►
From YouTube: IETF104-BCAUSE-20190326-1120
Description
BCAUSE meeting session at IETF104
2019/03/26 1120
https://datatracker.ietf.org/meeting/104/proceedings/
A
B
A
A
And
I'm
here
all
week
and
that's
quite
annoying
right,
so
this
is
a
buff,
but
all
of
them
note
well,
stuff
still
applies.
You
are
contributing
to
the
ietf
if
you
speak,
breathe,
twitch
or
anything
else
there.
If
you're
unsure
what
this
means,
please
go
and
read
about
it
or
ask
someone
and
take
advice
from
your
company's
agenda,
so
the
agendas
online
and
was
circulated.
We
haven't
made
any
changes
we'll
come
to
that
in
a
minute.
A
A
We
will
cover
what
the
purpose
of
the
prof
is
in
a
minute
and
our
ad
will
talk
to
us.
Our
blue
sheets
are
circulating
because
the
room
is
very
full
when
the
blue
sheets
reach
the
back,
could
you
pass
them
back
forwards
and
keep
them
circulating,
because
people
have
come
in
and
filled
gaps
when
you
speak
at
the
microphone
and
remember
that
other
people
do
not
know
who
you
are,
and
you
need
to
say
your
name
slowly
and
clearly
on
which
I
am
Adriene
and
Mrs
Barbara.
We
are
easily
confused.
A
No
no
I
mean
we
are
easily
confused,
not
you
are
confused
as
to
who
we
are
right.
It's
a
fairly
hot
topic.
People
have
strong
views,
please
listen
to
yourself
and
be
civil.
Please
listen
to
other
people
and
be
civil.
Someone
may
not
understand
what
you're
saying.
That's
your
fault,
not
their
fault.
Please
help
them
understand
speakers.
Please
stick
to
your
time.
Limits.
I!
Think
we're
not
going
to
have
a
problem
with
that
and
we
have
discussion
time
at
the
end
of
the
agenda.
C
B
D
D
The
main
purpose
of
the
buff
is
that
those
that
come
on
the
mic-
those
that
we'll
present
please
have
the
room.
The
channels
and
me
understand
your
needs,
because
that's
on
that
basis
that
we
will
try
to
uncover
work
if
any
that
the
ietf
would
do
and
when
I
say
work,
I
really
mean
meaningful
work
for
the
industry,
we're
not
here
to
discuss
how
we
could
lose
the
time
of
the
iesg,
the
FCA
detail
and
so
on.
So
second
point
is
a
plea
to
all
of
you:
oh
yeah
I've.
D
D
D
It
is
a
very
important
context
to
integrate
and
I
have
been
very
clear
about
that
in
the
first
email
I've
sent
on
the
on
the
on
the
because
list,
providing
you
with
the
reason
J
zones
will
have
a
five
minute
talk
or
ten
minute
talks
from
David
to
give
you
more
information
on
that.
But
it
is
important
that
whenever
you
express
your
opinions,
you
intake
like
that.
In
that
context,
I'll
say
word
on
Barbara's
email,
so
I
for
putting
you
on
the
spot.
But
I
guess
you
did
yourself
so
I
told
Barbara.
D
D
A
So
this
slide
is
just
to
remind
you.
This
is
a
buff
and
there
is
an
RFC
about
buffs
you.
It
raises
a
number
of
questions
that
you
might
want
to
have
in
your
mind.
We
will
not
dwell
on
these
questions.
You
can
go
and
look
at
the
RFC,
but
these
are
questions.
We
will
dwell
on
a
little
bit
so
on
the
agenda
after
our
three
speakers.
We're
going
to
come
back
to
use
this
slide
for
open
discussion
and
we
don't
need
to
go
through
the
questions
one
by
one.
A
If
one
of
the
questions
is
something
that
sticks
in
your
mind,
then
come
to
the
microphone
and
and
talk
about
that
question.
Otherwise,
we'll
find
that
everybody
wanted
to
talk
about
the
last
question,
but
we
won't
get
to
it
in
time.
So
this
is
a
trigger
to
discussion,
and
the
answers
to
these
questions
will
help
the
ad
understand
what
to
do
and
what
the
feeling
of
the
room
is
too
many
for
us
to
read
in
the
time
and
I
think
we
would
be
better.
E
So,
just
to
give
you
a
background,
if
you
don't
know
what
a
BMG
is
is
basically
the
exits
router
for
residential
users
and
it's
basically
the
first
hop
for
residential
broadband
subscribers
in
China
most
network.
What
we
call
it
internal
mobile,
the
Internet
network
in
China,
Mobile
and
China
Mobile
was
a
cellular
network.
Centric.
E
Channel
Mobile
versus
seller,
now
it's
eccentric
operator,
but
we
have
150
million
residential
subscribers
that
has
that
figure
has
grown
up
from
about
15
million
and
in
2013
to
the
end
of
last
year,
the
figure
has
grown
to
150
million.
So
that's
a
huge
rolls
in
terms
of
a
subscriber
use
case,
a
subscriber
base
and
personally
I
actually
joined
China
Mobile
in
2013
in
this
home,
a
broadband
service
service,
I'm
team.
E
So
what
we
see
has
China
Mobile
has
actually
distributed
the
be
it
be
ng
or
beer
us
equipment
to
the
district
level
or
central
office
level
in
a
very
rapidly
growth
rate,
and
we
have
more
than
10,000
deployed
BMG
devices
and
most
of
our
subscribers
are
based
on
G
pong
technology,
95%
of
them
and
the
other
5%
is
not
a
DSL
or
anything
copper.
It's
a
pump
so
mostly
100%
our
fiber
optics.
E
E
Also,
we
see
and
the
address
management
has
caused
a
lot
of
time-consuming
maintenance
work
for
our
professional
companies,
its
ipv4
Bay,
so
we
have
inadequate
address
and
normally
what
we
do
is
to
give
advanced
planning
for
the
address,
pool
that
who
will
be
configured
to
the
distributed,
be
ng
devices,
but
we
have
really
poor
predictions
as
we
see
and
the
growth
rate
is
really
high
of
the
roasting
user
base.
So
a
lot
of
we
can't
actually
restore
the
the
IP
address
back
to
the
EMS
or
BMG
management
plane.
E
If
you
have
been
involved
with
the
because
mailing
this
discussion,
you
will
be
very
familiar
with
the
architecture,
basically
separate.
The
what
we
call
controlling
out
PNG,
but
it's
more
or
less
at
manage
play,
so
we
visually
separate
the
control
plane
and
the
data
plane
we're
in
the
in
the
centralized
control
playing
you
deal
with
the
session
management.
E
You
deal
with
address
management
and
the
policies
wherein
the
user
playing
the
distributed
user
playing
device
either
it
is
a
virtualized
function
or
it
is
a
physical
function,
deal
with
the
forwarding
plane
and
deal
with
the
pppoe
forwarding
and
etc,
etc,
and
in
Beijing
we
deployed
more
than
200
you
user,
plane.
Bng
function
managed
spend
only
one
control
play,
but
this
is
for
redundancy.
We
deploy
to.
E
So
China
Mobile
has
been
doing
this
since
2017,
and
some
standardized
work
has
been
carried
on
here
in
ITF
in
RT
gwg.
So
in
the
trial
we
carried
out
last
year,
we
deployed
V&F
for
C.
That's
of
course,
four
four,
four
control
playing
and
for
you
circling
we
have
two
versions:
one
is
for
the
session
greedy
one
because
it's
computing
greedy.
We
use
common
hardware
and
we
use
software
based
solution
for
that
controller.
E
We
see
a
40%,
better
performance
of
IP,
address
utilization
and,
of
course,
if
you
have
a
centralized
control
plane
where
you
can
manage
hundreds
of
your
user
plan
bng,
you
will
see
a
session
capacity
with
a
huge
growth
and
also
we
see
the
SUBSCRIBE
activation
rate
has
has
been
improved
by
10
times
and,
of
course,
the
provisioning
of
new
services
are
much
easier
and
one
thing
to
mention
that
is.
If
you
have
a
centralized
control
playing
with
software
solution,
you
have
even
you
have
a
better
chance
to
have
a
simpler.
E
E
So
next
step
we're
going
to
deploy
these
distributed
or
desegregated
BMG
device
in
this
year
and
at
this
deployment
is
commercial
deployment
and
we
will
upgrade
or
legacy
BMG
devices
in
the
coming
years.
We
would
expect
fifty
percent
to
be
upgraded
in
few
years.
So
when
we
talk
about
mass
commercialized
deployment,
where
we
talk
about
inter
cooperation
interoperation
between
vendors,
so
that's
why
we
are
here
in
9cf
to
discuss
the
possibility
of
bringing
the
standardized
work
to
this
interface
between
the
use
of
user
play
and
control
playing
of
the
a
BMG
and
the
progress.
E
F
F
But
what
we
prefer
is
that,
but
watch
nothing
what,
but
what
we
prefer
is
that
we
that
this
interpretation
is
open
in
with
the
inter
standards
within
providers
is
a
converged
well
fixed
wireless
or
mobile
networks,
and
it's
based
on
infrastructure,
whatever
this
will
be
in
the
future,
I
suppose
more
or
less
software-defined
Hardware
operations,
okay,
and
we
have
only
three
slides.
So
the
first
slide
we
want
to
show
I
have
a.
B
B
F
F
F
So
we
suppose
that
the
infrastructure
will
be
used
for
SS
or
SH
Y
aligned,
fixed
and
mobile
access
based
all
on
the
same
hardware.
So
it
should
support
a
very
low
layer
like
native
I,
a
the
net
access
and
all
of
them,
the
three
types
or
even
fam.
What
should
you
use
them?
For?
We
said
me
to
Ethernet
services
and
especially
what
my
background
is
the
wireless
mobile
part.
F
F
F
What
is
we,
that
is
a
TNT
and
also
a
dodgy
telecom,
are
working
on
proof
of
concepts
for
that
and
have
prototypes
not
so
many.
At
least
we
have
not
so
many
as
in
China,
because
we
have
much
less
subscribers,
but
we
try
to
build
and
prove
something
up
and
see
this
convergence
is
well.
We
talked
about
that
for
20
years
or
more.
This
is
a
slow
roll
app,
not
something
which
happens.
F
It
tend
to
work,
they
are
slower
yeah
and
what
we
think
that
it's
on
the
user
planned
that
do
you
supplying
functions
of
course
use
the
plane
is
where
the
most
traffic
is,
we
already
heard,
and
what
the
users
interested
in
that
the
traffic
is
transported
to
them
at
different
different
relationships,
different
service
level
agreements,
of
course,
also
doing
costs
and
prices
and
regulations
also
different
in
different
countries.
F
So,
therefore,
we
are
not
able
to
huge,
steady
change
in
the
user
plane.
This
should
be
something
which
is
constant
and
just
work
and
provide
the
service
yeah.
Okay,
when
this
is
the
deployment
next
year
or
after
next
year,
at
least,
we
want
to
have
a
solution
very
soon,
because
operators
are.
We
are
also
end
up
kind
of
focused
pressure
and
want
to
have
a
common
solution
and
then
there's
a
figure.
How
could
such
a
deployment
look
like
very
rough
figure
showing
different
services?
All
is
based
on
how
to
say
margin.
F
F
Maybe
there's
Rudy
aeration,
there's
layer,
2,
tunneling
and
management,
but
then
we
have
the
PNG
with
user
pen
and
control
plane,
splitted,
perhaps
other
Phung
other
entities
have
also
this
split
and
5g
control,
plane
function
and
choose
the
plane
function
of
services.
Providing
such
functionality
should
be
able,
and
ideally
this
should
all
work
on
the
same
hardware,
preps
with
same
protocols
and
a
pass
of
course.
Iolaus
for
operation
support
systems,
business
of
court
system
in
doing
the
remote.
F
And
anything
more
and
what
may
I
have
forgotten.
So,
of
course,
a
slicing
is
also
an
important
topic
to
differentiate
between
different
requirements
for
different
services,
which
would
be
odd
provider
for
same
infrastructure,
yeah
and,
of
course
there
should
be
some
business
value
in
because
we
won't
get
it
flown,
so
I
think,
but
that
was
all
I
want
to
say.
So
we
are
interested
in
some
standardized
work,
which
is
usable
by
as
much
as
applications
as
possible
and
something
more
no
that
was
it
yeah.
A
B
A
I
J
A
Dave
Simic
rope
is,
has
the
joy
of
being
bi-directional
liaison
manager
between
ITF
and
BBF,
and
it's
seems
to
us
pretty
important
to
understand
what
BBF
is
doing
in
this
space
and
to
inform
those
of
you
who
are
IETF,
only
participants
how
the
BBF
works
and
and
so
on.
So
thank
you.
Dave
thanks
thanks
Hadrian,
so.
K
There
wasn't
another
presentation
from
a
carrier
before
this:
no
okay,
okay,
so
as
Adrian
said,
I
there
are
just
to
be
clear.
Is
for
this
meeting.
I
am
speaking
from
the
role
of
the
IETF
liaison
manager
to
broadband
forum.
There
is
a
by
two
right:
there
is
another
direction
to
that:
I'm
leaving
that
off
the
table.
So
this
is
the
only
hat
that
will
be
seen
today
and
if
you're
wondering
what
an
IETF
liaison
manager
is
and
does
there's
a
link
off
the
main
IETF
that
work
page.
K
That
explains
the
concept
of
liaison
relationship
between
the
IETF
and
other
stos
standards,
development
organizations
and
then
what
a
liaison
manager
does
from
there.
So
specifically
with
regard
to
disaggregated
bng,
the
broadband
forum
and
the
ITF,
so
I
put
this
in
the
form
of
questions
and
answers,
because
it's
just
easier
to
think
about
it.
That
way,
but
that's
a
bit
tiny.
K
K
Okay,
there
we
go
all
right
so
on
this
slide.
Basically,
the
bottom
line
message
here
is
you
know
what
is
BBF
been
doing
on
B
NGS,
that
list
long
list
of
TRS
goes
back
20-something
years
with
that
they've
been
defining
the
architecture
and
requirements
on
broadband
network
gateways,
so
chances
are,
if
you're,
using
a
G
pond
or
DSL
service
in
the
world.
Today,
it's
running
over
one
of
those
specific
a
box.
That's
using
one
of
those
specifications.
A
K
Okay,
so
that's
what
BBF
has
done
in
the
past?
What
do
they
have
now,
specifically
in
the
space
that
disaggregated
bng?
There
are
two
projects
underway.
One
is
WT.
Working
text
is
what
that
stands
for.
One
is
specifically
on
disaggregated
bng
that
was
started
recently
in
first
quarter,
and
it's
all.
This
is
coming
from
the
liaisons
that
you
can
get
to
on
the
lays
on
statements
page,
but
in
a
synopsis
they
started
work
on
the
control,
plane,
user,
plane,
separation,
they're,
defining
an
architecture
and
the
nodal
requirements
on
that
on
that
project.
K
L
K
K
So
what
if
so,
the
next
thing
is?
What
is
the
BBF
communicated
between?
How
is
the
BBF
and
commuted
and
I
each
have
communicated
back
and
forth
on
these
topics,
and
here
the
list
lays
on
statements
on
the
IETF
statements
pages
that
record
the
dialogue.
So
it
started
last
year,
March
15th,
with
publication
at
TR
384
and
has
progressed
from
there.
K
M
K
K
Number,
the
Lynx
fine
print
down
here,
the
Lynx
that
the
lay
isn't
supposed
to
be
linked
together.
The
links
are
broken,
I'm
still
working
with
Secretariat
on
that
so
Adrian
mentioned
before
that
the
BBF
is
a
membership
organization,
as
many
many
of
the
stos
are
in
the
industry.
This
means
that
a
company
joins
the
BBF.
They
pay
dues
right,
it's
not
so
different
from
ietf,
except
that
the
company
is
a
member
as
opposed
to
individual
participation.
K
What
does
that
mean,
though?
It
means
that
the
contribution
and
the
work-in-progress
are
restricted
to
members?
Okay,
except
where
the
organization
makes
things
available
work-in-progress
available
by
liaison,
so
we,
the
BBF
in
the
ITF,
have
a
long
history
of
doing
this,
as
is
in
the
IETF
with
other
stos.
Has
a
long
history
of
doing
this
with
membership
organizations.
We're
work
in
progress
from
the
membership
organization
is
liaised
to
the
IETF.
The
full
ITF
body
can
look
at
it,
write
any
feedback.
K
Any
comments
actually
get
put
back
in
the
liaison
and
sent
back
to
the
organisation,
the
other,
so
that
that
would
be
how
the
communication
is
facilitated
across
the
stos
from
the
membership
to
the
ie
from
the
membership
work
to
the
IETF
from
the
IETF
back.
It's
very
simple
because
all
of
the
IETF
facilities
are
open
right,
so
the
the
other
membership
organization
is
usually
keeping
track
of.
What's
going
on
in
the
IETF
by
looking
at
data
tracker
and
email
lists.
The
key
here,
though,
is
mutual
participation.
K
Okay,
usually
there's
people
on
both
sides
in
both
organizations
that
are
progressing
the
work
in
both
organizations
and
to
date
there's
been
no
I
mean
this
is
business
as
usual,
like,
for
example,
I
did
a
little
data
mining
on
the
because
list
before
I
got
to
the
mic,
except
for
two
folks,
whose
affiliation
I
really
can't
determine
from
their
email
addresses
and
one
of
them's
him
everybody,
everybody
that's
been
vocal
on.
The
lists
is
affiliated
with
a
BBF
member
company
right,
so
this
mutual
participation
is
definitely
possible.
K
So
what
does
it
mean
in
practice?
And-
and
this
is
very
key,
if
the
if
work
is
done
in
the
ITF
okay,
then
work
is
done
in
the
ITF
according
to
the
IHF
process
on
ITF
facilities,
if
work
is
done
in
the
BBF,
work
is
done
in
the
BBF.
According
to
the
BBF
process
on
BBF
facilities,
okay
and
I
can't
make
that
more
clear
if
number
of
times
not
just
on
this
topic,
but
other
topics-
people
try
to
send.
K
You
know,
lays
on
comet
every
technical
comments
and
liaisons.
It
works
much
better
if
people
just
come
to
the
IETF
list
and
put
their
comments
right
in
the
other
direction,
if
their
member
companies,
if
they
belong
to
member
companies,
it's
easier
for
them
to
just
go
comment
and
within
the
BBF
or
within
the
membership
organization,
so
for
work.
I
already
said
that
for
working
the
BBF,
those
participating
who
want
to
participate
in
the
BBF
work.
Okay,
those
members
participate
directly.
K
Okay,
if
you're,
not
members,
then
you
can
still
provide
input
and
comment
and
feedback
on
the
liaised
work
vilely,
as
on
exchange.
It's
that
simple.
If
you're,
not
members,
of
course,
there's
always
that
option
of
joining
the
BBF,
but
this
is
not
a
public
service
announcement
for
BBF
and
then,
like
things
like
coordination,
add
dates
and
project
scope,
dependencies
question
in
such
etc.
Temperature
are
usually
handled
by
liaison,
so
this
is
one
organization
formally
contacting
the
other.
I
can't
stress
this
enough.
This
is
norm.
This
is
business
as
usual.
K
L
Again
in
zone
from
flowey,
this
is
not
question.
Jessica
people
may
be
more
vocal
background.
You
go
back
to
the
first
yeah
I
think
it
was
a
okay
there's.
A
lot
of
you
know
TR
publisher,
standard
in
PBR
for
PNG,
but
I
would
highlight
that,
in
my
opinion.
As
far
as
I
know,
a
TR
384
is
the
latest
and
probably
the
only
one
do
dealing
with
see.
You
see
your
separation
right,
TR.
O
P
L
H
H
It's
all
fine,
but
in
the
context
of
this
work
right,
this
proposed
this
both
are
there
any
specific
technical
problems
in
the
last
three
presentations
I'm
trying
to
understand
what
exactly
is
a
technical
problem
in
general
its
given
that
we
want
to
subject
control
and
use
a
plain
b
ng
I
work
for
my
my
employer,
cisco.
We
do
have
you
know,
support
such
scheme.
So
what
I'm
trying
to
understand
is
what
exactly
is
a
standardization
aspect
here?
Is
it
the
control,
plane,
controller
to
data
plane
interface?
K
I
I
can't
I
can't
tell
you
in
my
ways,
are
manager.
Boss
is
in
back
of
you
by
the
way,
so
I
failed
to
mention
liaison
manager's
report
to
the
IAB
and
ni,
be
Shepherd
Brian's,
the
IEP
Shepherd
so
I.
What
I
can
tell
you
and
you're
probably
going
to
be
dissatisfied
at
this
answer,
because
it
secretly
level
is
a
lot
of
the
questions
and
materials
that
are
being
shown
here,
I'm,
seeing
extraordinarily
similar
questions
and
materials
being
shown
in
PDF.
K
L
K
P
Like
I
actually
came
in
here
with
my
my
Dave's
boss,
hat
on
but
great
job,
this
isn't
very
thank
you
for
this
presentation
from.
P
Brian
Channel
I'm
wearing
the
IAB
liaison
manager
or
manager
manager,
manager
hat,
it's
just
a
whole
bunch,
a
stack
on
top
of
each
other
yeah.
So
my
specific
question
was
so
primarily
you
would
see
the
interaction
between
ongoing
work
and
PBF
and
ongoing
work
in
this
working
group,
if
chartered
being
with
respect
to
T
or
384
there.
Other
things
coming
down
the
pike
that
would
be
I
mean
like
so.
P
The
work
here
would
build
on
top
of
a
lot
of
this
stuff
right,
but
that's
all
well
and
good
that
the
liaison,
though
he
is
on
no
other
way
where
you
know
running
us
to
50.
So,
okay,.
P
K
K
The
work
that
would
be
coordinated
between
the
group
to
give
if
the
work
young
group
is
chartered
yeah,
be
between
these
two
projects
and
they're
follow-ons.
So.
D
P
K
P
So
the
stuff
in
here
is
probably
going
to
have
to
follow.
What's
going
on,
w2
459
and
then
like
iterate
is
the
like,
so
anything
that
had
anything
that
happens
here
that
would
it
would
follow
you
into
a
update
to
459
or
to
adult
work.
That's
built
on
top
before
39
in
BB.
F
would
just
be
another
another
piece
of
work,
there's
another
number
on
it.
We
probably
take
this
off
point.
Okay,.
N
A
Thanks
Dave
can
I
ask
a
question,
because
the
ITF
has
sometimes
done
joint
work
with
other
bodies
and
sometimes
other
bodies.
All
the
ITF
has
opened
up
for
participation
on
a
specific
topic
in
a
more
technical
day-by-day
way
than
is
allowed
by
formal
liaison.
So
you
know
for
more
liaisons
tend
to
get
a
bit
heavy
and
everybody
has
to
sign
them
before
they
progress
so
yeah.
Sometimes
this
is
like
joint
interim
meetings
or
our
meetings
going
to
be
talking
about
this.
Even
if
you're,
not
a
member,
come
along.
K
K
K
K
Especially
when
there's
lawyers
involved
yes
know
the
difference
being
is
coordination
is
a
rough
loose
kind
of
tie
to
to
work
items
or
to
pieces
of
work
and
to
SDOs
collaboration
is
much
more
technically
integrated,
I'm.
Sorry,
cooperation
is
much
more
technically.
Integrated
collaboration
implies
a
joint
deliverable
according
to
joint
procedures
and
joint
IPR
policies,
and
so
now
you
see
what
the
lawyers
come
in
right
and
I
I
would
recommend
and
I
usually
recommend
staying
away
from
that
one.
Q
Then
it's
a
I
think
in
the
context
of
this
discussion.
It
will
be
interesting
to
understand,
historically
how
the
division
of
work
has
been
done
between
ITF
and
BBF,
because
I
think
we
can
talk
all
day
about
how
great
collaboration
can
be.
But
the
point
is
that
you
gotta
have
to
have
some
distinct
separation
of
concern.
You
use
the
resources
with
those
most
knowledge
and
you
kind
of
just
have
touch
base.
Yes
and
I'd
like
to
understand
that,
because
this
is
this,
this
should
be
a
good
feed
into
you
know
what
should
be
next.
Q
It
makes
no
sense
for
people
to
do
swivel
chair
to
go
to
different
meetings
or
to
all
sorts
of
other
things.
As
long
as
you
know,
they
know
exactly
what
the
responsibilities
are
around
both
sides.
You
don't
need
to
be
in
two
places
and-
and-
and
you
know
you
maybe
define
some
interface
I-
don't
know
I
just
right.
K
Yes,
and
how
is
that
definition
of
so
in
the
work
so
I
made
mention
that
there's
a
long
history
right
in
particularly
between
the
routing
area
and
the
internet
area
and
the
program
forum
and
and
I
need
to
very
stress
very
strongly
that
past
performance
is
not
an
indicator,
an
indication
of
like
future
condition.
Okay,
so
what's
happened
in
the
past,
has
been
that,
for
example,
on
a
number
of
specs,
the
broadband
forum
has
defined
the
architecture
and
the
nodal
requirements,
the
equipment
requirements
and
put
as
part
of
the
equipment
requirements
they
actually
specify
protocol.
K
Now,
I,
don't
mean
design.
Protocol
I
mean
specify
protocol
right,
so
we're
going
to
use
RFC
XYZ
and
it's
mandatory
or
we're
going
to
use
this
part
of
our
XE.
You
know
ABC,
and
it's
optional
right
where
there
is
the
BB
FN
in
general,
has
tended
not
to
is
tended
to
try
to
reuse
protocol
that
exists
right
and
not
to
find
it
themselves.
However,
where
they
have
found
gaps,
they
have
certainly
defined
extensions.
They
have,
in
some
cases,
found
the
gap
so
large
that
they
actually
needed
to
define
their
own
protocol.
K
There
TR
69
protocol
is
an
example
of
that.
Okay,
when
we
get
into
data
models
that
becomes
even
more
so
where
they
found
gaps
in
the
modeling
they've
gone
ahead
and
and
constituted
their
own
models
and
then,
in
fact,
in
some
cases,
they've
they've
seen
that
like,
for
example,
work
and
I
Triple
E
on
a
doe
2.1
CFM
has
been
lagging
so
they
get
an
interim
and
then
they're
gonna
come
back
and
sync
up
with
the
I
Triple
E
model.
So
the
desire
to
work
with
those
defining
the
protocols
is
definitely
there.
K
A
Let
me
say
that
brilliant
thanks,
Dave
I,
know
you're.
In
a
few
can
we
can
we
transition
to
the
general
stuff
so
that
we'd
we
we
get
a
bit
of
discussion
before
the
end,
but
you're
in
front
of
the
queue.
So
you
can
ask
any
or
make
any
point
you
want,
but
not
today,
yeah
so
for
this
bit
say
who
you
are
try
to
be
brief.
Don't
give
a
lecture
because
other
people
may
want
to
make
comments
as
well
go
for
it.
Yeah.
O
Easier
gone,
who
are
we?
This
is
not
new.
We
have
done
this
before
many
times,
stos,
where
BBF
is
one
of
them
on
3gpp,
much
larger
interaction
with
IETF
through
the
liaison
that
they've
mentioned,
so
the
requirement
from
BBF
can
come
here
to
really
lay
the
foundation
for
the
protocol
or
protocol
extension
work.
So
this
is
nothing
new.
It's
a
well-established
process
in
place.
I
guess
that's
the
reason
that
they
was
here
to
basically
follow
on
their
existing
process.
This
is
nothing
that
fear.
We
are
not
reinventing
the
wheel.
The
process
is
a
very
mature.
O
The
only
thing
is
the
shorter
of
their
future
working
group
has
to
really
focus
on
the
set
of
requirements
that
are
indeed
part
of
the
liaison
communication
from
BBF
to
IETF.
If
the
BBF
they
have
a
set
of
core
requirement
for
defining
a
new
protocol
or
new
protocol
extension
that
makes
even
IETF
look
good
and
make
IETF
more
relevant
to
their
bigger
industry.
Thank
you.
I.
J
Have
an
opinion,
romantics,
Nokia
and
so
I'm
mainly
going
to
focus
on
the
the
questions
being
asked.
I
think
I.
Several
operators
have
I
requested
their
requirements
because
some
of
these
are
fixed
only
I.
Think
some
of
used
is
our
combination
of
the
tools,
some
of
I.
So
you
see
that
it's
clear
that
there
is
different,
let's
say,
deployment
models
that
are
asking
for,
let's
say
similar
solution
to
the
problem,
so
I
think
I
in
general.
We
should
look
at
all
of
these
and
figure
out
what
is
the
best
way
forward
to
address
that
now?
J
If
you
look
at
what
work
belongs
in
IETF,
I
personally,
I
believe
if
you
look
to
what
they've
presented
BBF
has
the
knowledge
of
what
a
bng
and
a
subscriber
h
is
all
about
right.
So
it's
a
bit
strange
what
what
we
are
doing
right
now.
We
are
trying
to
do
protocol
work,
but
we
don't
understand
in
ITF
exactly
the
the
use
case
or
the
what
a
B
and
G
actually
is.
If
you
look
I
scanned
all
the
ITF
documents
you
actually,
people,
B
and
G
actually
doesn't
exist
in
ITF
as
terminology
I.
J
Think
what
you
can
find
is
nuts,
or
something
like
that
right
so
as
such
I
believe
that
we
are
better
off
in
general
to
have
and
help
and
focus
all
our
efforts
in
BB
F
and
make
sure
we
understand
exactly
the
requirements.
We
look
at
the
protocol,
which
exists
and
I
think
we
are
not
going
to
discuss
the
protocol
at
such
right
now,
because
I
think
that's
probably
contentious,
but
have
the
BB
F
focus
on
what
is
available.
What
is
the
protocol
and
whether
they
need
help
to
have
extensions
or
not
right?
J
And
if
so,
then
they
should
come
to
the
ITF
to
help
with
potential
protocol
words
or
not
or
potentially
do
it
themselves
because
they
have
knowledge,
as
they've
already
was
saying
like
they
have
extended
existing
protocols
with
information
elements
or
they
have
defined
new
right
so
and
I
think
that
should
be
the
mode
of
operation.
In
my
view,
for
this
scenario
is
have
the
BB
F
focus
on
the
requirements
and
they
started
that
work
and
I
think
there
are.
There
is
people
who
are
in
a
hurry
to
get
a
solution
as
quickly
as
possible.
J
R
Mark
I
think
part
of
the
reason
you
didn't
get
much
feedback
was
because
you
weren't
proposing
anything.
So
the
question
should
something
be
done
in
the
idea
of
it's
like
well,
no
one
has
said
here's
a
proposal
for
here's.
What
the
actual
technical
content
is
of
this
interface
between
your
ATF
funding
and
the
BBF.
Does
this
make
sense?
If
you
ask
that
question,
you
would
have
gotten
people
to
sign
up
a
lot
quicker.
This
feels
extremely
nebulous
and
from
a
manner
so
there's
a
technical
sort
of
question
about
the
content.
R
S
H
S
N
So
Sunday
what
were
Nokia
so
just
continuing
what
sin
Dave
cynic
represented.
He
showed
an
eye
chart
of
all
those
TRS.
So
if
you
notice,
if
you
notice
every
media
transition
that
has
happened
on
the
bng
BBF
has
come
up
with
a
set
of
requirements
right.
If
you
look
at
here,
101
starting
way
back
ATM
to
Ethernet
v6
introduction,
IPO
II
sessions
converged
policy
on
and
on
and
on,
and
this
is
again
a
bigger
architectural
change.
I
would
say
so.
Where
are
those
requirements?
That's
why
WP
459
has
been
spawned
so
I
mean
putting.
N
T
N
Q
T
Well,
ok,
I'm
surprised
that
people,
one
of
the
co-authors
of
the
requirement
documents
now
says
that
there
is
no
ITF
document.
Oh
it's
not
ideal
document
the
original
draft
with
the
requirements.
No,
we
have
even
two
individual
documents
requirements
that
list
requirements
and
we
do
have
material
that
we
can
work
with
whether
it's
well
captured
requirements.
Well,
when
we
start
working
with
them,
we'll
find
out
and
I
hope
that,
as
data
pointed
out,
we
have
very
good
cross-pollination
between
two
zeros.
T
We
can
find
the
right
level
of
interaction
between
organizations,
formal
and
informal,
to
progress
it.
If
we
decide
to
do
that
so
I
don't
see
the
problem.
That
ITF
doesn't
have
rich
expertise
on
being
G,
because
we
do
have
a
starting
point
to
work
with.
What
is
clear
is
that
saying
that?
Well,
we
need
to
wait
for
some
documents
to
mature
and
then
we
can
start
analyzing.
It
I.
Don't
think
that
it's
a
right
approach,
especially
because
in
the
first
presentation
we
were
given
a
timeline
of
operator
that
is,
they
have
their
own
timeline.
Okay,.
C
U
Well
rested:
the
oberliga
came
with
many
ideas
and
the
arguments,
and
it
was
surprised
that
dork
was
stealing
all
of
them
in
their
presentation.
So
in
this
part
of
whether
we
know
where
the
party
is
or
not,
I
think
that
the
general
problem
of
addressing
how
to
deal
with
access
notes,
how
to
decouple.
Q
The
end
rota,
based
on
what
they've
explained
the
relationship
and
collaboration
I,
truly
think
that
what
BBF
started
is
it
good
as
a
good
model?
And
you
know
they
should
be
able
to
identify
the
gaps
and
if
there
are
gaps,
come
back
to
ITF
with
the
protocol
requirements
very
specific,
very
clear
work
done.
This
is
not
greenfield.
The
development
start.
The
requirements
here
in.
L
From
highway,
I'm
gonna
be
the
concern
if
people
claim
that
okay,
this
kind
of
work
for
the
owning
town
country
here
in
this
organization
or
or
other
another
one,
have
to
wait
until
some
concrete
outcome
here,
because
I
wanna,
clear
see
that
is
the
same
group
of
people
contribute
both
to
IETF
and
the
ppl.
For
example,
we
have
a
bunch
of
drugs,
both
from
you
know,
Xavier
and
Nokia.
L
We
have
also
in
PBF
organized
some
conference,
call
to
reveal
the
IETF
draft
to
to
happy
the
crew
crest
so
also
under
the
age,
and
we
think
that
PBF
is
not
blocking.
You
know:
I
trip
to
gaining
critical
work,
so
I
think
we
are.
You
know
in
a
good
corporation
way,
so
so
I
so
yeah
that
I
can
see
the
energy
here
so.
K
H
K
Need
to
make
a
quick
intervention
here,
yes,
in
one
of
the
liaisons
Davidson
acro
ITF
liaison
manager,
BBF
in
one
of
the
liaisons
from
BBF,
it
says
says
very
clearly
that
the
BBF
does
not
object
to
the
IETF
continuing
work.
It
needs
to
be
pointed
out
that
no
organization
should
object
to
any
other
organization.
Doing
anything,
that's
more
cordiality.
M
Hi
Dave
Alan
Erickson,
but
I'm
speaking
with
wearing
the
hat
of
wireless
wireline
convergence
area
director
at
broadband
forum,
we're
currently
working
on
fixed
mobile
convergence,
which
is
providing
broadband
forum
defined
access
to
the
5g
core.
We
have
been
doing
this
in
cooperation
with
3gpp
sa
to
the
genesis
was
in
2016
when
we
had
some
10
major
operators
come
to
us
and
ask
us
specifically
to
do
this
now.
At
the
moment,
we
have
a
pairwise
relationship
between
the
broadband
forum
and
3gpp
that
has
been
working
very
well.
M
We
are
in
the
middle
of
doing
normative
work
in
the
release.
16
time
frame,
my
observation
would
be,
and
yes,
we
are
looking
at
cups
as
applied
to
the
broadband
forum
network
and
supporting
and
providing
support
for
legacy
devices.
Our
current
intention
with
working
text
for
58
was
to
use
PFC
P
for
this
and
for
at
least
the
work
that
my
working
group
is
doing.
M
It's
not
in
my
working
groups,
interest
to
turn
this
very
nice
pairwise
relationship
into
a
threesome
and-
and
that
presents
a
bit
of
a
conundrum
now,
because
the
fact
that
we
want
legacy
device
support
as
part
of
perhaps
a
carrier
transition
strategy.
Migration
allow
people
to
bring
their
own
devices
to
the
party.
That
means
we're
going
to
have
a
lot
of
duplicated
functionality
as
far
as
the
access
facing
aspects
are
concerned
and
I,
don't
know
how
we
avoid
getting
multiple
solutions
in
the
industry
to
the
same
problem.
M
A
D
It's
a
pity
that
we
didn't
share
the
only
enough
time
to
have
a
longer
discussion,
but
what
I've
heard
at
least
being
said
across
the
room
or
by
the
different
speakers
is
that
there
is
a
tendency
to
prefer
waiting
on
BBF
to
complete
its
work
on
requirements
before
we
can
reasonably
envies
aged
work.
I
also
understand
that
waiting
is
not
really
a
notion
in
some
cases,
but
Dave
reassured
us
that
the
target
was
real,
so
so
I
think
we're
talking
about
a
five-month
delay
before
we
can
have
a
serious
discussion
on
protocol
work.