►
From YouTube: IETF106-BABEL-20191119-1330
Description
BABEL meeting session at IETF106
2019/11/19 1330
https://datatracker.ietf.org/meeting/106/proceedings/
B
C
D
A
Okay,
so
we
don't
actually
have
too
much
to
do
and
I
got
early,
so
I
guess
I'll
go
ahead
really
rapidly.
Here,
it's
not
Donald
Eastlake
with
future
way.
Russ
White
was
unable
to
make
this
meeting
Martin
art
area
director
is
sitting
in
front
here,
so
this
okay
people
are
presumably
aware
of
the
note
well
and
the
fact
that
you
want
quality
documents.
You
should
review
other
documents
and
if
you
do
that,
people
probably
review
your
documents.
A
So
here's
the
proposed
agenda
so
just
do
some
update
on
status
just
briefly
touched
on
the
yang
model
document,
so
that
block
meant
that
graft.
The
yang
model
Lachman
is
in
last
call
ending
at
this
meeting.
So
we'll
see
what
there
haven't
been
any
negative
comments.
There
have
been
some
positive
comments
on
the
list
here
in
Mabel
presentation
by
Sandy,
Jiang
and
I
just
have
a
presentation
on
possible
DVF.
A
Maybe
they
will
futures
which
I
can
go
over
well,
what
seems
like
things
we
might
want
to
work
on
in
the
future,
mostly
for
discussion
on
the
mailing
list
and
showing
anybody
has
discussion
here
or
ideas
here
we
can
do
that.
Does
anybody
want
any
changes
in
this
agenda
additions
deletions
whatever
okay?
A
So
we
have
two
documents
that
have
been
food
with
the
iesg
inner,
all
green
in
the
Ballad,
so
they're
two
documents
which
have
discusses,
though,
need
to
be
resolved.
The
H
max
security
document
and
the
new
base
protocol
document.
There's
two
documents
with
publication:
requested
source
specific
information
model,
as
I
said
that
will
yang
models
in
working
group
last
call.
There
is
also
a
working
group
draft
on
the
round-trip
time
extension,
which
was
presented
at
the
last
etf
meeting
and
there's
a
couple
of.
A
Related
drafts,
which
is
the
home
that
draft,
which
is
waiting
on
the
protocol
draft
and
Sandy's
draft
on
beer
papal
extensions.
So
we've
really
done
pretty
well
on
the
milestones
that
are
in
the
Charter.
Really,
the
only
one
left
is
to
submit
the
yang
document
to
the
iesg,
which
perhaps
we
can
do
later
today,
which
at
that
point
there
would
be
all
the
milestones
in
the
current
milestone.
A
Listen,
the
Charter
would
be
completed,
which
seems
good,
so
that's
it
for
these
agenda
and
status,
slides
so
I
guess
on
the
yang
model
draft
the
only
question
I
really
had
was
it
currently
there's
an
orangey
yang
symbol,
which
means
that
the
automated
yang
checker
isn't
completely
happy
with
it?
Is
that
a
problem
or
can
they
I
mean
because
there's
certainly
most
people
think
it's
an
excellent
condition
and
we
could
just
go.
E
A
C
So
I
guess
what
you're
referring
to
is
on
data
tracker,
yeah
saying
that
yeah
well,
there
are
some
issues
with
the
tools
themselves
that
sometimes
lead
to
that
error,
and
it's
good
is
as
as
simple
as
saying
that
one
of
the
included
files
that
I
might
have
referred
to
is
something
it
can't
find.
I
can
suddenly
take
a
look
at
it
and
send
a
note
to
the
mailing
list
if
need
be
to
let
have
even
know,
maybe
I
see
Martinez
pull
up.
Oh.
C
Believe
that
that
is
a
bug
in
the
tool,
the
a
feature
requirement
for
it
to
be
mechanical
order,
although
the
a
B
and
F
grammar
for
the
language
says
that
that
is
the
right
order.
For
some
reason,
it's
complaining,
but
anyway,
as
I
said,
I'll
take
a
look
at
it
and
send
a
note
to
the
mailing
list.
Okay,
confirming.
A
Is
there
any
objection
after
that
notice
posted
to
declaring
the
draft
to
a
past
working
group?
Last
fall,
it's
been
I
have
noticed
out
for
a
while,
like
I
say
this
with
no
objections
and
some
support
so
hearing
nothing
as
soon
as
you
post
that
note
or
unless
you
say
you
don't
want
to
I
need
to
head.
In
that
note,
then
I'll
post,
a
message
saying
it's
fast
working
group
last
call
and
I
can
start
working
on
the
publication
request.
A
A
F
F
We,
our
intent,
is
to
use
paper
protocol
Esbjerg
underlay
protocol
to
build
a
beer
forwarding
there.
Let's,
let
me
introduce
peer
technology
here
and
briefly.
You
know
that
beer
is
a
new
multicast,
a
technology,
and
it
can.
You
can
build
multicast
forwarding
there
without
it
is
explicitly
multicast
distributing
trees
building
and
it
doesn't
require
any
inter
mediate
in
those
to
maintain
any
per
flow
stage.
So
it's
a
new
multi
has
the
technology
and
the
other
can
bring
many
advantage
of
what
it
has.
The
forwarding
and
the
beer
is
split.
F
The
networking
into
three
parts
and
the
first
one
is
a
core
layer.
It
me
is
called
peer,
forwarding,
layer
and
needing
this
forwarding
there.
The
vector
can
be
forward
by
multicast
duplicate
and
it
can
reach
around
her
without
the
SG
or
much
much
e
in
the
intermediary
network
and
the
packet
can
be
encapsulated
with
her
header.
F
Now,
mr.
processor
I'm
curious
function,
Ethernet
function
and
a
v6
function
is
ongoing,
but
in
order
to
build
as
a
beautiful
loading
there,
we
must
rely
on
some
under
a
protocol
to
build
forwarding.
So
we
depends
on
the
ITP
protocol
and
the
beach
P
protocol
and
the
variable
to
to
build
a
the
routing
under
day.
So
we
use
the
exchanging
of
this
routing
protocol
like
a
Bible
like
her
ITP
to
forward
querida.
F
We
are
information
and
spread
it
into
the
networks
and
all
the
computer
knows
in
the
network
can
calculated
unicast
routing
and
the
can
also
calculate
as
a
beer
forwarding
table.
So
as
a
beautiful
wedding
layer
can
be
built
from
the
IDP
extension
and
after
the
forwarding
layer
is
built
and
the
overlay
is
used
to
corrected
the
podcast
information
about
the
numerous
router
and
the
egress
wrote
our
search
for
some
specific
multicast
the
flow.
So
the
be
the
universe.
Rotor
like
this
BF
IR
can
run
wanna,
be
PFI
all
receive
some
specific
multicast
flow.
F
It
can
encapsulate
the
packet
with,
according
to
a
pair
header
and
the
forwarded
into
the
network,
then
the
the
intermediator
notes
came
forward.
This
package
and
the
duplicated
this
factor
according
to
the
previous
calculation
result
and
push
this
packet
to
the
edge
routers,
so
the
parameter
can
be
forwarded
in
this
network
and
using
be
peer,
multicast,
the
technology.
F
So
in
some
network
we
know
that
we
may
only
use
bebo
is
routing
protocol.
So
every
day,
like
a
fool,
the
bro
I
appear
in
these
networks.
We
can't
depends
on
SPF
extension,
ice,
I,
say,
extension
or
PD
extension.
We
can
only
rely
on
Bebo
extension,
so
we'd
like
to
make
some
extensions
in
Babel
protocol
to
build
as
a
beer
forwarding
layer.
So
it's
our
motivation
so
so
briefly,
summary
of
protocol.
F
We
know
that
bebo
is
distance,
vector
routing
protocol,
it
can
be
used
the
young
world
and
the
worldÃs
mass
networks.
So
it's
a
very
good
routing
protocol
can
be
used
in
well
various
network,
it's
our
intent
because
it
can
extend
the
beer
deployment
in
the
wireless
network,
and
we
know
that
people
has
made
some
heavy
extension
to
carry
the
indigenous
additional
information.
So
we
think
that
people
can
also
use
a
new
sub
T
of
V
took
away
the
beer
information.
F
So
we
with
racket
will
use
peer
extension
to
carry
the
beer
parameter
in
the
network,
so
it's
an
example
of
implementation.
We
know
that
suppose
that
the
node
a
is
a
universe,
brought
her
officer
network
and
there's
other
routers
cow,
also
wrong
the
Babel
protocol
and
when
people
protocol
is
used
to
advertise,
the
prefix
of
every
node,
The
Associated
appear
information
such
as
sub
domain
and
the
bf
ID,
and
some
other
information
can
also
be
carried
with
the
prefix
advertisement.
F
So
one
Babel
protocol
is
running
the
network
and
as
long
as
the
prefix
information
has
spread
in
the
network,
peer
information
can
also
be
spreading
this
network,
so
one
pebble
protocol
calculated
the
unique
hostile
route.
It
can
also
calculate
as
a
peer
forwarding
haber,
so
peer
forwarding
layer
can
be
built
from
this.
F
Mrb
is
an
example
of
peer
or
a
protocol,
so
in
this
net
we
suppose
that
the
MRD
is
used
to
convey
the
AG
rotors
information
for
some
specific,
her
multicast
flow.
So
the
for
example
know
the
a
hanging
capsulate,
the
associated
Heather
into
the
in
the
the
motive
has
to
flow
packet
and
then
forward,
adding
the
network
and
all
the
net.
F
The
noting
this
networker
come
forward
the
packet
correctly
according
to
the
pier
forwarding
table,
and
we
know
that
because
the
pier
capability
in
this
network
may
be
an
optional,
so
we
you
define
some
an
entire
Beach
of
the
extended
sub
jovi
to
zero.
If
a
router
cannot
recognize
the
sake
of
e,
it
can,
it
must
even
know
the
unknown
sub
shell
V
and
it
cannot
processing
the
other
key
fields
in
this
extension,
it's
for
matter
of
a
post-secondary
and
that
we,
the
second
only
profits,
is
aligned
with
the
definition
in
SPF
and
ice
is
so.
F
We
also
defined
at
the
same
fields
in
the
extension,
but
we
may
not
use
it
in
be
able
to
because
the
such
as
PA
are
bar
and
the
IP
ie
views
may
not
have
special
meaning
in
paper
protocol.
So
we
just
define
and
the
to
align
with
advertisement
with
IP
protocol,
but
it
may
be,
may
not
be
used
for
now,
but
maybe
in
some
footer
we
can
find
some
use
of
this
field
and
also,
as
we
defined
a
sub
sub
activity
of
MPs
encapsulation.
F
F
F
So
when
it
calculates
the
roads
and
routes,
maybe
we
know
that
because
it's
only
one
hop
away,
so
it
may
define
some
tunnels
or
we
use
special
method
to
build
a
tunnel
to
across
the
router,
which
cannot
support
a
beer
for
body,
and
there
is
an
implementation
of
the
control
plane
of
beer
in
Babel
you
can.
Maybe
you
can
find
a
some
something
where
you
being
this
link.
Ok,
that's
all
for
today.
Thank
you.
Ok,.
A
A
G
A
A
A
So
I
have
a
little
presentation
on
possible
futures,
so
you
know
starts
by
going
over
what
the
work
items
are
in
our
Charter
and
their
status.
So
Green
is
basically
a
good
color
indicates
we've
there,
so
the
stability
segment
is
explicitly
called
out
in
the
Charter
and
that's
that's
the
iesg.
There's
the
proposed
standard
Nevada
based
protocol
standard
there's,
a
security
needs
need
to
be
addressed
and
we
have
two
drafts
that
do
that,
one
of
which
is
it's
fully
approved
and
one
of
which
has
a
one
discuss
left
needs
to
be
resolved.
A
The
Charter
says
we
should
document
implementation
experience,
but
it
explicitly
says
it
doesn't
need
to
be
an
RFC
or
a
document
or
something
like
that.
So
we've
basically
been
doing
that
by
doing
reports
at
meetings.
There
is
not
such
a
report
at
this
meeting,
but
at
last
meeting
there
was
a
report
on
the
implementation,
so
we've
sort
of
been
doing
that
by
my
reports
to
the
meetings
pretty
much
meets
that
item.
A
There
are
two
optional
things
in
the
existing
Charter
One
with
sort
specific
routing
and
there's
a
graph
on
that
which
is
in
publication,
requested
state
and
the
other
is
multicast.
So
we
haven't
made
nearly
that
progress
on
multicast
I
think
it's
more
complicated
topic,
but
we
do
have
this
beer
available
extensions
draft
and
it's
something
which
we
can.
Our
charted
to
work
on
with
our
current
Charter
is
an
optional
item.
A
So
I
came
up
with
a
couple
of
other
things:
the
list
as
possible
work
items
one
would
be
type
of
service
specific
routing,
so
you
have
different
routing
for
different
types
of
service.
There
is
an
expired
graft
on
that
topic
for
our
working
group,
which
hasn't
been
worked
on
in
a
while,
but
that
that
could
be
revived
and
I
think
it's
mara
way
so
straight
forward.
Another
possibility,
I
thought
of
is
doing
an
802
11
mesh.
It's
a
Wi-Fi
mesh
pass
selection
protocol.
A
So
Wi-Fi
mesh
assumes,
of
course,
all
the
nodes
and
your
Wi-Fi
mesh
have
the
same
path:
selection,
protocol,
routing
protocol
II,
it's
coming
to
call
it
a
past
election
protocol
and
their
provisions
are
having
different
ones
but
I
say
any
particular
mesh.
It
has
to
be
consistent,
there's
even
a
provision
for
vendors,
specific
path,
selection
protocols.
So
we
could,
you
know
it's
designed
so
that
other
organizations
could
make
up
a
selection
protocols
and
use
their
UI.
A
We
use
the
Eno
aur
to
come
up
with
our
own
designation
I've
found
ATAR
to
11
to
be
somewhat
receptive
to
this
sort
of
thing.
There
was
a
draft
done
in
the
IETF
which,
down
at
the
bottom
of
this
slide,
called
secure,
hybrid
wireless
mesh
protocol,
and
this
draft
was
liaison
to
802,
11,
I,
think
and
a
neutral
Evan
said,
oh
sure,
and
so
they
allocated
a
code
point
in
the
802
11
space.
A
For
that,
so
you
can
identify
the
past
election
protocol
either
by
a
single
byte
or
by
oh,
oh
you
and
a
sub
selector
for
a
particular
vendor.
So
the
802
11
has
its
own
assigned
number
Authority
and
I
just
went
and
looked
yesterday
in
there
and
there's
two
values
in
for
past
selected
protocols.
One
is
the
basic
one
in
the
802
11
standard
and
the
other.
Is
this
secure,
hybrid
wireless
mesh
protocol?
A
So
they
allocated
point
to
a
while
ago,
and
it's
of
course
it's
still
allocated
and
I
suspect
that
if
we
proposed
a
path
selection
protocol
using
babel
for
it
or
211
mesh
that
they'd
be
happy
to
allocate
a
value
3
in
their
space.
Well,
we
could
use
our
own
and
you
know
if
there
were
some
extension
somehow
required.
It's
it's
possible.
They
would
do
that
so
I,
don't
think.
A
There's
much
of
an
inter
standards,
development
organization
problem
there
and
it
seems
like
the
kind
of
thing
that
that
will
should
be
good
for
because
you
have
obviously
highly
variable
link
metrics
in
an
802
11
mesh,
so
I
have
another
slide.
I
was
wondering:
could
anybody
in
the
room
or
remotely
or
whatever
has
any
ideas
for
other
possible
work
items
they'd
like
to
to
finish
possible
future
or
new
work
items?
They
would
like
to
have
added
to
this
list
of
stuff
here
and
if
they
don't
want
to
speak
up
now,
they
can
act.
B
I
Barbara
Stark
I,
don't
see
Julia
stepping
up
to
the
mic
and
I
know.
David
is
because
I
just
heard
him
at
the
microphone
over
two
doors
down.
I
think
we
may
be
just
a
little
bit
distracted
in
this
session
right
now
and
it
might
be
better.
A
A
Reasonable
to
me
I
think
we
should,
if
nobody
had,
regardless
of
whether
people
had
things
to
say
here,
there's
a
lot
of
people
jumped
up
and
they
all
said
there
should
be
absolutely
no
new
work.
Oh
that
doesn't
seem
to
be
happening,
I'm
pretty
sure
we
should
discuss
it
on
the
list
and
perhaps
work
on
a
draft
new
charter
language.
There
was.
D
G
A
A
J
So
the
idea
of
a
routing
protocol
that
works
well
with
dynamic
link
costs
as
a
particular
interest
to
to
us
fiber,
probably
varies
on
the
granularity
on
the
order
of
an
hour.
Butbut
free
space
links
or
satellite
links
could
be
minutes
or
below
so
I'm
curious.
You
know
just
completely
off
the
top
of
my
head.
Whether
or
not
Babel
is
something
we
should
look
at
as
a
routing
protocol
for
those
kinds
of
networks.
J
A
A
In
the
in
this
presentation-
okay,
so
sorry
yeah,
it
should
be
down
here,
I,
guess
it's!
It's
not
something
explicitly
called
out
in
the
current
Charter
and
it's
something
that
is
already
it's
already
a
working
group
draft.
It
seems
like
it's
sort
of
ascetic,
it's
a
kind
of
part
of
the
protocol,
draft
I,
think
I,
guess
I
considered
it
to
kind
of
fall
under
under
the
existing
work.
It
should
have
been
listed
there.
It
is
a
let's
see
here,
go
back
to
the
status
of
documents
here,
so
so
yeah.
So
that's
down
there
is.
A
A
A
H
A
Else
from
anybody,
okay,
we
measure
sent
the
emails
and
Nick
glance
and
we
send
out
a
declaration
that
the
yang
model
has
passed
working
request,
call
a
little
bit
later
today
and
I
think
with
that.
That
concludes
about
the
working
group
meeting
for
this
time.
So,
thanks
for
barbara
for
taking
minutes
and
see
you
on
the
list
and
the
next
IDF
meeting.