about
just
formatting
in
the
in
the
text
of
the
draft.
That's
you
know
just
sort
of
editorial
how
we
would
you
know,
interact
better
with
the
with
the
tooling
and
the
diff
process
and
stuff
in
the
in
the
markdown,
but
the
rest
of
these
are
about
content
that
we
intend
to
get
folded
into
this
draft
in
some
form
the
one
that
stewart
mentioned,
spencer,
one
that
spencer
mentioned.
B
There
right
sure,
apparently
it's
even
earlier
than
I
thought,
yeah
one
that
spencer
mentioned
he
he
would
love
to
raise.
Ask
if
anybody
just
sort
of
has
an
easy
path
here
is
when
spencer
added
a
section
on
you
know
the
the
kind
of
unexpected
outcome
of
bandwidth
when
covid
lockdowns
kind
of
started
happening.
He
referenced
a
page
documenting
some
of
the
issues
there.
That
was
just
sort
of
put
up
by
at
t,
but
the
part
that
he
was
referencing
was
like
way
down
the
page.
B
It
was
a
very
long
document
with
new
things
being
added
in
the
front,
and
you
know
there
weren't
sort
of
subsection
references
that
we
could
insert
there.
So
you
know
he
said
that
he's
not
sure
he
can
pull
that
same
reference
without
getting
att
to
kind
of
tweak
that
page
a
little
bit
to
make
it
easier.
It
also
might
be
possible
to
just
sort
of
quote
some
text
to
look
for
and
and
leave
it
alone.
B
We're
also
not
really
sure
like
how
stable
is
this
reference?
I
think
so.
I
guess
the
main
question
was
like
anybody
know
who
we
can
where
to
find
somebody,
and
I
think
the
goal
there
would
be
to
save
some
of
us
authors
a
few
days
worth
of
trying
to
track
down
the
right
person
to
talk
to
or
just
give
up.
We
might
just
decline
it.
I
guess
the
issue
anyway.
If
anybody
knows
the
right
contact,
please
send
it
to
to
mayor
spencer.
B
You
know
to
yeah
authors
list
or
something
like
that
offline,
and
these
are
the
open
issues
we
think
once
we
get
these
taken
care
of,
then
the
the
goal
is
that
the
doc
would
be
ready
to
go
and
we
don't
know
of
any
others
that
we
need
to
cover,
although
that
may
of
course,
change
as
we,
you
know,
try
to
write
things
and
find
out
we're
missing
other
things,
further
yeah.
So
that's
current
status.
Any
questions
or
comments
about
this.
J
H
B
J
No
well
so
I
guess
I
I
haven't
looked
at
the
source
material,
so
I
don't
know
how
outrageous
this
suggestion
actually
is.
I
mean
my
thinking
was
as
much
like
I
mean
it
can
even
be
referenced,
it's
not
like
it
and
I'm
not
suggesting
theft,
but
if,
if
you
want
to
actually
incorporate
perspectives
that
have
been
perceived
as
I
don't
know,
and
and
then
maybe
it
becomes
the
thing,
that's
defeats
the
whole
document.
So
maybe
it's
not
worth
it.
B
B
B
B
I
think
it
was
citing
some
statistics
that
came
out
right,
so
it
the
way
that
it
jumped
from
from
a
smaller
bandwidth
utilization
average
to
a
bigger
one
when
the
lockdown
started
or
something
like
this
right.
So
there's
like
a
particular
date
that
was
picked
and
by
now
I
think
it
it
may
have
scrolled
off
the
page.
Perhaps
so.
This
is
the
kind
of
like
unstable
reference
that
we're
kind
of
looking
at
here.
So
we're
looking
to
get
a
more
stable
reference
and
we
think
april.
J
B
C
B
Yeah,
I
I
think
the
the
section
that's
there
just
went
through
several
different
sort
of
separate
references
and
gave
them
as
as
examples
of
things
that
have
been
observed
and
publicly
reported,
and
just
you
know,
providing
the
link
to
it.
Anybody
that
wants
to
follow
the
link
is
going
to
have
a
hard
time
looking
at
it,
but
I
guess
it
doesn't
really
matter.
We
can
just
say
what
we
were
pulling
out
of
it,
give
our
source
and
that's
that's
good
enough.
Okay,
that's
fine!.
H
B
J
B
J
B
B
B
You
know,
but
I
haven't
heard
whether
he's
you
know
willing
to
take
on
the
rest
of
the
author's
stuff,
I'm
kind
of
assuming
the
absence
of
follow-up,
probably
not,
but
nonetheless
I
think
it's
still
a
good
idea
to
add
a
section
on
ads
and
we'll
probably
be
doing
that.
But
I
don't
know
I'll
reach
out
to
him
again.
I
think,
but.
B
J
I
think
that
that's
always
a
slippery
slope
and-
and
I
would
suggest
for
practical
purposes-
is
just
personal
suggestion
and
my
co-chair
is
welcome
to
disagree
with
me
and,
and
anyone
else
take
the
suggestions
if
they
look
that,
like
they're
useful
and
put
him
in
the
acknowledgement
section.
If,
if
you
succeed
in
reaching
out
to
him
and
find
that
he's
willing
to
share
the
load
of
authorship,
not
just
of
this
section,
then
that
could
be
a
separate
discussion.
B
Okay,
that
sounds
good
yeah.
I
certainly
think
that
that
there's
a
lot
of
value
in
adding
something
and
going
over
the
the
you
know
the
considerations
about
advertising
with
regard
to
operating
video
services,
and
especially
you
know
where
it
impacts
networking.
I
think
that's
a
great
suggestion.
Absolutely
the
text
of
it,
I
think
needs
needs
some
work,
a
little
bit,
the
one
that's
been
posted,
but
I'll
just
kind
of
take
the
input
as
advice.
I
think
it
falls
under
the
note.
E
B
Okay,
yeah,
I
I
don't
think
I've
met
him.
I
think
it's
a
it's.
You
know
valuable
discussion
worthwhile
point,
I'm
not
sure
you
know
we
we'd
have
to
have
a.
I
think
we
do
have
space
for
another
author.
If
he's
really
interested
in
becoming
an
author,
I
don't
know
with
the
co-authors
whether
you
know
we'd
have
to
we'd
have
to
come
to
some
kind
of
consensus.
So
if,
if
he
wants
to
do
that,
we
should
have
an
offline
discussion
about
it.
B
Further,
that's
one
of
the
stopping
points
on
the
pull
request
as
it
stands,
but
there
would
be,
but
there's
another
couple
of
things
to
to.
You
know
change
about
the
text,
but
yeah
I
mean
the
the
contributions
are
very
welcome
and
much
appreciated.
Certainly
so
I
appreciate
you
passing
that
along.
Thank
you
mike
yep,
hey
all
right!
So
if
there's
nothing
else,
then
I'm
done
with
this
bit.
J
J
B
J
B
J
J
H
J
H
Yes
and
it
will
be
and
we'll
be
presenting
results
from
comcast
into
it
right
by
the
way,
video
plays
a
pretty
big
role
in
the
networks.
No
surprise
since
coveted
it
remains
the
number
one
transport
you
know
the
number
one
thing
transport,
even
when
you
take
into
account
the
broken
you
know
stuff
like
what
you
know.
Virtual
meetings.
J
J
I
I
I
or
kyle
have
yet
to
reach
out
to
them
to
see
if
they
would
be
willing
to
talk
about
it,
they
seemed
interested
when
we
were
in
the
throes
of
the
108
session.
So
I'm
optimistic
that
will
happen
and
then
also
from
108.
We
have
the
media
operations
use
case
for
an
augmented
reality:
application
on
edge
computing
infrastructure
by
rinan,
krishna
draft
krishna
mob,
say
our
use
case.
J
So
I'm
I'm
hopeful
we'll
get
that
on
the
agenda.
So,
as
eric
said
in
the
chat,
it's
we're
certainly
open
to
people
bringing
in
descriptions
of
work
being
done
elsewhere
or
other
experiences
in
video,
so
on
and
so
forth.
So
yeah
be
creative,
make
suggestions
and
happy
to
hear
any
thoughts
on
how
we
can
best
shape
the
agenda
for
109.
B
B
J
F
Do
you
do
we
want
to
put
some
time
for
new
groups
at
the
ietf
food
charter
could
be
of
interaction
with
too
much.
I'm
thinking
specifically
of
the
new
s
frame
group,
which
is
media
encryption
independently
of
the
transport
protocol,
or
maybe
web
transport
that
just
been
created
with
the
willow
from
akamai
as
a
chair.
Is
that
some
something
that
would
be
interesting
to
the
group.