►
From YouTube: IETF100-IPSECME-20171113-0930
Description
IPSECME meeting session at IETF100
2017/11/13 0930
https://datatracker.ietf.org/meeting/100/proceedings/
A
So
so
this
session
we're
actually
we're
sharing
the
room
with
lamps,
so
so
this
session
will
run
from
from
now
until
11:00,
and
then
lamps
will
run
from
11:00
to
I
believe
noon
so
we'll
have
to
end
promptly
we'd.
Ask
you
not
to
stay
in
the
room
and
have
conversations
if
you
want
to
talk.
You
know
please
do
so
out
in
the
hall
or
and
stay
and
enjoy
lamps
next
slide.
Please.
A
So
this
is
the
nail
well
you've
agreed
to
this
as
part
of
your
registration.
This
is
just
a
reminder
that,
but
any
anything
that
you
say
at
the
mic
or
I'm
any
participation
in
the
IETF
does
the
contribution
to
the
idea
next
slide.
Please
logistics,
I'm
I've
got
the
blue
sheets
around,
so
they
should
be
on
their
way.
We
have
Tommy's
agreed
to
be
a
note-taker.
Can
I
get
a
second
volunteer,
we're
taking
notes
and
etherpad,
so
we
were
just
be
useful
to
have
one
other
person
watching
and
helping
out
there.
Any
volunteers,
I.
A
A
A
B
A
We've
had
actually
a
fairly
productive
period
between
now
and
the
last
IETF.
We've
actually
had
three
drafts
published
as
RFC
sr2,
crypto,
algorithm,
related
drafts
and
tcp
and
caps.
So
thank
you
to
to
the
authors
and
reviewers
for
all
of
your
hard
work
on
those
drafts.
It's
good
to
have
those
done.
A
The
EDD
essay
draft
is
I
believe
pretty
much
ready
to
to
progress.
We've
recently
had
a
update,
publish
so
I
think
at
this
time
we're
just
waiting
for
the
ad
go
ahead
in
order
to
progress
that
that
draft
forward
and
we've
just
concluded
a
working
grab.
Working
group
last
call
on
split
DNS
and
implicit
Ivy.
We've
gotten
some
feedback
on
those
drafts.
A
I
believe
that
the
authors
are
working
on
on
resolving
those
comments,
and
we
may
have
some
discussion
on
some
outstanding
issues
today
and
then
the
QR
work
is
is
I,
think
doing
fairly
well,
where
we're
getting
pretty
close
on
on
that
I
think
there
were
some
recent
changes
in
the
in
the
the
latest
update.
So
it
would
be
great
to
you
know,
to
focus
the
group
around
some
more
review
on
that
draft.
So
we
can
get
that
wrapped
up
I'm
here
anything
else.
Bad!
A
C
Is
okay,
so
this
Balcombe
speaking
and
so
I
just
submitted
0-3
yesterday?
That
was
all
the
comments
in
regards
to
Valerie's
comments
and
as
far
as
I
know,
there's
nothing
else
pending.
So
it
would
be
nice
to
get
a
few
more
reviews,
but
I
think
we've
gone
over
this
especially
last
idea.
If
you
had
a
big
discussion
about
how
to
move
forward
with
the
format,
so
I
think
it's
in
pretty
good
shape
and
we
can
move
on.
Okay,.
A
D
So
we
received
a
few
comments.
We
addressed
those
and
the
current
status
is
that
using
the
implicit
IV
can
Holloway
is
not
compatible
with
all
the
Ike
extensions.
So
what
we
are
well,
the
current
tax
is
saying
that
we
should
not
use
it
with
Ike
v2,
so
other
another
way
to
do
would
be
to
say,
don't
use
it
with
these
extensions.
I
assume
you
mean
usually
encrypt
I
person
to
traffic
yeah.
C
Yes,
please,
okay,
so
how
about
so
so
I?
Would
it
would
be
nice
because
the
this
draft
is
all
about
cutting
out
a
few
bytes
left
and
right
for
the
IOT
people.
It
would
be
nice
if
you
can
actually
make
it
work
with
IP
and
just
saying,
like
all,
we
found
it
difficulty
night
v.
Let's
is
not
user
directly
seems
a
shame
like
it
seems
like
a
defeating
half
of
the
purpose,
and
even
though
the
draft
says
like
oh
well,
it's
only
I
guess
only
a
few
packets
as
many
more
you
speed,
packets,
I,
think.
E
After
you
know
no
chair,
so
so
I
was
actually
think
about.
We
have
a
couple
of
other
items
coming
with
ESP
and
we
have
also
have
an
compressed
I
question
to
stuff
and
I
was
thinking
about.
It
might
be
a
better
option
to
do
the
compressed
I
question
to
then
to
start
saving
the
implicit
iv9
person
to
do
stuff
because
I
could,
let's
do
it
do
have
it?
You
know
the
message
IDs
and
so
on,
and
it's
a
little
bit
kicking.
It
actually
gets
quite
complicated,
so
I
was
actually
hoping.
E
We
could
actually
get
this
one
out
and
they're
actually
separate.
You
know
we
did.
There
is
no
need
to
play
them
to
be
in
the
same
document.
Even
if
you
start
working
on
this
after
after
it
weakens
put
this
implicit
IV
for
ESP
first
out,
and
then
we
start
working
on
like
parts
and
that's
something
that
we
could
discuss
on
the
Reese
artery,
and
that
was
just
a
chair.
G
F
G
F
F
G
G
D
A
E
A
F
F
While
I
could
I
know
it
is
draft
ATF
I'd
be
second
year
next,
please
so
in
case
you
forgot
what
it
is
draft
about
some
background,
so
currently
I
protected
traffic
can
be
stored
and
later,
when
quantum
fully
function,
quantum
computers
is
available
can
be
decrypted
and,
as
the
draft
proposes
a
short-term
solution
for
this
problem.
This
problem,
by
introducing
what
quantum
a
Schottky
as
a
tree,
steered
into
the
calculation
of
keys
and
when
used
it
protects
communication
from
later
decryption
using
quantum
computers.
So
next,
please.
F
F
So
in
this
case
the
responder
has
no
ability
to
continue
exchange
and
only
has
to
abort
exchange,
even
if
using
PPK
is
not
mandatory
for
both
so
because
our
load
is
calculated,
it's
already
calculated
using
typically,
so
we
propose
a
solution
to
this
problem
that,
in
case
as
initiator,
using
be
careful,
the
initiator
is
not
mandatory
in
it.
The
initiator
may
provide
an
optional
optional.
F
F
G
F
F
F
F
F
So,
for
that
reason
we
recommend
that
initiate
a
case
negative
result
and
at
least
doesn't
try
to
too
often
to
continue
to
to
to
raise
stability,
so
you
can
say
because
otherwise
the
respondent
may
think
that
it
is
under
current.
Take
some
measures
like
looking
at
the
address
end.
So
then
it
is
possible
with
another
attack
is
a
downgrade
attacks.
It
is
possible
if
attack
is
complete,
is
equipped
with
fully
functional
quantum
computer
and
is
possible
to
break
digital
signature
in
time.
It
can
just
in
this
case
and
can
eliminate
the
case.
Opponent.
F
Unification
is
an
Forge
digital
signature.
So
the
reason
the
solution,
the
our
recommendation-
is
to
make
epic
a
using
mandatory.
In
this
case
the
attack
will
be
detected
and
it's
just
like
I
said,
won't,
be
established
or
use
shared
key
authentication.
The
scales
attack
will
be
detected
too
well,
and
the
third
attack
is.
F
If
a
token
is
able
to
each
drop
and
injects
packingtown
to
the
network,
it
can
respond,
it
can
mimic,
they
respond.
The
response
for
either
and,
for
example,
well,
if
using
PPK
is
mandatory
for
initiator
and
the
turkey
is
able
to
inject
packet
without
PPK
supports
the
machete
will
abroad
the
exchange
before
the
genuine
response
from
the
response,
even
if
respondent
sports
prickly.
So
we
recommends
that
insert
should
wait
for
more
responses.
If
it
receives
negative
responses,
it
should
wait
for
more
responses
in
case.
F
Doesn't
look
so
we
expanded
security
consideration,
consideration
section
describing
how
typically
are
distributed
described
in
using
grown
epic,
a
is
keys
and
the
key
on
those
indications
of
became
on
the
syndication.
His
situation
is
the
case
when
PPK
is
used
with
now
our
syndication
method,
so
it
eliminates
a
second
set
of
credential
for
initiating
responder
so
draft
in
its.
F
Shooter
and
responder
have
two
sets
of
security
credential
and
be
the
case
and
conventional
credentials
like
certificates.
So
we
can
get
rid
of
certificates
because
with
quantum
computers,
useless
and
use
PK
only
authentication
and
by
using
knowledge,
syndication
method
with
BBK,
so
I
grew
up
appliques
on
the
describe,
but
it
is
not
recommended
and
just
all
tweaks
are
disadvantages
are
explained,
so
we
also
tweak
duplicate
if
our
mod
shifting
all
really.
C
So
yes,
I,
agree
to
be
standards
track.
The
leaves
and
implementation
implements.
The
flavor
drafts
are
not
the
least
the
latest
revision
yet,
but
maybe
we
can
brush
it
up
this
week
and
make
it
work.
So
if
anyone
wants
to
do
some
intro
testing
and
we
have
a
server
online
to
do
some
interrupt
testing,
this
I'll
Center
details
of
the
configuration
to
the
list,
so
people
can
China
and
and
interoperate
I
had
one
question
of
what
one
concern
and
and
even
though
I'm
not
a
cryptographer
and
Scott
is
I.
C
Just
I
would
like
some
more
feedback
from
cryptographers
about
doing
to
encryptions,
with
the
same
information
that
only
differs
by
a
pea
piqué
and
how
much
that
could
possibly
open
us
up
to
an
attack.
So
if
an
attacker
can
see
both
the
encryption
with
the
PPK
and
the
encryption
of
the
auth
payload
without
the
PPK,
does
that
open
any
kind
of
attack
and
I'm
a
little
nervous
by
that?
H
Mark
McFadden
I
have
a
question
in
in
the
operational
consideration.
First
of
all,
let
me
say
that
I'm
in
support
of
this
is
being
a
standard
four
track
document,
but
in
the
operational
considerations
section
you
talk
about
PPK
distribution
and
you
talk
about
a
possible
way
to
actually
do
the
distribution
right
and
what
I
was
thinking
about,
and
this
actually
goes
to
the
discussion
of
charter
later-
is
that
it
seems
like
that
operational
consideration.
H
H
It's
understood,
understood
I,
guess
I'd
say
to
the
chairs
that
when
we
come
to
the
point,
are
we
actually
talking
about
chartering
again?
I
may
bring
this
up
again
because
I
think
that
in
I
think
the
operational
consideration
sections
is
excellent,
but
I
think
that
particular
example.
The
distribution
issue
ought
to
be
something
that
might
be
a
focus
of
another
document.
Thank
you.
I.
I
I
I
J
C
Scott
this
ball.
Yes,
it
came
out
clear,
I'll,
just
maybe
repeat
it,
because
this
microphone
seems
or
better
Scott
is
saying:
if
you
have
a
quantum
computer
you
can,
you
can
basically
attacked
it
if
you
Helmand,
get
all
the
components
of
the
I
cough
classic
payload
and
the
fact
that
only
the
PPK
is
different
doesn't
help
you
determining
the
PPK
after
if
you
got
both
calculations,
so
we
were
just
silent
because
we
were
like
we
don't
understand
this.
We
are
not
cryptographers.
A
E
One
of
the
things
that
the
by
its
information
on
is
that
when
you
put
a
new
draft
there,
it's
always
an
information
until
we
changed
it
to
standard
track,
usually
yeah
I
think.
Actually
the
document
default
is
an
informational.
So
that's
why
people
always
put
an
informational
first
but
I.
Yes,
I
think
it
should
be
a
standard
track,
because
this
is
something
that
is
modifying
it.
You
know
I
can
actually
would
next
question.
Do
we
actually
want
it
to
be
updating?
E
C
A
Okay,
so
so
the
question
for
the
hum
is:
should
we
should
we
promote
this
document
to
be
a
standards
track?
Yes
or
no?
So
for
yes,
please
hum
and
now,
okay,
I
think
that's
pretty
clear
consensus
around
standards
track
and
what
was
the
question
so.
E
The
next
question
is
that
do
we
actually,
because
this
is
actually
something
that
is
vague
in
most
of
the
IETF
principles,
we
have
sometimes
update
the
base
document,
and
sometimes
we
don't,
depending
on
what
we
do
in
some
cases,
there
are
actually
had
been
confusions
with
this
update
or
not
and
I
think
it
currently
already
had
one
document
that
updates
sub-base
I
conversion.
That's
a
signature,
authentication
product
I
think
this
belongs
in
the
same
category
that
it
should
be
updating,
even
if
this
is
optional,
but
it's
something
that
they,
you
know
modifies
the
basic
understanding.
E
C
So
as
well
think
that
it
should
update
the
base
document,
but
my
reasoning
is
more
of
that,
even
though
this
is
in
a
separate
document,
it
does
update
a
core
document
because
we
really
want
people
to
use
this.
This
is
not
like
some
optional
feature
that
gives
you
another
nice
shiny
flag.
This
is
something
we
think
is
the
core
security
of
the
IKE
protocol,
so
in
that
sense,
I
think
it
really
well
it's
not
mandatory,
but
like
we
strongly
recommend
that
you
do
this.
K
Tommy
Polly
Apple,
so
I'm
fine
with
it
updating
I,
think
it's
very
important
to
make
sure
that
people
are
aware
that,
when
they're
looking
at
like
free
to
spec,
they
see
this
in
the
link
at
the
top.
The
one
question
I
have
just
around
process
is
because
we
think
this
is
a
band-aid
for
now
and
then
at
some
point
that
will
get
replaced
is
so
presume
we
would
have
a
document
that
replaces
this
RFC
with,
like
here's,
the
right
way
to
do
quantum
resistant
long
term.
E
Yeah,
that's
it
yeah
I,
add
at
least
a
fine
idea,
if
more
about
that.
If
he
said
something
that
updates
something
it's
something
that
may
actually
consider
too
putting
into
the
base
document
the
next
time
in
some
cases.
So
it's
something
that
actually
modifies
for
except
the
signature
on
the
middle
craft
is
something
that
or
RFC
was
something
that
we
put
have
put
in
in
the
base
right
we're
into
document.
But
it
wasn't
ready
at
that
point.
So
that's
why
we
left
it
out.
E
That's
a
shame
is
there
are
say
you
know
raw
RSA
keys
and
something
like
that.
You
have
a
couple
of
these
cases
very
accelerate
or
so
so
we
we
are
think
about
okay,
but
also,
if
it's
a,
if
it
doesn't
say,
update.
That
means
that
okay,
people
doesn't
need
to
read
this
when
implementing
a
business
to
and
I
think
this
actually
goes
to
that
category.
E
You
don't
need
to
read
this
to
be
implementing
base
I
person
to
if
you
don't
care
about
for
the
rest
of
if
it's
something
that
modifies
something
in
I
would
say
that.
Okay,
if
you
implement
I
person
to-
and
this
one
says
you
must
not
do
that
in
the
base.
I
personally
and
it's
query,
updating
it's
something
that
modifies
the
base
I
person
to,
and
you
must
read
this
document
to
be
able
to
do
proper.
You
know
like
we
had
this
verification
of
the
diffie-hellman
parameters.
K
E
M
M
So
with
the
the
the
drafters
talking
about
the
previous
meeting
that
described
a
protocol
that
you
could
use
but
into
which
she
could
slot
any
asymmetric
algorithm.
So
I
guess
what
I'd
like
to
say
is
that
I'd
like
to
see
at
least
the
same
status
given
to
asymmetric
algorithms
as
opposed
to
the
symmetric
solution
with
ppk/s
and
yeah?
Given
that
ppk/s
are
seen
as
an
interim
solution,
then
just
to
you
know,
question
whether
it's
appropriate.
M
E
So
so
we
actually
have
a
charter
items
that
there
are
proposed
that
they
actually
start
working
on,
that
you
know
more
from
out
from
the
asymmetry
or
or
different
matters
of
fact,
to
real.
Imagine
for
the
point
of
resistant
versions
of
the
authentication,
substitution,
PPK,
but
actually
one
of
the
things
I
think
it
up.
E
People
have
been
saying
that
they
are
off
with
a
bit
early
so
that
that
it
might
be
actually
possible
that
those
are
actually
an
experiment
off
track
in
the
beginning,
depending
on
how
far
they
are
advanced
when
they
actually
published
they
might
be
standard
tracked.
If
you
think
that
they
are
ready,
but
I
mean
I
mean
that's
something
that
we
don't
have
to
discuss
me
to
come
to
discuss
that
after
we
get.
You
know
the
Charter
disgustin
got
you
thought
very
soon,
but
anyway,
so
so
Mike
as
I
said.
E
This
is
something
that
we
can
iterate
and
I.
Think
we
actually
probably
also
want
to
talk
with
a
lady
about
this
way.
If
it's
going
to
be
updated
on
or
something,
but
it's
something
that
I
want
to
people
to
think
about,
and
and
we
could
actually
take
a
ham
now,
but
I
have
now
so
so
two
up
since
is
take
disasters
make
this
so
that
it
actually
updates
the
big
face
like
person
to
document
and
the
adoption
is
that
we
just
leave
it
as
a
standard
track
document,
but
not
updating
the
current.
A
So
so
next
we're
Tara
is
going
to
lead
us
through
the
reach
are
during
discussion,
so
we
we
need
to
reach
Artur,
because
our
our
current
Charter
is
marked
to
expire
in
in
December
of
this
year,
we've
been
we've
been
having
a
number
of
discussions
on
the
list
about
potential
new
work
items
here
seems
to
be
interested
in
continuing
some
work
in
this
in
this
space.
So
Tara
is
going
to
lead
us
through
a
conversation
about
the
the
work
items
that
we've
received
to
date.
We're
also
open
to
some
discussion
of
of
additional
work
items.
G
E
Time
anyway,
because
we
have
this
deadline
in
our
throat
are
saying
that
we
have
to
reach
charter
by
the
end
of
2017.
Last
time
we
were
about
halfway,
you're
right,
our
charter
date.
You
need
to
reach
our
trip
before
this
and
we
were
hot
for
you
then,
after
that
you
know
this.
Oh,
we
hope
should
have
restarted
last
time,
so
we
started
doing
it
the
next
item.
So
now
we
are
trying
to
actually
do
this
on
time.
E
Okay
connects
fine,
okay,
so
we
have
a
items
in
the
Charter.
That's
already
done
the
DDoS
protection
operating
and
monitoring,
implement
algorithms,
adding
new
algorithms,
and
then
we
have
this
TCP
encapsulation
ECDSA
marked,
as
here
I,
said,
almost
done,
because
it's
rtq
so
much
next
time,
and
then
we
have
a
couple
of
items
that
are
still
in
there.
You
know
inverter
work
of
process
like
want
to
resistance
in
that
question,
to
split
DNS
and
implicit
IV
next
one.
So
this
is
the
Bay
boy
loop
right.
E
The
first
two
paragraph
of
tower
charter
and
I
actually
put
it
here
because
I
want
people
to
actually
read
it
and
see
if
there's
something
that
we
did.
You
modify
here,
because
this
is
something
that
has
been
staying
for
some
time
already
and
I
think
we
might
actually
useful
to
every
now
and
then
a
check
if
this
is
to
no
valid
so
ever
two
people
to
think
about
and
saying,
if
you
have
any
comments
about
that
story,
if
it's
fine
and
if
it's
fine,
okay,
it's
got
good
thumbs.
E
We
have
made
a
decision
earlier.
A
couple
of
five
years
show
called
back
on
that
we
are
not
going
to
be
do
any
work
based
on
a
question
one,
but
we
still
left
it
in
the
Charter,
because
if
there
is
something
work
it
could
be
done
on
like
version
one
I
want
it
to
be
happening
here,
not
in
some
other
place.
So
that's
why
I
have
you
actually
think
about?
It
might
be
a
good
idea
to
keep
it
here.
N
O
K
K
E
Right
so
so
what
I
have
think
people
aren't
saying
that?
Ok,
we
should
probably
actually
in
the
first
paragraph,
which
we
say
talk
version
one.
It's
still
actually
is
part
of
the
IPSec
protocols.
We
could
actually
put.
You
know
currently
already
obsolete
it,
because
that
it's
actually
so
credited
already.
The
right
person
won
by
first
two,
so
we
could
actually
add
it
in
there
baby.
Please
start
off
see.
E
C
Party
so
one
question
I,
guess
formally
speaking,
and
there
should
really
be
a
an
IPSec,
ops
and
IPSec
extensions
crew
like
technically.
We
should
split
it
into
now.
This
is
a
small
room.
We
have
doubled
in
size
over
the
last
year,
which
you
know
yay
I
think
it's
a
little
silly
to
split.
It
I
think
we
we
don't
have
enough
work
to
actually
have
like
an
option
and
an
extension
Rupert,
but
I
just
wanted
to
look
at
the
80s,
because
formally
that's
how
you
would
like
solve
some
of
these
issues.
E
P
Every
scroll
is
80
I'm
happy
to
leave
it
in
one
group.
Maybe
the
office
person
floating
around,
but
it
that
they
can
speak
up
at
some
point.
But
it
seems
small
and.
E
Anyway,
as
I
was
saying,
all
of
these
that
work
we
are
changing
here
are
something
that
we
actually
going
to
talk
about
in
the
list
anyway,
and
we
have
to
from
all
of
these
changes
in
the
list
first
and
then
10
it
will
do
the
IRC
and
or
ad
first
and
then
I
G,
and
so,
and
so
that's
why
I've
been
doing
live,
and
it's
here
isn't
doesn't
really
help,
because
we
have
to
still
call
did
I.
Do
the
make
a
list
go
to
the
next
part.
E
E
So-
and
this
is
something
that
we
are
talking
now
and
at
the
current
work
item
we
have
is
just
to
do
this
and
I
think
we
are
always
ready
another
eye.
That's
actually
not
really
covering
there.
You
know
the
later
stuff.
So
that's
why
this
is
just
to
pay.
First,
you
know
PPK
paste
the
stuff.
Anybody
has
any
comment
on
this.
E
Okay.
Next
one
split
the
a
nurse.
This
is
old
and
I
think
it
is
something
that
this
very
well
it's
already
done,
so
I,
don't
think
we
have
to
do
a
lot.
Okay,
next
implicit
IV.
This
is
also
all
old,
stuff
and
and-
and
we
actually
only
say,
ESP
format,
arguing
that
is
before,
when
this
used
to
show
that
this
actually
only
covers
ESP
as
it
is
now.
E
This
is
stuff
that
has
been
calling
around
a
long
time,
and
this
is
here
mostly
because
the
this
m/sec
working
group
was
closed
down,
so
it
doesn't
have
a
home
now
and
because
it's
still
too
late
to
drive
person
to
and
it's
related
IPSec.
So
that's
why
I
like
looking
on
your
home
to
actually
get
it
published
at
some
point
Brian?
How
old
is
this?
You
can
actually
tell
a
little
bit
about
this.
Are
your
etiquette
comments?
Brian?
Why
see?
What
was
the
question?
How.
Q
Old
is
it
yeah
yeah
we've
had
implications
for
in
the
drafts,
or
sometime
there's
now
at
least
two
partial
plantations
and
where
other
implications
I'm
aware
of
I
should
just
say
while
I'm
here
in
the
in
the
spirit
of
making
like
people
and
historic,
we
really
do
need
a
group
key
management
protocol.
It's
basically
like
me
to
to
replace.
So
we
have
to
do
anything
more
of
GOI
right.
Q
E
And
this
is
this:
is
a
new
item
so
now
I'm
going
to
first
get
information?
How
many
people
actually
think
that
this
is
this?
What
is
said
here?
What
it
is
is
clear
and
in
enough
for
the
described
as
a
charter
item,
so
is
there
anything
that
we
need
to
change
here
at
you
actually
actually
know
do
not
know
the
words
if
you
actually
think
this
is
a
work
item
that
you're
going
to
be
doing
is
clear
enough
that
we
can
actually
say
if
we
want
to
eat
this
as
a
charter.
I
think
not.
E
G
G
R
Draft
is
supposed
to
replace
the
G
Dorian
Dido
is
used
for
both
multicast
and
unicast,
and
we
don't
see
any
reason
why
why
the
new
thing
should
use
also
for
both
multicast
and
units
for
unicast
with
group
key
but
still
a
unicast,
and
so
that's
response
to
the
question
and
as
far
as
why
we
should
do
this.
Well,
m/sec
is
closed.
So
where
else
was
will
this
thing
go.
Q
E
So,
let's
go
to
the
humps
or
so
first
do
people
understand
what
we
are
trying
to
do
here
in
the
in
the
sense
that
you
can
actually
decide.
If
you
come
to
include
this
in
a
working
group
item
or
not.
So
if
you
think
you
understand
enough
to
be
able
to
not
actually
wear
it,
but
you
know
what
we
are
actually
doing
hum
now.
E
E
P
E
That's
actually
one
of
the
questions
yeah
so
so
the
next
question
is:
who
is
actually
going
to
be
willing
to
do
that?
Actually,
the
draft
authors
are
already
there
I
think.
Actually
they
are
still
willing
to
continue,
but
who
is
actually
willing
to
review
this
document
and
and
and
pass
work
on
it.
So
so
I
actually
want
to
see
a
show
of
hands
now
so
who
who's
willing
to
review
this
document.
So,
okay,
there
is
about
the
four
usual
suspects.
E
So
it's
it's
not
big
group,
but
it's
actually
currently
good.
That
was
better
than
normally,
because
that
wasn't
there's
no
I,
don't
think
there
is
any
authors
in
that
list.
Usually
we
have
those
outsource
our
order,
since
they
aim
for
people
who
are
reviewing
everything
alright,
so
okay
go
to
the
next
slide
next
item.
So
this
is
the
Valerie's
stuff.
E
Responder
mobile
and
TC
has
a
little
bit
more
questions
because
there
have
been
already
some
discussion
about
is
how
can
we
do
this
another
way
so
I'd
like
to
doing
the
responder
redirect
or
something
like
that,
and
that's
something
that
is
a
little
bit
more
unclear.
So
actually
I
was
thinking
about.
We
should
probably
skip
this
now
and
do
that
in
the
end,
because
I
think
this
is
going
to
be
having
more
discussion
in
in
whatever
that
we
are
actually
going
to
be
doing.
E
So,
if
you
go
to
the
next
part-
and
we
skip
come
back
to
the
here
later
so
post
quantum.
My
question
too,
though,
so
this
is
now
doing
first
quantum
in
in
a
sense
using
the
new
algorithm.
It's
not
doing
you
know
the
PPK,
but
using
newer
grimaces
also
includes
the
stuff
like
the
how
to
transfer
to
pick
stuff
in
between
the
like
out
or
before
dragged
out,
because
this
is
you
know
how
to
transport
those
keys,
because
there
are
bigger
than
time
you
can
fit
in
the
pace
and
so
message.
E
G
Queen,
dang
goodness
this
is
going
to
be
really
good
work.
However,
at
this
moment
we
think
at
least
we
will
receive
like
37
or
more
submissions,
so
I
don't
know
we
want
to
test
them
all
or
would
be
better
to
wait
for
like
in
a
year
or
so.
When
we
cut
down
these
submissions
to
a
reasonable
good
list
of
candidates,
and
then
we
can,
we
can
do
to
walk
over
here
to
test
those
out,
because
why
now
we
have
too
many
actually.
E
This
is
this
is
something
that
we
have
to
do.
First,
also
that
meaning
there's
this:
how
does
how
do
actually
transport
those
Peaks,
because
we
see
that
all
of
these
algorithms
are
going
to
be
requiring
bigger
payloads
that
we
can
actually
charge,
but
now?
But
so
so,
this
is
actually
something
that
we
can
ask
and
in
normal
case
I,
don't
know
I,
don't
think
we
actually
we
tried
here
not
to
fix
any
specific
algorithm
is,
and
this
couple
are
probably
going
to
take
couple
of
years
anyway.
E
G
E
E
S
G
G
So
so
so
so
in
order
to
do
this,
good
I
think
we
need
to
define,
which
ones
are
we
going
to
test
out,
but
but
in
order
to
do
that
at
least
I
think
we
would
pay
attention
to
the
NIST
submit
at
once
and
and
by
down
a
row.
We
know
you
know
at
least
like
10
or
15,
or
something
right
now
we
have
too
many
and
but.
E
What
we
have
there?
The
first
item
is
the
working
group
which
analyzed
the
possible
problems,
which
is
something
that
we
need
to
actually
analyze
them.
We
need
to
project
out
different
protocols
check
out
what
are
what
are
the
problems
if
you
try
to
use
them
like
okay,
this
one
you'd
requires
to
make
two
megabyte
messages,
so
we
have
to
do
something
for
that.
This
one
requires.
You
know
something
like
so
so.
This
is
something
like
this
is
like
very
generic
trying
to
get
okay.
Do
we
actually
want
to
work
on
this
wait,
but
so
yeah.
F
G
B
F
F
P
Air
Patrol
hat
off
I
mean
so
III
think
what
Richard
said
is
fairly
persuasive.
You
know,
if
you
I,
think
that
we're
cutting
here
that
seems
like
it
might
be
relevant.
Now
we
used
to
get
some
sense
of
the
impact
on
Ike.
If
you
had
to
move
to
larger
keys,
much
much
larger
keys
right
like
I,
don't
know
and
I,
don't
know
how
much
work
that
issue
actually
is
I,
don't
know
how
the
impact
on
the
performance
or
behavior
is
Mike
would
be.
If
you
had
to
like,
have
multi
megabyte
keys.
P
It
may
be
the
case.
We
don't
need
that
so,
like
si
D
H
has
keys
that
are
comparable
size
that
if
you
come
in
first,
it's
now
I
had
on
on
I
on
I
would
not
expect
the
this
working
group
to
be
so
picking.
Algorithms,
I
respect,
CFR
G
to
be
selecting
our
rooms
and
us
to
be
university'
them
across
ITF.
U
Anyway,
so
I
just
want
to
point
out
that
this
is
key
exchange,
not
signature,
yeah,
so
hash
key
signatures
doesn't
apply
so
I
personally,
not
personally,
but
I
do
think.
This
does
need
to
be
worked
on
right,
because
that
key
exchange
will
change
because
of
the
size
and
it's
best
to
start
working
creatively
on
how
you're
gonna
do
that
with
larger
key
sizes.
Now
then
wait
yeah,
yeah.
C
U
E
So
so
I
think.
The
reason
we
apply
we're
discussing
here
is
because
there
are
some
people
who
are
actually
doing
some
work
on
this
area
and
they
want
to
see
some
solutions
here
and
they
want.
You
know,
get
something
better
than
PPK
something
later
than
six
preset
keys
fit
for
quantum
registers,
and
this
is
something
that
is
and
I
expect
this
to
happen
to
take.
E
You
know
some
time:
it's
not
going
to
be
finished,
you
know
next
year
it
was
probably
going
to
be,
it
might
be
two
dozen
you
know
19
or
something
like
that
so
and
it,
but
there
is
so,
of
course,
one
of
the
things
that
we
can't
completely
separate
a
big.
You
know
payload
transport
from
this,
because
we
need
to
know
when
to
do
it,
because
there
is
an
option
or
we
could
have.
E
E
So
that's
why
I
don't
think
we
actually
got
separated
completely
as
oh,
we
do
the
pic
T
transport
first
and
then
we
start
working
on
this,
so
I
think
we
need
to,
but
we
need
to
check
out
first
what
what
is
required
for
each
of
these
okay.
This
one
requires
these
pics
algorithm
course.
This
information,
this
one,
because
the
how
many
you
know
message
sees
and
so,
and
so
we
can
actually
start
doing
the
modifications
right
protocol,
so
we
can
actually
plug
in
the
algorithms
later.
M
Markovski
and
CSE
I
support
this
as
a
chart.
Item
I
think
it's
important
that
we
should
be
working
on
it.
Maybe
some
of
the
concerns
that
have
been
raised
here
today
could
be
address
if
that,
if
it
was
slightly
reworded
to
make
it
explicit
that
that
it
that
it
is
going
to
be
algorithm
agnostic,
it
is
only
like
you're
developing
a
protocol
that
can
they
can
cope
with
the
longer
key
sizes,
and
it's
it's
nothing
to
do
about
selecting
algorithms.
E
Okay,
so
so
now
we
have
had
discussing
about
this
charter
item
itself,
so
how
many
people
actually
have
some
kind
of
understanding
what
we
are
trying
to
do
here
in
this
chart
or
either
so
we
can
actually
should
we
actually
know?
Can
we
actually
decide
whether
we
take
this
as
charger
item
a
note?
So
if
you
think
that
we
can
decide
now
so
hum
now,
okay,
if
you
think
that
we
need
more
more
information
and
can't
decide
now
hum
now,
okay,
so
I
think
that
is
clear.
That
people
think
that
they
can
understand.
E
This
is
issue
clear
now.
So,
okay,
so
next
question:
do
we
take
this
item
as
this
work
item
as
a
charter
item
in
the
IPSec
AMI
and
start
working
on
it?
So
the
other
option
is
that
we
leave
it
for
later
or
some
other
working
group.
So
so
the
first
option
do
we
work?
Do
we
add
this,
as
a
charter
item
to
my
question,
to
this
game?
A
if
you
think
so
hum
now,
if
you
think
that
this
would
be,
should
not
be
added
hum
now.
E
Okay,
I
think
there
is
clear
concepts
that
people
want
to
do
work
on
this
and
we
have
to,
and
we
are
probably
going
to
get
you
know
we
have
to
be
very
careful
about
what
we
write
here,
because
I
think
the
ad
said
so
or
I
could
be
concerned
about
that
we
don't.
We
are
not
going
to
be
defining
and
here,
okay,
next
one,
are
we
actually
at
here?
Okay,
how
many
people
actually
are
willing
to
work
on
this
review?
Draft
write
draft.
Do
the
analysis
on
algorithms
and
so
on?
E
Oh
okay,
how
many
people
are
willing
to
review
tryouts?
Okay,
we
have
more
than
normally
how
many
are
willing
to
help
editing
or
writing
troughs
okay.
We
have
a
little
bit
less,
but
I
think
that's
good,
so
the
next
time
piety
SP.
This
is
you
know
there
has
been
discussed
discussing
about
this-
that
we
should
be.
E
Let
me
see
if
the
compression
is
like.
Listen
to
compression
is
also
here,
so
this
is
not
only
died
at
the
ESPYs.
Also,
you
know
diet
I
Persian
to
so
how
to
compress
ESP
and
I
can
in
a
suitable
way
to
do
some
IOT
like
of
things.
There
are
some
cases
which
has
been
you
know,
discussed
and,
and
there's
a
couple
of
proposed
troughs
that
we
can
start
working
on.
E
So
is
there
anything
any
questions
about
this
rectum
people
actually
want
to
ask
before
we
actually
ask
if
you
understand
what
disease
okay,
so
how
many
people
are?
So,
let's
do
a
hum
against
directly.
So
so,
if
you
think
that
you
understand
that
the
vortices
and
we
can
decide
whether
we
take
a
stratum
hum
now.
E
Okay,
if
you
think
you
have
any
deeper
more
information
before
you
can
decide
hum
now,
okay,
so
I
think
we
still
have
more
people
understanding
what
what
we
are
actually
doing
here.
So
the
next
question
is:
do
we
actually
want
to
take
this
as
an
working
group
item
or
work
item?
Sorry,
work
item
for
our
charter?
So
if
you
think
that
we
should
take
this
as
a
work
item
hum
now,
if
you
don't
think
that
it
should
be
done
somewhere
else,
our
not
at
all
hum
now.
J
E
But
okay,
but
I
think
we
still
think
something.
We
have
more
people,
okay,
how
many
people
are
willing
to
work
on
this
meaning?
The
review
drops
out.
Let's
start
with
the
review
troughs
alright,
so
we
have
a
usual
five
or
six
abstract
suspect
and
how
many
of
you
are
willing
to
try
it
right.
The
drafts
and/or
edit
them.
Oh,
we
have
about
same
to
group
of.
C
People
don't
so
one
thing
I
would
like
to
see
is
more
people
from
the
IOT
space
actually
come
here
so
that
they
can
explain
better
to
us.
What
is
important,
because
we
can
do
all
these
things
about
compression
and
we
can
save
on
bite
and
double
the
CPU
usage
and
they
might
come
back
to
us
and
say
yeah.
That's
not
only
one
actually.
E
E
Okay,
let's
go
to
the
next
one
okay,
so
this
is
something
that
is
signature,
algorithm
negotiation.
So
we
added
big
enough
signatures.
We
add
negotiate
the
hash
algorithms
there,
but
we
don't
negotiate
the
actual
keith
we
are
going
to
be
using
because
in
most
of
the
cases
when
I
was
writing
that
specification,
I
was
thinking
about.
E
Yes,
you
have,
you
have
a
configuration
which
says
that
when
I'm
talking
to
Paul
I'm
using
this
key
when
I'm
talking
to
Valerie
I'm
using
this
key,
it's
the
same
thing
with
the
preset
key
same
thing
with
you
know:
certificates
and
RSA
keys,
I
configure
the
key
that
I'm
good
using
I
authentication
from
I'm,
not
so
say.
Okay
here
is
list
of
keys
use,
whatever
key
that
I
have
to
talk
to
whatever
people
you
have
the
out
there.
That's!
E
E
We
and
RSA
pkcs1
with
the
same
key
you're,
always
supposed
to
have
two
different
for
one
of
one
for
another
I
know
people
are
doing
this
using
the
same
key
for
same
things
or
different
algorithms,
but
you
are
not
supposed
to
do
that.
So
that's
why
I
have
a
original
idea.
Why
I
deviant
this
is,
or
we
actually
have
a
lot
discussed
on
rotation
to
get
a
signature
or
FC
reversed.
You
know
working
on
so,
but
he
says
has
been
issued
with
some
people.
Perhaps
me
saying
that
this
is
operationally
so
for
Tim.
G
G
F
G
C
Apparently
it's
22:6
for
the
Patriots
box,
so
I
do
think
that
we
need
a
little
bit
more
freedom
here
to
negotiate
or
indicate
the
support
of
everything's,
because
while
you
are
clearly
thinking
of
the
pre-configuration
predetermined,
you
know
a
server
client
case.
There's
many
opportunistic
in
other
cases
where
we
don't
know
who
we're
talking
to
him.
We
don't
know
their
capabilities
in
advance.
We
haven't
pre-negotiated
them
and
so
I
think
that
there's
a
problem
here.
C
As
for
the
RSA
case,
in
theory,
you
you're
right
that
it's
easily
salt,
because
the
what
is
that
rc8,
3
3
1
now
says
you
must
support
the
SSO
everybody
to
just
support
the
assassin
and
this
case
closed.
Unfortunately,
the
most
widely
available,
crypto
libraries
and
like
implementations,
don't
support
PSS,
yet
so
they're
all
violating
this
shiny,
new
RFC.
P
Yeah
arrow
scroll,
head
off,
yeah
I
mean
I,
guess
I'm,
not
super
pessimistic
that
we're
gonna
have
situations
where
the
same
key
will
be
used.
Multiple
rhythms
again,
we
said
hopefully,
you've
learned
a
lesson
there
is
you
Paul
scoffing,
but
you
know
maybe,
but
it
does
seem
like
it's
the
case.
P
That's
a
problem
for
solving,
even
if
you
don't
have
to
solve
this
hash
key
asynchrony
I
mean
so
certainly
I
guess
one
thing
I'll
say
you
know
that
Rupa
I
used
to
have
is
it's
the
same
thing,
but
over
in
TLS
we
sort
of
come
down
on,
like
you
know,
signature,
algorithms
and
hash
algorithms
are
all
just
one
giant
bundle
we're
not
gonna.
Let
you
in
I
you
know
on
you
know
this
century
are
when
this
hash
algorithm.
P
E
Done
all
right,
so
okay
I
mean
don't
do
the
normal
hums.
So
so
who
thinks
that
we
actually
understand
what
this
issue
is
and
think
that
we
should?
You
know
we
can
actually
decide
now
whether
to
take
the
starter
ons.
So
if
you
think
so
now,
okay,
who
thinks
that
we
actually
need
more
information
before
we
can
decide
hum
now.
E
E
So
if
you
think
so
hum
now,
okay,
if
you
think
that
we
shouldn't
add
this
as
a
work
item
hum
now,
if
it's
the
first
time
have
a
little
bit
more
resistance
but
but
I
still,
ver
still
still
have
made
clear,
my
rarity
on
saying
that
we
should
add
this
s.
Item
alright.
So
next
item:
okay,
now
we
go
back
to
the
Valerie's
cope.
E
This
one
because
we
skip
this
one
because
I
was
expecting
that
it
might
take
some
time
because
we
have
you
know
20
minutes
now
and
we
can
actually
tell
so.
This
is
assume
that
we
already
had
discussion
in
the
mailing
list.
There
never
say
you
know
people
asking
why
wait?
Why
can't
use
this?
Why
can't
you
use
that
and
that's
one
of
the
very
cool
when
I
was
to
looking
at
this
I
had
the
same
dream.
You
know
idea
that
I
was
you
know
you
can
use
you
can
use
you
know,
Rick
reaches.
E
You
know
redirect
as
an
information
exchange
after
they
say
it's
up,
so
you
can
actually
move
to
other
people.
Otherwise,
so
why
can't
you
use
that
and
I
have
no
idea
why
you
can't,
but
Valerie
clearly
has
some
idea.
Why
it
you
can't
you,
so
perhaps
you
can
actually
explain
a
little
bit
more
about
this
use
case.
E
What
other
issues
why
you
can't
use
the
redirect
mechanism,
because
this
is
something
that
we
have
to
you
know
know
if
you
can
use
redirect
it's,
but
still
a
good
idea
to
have
a
draft
or
RFC
saying
that
you
couldn't
use
it
this
way.
Just
if
we
don't
actually
add
such
new
protocol,
we
just
described
how
what
you
can
use
with
the
current
protocol
by
combining
them
together,
but.
F
So
redirect
is
clearly
a
possible
solution,
but
it
has
a
marginal
drawbacks
that
each
requires
the
fullness
depletion
of
focusing.
So
it's
big
penalty,
big
penalty,
because
it's
not
an
Olympic
penalty.
If
we
talk
about
resource
it's
only
its
some
cases,
it
may
be
inappropriate
because
for
for
the
client,
it
is
a
certain
well
if,
if
client
needs
to
enter
some
directive,
information
like
be
know
like,
if
password
to
authenticate
it
will
sudden
for
him
when
server
cluster
decides
to
switch
it
to
another.
G
G
F
E
Think
about
it,
I,
don't
actually
remember
if
more
Walker
cell
limits
it
so
that
you
always
has
to
be
in
a
theater
I
think
it's!
It
actually
allows
you
to
do
that
anyway,
but
anyway,
so
this
is
sort
of
the
exactor.
All
other
proposals
is
to
use
the
redirect
and
resumption,
so
you
get
rid
of
the
other,
but
anyway
I
think
you
actually
now
probably
understand
a
little
bit
more
about
this
question
and
it
seems
to
be
that
there
is
lots
of
different
ways.
E
We
can
actually
solve
this
issue
and
I
mean
you're
not
trying
to
solve
this
issue.
No,
we
are
trying
to
think
about.
If
you
actually
take
this
and
I
think
there
is
an.
You
can
clearly
see
that
there
is
solutions
out
there.
So
this
is
not
something
that
we
don't
know.
If
you
actually
thought
it
can't
be
solved
with
Gerstel,
you
see
how
easy
and
what
kind
of
product
race
and
ID
says
for
writer,
but
for
parties.
C
C
You
actually
do
the
actual
mobike
update,
and
I
found
that
basically
everybody
in
a
wild
skips
this
this
step
and
just
sends
an
update,
even
if
it's
not
updated,
because
you
know
it
saves
you
some
latency
in
a
roundtrip,
so
I
was
wondering
maybe
to
actually
expand
this
charter
item
and
say:
let's
just
do
an
update
on
mah
bike.
Let's
look
at
all
the
mob,
Ike
stuff,
that's
that's
out
there
and
actually
do
a
sort
of
clarification
updates
document.
C
That
also
takes
this
issue
into
account,
but
also
maybe
look
look
at
other
things
and
one
other
note
I
wanted
to
make
Spencer
the
Debian
and
Ubuntu
kernels
disable
the
access
from
migrate,
which
is
used
by
mobile
to
actually
migrate.
Yes,
a
to
another
end
point
because
of
security
reasons.
So
maybe
we
should
somehow
also
have
another
document
that
tells
those
people
that
you
know.
No.
C
D
So
Daniel
me
go
I.
Think
issue
to
solve
is
a
rather
important,
be
able
to
switch
from
one
context
to
inform
one
machine
to
the
other
that
we
use
mobile
or
redirect
is
I
would
say
me,
call
I,
don't
know,
don't
really
care,
but
that
we
define
the
context
and
that
we
able
to
transfer
that
contract
from
one
house
to
the
other
is
something
important.
I
guess
and
I'd
like
to
see
that
in
each
other,
all
right,
okay,.
E
Right,
so
to
extend
this
to
be
more
mobile,
clarification
document,
and
just
only
you
know
this,
this
kind
of
this
once
more
work
item.
This
is
actually
said
yeah.
This
is
something
that
we
can
actually
think
about,
and
this
is
something
of
course.
We
need
to
this
text
isn't
suitable
for
that
Commission's.
We
need
to.
We
have
yeah,
but
we
need
to
be
cut,
discuss
about
this
a
meaningless
but
anyway.
So
so,
let's
see
let's
make
our
home.
E
So
so,
if
you
think
that
we
have
enough
information
that
we
can
decide
whether
what
kind
of
we
have
enough
information,
whether
we
want
to
do
some
kind
of
mobike
related
work
item
in
the
the
charter,
not
necessarily
you
know
this,
but
it
might
be
also
the
other
one
I
was
thinking
about
having
you
know,
you
know
some.
So
if
you
think
of
the
beat,
you
understand
problem
well
enough
that
we
can
decide
whether
we
take
a
mobile
bike
related
working
out
or
not.
E
So
if
you
think
that
you
understand
hum
now,
if
you
think
that
we
need
to
get
more
information
in
the
mailing
list
or
something
like
that
before
we
can
decide
hum
now:
okay,
people,
we
have
really
smart
people
here.
They
all
know
everything,
okay.
So
the
next
question
is
then:
do
we
want
to
take
some
kind
of
mobile
related?
You
know
I
thought
this
is
what
verbal
but
I'm
going
to
go.
E
Ask
the
specific
question
also
to
be
able
to
take
this
exactly
or
do
we
actually
want
to
have
just
generic
mobile,
but
I
want
to
I
want
to
see
if
there
is
any
conscious
that
they
actually
want
to
update
my
bike
in
in
some
way
or
not
before
we
actually
go
in
what
maybe
we
are
going
to
be
changing?
You
know
this
is
road.
We
hams
after
humph
after
this
this
just
you
know
they
do.
We
actually
want
to
take
something
that
operates
more
like
a
new
working
group.
E
E
So
so
I
think.
Actually
it
is
something
that
yeah.
Let's
me
see,
we
have
ten
minutes.
So
I
think
this
is
something
that
we
need
to
go
back
to
the
list
and
start
working
on
mailing
list,
because
we
will
no,
no,
no
lamps
start
here
in
11.
So
so
so
that's
why
I
was
a
to
think
about.
I
think
this
is
something
that
we
can't
decide
now,
so
we
actually
go
through
code
back
to
the
mailing
list
and
we
probably
need
to
have
a
poll
right
she's.
You
know
proposal
what
is
called
to
be.
E
F
I
think
that
we
are
a
little
a
little
bit
more
than
needed,
concentrated
on
my
bike,
as
a
real
question
is
how
to
make
a
lot
share
in
cluster
with
a
basic
and
it
can
be
well.
My
bike
is
just
one
a
solution
for
I
mean
my
draft
is
just
one
attempt.
It
just
causes
problem,
but
the
problem
can
be
solved
in
different
ways,
and
it's.
G
E
We
have
done
some
work
on
load,
balancing
and
high
availability
earlier,
and
most
of
those
have
been
going
quite
badly,
I
think
because
we
have
we
haven't,
we
have
had
issues
there,
because
those
calls
so
deeply
into
the
implementation
in
most
of
the
cases
that
it's
people
have
a
different
solution.
So
quite
often
it's
actually
is
more
implementation.
Detail
stand,
you
know
actual
protocol
issues
but
anyway,
so
so
we
come
back
to
this
on
the
list.
So
I
think
there
was
a
couple
of
more
with
that.
E
Bush
didn't
have
a
charter
text
ready,
so
there
is
this
ipv6
ipv4
address
failure
issues.
This
is
the
new
draft
that
was
sent
the
mailing
list
couple
of
well
because
it
came
to
me
as
an
eye
and
eye
expert
and
they
said
okay,
we
publish
this
document
two
weeks
ago
and
we
want
to
get
on
a
number
for
it,
and
this
was
the
first
time
I
ever
saw
that
trough.
So
then
I
said
no.
E
We
want
to
you
want
to
bring
it
first
today
you
know
IPSec
Amy
and
if
IPSec
Amy
says
no,
they
need
to
have
core,
of
course,
code.
The
IDs
and
AD
bilko
and
and
I'm
not
sure
if
the
traffic
actually
I'm
quite
sure
that
draft
isn't
the
draft
isn't
stable
enough
that
there
might
be
some
changes
to
that
in
the
last
call.
So
that's
why
I
don't
want
to
keep
numbers
before
they
are
stable
enough.
So
I
want
to.
You
know
people
accidentally.
There
were
some
interview
on
that,
but
it's
actually
something
that.
E
Should
we
take
it
to
their
work
item
or
not
it's
the
outer
of
the
craft
by
the
way
here,
somewhere,
no
okay,
so
I
was
think
about
because
yeah.
So
so
this
is
coming
from
the
3gpp
side.
They
have
it,
they
have
a
menu.
Do
you
know
3gpp
authentication,
you
have
this,
do
ipv6
hundreds
ipv4
address
and
you
may
get
an
error
for
both
of
those
or
you
get
one
or
only
you
know
other
one
or
something
like
that,
and
that
error
is
something
that
they
want.
You
know
propagated
all
sort
of
the
IPSec.
E
It's
only
if
you're,
you
know
if
you
didn't
get
IPPC
ipv4
address
at
all
from
the
public
mobile
network
when
you
just
want
to
be
also
sent
that
that
IP,
second,
that
okay
I
didn't
get
I've
been
before
it.
So
that's,
that's
I!
Think
it's
the
best
idea,
but
crafty
skin
Eric
saying
that.
Okay,
we
can
do
this
in
any
case
system,
where
you
have
an
address
allocation,
you
say
that,
okay,
no,
you
can't
get
typing
before
you
get
on
ipv6,
but
you
didn't
request
it
so
come
on.
We
got
it
again.
E
So
so
do
people
understand
what
this
item
is.
If
you
send
people
noting
I,
don't
think
we
can
actually
decide
on
on
the
Tartar
item,
because
we
don't
have
any
text
here,
but
I
would
like
to
get
through.
You
know
something
a
little
bit
concerned,
or
do
you
think
that
this
is
something
that
we
should
start
working
on
to
added
charter
items?
Also
we
actually
or
do
we
just
say:
okay,
go
to
individual
life,
you
know
submission
or
or
go
to
the
ad
sponsored
our
draft.
A
E
Okay,
so
I
think
we
need
to
get
more
really
preview
on
that,
and
then
you
can
actually
decide
okay.
So
let's
come
to
that.
So
it's
not
the
night
became
a
prophet
and
it's
not
the
worker
out
there,
but
I
want
the
people
to
actually
read
and
comment
okay.
So
what
would
a
please
ipsec
that
was
Paul's
late
minute
addition
with
no
hats,
I.
C
Was
ready
for
straight
hat?
Actually,
yes,
this
was
redhead.
I
get
just
did
answer
me.
I
did
send
text
to
the
mailing
list
for
the
Charter,
but
it's
basically
we
have
security
context
and
we
want
to
be
able
to
negotiate
them.
So
we
can
sort
of
chain
machines
in
with
different
security
levels.
Together
we
had
this
an
IKE,
v1
I
was
badly
hacked.
Then
it
was
slightly
less
badly
hacked
and
I'm
hoping
we
can
do
a
really
nice
solution.
Fight
with
you.
R
E
So
so,
but
I
I,
don't
think
we
actually
can
make
any
decisions
about
this
here,
but
I
think
that
the
subject
that
talk
we
should
take
at
least
and
and
and
think
about
and
foremost
read
the
Provos,
your
proposal
and
it's
actually
something
that
we
actually
might
want
to
have
a
design
team.
Think
about
you
know
what
actually
protocol
it's
good,
because
there
are
so
many
different
ways
to
doing
this,
and
and
and
and
it's
not
it's
also,
as
you
said,
we
don't
really
use
it.
E
So
how
do
we
know
what
what's
problem
you
are
trying
to
solve
and
if
we,
if
you
don't
know
what
problem
you're
solving,
if
you
know
that
our
solution
is
good
to
be
suitable,
so
it
might
be
useful
to
first
get.
You
know,
people
who
actually
use
these
to
actually
tell
us
what
they
actually
want
before
we
actually
start
working
on
it
all
right.
So
let's
go
to
the
others
yep.
So.
K
Tommy
Paulie,
Apple
and
I'm
speaking
on
behalf
of
what
David
Scotty,
also
from
Apple,
sent
out
earlier
this
morning
as
proposed
order
fixed
at
seven
thirty
nine.
So
you
had
plenty
of
time
to
read
it,
and
so
this
is
it's
essentially
trying
to
say.
Should
we
be
addressing
some
ability
to
mitigates
privacy
concerns
around
certain
deployments,
so
he'd
send
out
a
proposal,
I
think
on
the
list
a
little
bit
ago.
K
E
This
reminds
me
one
of
the
things
that
we
forgot
from
the
latest
is
the
pseudonyms
we
have.
We
were
having
this
discussion
in
in
PPK
context.
We
were
saying:
okay,
we
don't
want
to
actually
do
the
ID
protection
now,
because
we
can
actually
do
it.
So
the
names
and
I
think
children
will
probably
solve
that
issues.
Meeting
me:
plural,
meaning
that
in
the
ID
payload
we
actually
trust
random
thing,
and
when
we
actually
get
you
know
the
authenticated
thing,
then
we
actually
update
the
sailed
along.
E
So
every
time
we
have
I
have
a
list
of
ten
of
an
external
be
using
and
and
then
you
know,
when
we
run
out,
we
actually
get
new.
Okay.
Give
me
ten
new
random.
You
know
IDs
for
me
that
I
can
use
so
I,
never
tell
you
about
Real
ID
at
the
end,
or
you
know,
only
the
one
who
actually
gave
gives
me
those
not
a
list
of
children
who
will
be
able
to
map
it
or
allow
the
ID
so.
K
E
That
that's
actually
something
that
yes,
because
I
think
the
children
who
issue
something
that
would
be
actually
interest
for,
at
least
for
me
and
I,
think
there
was
a
puppy.
Other
people
also
wanted
to
solve
this
issue.
So
it
might
be
actually
told
that
me
if
the
children
is
probably
solving
his
problem.
Also,
that
might
be
actually
done.
I
didn't
order.
H
Is
just
following
up
on
something
I
said
before
I
think
for
the
Charter
one
of
the
things
that
we
should
add
are
some
operational
documents
so
basically
either
be
CPS
or
here's
how
you
actually
implement
what
it
is,
we're
proposing
and
I
give
them.
My
exam
I'm
gonna,
give
two
examples
here
that
I
think
we
should
have
documents
for,
and
one
is
the
PPK
distribution
example
right,
a
short
document,
but
that
describes
the
mechanics
of
actually
doing
the
distribution.
The
second
one,
I
think
is
the
algorithm.
No
negotiation
is
an
example
of
here's.
H
E
Do
okay,
okay,
so
so
that
was
to
last
other
items.
So
so
what
we
are
next
our
plan
for
this
is
actually
now
it's
going
to
be
working
that
well,
but
our
plan
was
first
shown
so
that
we
actually
as
a
chairs,
we
work
on
that.
You
know
charter
text
with
people
who
are
on
Excel.
It
said
by
the
end
of
the
week.
We
would
have
something
that
would
be
something
that
we
could
actually
discuss
and
say
that
this
is
the
Charter.
E
But
okay,
we
have
now
a
couple
of
items
there,
of
course,
which
still
require
more
work.
But
if
you
try
to
get
you
know
the
base
stuff
that
we
already
agreed
on
in
a
format
that
we
can
actually
say
that,
okay-
or
if
these
are,
you
know,
you
know
suitable
for
everybody,
okay.
If
so,
then
we
can
actually
forward,
ladies,
which
will
then
say
there.
You
know
birds
on
that,
but
but
then,
then
we
have
a
couple
of
these
things
that
we
I
would
like
to
create
very
quickly.
These.
E
V
P
I
mean
you
know:
III
go
back
to
Richard
Barnes's
always
be
closing.
I
mean
like
well.
I
want
to
see
is
like
work
happening.
That
is
valuable
work.
That
is
like
that
people
want
to
do
so.
That's
why
I
asked
people
want
to
work
on
things
so
so
my
perspective
as
long
as
people
want
to
keep
working
on
it
and
I
guess
I,
nice
and
future
me
I
want
to
ask
who's
gonna
implement,
but
as
long
as
those
things
are
happening,
I'm
I'm
happy
to
have
this
continue
when
this
is
aren't
happening.