►
From YouTube: IETF98-RTCWEB-20170328-1640
Description
RTCWEB meeting session at IETF98
2017/03/28 1640
A
F
For
those
of
you
who
were
not
expecting
that
announcement,
we're
hoping
to
close
this
working
group
in
Prague,
so
please
get
ready
for
the
ride
between
now
and
Prague
to
involve
you
reviewing
some
documents.
You
applauding
our
document,
authors
and
editors
and
you're,
generally
speaking,
being
ready
to
finish
this
up
so
you're
in
chicago.
That's
Chicago!
So
are
you
if
you're
not
meaning
to
be
in
Chicago
or
in
the
RTC
web
meeting?
Now
would
be
a
moment
to
head
to
the
exits.
F
This
is
the
note.
Well,
if
you
don't
understand
the
note
well,
please
find
your
friendly
neighborhood
lawyer
or
your
corporate
lawyer,
your
choice
and
have
them
explain
it
to
you?
Basically,
if
you
don't
understand
it
getting
up
at
the
microphone
likely,
not
a
good
idea,
you
may
also
find
getting
up
at
the
microphone,
not
a
good
idea.
In
other
circumstances,
that's
not
the
problem
of
the
note.
F
Well,
the
working
group
status
update
look
two
thumbs
up
on
draft
RTC
web
overview
going
to
the
iesg
boo
and
we've
had
two
working
group
last
call
versions
of
RTC
web
GF,
which
is
our
major
topic
for
today,
two
thumbs
up
for
the
two
new
versions
as
well:
here's
our
agenda,
administrivia-
you
are
here
before
we
move
on
from
here,
guess
what
we're
going
to
need?
No
takers,
jabber
relays,
possibly
someone
to
meme
the
meeting
if
you'd
like
to
volunteer
for
any
of
these
roles,
please
raise
your
hand
now.
F
B
F
B
So
the
good
news
of
our
dependencies
is
that
list
is
about
half
the
length
that
we
were
lost,
the
time,
okay
and
so
most
of
the
documents
so
that
we,
actually
you
know,
normatively
depend
on
across
several
groups,
have
moved
to
the
state
of
you
know
up
requested
or
are
further
along,
which
is
really
good.
B
That's
the
status
of
other
things
that
we
depend
on
that
are
not
yet
rfcs,
but
you'll
notice
that
a
crapload
of
those
are
in
the
RFC
editor
queue
in
the
sense
that
they
are
approved
and
just
basically
blocked
waiting
for
references.
So
once
the
rest
of
the
stuff
approves,
you
know
the
the
the
cluster
from
area,
41
will
Buster.
F
G
B
H
C
The
next
one
is
the
stick,
that's
very
true
che
tu,
che
yeah
I,
think
I
really
needed.
Actually
the
sticks
more.
I
C
F
D
So
this
view
of
her
and
tell
us
will
remember
how
woefully
unprepared
I
was
for
that
on
imagine
home
prepared.
I
am
for
this,
given
that
most
of
my
time
doing,
TLS
slides
one
of
these
slides
I
fact
only
saw
about
two
minutes
ago.
D
Exactly
next
slide,
so
we
did
19
we're
at
19.
We
had
18,
you
see
how
this
works.
There
was
a
bunch
of
working
of
us
call
comments.
We
made
a
bunch
of
changes
I'm
here
to
summarize
them
I,
don't
believe,
actually
any
open
issues
to
be
open
about
so
I'm
really
summarizing
on
these
things
were
not.
Some
of
these
things
were
numb
who
could
list
some
work,
your
job
to
figure
out
which
ones
I
will
stop
briefly
after
each
thing
to
see.
D
If
anybody
is
very
upset
next
slide,
we
will
not
be
discussing
appendix
B.
Appendix
B
has
been
by
consensus
of
people
who
care
kicked
over
to
predict
art
and
music.
He
merges
into
bundle
and
there
and
wordsmith
there
uh-huh.
So,
if
you're
hoping
to
see
me
and
Magnus
argue
about
our
DP
DX
muxing
you're
in
the
wrong
place,
that
will
be
somewhat
other
time.
So
next
slide,
so
there
was
some.
There
were
some
questions
about
these.
D
The
right
answer
was
that
for
consistency,
and
also
for
simplicity
that
you
basically
only
said,
bundle
only
a
MUX,
only
a
new
em
sections
and
there's,
and
so,
however,
you
do
say
on
weight
on
in
this,
and
the
answer
says:
bundle
or
box,
as
the
case
may
be,
he
doesn't
say
only
ever
on
so
on.
This
was
included
as
best
thing
to
do,
and
this
is
what
we
have
done.
D
D
Yes,
okay,
yes,
basically,
if
there
is
a
that
the
right
way
to
think
about
this
is
if
there
is
actually
some
question
about
whether
this
thing
could
be
treated
differently,
we
will
tell
you
and
if
there's
not,
we
will
not
next
slide,
also
in
the
bundle
theme
I'm,
not
sure
if
these
are
actually
sorted
by
theme
or
just
happened
to
be
this
way,
but
this
is
how
they
are
on.
D
So
there
is.
This
question
appeared
about
whether
how
I
should
handle
on
our
attributes
the
only
made
sense
for
media
sections
like
RTC
mocks,
but
that
appeared
in
situations
where
you're
a
bundling
the
thing,
a
media
section
onto
a
data
section.
So
one
way
that's
going
to
happen
is,
if
you
start
out
by
negotiating
a
data
section
as
you
have,
the
data
session
could
be
negotiating,
you
add,
say
video
m
line,
but
you
wish
to
bundle
it
onto
the
on.
D
You
wish
to
bundle
it
onto
the
datum
section
and
I
guess
by
us
already
box
is
irrelevant,
but
others
letarte
on,
and
so
we
say
otherwise
that
these
meet
specific.
These
these
transferred
end
of
attributes
should
only
go
on
the
thing
that
is
being
bundled
into
a
nowhere
else,
but
it's
just
kind
of
goofy
that
there
on
the
data
section,
we
concluded
that
goofy
was
better
than
non
goofy,
and
so
we
should
do
that.
There's
a
bit
as
a
discussion
on
the
music
list
about
this
topic.
C
D
Agreeing
that
we
could
do
this,
we
will
need
to
presume
text
updates
in
a
variety
of
documents
which
will
notice
the
last
call's.
We
go
forward
noticing
that
says
things
that
contradict
this
arm,
but
that
was
the
way
we
did
in
jsf
with
the
plan
that
we
do
that.
D
At
that
point
you
were
like
could
only
do
B
and
you
couldn't
offer
see
and
and
and
a
riad
right-
and
this
isn't
had
to
be
simplicity,
but
apparently
make
people
sad
day
had
situations
where
they
wanted
to
add
them
back
it's
permitted
by
30
days
before
so
we
weren't,
like
you
know
we
weren't
doing
anything
goofy
here
by
allowing
it
on
so
we
changed
that
back
on,
but
the
rule
is,
you
have
to
add
them
at
the
end.
The
reason
that
I'm
at
the
end
is
to
pro
stability
of
the
negotiated
mechanism.
L
D
There
is
no
prob
for
it,
I
don't
believe.
Are
you
try?
Twitter?
Yes,
so
if
it
I
believe,
if
you
were
the
person
who
is
responsible
for
removing
it,
then
set
color
preferences
might
have
some
effect
on.
However,
the
other
person
was
that
it
will
not
on
so
this
should
have
a
frog
for
it
on
you,
as
the
person
making
the
offer
at
this
point,
no
matter
who
was
the
original
offer
answer.
D
D
D
D
Okay,
this
is
my
last
working
for
the
day
and,
let's
just
say,
this
ad
thing
is
tiring.
Okay,
right,
I
think
this
is
the
one
that
actually
like
Colin,
just
added
exciting
to
do
it
right
on
so
right.
We're
now
allowing
you
to
on
to
do
credit
references
and
have
them
to
have
them
apply
in
the
middle
of
a
thing
as
previous
my
colloquy
Bernard
previously
indicated,
we
didn't
just
allow
that
we
thought
about
it
for
a
while
glue
to
probably
work.
D
Now
we
allow
it
that's
what
I
like
to
say
about
that.
D
Right
camera
raise,
this
may
be.
D
His
old
mate
was
/
nard,
one
of
those
junior
pickers
that
that,
basically,
what
happens
when
an
API
call
fails,
and
you
know
what
what
guarantees
do
deliver
and
let
me
tell
you
as
someone
who
actually
wrote
these
things,
the
guarantees
we
really
wanted
to
liver
were
none
boys,
the
CC,
an
undefined
behavior,
but
we
eventually,
after
some
drinking
and
coffee
included
that
really
that
wasn't
very
cool,
and
so
it
now
says
something
the
implementers
are
going
to
hate,
I'm
sure
the
people
who
do
hurry
this
code-
it
was
a
lot-
will
come,
come
we
later,
which
is
that
if
these
calls
fail,
you
have
to
go
back
to
the
previous
tape,
so
it
has
to
be
basing
is
if
the
thing
never
happened.
D
D
May
be
the
case,
I
I,
he
does
know
where
I
sit.
Unfortunately,
ok
next
slide.
Is
that
really
it
well?
I
was
like
pretty
good
we're
like
16
minutes
in
so
I
guess
the
question
becomes
now.
Are
there
any
other
issues
anybody
like
to
worry
about
jcf,
because
if
not,
we
will
have
held
that
you
forfeited
the
right
to
raise
those?
No
no
wait.
Wait.
First,.
F
We're
just
now
we're
just
trying
to
make
sure
you're
paying
attention
here
right,
because
in
those
16
minutes
some
of
you
drifted
off,
I
could
see
yogurt
being
consumed,
we're
really
in
a
serious
moment
in
those
people.
Despite
the
attitude
of
all
four
of
us
in
front,
this
was,
after
a
working
group,
last
call
after
the
second
working
group
last
call.
This
is
the
set
of
issues
that
was
raised
in
the
second
working
group.
Last
call
the
authors
and
the
chairs
believe
them
to
be
all
resolved.
F
F
I
Ok,
this
is
Easter
Mike,
ok
shortly.
Ok,
this
is
for
the
stp
example
staffed
and
I.
Have
this
next
slide
show
yeah?
We
know
like
this
is
about
trying
to
come
up
with
some
examples
that
uses
Jessup
and
try
to
be
as
close
to
Jessup
as
possible
in
terms
of
stp
construction,
and
it
provides
someone
who
is
trying
to
read
how
the
stp
is
used.
Some
references
to
artificer
that
they
can
figure
out
how
a
particular
part
in
sdp
is
used
next
slide.
I
Ok,
what
happened
between
0
to
10
for
ocean,
so
03
was
just
an
update
to
the
ocean
because
it
was
expiring.
So
04
is
the
thing
where
a
lot
of
things
went
in.
So
we,
like
Cullen,
had
presented
an
idea
of
96
cup
of
open
issues
on
how
some
of
the
identical
on
category
and
transport
category
attribute
should
appear
in
stp
and
a
gist
of
19
and
of
clarifies
every
of
those
things.
So
that's
good.
I
L
I
L
I
The
offers
and
ancestors
added
and
then
one
example
has
been
added
to
also
show
how
you
use
identity
attribute
in
stp.
Apart
from
that
new
edition,
that
happened
was
we
created
in
as
sdp
attributes
checklist
appendix
that
way?
What
it
basically
does
is
it
go?
It
went
through
J
sub,
section,
J,
sub,
section,
5
and
compiled
list
of
requirements
that
we
feel
the
stp
attributes
in
the
example
should
match.
So
this
was
basically
done
so
that
we
can
align
to
Jessup
19.
I
F
I
F
F
No
two
more
reviewers
paul
can
review
again
if
he
wants,
but
I
want
to
reviewers
of
the
current
version.
So
we
have
like
20
minutes
of
silence
to
fill.
Your
chairs
will
sit
here
and
stare
at
each
one
of
you
until
two
hands
go
up
either
electronically
or
in
the
room
could
could
you
bring
up
the
jeopardy
theme
for
us?
Please
no.
I
mean
if
somebody
puts
their
little
hand
up
we're
willing
to
press
the
red
button.
If
what
they're
gonna
say,
as
we
volunteer
to
review
Oh.
F
J
F
Let
the
note-taker
show
that
two
volunteers
from
Mozilla
have
agreed
to
review
this
document
and
have
the
thanks
of
the
working
group
for
same.
Thank
you
very
much.
That
doesn't
mean
the
rest
of
you
are
going
to
get
off
of
it
because,
as
soon
as
they
put
their
review
in
it's
going
to
come
to
working
group
last
call,
which
means
you
all
get
your
wack.