►
From YouTube: LAKE WG Interim Meeting, 2020-08-20
Description
LAKE WG Interim Meeting, 2020-08-20
A
Okay,
I
think
it's
time
to
get
started.
At
least
here.
It's
it's
pretty
late,
past
midnight,
so
this
is
yari
and
I
am
lake
working
with
pika
stephen
has
a
family
issue
in
china
slate.
A
So
I
started
the
call
and
steven
has
posted
an
agenda
which
we
will
follow.
Of
course,
if
you
have
comments
on
that
sentence,
we
can
reshuffle
that
and
I
have
turned
on
recording.
A
A
So
stevens
agenda
said
we
would
do
basically
three
things:
one
do
around
the
table:
status,
updates
or
introductions,
not
that
we
need
introductions
per
se.
But
if
you
have
an
update
on
what
you've
been
up
to
recently
in
recent
weeks,
then
please
give
us
that
or
a
burning
thing
in
your
mind
that
we
have
to
solve
this
at
least
to
us.
A
Then
we
should
talk
about
the
way
forward,
in
particular
like
how
do
we?
We
have
lots
of
text
and
lots
of
people
writing
different
things.
How
do
we
move
forward
and
actually
get
something
done
that
I
think
will
be
useful
to
talk
about
and
then
what
to
do
next
in
terms
of
meetings
and
so
forth,
so
starting
with
the
intro
status
updates,
and
I
actually
try
and
do
this
call
people's
names
make
sure
that
we
go
through
everybody.
A
You
don't
have
to
say
something
if
you
have
nothing
to
report,
but
I'll
just
go
by
the
order
that
I
have
on
my
screen
and
we'll
start
with
coming.
C
Hi
dominique
it's
august
in
the
uk,
so
most
of
us
are
on
holiday.
So
not
a
lot
has
been
done
yet,
but
I
just
wanted
to
thank
you
yari
for
the
comments
that
I
just
received
on
the
on
the
email
list.
I
noticed
you
sent
out
a
number
of
comments,
including
on
your
own
document,
so
that
was
good
and
it
was.
It
was
really
good,
actually
really
really
helpful
for
me
and
just
want
to
make
sure
that
we,
you
know
kind
of
figure
out
how
those
documents
fit
into
moving
forward.
D
Sorry,
finding
the
mute
ira
mcdonald
consultant
in
the
toyota
global
vehicle
security
team
and
I'm
like
colin
and
some
others,
probably
listening
in
to
learn
more
about
this.
I've
been
following
the
mailing
list
and
reading
some
of
the
documents,
as
I
could.
A
Thanks
and
and
then.
E
A
Great
welcome
and
I'll,
if
I'll
jump
back
to
christian,
oh
wait,
you
may
not
be.
Yes.
Are
we
able
to
speak
it?
Yes,
you
are.
F
Sorry
I
was
almost
ready
to
get
a
lunch
and
then
I
realized
that
I
had
to
join
you,
so
I'm
here.
Yes,.
A
Sorry,
sorry
about
your
lunch
yeah,
so
we're
in
a
sort
of
status
report
around.
So
if
you
have
any
particular
things
you
want
to
mention,
that
would
be
a
good.
F
Oh
okay,
I'm
cushioned
with
emma
I've
been
following.
The
model
t
work
from
the
beginning
and
I
am
not
spending
as
much
time
as
I
would
like
on
it,
but
I'm
still
following
and
trying
to
to
get
there.
A
Okay
thanks
and
then
joe
s
or
josh.
This
is
joe.
G
Yeah
this
is
joe
falloway
yeah.
Sorry
I,
my
webex
name
is
not
as
clear
as
it
should
be
yeah.
I
have
nothing
new
to
report
just
trying
to
catch
up
on.
What's
going
on
here.
H
All
right
and
then
mark
hi,
it's
mark
mcfadden.
I
have
a
couple
updated
drafts,
one
on
changes
to
the
threat
model
and
another
one
that
I
sent
to
the
map
research
group.
It's
a
piece
of
work
that
does
a
statistical
analysis
of
words
and
security,
consideration,
sections
and
it's
an
interesting
piece
of
work.
That
sort
of,
I
think,
speaks
to
the
model,
t
work
but
kind
of
obliquely.
H
So
those
are
the
two
updates.
I
have
thanks
gary.
A
Thank
you.
We,
the
the
other
draft
you
had
to
change
these
on,
but
the
other
one.
A
H
So
there's
really
three
drafts
in
play.
There's
one!
That's
about
the
changes
to
the
threatened
environment!
There's
one
on
the
endpoint
taxonomy
draft
that
I
did
update
in
july
and
then
I
was
just
calling
to
your
attention
and
I
won't
talk
a
lot
about
it.
But
a
piece
of
work
that
co-author
and
I
did
on
taking
a
look
at
word-
counts
in
security,
consideration
sections
and
seeing
how
they've
changed
since
2003..
A
All
right,
thanks
and
martin.
I
Martin
thompson
I've
been
following
since
the
beginning:
I've
got
a
draft
that
I
submitted.
We've
discussed
a
little
bit,
but
nothing's
happened
since
we
last
met,
and
unfortunately
I
was
sleeping
when
you
already
said
his
email.
So
I
haven't
read
them
yet.
A
All
right
and
russ.
J
So
I've
also
been
following
this
group
since
the
beginning.
I
have
not
yet
read
er
gary's
emails
from
not
that
long
ago,
but
otherwise
I
think
I'm
up
to
date
on
the
list,
but
nothing
new
to
report.
A
All
right
and
tommy.
K
Yeah
hi
yeah,
so
I'm
I'm
tommy
c
from
ncsc,
so
I've
been
following
this
work
from
the
beginning,
I'm
interested
to
see
where
it
is
developing
from
today,
and
I'm
also
looking
how
ncsc
can
bring
more
of
our
work
on
attacked
events
in
to
build
on
the
indicators
of
compromise
work
that
we
have
done
this
far.
A
Thanks
and
wendy.
L
Yes,
wendy
seltzer
w3c
following
this
work,
mostly
to
and
working
to
to
bring
it
to
w3c
and
ideas
from
there
here.
Vice
versa,
including
the
the
work
that
jeffrey
is,
is
doing
on
threat
models
and
we've
recently
seen
some
new
work
on
privacy
protective
features
incubating
in
a
relatively
new
privacy
community
group
at.
L
A
Anybody
guess
not
so
updates
yeah
there's
been
some
some
email
updates.
Perhaps
so
I
did
some
reviews
and
me
mostly
thinking
about
like
how
to
move
forward
and
as
useful
with
the
itf.
It's
maybe
not
so
much
about
like
adding
stuff
but
making
sure
that
that
you
actually
have
a
sort
of
consistent
story,
piece
that
could
be
published
and
has
the
necessary
pieces,
but
not
not
too
much
else
and
yeah.
I
guess
we
can
talk
about
that
in
in
a
bit
and
otherwise
not
much
on.
A
So
so
maybe
the
next
item,
then,
is
what
steven
had
suggested
that
we
talk
about
the
way
forward
and-
and
I
guess
his
point
of
view-
was
that
that
we're
heading
towards
two
different
things:
one
trying
to
offer
the
idea
some
text
as
a
possible
update
of
pcb38-
or
maybe
it's
not
a
document
too,
and
secondly,
writing
some
background
documents
that
explain
the
motivation
for
that
or
or
maybe
it's
just
motivation,
but
it
could
also
be
other
things
like
guidelines
on
how
to
deal
with
specific
issues
and
so
forth.
A
But
the
general
trust
seems
to
be
that
we
would
have
like
two
two
separate
pieces
of
work.
One
one
is
something
that
we
could
send
off
to
the
ietf.
We
don't
decide
on
on
these
pcb
changes.
Obviously
it's
the
itf's
turf,
but
we
could
certainly
send
a
document
or
individuals
could
send
the
documents
and
say
hey.
Let's,
let's
look
at
this
on
the
idf
side,
so
that's
one
and
the
second
one
would
be
this
background
information.
A
It
may
be
a
summary
of
of
this
or
summary
of
of
different
kinds
of
issues
to
take
into
account
or
or
something
like
that
and
of
which
we
have
several
candidates
right
now.
Several
documents
have
in
that
space.
A
H
As
far
as
a
two-pronged
approach
goes,
I
I
actually
think
that's
a
good
idea,
because
I
think
the
authors
of
some
of
the
drafts
that
talk
about
the
threat
models
can
sort
of
meet
one
of
the
goals
that
was
in
the
charter
right
and
that
the
second
bullet
of
deliverables
in
the
charter
says
that
one
of
the
things
we're
going
to
do
is
produce
a
document
that
takes
a
look
at
how
bcp
72
either
matches
or
doesn't
match
today's
reality.
H
H
I
don't
think
we're
in
a
position
and
I
would
argue
against
offering
text
to
the
itf
at
this
point.
That
is
an
actual
update
to
bcp
72.
I
don't,
I
don't
think
we're
even
close
to
that
yet,
but
the
two-part
approach
seems
reasonable
to
me
and
the
the
second
thing
I
would
say
about
it
is
that
I
think
we
have
some
documents
already
that
if
we
brought
them
together,
edited
edited
them
so
that
they
were,
let's
say
a
little
a
little
stricter
a
little
a
little
more
concise.
A
C
Yeah
yeah
yeah-
this
is
dominique.
I
was
gonna,
I
was
unmuting
as
mark
was
unmuting,
I
yeah
I
do.
I
do
support
the
two-prong
approach
and
similarly,
just
I
only
briefly
looked
at
your
comments
because
you
sent
them
about
20
minutes
before
and
I
too
was
resting
before
the
call,
but
basically
I
think
the
the
the
content.
C
We
have
quite
a
lot
of
content
and-
and
I
think
one
of
the
comments
that
you
had
was
was
on
focusing
and
I
think,
along
the
same
lines
as
mark's
comment,
if
we
take
all
of
that
together
and
focus
a
lot
of
the
the
research
and
the
input
and
the
information
and
the
editing
that
needs
to
be
done,
that
can
inform
the
second
part
of
it,
but
I
think
you
know
we're
we're
not
quite
yet
at
that
point
where
we
can
do
an
update
to
bcp
72,
but
I
do
think
we
have
a
lot
of.
C
We
have
a
lot
of
information
and
we
have
a
lot
of
you
know,
sort
of
people.
Thinking
about
it.
We
have
seven
drafts,
obviously
as
well
and
now
also
the
w3c
is
participating.
Thank
you
and
I'm
looking
forward
to
to
hearing
more
from
them,
but
I
think
we're
at
a
point
where
we're
gathering
and
consolidating
information
in
a
way
that
reflects
the
current
security
information
that
can
then
inform
the
next
step.
After
that,
thanks.
F
Question
here
please
yeah,
I
mean,
I
think,
having
a
taxonomy
of
the
threats
is
a
really
good
idea,
and
I,
like
dominic's
draft
on
that,
because
I
mean
things
like
botnet
fishing
network
penetrations
is
something
that
we
have
to
be
concerned
and
that
have
to
be
out
there
too.
Much
of
the
threat
model
that
we
have
now
is
focused
on
just
addressing
the
direct
communication
between
two
points
and
not
looking
at
the
network
effects
is,
is
a
bad
idea.
F
What
seems
to
be,
I
mean
there
are
little
details
in
dominic's
draft
that
I
I
would
like
to
to
see.
I've
been
a
bit,
I
would
say
expanded
on.
F
One
is
our
use
of
network
attack,
which
is
very
ambiguous
in
an
atf
context,
because
network
attack
in
in
an
itf
context
means
attack
on
the
waters
typically,
and
that's
not
what
we're
speaking
about.
We
are
speaking
about
typically,
an
attack
that
culminates
in
something
like
bridging
the
active
directory
for
the
site,
or
something
like
that.
It's
basically
attacking
a
community
rather
than
attacking
a
piece
of
hardware
and,
and
so
that
should
probably
be
expanded
a
little
bit
on.
F
The
other
thing
that
I
I
like
is
about
the
fishing
part,
but
we
should
probably
have
a
description
of
the
relation
between
all
that.
Typically,
the
fact
that
an
attack
starts
with
a
bridge
head
which
is
typically
obtained
by
fishing,
but
can
be
obtained
by
something
else
and
then
goes
into
expansion
to
eventually
try
to
control
the
network,
and
that
is
something
that
is
important
for
me
and
I'm
guilty
there
of
not
giving
feedback
early
and
providing
text
to
dominic.
F
For
example,
if
we
follow
the
attack
model
in
which
fishing
leads
to
a
beach
head,
lead
to
a
lateral
propagation
then
leads
to
a
network
compromise
and
then
leads
to
all
kinds
of
exploitation.
F
There
are
different
defenses
that
apply
at
different
points
and
so
it'd
be
very
interesting
to
get
from
the
first
taxonomy
of
attacks,
a
second
one,
which
is
a
taxonomy
of
defense.
But
for
me
that
should
be
a
second
step,
and
I've
already
spoken
too
long
so
give
the
floor
to
whoever
wants
to
speak
next.
I
I've
got
a
couple
of
questions
and
I'm
wondering
if
you
can
hear
me,
because
this
setup
is
weird.
I
Oh
excellent,
so
I
think
the
the
broad
spectrum
of
threats
that
we've
sort
of
seen
documented
in
these
drafts
are
interesting
and
it.
I
The
question
that
I
have
and
the
question
I've
been
struggling
with,
is
what
does
the
ietf
do
about
some
of
these
and
I
think,
there's
there's
various
contexts
in
which
we
do
understand
what
happens,
but
the
the
protocol
design
aspects
of
of
this
are
much
less
clear
to
me,
and
so
I'm
I'm
interested
in
a
concise
listing
of
things
that
we
might
concretely
suggest.
As
as
it
goes
to
your
first
side
of
things
when
it
comes
to
something
like
fishing
and.
I
Establishing
a
beach
head
and
lateral,
propagation
and
and
those
sorts
of
things
I
think
there
is
work
that's
going
on,
but
I
haven't
seen
a
very
concise,
clear
summary
of
the
of
the
actions
that
we
might
then
take
as
a
result
of
that.
But
do
you
understand
that.
F
But
martin,
I
I
think
I
agree
with
you
and
that's
the
reason
why
I
said
it
should
be
a
two-stage
approach.
First,
you
list,
I
mean
all
the
threat.
The
big
attacks
are
hanging
below
our
nose
and
we
we
absolutely
have
to
acknowledge
that
otherwise
I
mean
there's
no
credibility
in
the
work
and
then
there
will
be
defense
against
those
attacks
and
only
a
fraction
of
the
defense
is
something
that
is
in
the
ietf
purview,
but
we
have
to
do
that
in
the
two-step
approach.
First,.
I
I
I
think,
that's
reasonable,
the
the
I
guess
the
question
is
how
far
you
get
down
the
path
of
of
carefully
documenting
all
of
these
various
threats
before
you,
you
have
to
you
start
to
try
to
work
out
where
it
is
that
that
we
have
some
purchase
on
them
and
where
it
is
that
we've
just
decided
that
simply
these
are
out
of
scope
and
we'll
not
deal
with
them.
That
bit
is
what
what
bothers
me
most.
I
think.
A
Here
so
I
think
you
you're
asking
really
good
questions.
I
I
agree
with
christian
that
this
two-step
approach,
if
I
understand
his
approach
correctly,
is,
is
probably
the
right
answer
for
that
and
from
my
own
perspective,
I
think
it's
always
the
case
that
you
have
to
understand.
A
Record
recognition
of
this
is
important,
but
there's
a
other
side
of
your
question,
which
is
that
it's
not
like
one
thing
that
we're
documenting
it's
like
a
a
a
number
of
different
things
that
we
could
be
looking
at
and,
and
then
the
question
comes
and
which
ones
of
those
we
actually
want
to
dig
in
deeper.
A
I
think
that
that's
a
that's
a
good
question
to
ask,
and
I
don't
necessarily
have
an
answer,
but
we're
not
gonna,
don't
document
everything
that
has
changed
in
the
intent
since
you
know
20
years
ago,
and
that
that
would
be
too
much.
We
need
to
select
a
particular
subset.
That's
why
I've
been
calling
for
this
more
focused
approach.
F
That's
true
at
the
same
time,
there's
something
that
not
inside
about
the
tension
between
security
and
privacy,
but
what
we
see
in
a
number
of
those
big
attacks,
like
the
attack
against
the
office
of
budget
management
in
the
u.s,
is
the
the
fact
that
the
attacks,
the
goal
of
the
attack,
is
to
acquire
a
personal
identification
data
and
database
on
people
that
you
can
then
use
to
break
those
people's
privacy
and
also
to
mount
further
phishing
attacks.
And
things
like
that.
F
H
It's
mark
again
can
I
sort
of
follow
up
on
something
else
that
martin
said
and
that
that
there
there
would
be
sort
of
like
a
if,
if
we
think
about
christian's
two-step
approach
and
that
the
first
step
is
this
sort
of
taxonomy
and
I
think,
martin
and
not
putting
words
in
his
mouth.
But
I
think
one
of
the
things
martin
was
saying
is:
how
far
do
you
go
right?
H
H
We
would
only
come
up
with
the
things
that
we
thought
had
an
impact
on
protocol
design,
so
the
the
threats
and
attacks
that
we
we
want
to
highlight
that
give
guidance
to
protocol
designers
to
ensure
that
those
protocol
designers
are
taking
those
threats
and
risks
and
potentially
mitigations
into
consideration
and
that's
kind
of
how.
In
answer
to
martin's
question,
that's
kind
of
how
I
draw
the
line,
I'm
hand
waving
a
bit
right
because
I'm
not
telling
you
how
to
do
that.
I
Yeah,
I
think,
mark
I
think,
you've
you've
you've
basically
hit
it
there.
So
I
was
thinking
originally
that
this
would
be
exactly
that
sort
of
that
sort
of
advice
to
protocol
designer.
I
So
you
how
to
design
your
your
protocol
for
better
security,
but
when
you,
when
you
look
at
the
the
text,
that's
in
bcp,
38
and
3552
whatever
that
one
is
that
there's
a
lot
there
that
already
really
sort
of
covers
that
aspect
of
it,
but
the
the
thing
that
most
sort
of
interests
me
in
terms
of
my
own
blind
spot,
I
guess,
is
the
the
idea
that
we
potentially
design
new
protocols
in
response
to
particular
privacy
threats
or
security
threats
that
are
a
result
of
the
changing
landscape,
and
so
the
work
is
going
on
to
deal
with
incident
response,
for
instance,
is
an
example
of
that,
and-
and
so
I
think,
there's
there's
different
reactions
that
we
might
have,
and
some
of
those
are.
I
K
Tommy
here
I
think,
that's
right
and
to
just
pick
up
on
a
point
from
martin
and
christian,
I
think
listing
and
categorizing.
The
threats
is
important
and
I
do
accept
that
we
don't
really
know
as
protocol
designers,
what
impact
that
should
have
on
the
protocol
design,
and
it
may
be
that
it
actually
can
have
very
little
impact,
because
there's
not
much
that
the
protocol
design
can
do
about
it.
K
But
I
think
it
would
be
unwise
to
be
ignorant
of
those
threats,
because
it
may
be
that
the
protocol
design
can
actually
have
quite
a
lot
of
impact
on
how
effective
those
threats
can
be.
I
mean
just
taking
christians
example
of
establishing
a
bridgehead
and
then
an
attack
propagating
through
a
network,
that's
going
from
one
device
which
has
become
malicious,
two
other
ones
moving
through.
That
is
really
one
which
is
very
hard
to
stop
with
conventional
protocol
design.
K
So
yeah
I'd
support
sort
of
continuing
to
look
at
that
question
of
how
you
use
knowledge
of
these
threats
to
have
impacts
on
your
protocol
design
choices.
I
So
I
have
one
more
question
and
that
is:
is
it
possible
to
take
christian's
two-step
design
and
do
a
dry
run
or
a
trial
run
of
it
on
on
something?
That's
that's
relatively
well
understood.
I
So,
to
take
the
example
that
tommy
used
the
lateral
propagation
within
within
an
and
network,
you've
got
the
beach
head
and
you
you're
trying
to
propagate
across
into
into
other
nodes
within
that
network.
What
what?
What
is
it
that
we
might
do
in
that
context
and
do
a
deep
dive
on
something?
So
we
can
prove
that
that
that
is
tractable
or
we
can
prove
that
that
one
is
not
one
that
we're
interested
in.
I
suspect
that
we
are
interested
in
it,
but
that's
my
suggestion.
I
F
Yes,
I
think
that's
correct.
I
am
not
going
to
do
that
analysis
now,
but
stating
that
we
shall
have
a
document
going
in
depth
on
this
particular
attack.
I
mean
the
witch
head
problem,
it's
something
we
should
definitely
have
to
have
before
the
next.
C
Meeting
I
I
just
wanted
to
to
say
yeah
I'd,
be
happy
to
help
work
on
that
draft.
My
my
draft
on
the
protocol
design.
What
I
wanted
to
do
in
the
next
sort
of
iteration
of
it
is
to
actually
do
exactly
what
you
mentioned
and
what
you
were
talking
about,
martin
in,
and
that
is
to
really
look
in
more
detail
at
the
the
issues
and
that
what
is
impacted
right.
What
how
the
protocol
design
would
be
impacted
by
certain
security
and
privacy
threats
as
well.
C
So
it's
something
that
I've
had
on
my
mind
that
I
haven't
done
in
the
current
iteration,
but
it.
But
it's
definitely
something
that
I
want
to
do
so.
Yeah
happy
to
work
on.
C
H
A
possible
direction
forward
here
and
when
we
get
in
this
position,
sometimes
with
traditional
working
groups,
one
of
the
things
we
do
is
we
go
off
and
form
an
editing
team
to
go
away
and
and
try
to
meet
the
needs
of
multiple
authors
right,
and
I
think
we
could
take
martin's
suggestion
and
if
we
could
come
up
with
a
crisp,
a
crisp
description
of
what
we
were
asking
from
the
authors.
H
I'm
sure
that
we
would
have
volunteers
from
the
people
who
have
been
authoring
drafts
for
model
t
to
to
attempt
what
martin
martin
suggested,
that
sort
of
deep
dive
on
a
specific
tractable
problem
and
then
come
back
to
model
t
with
some
documentation
of
that.
That
comes
from
perhaps
a
a
consensus
or
unified
set
of
the
people
who
have
been
contributing
documents
to
model
t
over
time,
so
sort
of
like
forming
an
editing
team.
But
in
this
case,
to
meet
the
sort
of
specific
the
specific
goal
that
martin
talked
about.
H
I
do
think
we
need
to
somehow
capture
fairly
crisply.
What
how
we're
going
to
describe
that
for
the
editing
team.
But
that
would
just
be
a
straw
proposal
for
moving
forward.
A
Yes,
I
like
that
idea.
Actually
I'm
not
now
wondering
a
little
bit
if,
if
we
should
attempt
to
do
chris
definition
right
now
for
the
benefit
of
my
notes
or
for
the
recording
christian
or
martin
or
you
mark,
would
you
like
to
attempt
to
give
us
two
or
three
sentences
to
define
what
we
would
like
to
do.
I
I
I
don't
think
we
need
to
just
specify
which
item
that
we
dive
into.
I
think
that
maybe
the
bridgehead
problem
is
the
one
that
we
we
get,
but
I
would
like
to
see
a
a
very
concise
description
of
a
a
specific
type
of
problem
and
then
examine
the
sorts
of
responses
that
that
might
be
taken
to
that
problem
and
then
take
a
look
at
each
one
of
those
responses
and
say
what
effect
that
might
have
on
someone
who
might
choose
to
design
a
protocol
or
someone
who
is
designing
a
protocol.
H
I
I
think
I
would
say,
plus
one
martin's
suggestion
there.
I
I
think
we
could
leave
it
to
the
editing
team
to
pick
what
what
it
was
the
particular
threat
or
the
particular
risk
that
they
were
going
to
do.
The
deep
dive
on,
I
think,
what's
more
important
for
model
t
is
being
clear
about
sort
of
the
the
the
outline
of
what
we
expect
as
as
output,
and
I
agree
with
yari
that
it
could
just
be
two
or
three
sentences.
H
It
doesn't
have
to
be
more
than
that,
but
it
needs
to
be
crisp
enough
so
that
if
we
have
three
or
four
or
five
people
volunteering
to
try
to
produce
some
work
in
that
space,
that
they
have
a
common
view
of
what
it
is
they're
going
to
end
up.
F
Producing
the
reason
I
like
the
proposal
is
because
it's
concrete,
we
can
start
with
one
specific
problem
and
then
get
a
brainstorm
about
what
could
be
done.
Take
the
bridgehead
problem.
There
are
plenty
of
things
we
could
brainstorm
on.
I
mean
like
what
happens
with
protocols
that
assume
that,
if
you're
on
the
same
local
network,
your
authorized
things
like
that
and
just
go
in
one
problem,
not
try
to
overspecify
it
in
advance
and
see
what
it
gets.
It
seems
better
than
basically
you're
going
to
write
minutes
in
a
minute.
F
J
This
is
for
us,
I
don't
have
a
huge
amount
of
time
to
put
into
this,
but
I
would
be
willing
to
work
with
you
know
somebody
else
to
help
capture.
A
A
A
I
guess
not,
so
why
don't
we
move
to
the
next
item
on
that
center,
which
was
when
should
we
meet
again,
and
I
think
it
would
be
useful
for
a
small
team
to
indeed
get
some
results
before
the
next
call.
But
since
this
concrete
problem
and
not
a
huge
problem-
or
I
mean
it
can
be
hopefully
described
in
a
in
a
finite
amount
of
words,
then
maybe
that
could
actually
happen
reasonably
fast.
So
should
we
ask
steven
to
schedule
us
another
call
in
late
september?
A
Okay,
any
other
thoughts,
and
I
do
want
to
remind
people
that
we
do
need
to
I
mean
not
just
have
some
sort
of
people
do
stuff,
but
also
the
rest
of
us
have
to
review
their
their
results
and
also
review
the
other
of
the
documentation
that
we
have
and
in
in
itf
and
outside.
It
had
some
some
discussion
of
those
points.
So
this
work
that
remains
for
trying
to
understand
where
we
are
and
comment
on
documents
today,.
A
A
I
A
Thank
you,
I'm
not
going
to
think
about
what
what
your
time
zone
is
and
how
late
it
is
for
you,
probably
even
worse.
Okay,
I
think
we're
done.
Thank
you
all.
I
will
send
some
notes
as
we
process
the
call
recording
and
my
brief
notes
and
we'll
get
a
team
set
up
to
pursue
this
particular
practical
problem
and
then
we'll
see
where
we
go
from
there.