►
From YouTube: IETF115-ELEGY-20221109-1500
Description
ELEGY meeting session at IETF115
2022/11/09 1500
https://datatracker.ietf.org/meeting/115/proceedings/
A
Okay,
the
the
hands
are
now
in
the
appropriate
place
and
it
has
reached
the
top
of
the
hour.
Welcome
to
the
LG
working
group
at
ietf,
115
I
had
actually
checked
the
ietf
number
because
who
remembers
those
we
are
here
to
talk
about
the
nomcom
eligibility
criteria.
If
you
are
expecting
to
talk
about
routing
protocols
go
away,
foreign
Peter
is
actually
leaving
the
room.
Wow,
no
he's
just
closing
the
door
all
right.
So
let's
get
started
with
the
LG
working
group,
and
what's
that
and
yes,
we
now
have.
A
We
have
a
couple
of
remote
participants
now
who
came
on
at
just
the
right
time,
so
welcome
here's
where
we
go
here.
We
go
note
the
note
well
well,
this
was
supposed
to
be
larger
in
the
slides,
sorry,
John
I
know
you
can't
see
this,
neither
can
I
actually,
but
for
those
who
can
see
it,
if
you
have
any
IPR
associated
with
nam-com
eligibility
criteria,
please
declare
it
otherwise
we'll
move
on
from
that.
You
never
know.
A
Pete
says
these
slides
are
not
as
big
a
font
as
I
thought
they
were
going
to
be.
They
looked
much
bigger
on
my
computer
before
I
uploaded
them,
sorry,
folks,
so
the
regular
meeting
Logistics
please
do
wear
your
masks
if
you're
in
the
room
and
you're
not
me
up
here
at
the
front
Murray
that
includes
you
and
if
you,
if
you
need
to
speak
either
in
the
room
or
remotely,
please
use
the
app
to
get
yourself
in
the
queue
exactly
all
right.
A
Next,
we
have
the
agenda.
We
we
just
did
number
one
we're
about
to
do
number
two,
which
is
discussion
of
the
draft.
So
we
have
a
couple
of
things
there
that
are
specifically
on
the
list
of
what
to
discuss.
Robert
had
sent
a
message
about
data
that's
available
and
how
we
want
to
handle
that,
and
there
was
some
discussion
on
the
mailing
list
stemming
from
that.
So
we'll
continue
that
discussion,
that's
first,
because
I
think
that
will
be
a
little
quicker.
A
The
only
other
substantive
comments
we've
had
have
come
from
Donald
and
we've
had
a
good
discussion
of
that,
a
bit
on
the
list
so
far
and
we'll
continue
that
discussion
here
and
then
we'll
open
it
up
to
any
other
issues
that
people
want
to
raise
in
the
room.
I
hope
there
are
none
of
those
I
hope.
We've
had
the
discussion
we
needed
to
have
at
least
we've
started
the
discussions
we
need
to
have
on
the
list
and
then
we'll
bring
up
the
question
after
this
discussion.
A
A
D
A
C
C
So
it's
all
right
to
not
put
up
the
the
message.
Basically,
what
I
sent
to
the
list
was
a
a
recalculation
of
the
same
graphs
that
went
into
the
document
that
we're
missing,
showing
what
the
other
paths
had
we
had
them
in
place
would
have
qualified
in
addition
to
the
pass
that
we
had
before
and
it
we
started
the
conversation
noting
that,
well,
it
didn't
really
qualify
very
many
people,
many
other
people,
but
the
response
generally
was
so
what
we're
trying
to
make
it
provide
some
Runway,
some
inclusiveness
it.
C
C
E
C
A
So
I
will
I
will
seed
this.
This
is
Barry.
I
will
see
this
by
repeating
up
some
points
that
I
made
in
in
message
the
list
that
I
think
that
few
in
in
the
group
of
active
participants
who
are
not
participating
in
person
or
not
not
coming
to
the
meetings
whatever.
That
means
why
they're
not
volunteering
might
be
because
it's
a
new
thing
for
them
to
be
allowed
to
and
they're
not
used
to
it
yet
because
we
still
stress
that
in-person
participation
is
best
for
the
actual
seat
at
nam-com.
A
So
maybe
we
should
stop
saying
that
and
accept
that
the
nomcom
can
work
remotely
and
might
well
change
over
time
if
we
provide
the
options.
So
those
are
my
reasons
for
as
a
participant
for
preferring
to
leave
those
paths
on
the
list
so
seeding
the
discussion
there.
Does
anyone
want
to
say
something
about
that.
A
C
So
the
last
thing
that
I
had
in
the
message
that
I
sent
to
the
list
was
an
observation
that
it
does
add
complexity.
It
does
add
a
little
bit
of
risk
that
something
weird
goes
wrong
in
an
om-com,
because
the
calculations
could
be
messed
up
just
because
they're
not
super
easy,
they're
fuzzy
right
now,
because
the
data
tracker
has
a
mechanic
where
it
notes
when
people
become
chairs
positively.
C
So
we'll
make
changes
to
the
data
tracker
in
the
long
run,
so
that
these
intervals
are
a
little
bit
tighter,
but
I
think
the
spirit
of
the
document
for
the
implementation
we've
got.
We
would
potentially
overqualify
some
people
that
were
very,
very
qualify,
some
people
that
wouldn't
have
otherwise
just
because
we
didn't
get
a
cut
off
at
the
right
place
so
but
it'll
be
a
very
small
number.
So.
B
To
clarify
that
you're
saying
that
there
would
be,
for
instance,
people
that
cease
to
become
working
group
chairs
at
some
critical
point,
and
they
would
continue
to
be
qualified
for
nom-com
because
of
the
we
don't
know
when
they
exactly
they
stopped
right.
And
so
these
are
people
that
would
would
be
known
to
the
community.
They
wouldn't
be
complete
random
strangers
because
they
would
have
already
had
been
working
group
chairs
for
some
period
of
time
already.
B
A
F
C
G
So,
if
somebody
sort
of
you
know
if,
because
Robert
human,
they
tell
me
he'll
make
a
bug
and
if
somebody
says
wait
a
minute,
this
person
doesn't
really
qualify
because
when
we
list
the
candidates
we
will
say
under
what
the
committee
candidates
will
say
under
what
path
they
qualify,
we
can
go.
Oh
no
Richard
hasn't
been
You.
Know
Rich
hasn't
been
a
a
working
group
chair
for
three
years,
so
that
doesn't
qualify
so
I
think
that
the
risk
is
not
is
negligible.
G
I
A
J
So
I
don't
have
a
strong
feeling
about
any
of
this,
but
Donald
raised
a
number
of
objections
to
the
criteria
and
I
propose
that
we
kind
of
engage
from
the
start
with
the
first
order.
Objection
then
filter
down
to
the
relatively
small
stuff,
which
my
view
are
like
the
specific
objections
to
path
two
and
path,
three,
that
does
not
necessarily
relate
to
the
slides
that
I
put
up
but
and
of
course,
now
there's
a
GitHub
issue.
The
slides.
J
That
one
okay,
so
there
are
so
right.
So
there's
there's
a
basic
like
qualitative
argument
here
that
the
attendance
criteria
to
lacks
I
think
as
as
a
group
in
general,
the
vibe
has
been:
let's
try
to
get
as
big
a
fool
as
possible
and
so
I
think
the
two.
If
you
don't
mind
me
trying
to
characterize
the
argument,
there
are
two
two
separate
possible
objections
to
the
relatively
used
criteria.
J
One
is
that
they
open
up
an
attack
in
terms
of
nom-com
capture,
which
is
what's
kind
of
listed
in
the
security
considerations,
there's
kind
of
a
modeling
of
that
threat,
and
then
the
second
thing
is
that
just
like
the
quality
of
of
nam-com
member
will
be
reduced
because
you
have
people
who
you
know
come
in
for
one
session,
one
working
group:
they
don't
care
about
anything
else,
etc,
etc.
So
why
don't
I
give
Donald
a
chance
to
say
something
in
his
defense
and
I?
Have
some
comments.
A
B
K
The
problem
I
see
is
primarily
with
the
attendance
Criterion
here
and
the
the
reason
I
I
say
this
is
that
basically
publicly
facing
sort
of
virtual
systems,
like
is
effectively
suggested
here,
just
end
up
getting
attacked
sooner
or
later
it
seems
to
be
characteristic
of
the
of
the
internet.
You
know,
there's
always
this,
this
great
siren
song
that
everything
should
be.
You
know
more
open,
it
should
be
cheaper.
It
should
be
simpler,
more
open,
simpler,
cheaper
et
cetera,
et
cetera.
K
Everything
should
be
online,
it's
all
virtual,
it's
great,
and
it
works
quite
well
for
a
while
and
everybody's
happy
and
there's
a
sweetness
and
light.
But
you
know,
after
a
while,
somehow
something
goes
viral
somewhere
that
everybody
should
go
and
say
it's
a
vote
for
the
name
of
something
everybody
should
go
and
vote
for
some
absurd
or
obscene
name
or
something
like
that
or
say
something
like
this,
and
really
this
enables
fairly
easily.
Although
it
does
take
time,
it
takes
three
meetings.
K
Is
there
any
particular
reason
to
believe
where
you
couldn't
suddenly,
after
the
years
of
it,
working
fine
get
a
thousand
two
thousand
people
suddenly
in
the
pool,
because
the
effort
is
zero,
the
cost
is
zero.
I
mean
you
know.
Is
that
really
a
good
thing?
I'm
not
entirely
sure
what
the
defense
is
I
propose,
making
it
someone
more
cumbersome,
which
it
seems
like
it
would
help
I,
don't
even
know
if
it
helps
enough
and
that's
that's
sort
of
from
the
the
viral
meme
point
of
view.
K
There's
the
other
point
of
view
and
there's
actually
a
spectrum
between
these
that's
kind
of
a
a
a
a
typical
I,
don't
know
Enthusiast
or
something
like
that
point
of
view
or
troll
or
whatever,
and
the
other
point
of
view
is.
Is
there
a
technical
Advantage
if
some,
if
some
company
perceives
it
being
a
technical
advantage
to
being
able
to
influence
who
gets
selected
by
the
nam-com,
and
they
would
want
to
have
people
in
the
nom-com
and
the
past?
That
was
the
benefit
of
sending
people
to
the
IDF
which
will
help
the
itf2.
K
Of
course
the
people
are
available,
but
it
may
also
be
a
commercial
motivation
for
telling
some
of
their
Engineers
that
really
don't
have
much
to
do
anything
to
do
with
the
ITF.
You
know
you
should
really
just
log
into
a
session
some
of
these
things.
Maybe
you'll
learn
something,
and
that
way
you
know
we
will
have
you,
you
will
become
eligible
for
the
nom-com.
You
know
so
I'm
not
entirely
sure
what
the
answer
is,
but
I
I
just
am
quite
nervous
about
the
primarily
the
attendance
provision
in
the
current
draft.
K
I
mean
I
happen
to
all.
Bi.
Also
think
that
some
of
the
other
things
are
not
worth
it
because
they
only
qualify
a
few
people
and
why
make
it
more
complicated
and
so
on
and
so
forth,
but
but
really
they're,
not
so
bad,
and
it's
really
this
attendance
provision
virtual
attendance,
that's
the
I
see,
is
the
primary
problem.
L
Like
St
John's
I've
been
dipping
in
and
out
from
this,
so
I'm
not
spending
a
lot
of
time
on
the
details
here,
I
did
want
to
I,
saw
Don's
comment
in
the
tracker
and
and
re
and
responded
to
it.
The
main
thing
I'm
concerned
about
is
some
with
some
of
the
criteria
is
that
they're
that
there
are?
They
are
not
objectively
resolvable
during
a
challenge.
L
Okay
and
those
things
are
going
to
be
problematic
because
we've
already
seen
how
long
it
takes
once
the
challenge
period
starts
to
get
the
whole
thing
resolved
so
get
to
an
objective
standard
that
everybody
can
agree
on,
based
on
a
a
data
source
for,
however,
for
whatever
you
do.
Second,
there
has
to
be
a
cost
to
the
volunteer
in
some
way,
shape
or
form,
besides
just
clicking
on
just
clicking
on
something
or
else,
because
that's
how
you
deal
with
that's
how
you
deal
with
the
issues
of
jamming
into
jamming
into
the
system.
L
You
know
the
the
meme
that
says
oh
register
here
and
you
become,
and
you
can
select
the
next
next
ietf
people
and
it's
we
get
thousands
of
volunteers
from
we've
never
seen
before.
They
don't
actually
have
to
pay
any
money
for
remote
access,
because
we
don't
we've
got
that
as
on
our
way,
unlimited
waivers
by
the
time.
All
of
a
sudden
done.
You
need
to
think
about
this.
L
You
need
to
think
about
what
the
denial
of
service
attacks
are
on
this
thing
last
piece
of
the
puzzle
right
now,
if
you
are
a
large
group
of
people,
a
company,
an
organization,
whatever
the
way
the
math
works
on
how
we
select
the
nom-com
means
that
you
are
guaranteed
almost
guaranteed
to
have
two
members
in
the
pool
at
a
certain
level.
If
you
have
a
certain
proportion
of
the
pool,
you
have
a
certain.
You
have
you're
almost
guaranteed
to
have
your
max
high
probability,
yes
yeah,
but
it's
a
very
high
probability.
L
J
So
so
the
I
should
point
out
in
security
considerations,
actually
attempt
to
model
this
I
I
got
worried
about
just
letting
in
all
the
remote
attendees
too,
because
the
cost
does
go
to
zero
and
I.
Don't
want
to
say
the
security
consideration.
Just
say:
don't
worry
it's
not
a
problem,
but
what
I
can
say
if
you
work
through
the
probabilities
here
there
have
to
be.
There
have
to
be
a
lot
of
these
attendees
to
kind
of
really
have
the
probability
of
mom-com
capture
so
but.
J
All
right
well,
so
the
point.
The
point
is
that
again,
this
does
not
mean
there's
no
threat
here,
but
there's
an
early
warning
system
that,
if
all
of
a
sudden
for
three
itfs,
we
have
huge
numbers
of
attendees
who
don't
show
up
or
have
a
strange
affiliations
that
seem
very
suspect
Etc.
That
leadership
could
take
some
action
and,
unfortunately,
those
actions
are
probably
out
of
scope
of
this
document
and
what
this
group's
chartered
to
do,
but
there
like
a
complete
passive
leadership,
does
nothing
about
it
then.
J
A
So
I'll
do
a
chair
interrupt
there
I
think
the
defining
normatively.
What
we
do
about
it
is
out
of
scope,
but
discussing
the
issue
is
absolutely
not
out
of
scope
here
and
if
we
decide
in
that
discussion
that
we
can't
move
forward
with
this
because
of
that
threat,
then
we
may
need
to
recharter
to
to
be
more
specific
about.
J
L
You're
doing
changes
to
the
eligibility
commit
criteria
and
you
are,
you
ought
to
be
talking
about
the
selection
criteria
as
well.
At
the
same
point,
and
not
just
pretend
that
it
doesn't
matter.
For
example,
let's
say
your
proportion
of
three
three
meeting
wonders
versus
everybody
else:
who's
been,
you
know
who
say
we
have
800
here
and
we've
got
10
000
out
there,
okay
yeah,
you
might
notice
for
three
meetings,
but
again
it
takes
us
a
year
and
a
half
to
two
years
to
change
policies.
L
L
B
Speaking,
not
as
the
chairs,
but
my
belief
please
correct
me
if
I'm
wrong
at
this
point
was
that
remote
attendees
had
to
register
and
then
ask
for
a
fee
waiver
right.
So
it's
not
just
a
data
tracker
account
it's
a
it's
a
it's
a
three-step
process
and
we
would
see
in
those
fee
waivers
are
not
automatic.
B
B
C
Click
this
URL
exactly
yet
right.
If
we
get
to
the
point
and
good
mics
in
the
room.
If
we
get
to
the
point
where
we've
got
a
flood
of
people
that
are
asking
for
the
fee
waivers
we're
going
to
have
somebody
reasonable,
reasonably
stand
up
and
say:
hey,
wait!
This
isn't
right
and
stop
it
and
it's
not
going
to
go
past
one
meeting
right.
So
the
chances
of
this
getting
into
a
three
meeting
cycle
of
oh
here's,
a
flood.
We
didn't
notice
and,
oh
my
gosh.
C
Here's
this
massive
pool
of
people
that
we
weren't
expecting
to
have
to
deal
with
the
spirit
of
the
fee
waivers
is
going
to
you
know
as
long
as
we
get
a
reasonable
number
of
people
asking,
and
it's
not
worth
our
time
to
go
validate
that
somebody
really
can't
help
us
along
then
we'll
do
that.
But
when
it
gets
to
the
point
where
it's
taxing
our
resources,
then
limits
will
get
put
in
place.
A
M
Pete
Resnick,
so
I
I
think
to
put
a
very
fine
point
on
it
for
Mike's
comment.
What
Robert
just
said
indica
and
you
know
from
the
front
of
the
room.
What
it
says
to
me
is
this,
doesn't
go
on
for
longer
than
one
meeting
at
one
meeting
is
when
we
notice
that
there's
a
problem
and
then
we've
got
the
better
part
of
a
year
to
fix
it
if
it
needs
fixing.
M
The
other
thing
I
got
up
to
say
was
I.
Have
some
sympathy
for
what
Donald's
talking
about
I
think
we
could
play
with
the
criteria
a
bit
that
maybe
for
remote,
it's
not
three
out
of
five.
Maybe
it
requires
some
number
of
working
group
participations
or
time
of
working
with
participation.
I,
don't
know,
what's
automatable
at
that
level,
I
wouldn't
object
to
those,
but
I
think
the
point
that
there
is
a
circuit
breaker
of
just
this
is
going
to
be
exceedingly
obvious.
A
Don't
sit
down
yet
because
I
I
want
to
probe
one
thing
that
you
said
then
some
number
of
working
group
participations
suppose
someone
came
in
because
of
quick,
let's
say
just
picking
something
and
said:
I'm
going
to
spend
three
years
working
on
the
quick
documents.
But
that's
the
only
working
group
I'm
interested
in.
Would
you
be
in
favor
of
saying
that
person
is
not
nom-com
eligible
because
they
only
participated
in
one
working
group.
A
C
C
M
And
I
I
agree.
The
reason
it's
more
palatable
to
me
is
not
because
I
think
that
amount
of
time
is
a
good
criteria,
but
Donald's
concern
seems
to
be
centered
around
the
game,
ability
of
it.
The
game
ability
of
getting
your
ass
in
a
seat
in
the
room,
that
is
an
investment
of
there's
a
cost.
The
clicking
into
a
single
online
meeting
and
clicking
out
immediately
is
much
closer
to
zero
cost
having
to
sit
there
for
a
while.
Is
it
a
significantly
larger
cost?
M
C
J
D
Alex,
so
a
lot
of
secret
speaking
as
an
individual,
although
it's
a
LG
ID
there,
because
tools,
so
the
non-com
process
is
a
staggered
set
of
defenses
right,
so
don't
forget,
there's
a
confirmation
Step
at
the
very
end.
So
even
if
the
number
gets
captured
and
the
chair,
you
know
got
parachuted
in
right,
somebody
can
still
say
no.
D
It
causes
a
huge
headache
and
it's
causing
a
huge
delay,
but
given
that
it's
probably
going
to
be
a
rare
event
right
so
so
we
should
probably
think
about
how
do
we
add
to
those
dagger
defenses,
rather
than
trying
to
defend
against
everything
at
the
very
first
step
right
right.
So
that's
sort
of
I
think
that
the
principle
I
want
to
sort
of
strategic.
D
J
J
Think
you're,
like
a
quick
participant
example,
was
an
interesting
one,
because
that
kind
of
segues
to
the
next
Point
there's
the
actual
like
threat
model
in
the
security
considerations,
which
is
what,
if
someone
is
trying
to
like
spam,
the
system
to
get
to,
like
you
know,
take
over
the
nom-com,
which
is
not
what
you
were
talking
about.
The
that's
a
separate
question,
which
I
think
is
also
interesting:
that
Donald
implicitly
raised
which
is
like
what
is
a
good
nom-com
member.
Is
it
someone
that
is
focused
on
one
working
group?
J
So
this
is
not
gaming
system.
This
is
just
like.
What's
a
good
nom-com
member,
so
we
can
I,
don't
know
if
we
want
to
have
I
don't
know
if
people
are
like
from
largest
comment
like
comfortable
about
this
about
this
dos
problem
and
we
can
move
to
the
candidate
quality
problem
or
not,
I
mean
well
I.
Guess.
A
N
I
learned
about
the
problems
that
the
Ops
you
know,
Ops
area
ahead
and
as
an
unco
member
I,
wanted
to
address
those
problems,
making
sure
that
we
can
solve
those
problems
with
the
next
leadership.
You
know
that
was
coming
in
in
that
area,
so
this
is
an
option
that
if
a
person
worked
just
in
one
working
group
that
might
be
a
limiting
factor,
but
if
you're
working
across
an
area,
then
you
as
a
non-com
member
can
help
solve
some
problems.
N
Or
if
you
see
some
good
things,
you
can
help
make
a
decision
if
some
of
those
good
practices
can
be
applied
to
the
other
to
the
other
areas.
At
that
point,
the
the
security
area
had
a
really
good
process
of
moving
forward
a
draft
into
the
rfcs.
You
know
with
the
different
testing
of
a
software,
you
know
for
the
interoperability,
so
there
was
some
good
examples
that
could
be
shared
from
one
area
and
there
could
be
some
bad
examples
as
well
given
as
a
warning.
N
G
Yeah
hi
Rich
sauce.
So,
as
Laura
said,
and
one
of
you
guys
said
there
are
a
number
of
wait.
There
are
a
number
of
challenge
mechanisms
and
corrective
actions
that
can
be
taken
during
the
process,
and
so
pardon
me
for
if
this
becomes
a
concern,
when
we
first
publish
the
list
of
volunteers,
there
isn't
an
explicit
challenge
process
in
that.
G
But
if
we
decided
to
add
one
you
could
say:
hey
I
challenge
everybody
from
Akamai,
because
I
know
they're
all
just
sock
puppets
and
they
didn't
do
anything
or
you
could
look
at
the
final
list
and
say
well
wait
a
minute!
I
had
this
suspicion
about
Aquaman
now
4
out
of
ten
are
Akamai
employees,
so
you
can
challenge
it
that
way
too.
I
think
it's
a
concern.
I,
don't
think
it's
a
realistic
one
in
the
security
world.
It's
all
about
trade-offs.
G
H
Attorney,
thank
you.
Everybody
for
a
very
interesting
discussion.
I
think,
however,
we're
missing
a
little
bit
of
the
threat
model
that
I
saw
in
Donald's
email
and
that's
functionally
related
to
the
the
threat
which
occurred
in
a
different
sdo,
where
a
particular
organization
was
found
at
one
point
to
have
sent
a
whole
bunch
of
people
to
to
Garner
votes,
and
those
votes
were
then
important
to
the
success
of
that
organization.
H
H
Service
remotely
you,
you
could
end
up
in
a
situation
where
somebody
signs
up
20
times
for
three
three
times
so
that
they
are
in
in
effect,
20
entries
in
the
pool
not
as
good
as
the
threat
model
that
Mike
St
John's
point
out
but
entirely
possible,
and
you
would
not
have
any
way
at
the
end
of
it.
From
saying
whether
you
know
Ezekiel,
Elijah,
Ebenezer
or
Edward
was
the
real
person,
because
the
identifier
that
they
gave
was
freshly
minted.
H
I
wasn't
suggesting
that
I
was
suggesting
that
they
could
realistically
take
their
chance
of
being
selected
from
being
one
in
N
to
being
20
and
n
and
I.
Think
that's
not
something
that
the
current
set
of
defenses
in
depth
would
deal
with
very
well.
If
we
are
in
fact
going
to
encourage
people
to
participate
in
the
nomcom
remotely,
so
I
think
I
am
I'm
much
less
sanguine
about
the
current
set
of
defenses
in
depth.
H
Then
we
could
say:
okay,
we're
going
to
ask
nomcon
volunteers
to
have
a
video
chat
with
each
one
of
them
and
confirm
that
they
would
be
able
to
serve
on
an
omcon
because
they
have
their
their
appropriate
tooling
to
participate.
I
know
in
an
ITF
or
in
a
non-com
and
from
that
be
able
to
say.
Okay,
all
all
40
of
these
were
one
person.
H
B
But
I
think
that
the
other
interesting
valuable
part
of
what
you
just
said
is
that
if,
if
you
have
someone
who's
perpetually,
has
you
know
two-port
bandwidth
to
be
able
to
do
that?
Call
that
actually
would
affect
the
nomcom's
operation,
because
they
have
this
person
that
basically
can
never
properly
participate
on
a
remote
completely
remote
thing.
So
actually
doing
that
like
having
the
non-com
chair
do
that
vetting
right
is
actually
not
silly
because
otherwise,
you
know
like
like
they
just
don't
have
enough
bandwidth
to
be
able
to
participate.
H
Now
but
I
do
want
to
highlight
that
some
of
what
we're
talking
about
here
is
actually
a
new
enough
attack.
Given
the
change
in
participation
mechanism
that
the
defenses
in
depth
we
already
have
are
either
completely
untested
in.
In
the
case
of
the
utilization
of
the
veto
by
a
confirming
body
over
an
entire
slate.
J
E
Yeah
I'm,
actually
not
that
much
concerned
about
this.
This
trying
to
cheat
the
process
case
I,
don't
think
that
our
original
process
was
necessarily
signed
to
make
it
like
extremely
expensive
to
be
on
an
umcom
I.
Think
that
the
the
having
this
like
three
meeting
limit
was
mainly
designed
to
have
people
who
actually
know
something
about
our
community
I
think
it
does
help
you
on
the
nomcom.
If
you
know
the
candidates,
if
you
know
some
people
in
the
community
that
give
your
feedback
and
so
on,
that
does
help
you
with
an
job.
E
So
that
was
like
definitely
a
sign
criteria,
but
I
think
also
that
we
should
rethink
that
design
criteria
a
little
bit
because
I
think
it's
also
valuable
to
have
people
on
the
number,
maybe
with
like
some
fresh
eyes,
which
no
less
of
the
com
of
the
of
the
candidates
and
so
on,
and
can
bring
like
a
different
angle
and
I.
Don't
think
there
is
a
big
risk
that,
like
we
and
like
I,
think
the
one
failure
mode
we
would
have
is
like.
E
We
end
up
with
a
num
com
where
they're
all
kind
of
newbies
and
they
don't
know
nothing
but
I
think
this
is
very
unlikely
because
you
were
not
necessarily
volunteer
for
the
number.
If
you
have
like
no
idea,
I
think
that
will
not
happen
and
if
it
happened,
then
we
need
to
to
act
on
it.
So
I
think,
what's
more
important
is
really
kind
of
measure
in
the
future
a
little
bit
closer
about
these
expectations,
if
they
happen
or
not
to
be
able
to
react
quickly.
A
A
K
Yes,
Donald
he's
like
gonna,
say
something
real,
quick
I
was
sure
of
an
om-com
and
we
had
a
deadwood
member.
They
don't
I'm,
not
sure
if
they
even
voted
but
they're.
The
the
I
think
the
intent
was
the
mechanism
for
that
is
that,
but
the
nom-com
can
expel
a
member.
The
voting
member
is
going
to
expel
a
voting
member
by
two-thirds
vote.
No
true.
G
L
L
One
of
the
things
that
Ted
said
triggered
something
and
again
the
more
I
hear
the
discussions
here,
the
more
I'm
thinking
that
the
charter
is
too
narrow
for
what
we're
doing
that's
not
a
topic
for
today
probably
need
to
go
back
onto
the
mailing
list
and
have
that
chat.
But
the
particular
thing
he
said
was
about
how
bandwidth
and
such
affects
the
ability,
I'm
actually
thinking
the
number
of
remotes
might
actually
affect
the
ability.
I
remember
doing
various
I.
L
You
know
we
did
various
IAB
meetings
where
Ecker
was
you
know
back
in
the
other
country
and
trying
to
do
trying
to
actually
get
the
business
of
the
of
the
group.
Don
was
difficult
this,
where
it's
very
intensive
discussions
and
trying
to
get
everybody
together
and
and
deal
with
the
technology
issues
and
everything
else
is
it
possible.
J
D
Yes
speaking
as
an
individual
but
sort
of
keeping
the
the
motivation
and
timeline
for
this,
and
so
so
the
the
urgency
comes
from
having
had
fully
online
meetings
in
the
past
and
having
the
88
8989
experiment
be
the
thing
that
qualifies
all
of
us
that
were
remote
during
these
meetings
for
the
next
right.
So
that's
the
that's
the
immediate
fix,
because
I
think
the
next
non-com
will
have
a
small
pool.
If
we
don't
say
that
counted
any.
So
we
could
do
a
a
super.
D
But
yes,
especially,
sort
of
if
this
free
or
cheap
remote
participation
causes
this
sort
of
desire
to
change
things
further
down
the
nomcom
pipeline,
which
is
we're
currently
not
chartered
for
and
I'm
not
going
to
retire
you
for
it
at
this
point,
I
think
that
that's
a
different
kind
of
problem
that
then
we
set
out
to
solve
for
this
right
right.
We
might-
we
might
have
found
this
now
right
and
then
we
need
to
decide
what
to
do
about
it.
A
C
C
We'll
see
yeah
so
but
I
would
I
would
strongly
not
just
start
relaxing
saying,
okay!
Well,
let's
just
do
that.
Let's,
let's
go
ahead
and
have
the
the
hard
argument
and
I
really
think
that
it
is
within
our
grasp
to
do
something
that
is
going
to
be
longer,
lasting
and
and
I
want
to
fight
for
that.
I
agree
had
a
question
for.
L
G
B
A
A
G
G
G
J
What
I
haven't
heard
is
anyone
really
like
standing
up
for
the
all
remote
people?
I
mean
that
maybe
it
hasn't
been
on
the
table
out
of
Italy
but
like?
Let
me
just
at
the
risk
of
Designing
at
the
mic
like
when
we.
What
if
we
did
like
with
possible
exception
possible
grandfathering
for
the
like
the
the
post-pandemic
period
and
any
other
times,
we
don't
have
in-person
meetings.
What
if
we
did
like
one
in-person
meeting,
plus
some
remotes.
A
D
So
I'm
trying
to
swap
again
so
I,
would
ask
me
if
I
can
open
the
window
and
I
can't,
but
Paige
is
trying
to
the
CO2,
is
pretty
high
in
here
I.
Don't
remember
what
the
exact
Charter
check
says,
but
I
think
if
we
I
think
it
would
be
in
scope
for
us
to
talk
about
the
num-com
eligibility
criteria
and
also
about
a
remedy
Step
at
the
eligibility
phase
like
another
challenge
like
Rich
talked
about
or
something
right,
I
think.
A
D
A
A
H
Saturday,
and
just
in
case
this
wasn't
clear
because
I
was
talking
about
a
risk
that
doesn't
mean
I,
don't
think
the
benefits
outweigh
the
risks.
I
do
I
agree.
I
grew
up
in
a
place
where
people
used
suppression
to
keep
people
off
juries
to
keep
people
out
of
voting
things.
Anything
we
can
do
to
make
sure
that
we
keep
this
to
be.
An
open
process
is
important
because
it
means
that
the
participation
is
encouraged
for
the
broader
community
and
that's
a
great
thing,
so
I
I
think
the
benefits
outweigh
the
risks.
H
A
K
Sure,
sorry,
to
put
you
on
the
spot,
no
no
I
said
Donald
he's
like,
and
you
know,
I
think,
there's
problems
with
this.
If
the
current
draft,
if
the
current
graph
is
in
effect
for
lesser
time,
then
it
makes
the
problem
smaller,
but
but
I
I've.
Also,
someone
remembered
I
think
that
maybe
you
get
something
which
is
really
effective
for
longer.
K
If
you
I
mean
it
seems
like
it's
not
common,
that
we
have
to
have
virtual
meetings
because
of
a
global
pandemic,
I
mean
this
doesn't
seem
to
have
I
have
been
involved.
You
know
for
decades
with
the
ietf,
so
I
think
we
should
also
consider
at
least
for
a
while
on
the
mailing
list,
whatever
the
possibility
of
saying
that
in
this
attendance
window,
whatever
it
is
two
years,
it
means
that
if
there
are
physical
meetings,
you
have
to
have
gone
to
one
okay
and
the
other
ones
could
be
done
virtually.
A
K
I
agree
with
that,
but
I
think
that
attending
a
ietf
meeting
physically
is
a
qualitatively
different
thing
and
is
beneficial
to
somebody's
Service
as
a
nomcom.
If
they
do
it
and
the
question
is
how
much
Focus
do
you
put
on
some
poor
member
somewhere?
You
know
that
they're
you
got
to
be
ultra
fair
to
them.
Well,
I
didn't
mean
in
sense
of
money,
I
just
mean
disadvantaged
Stars.
K
C
Of
him
go
ahead,
Robert
direct
response
to
that.
There
are
large
companies
that
have
historically
sent
people
to
the
ITF
that
are
not
going
to
for
some
large
number
of
years,
starting
now,
they've
discovered
that
they
don't
have
to
pay
people
to
go
across
the
ocean
to
work
in
these
meetings
and
they're
not
going
to
send
them
anymore.
C
A
O
D
Individual
so
I
think
the
the
key
word
he
has
participant
right,
so
I'm
not
actually
very
worried
about
remote
participants
that
are
actually
like
participating
in
the
work
being
remote,
because
you
know
also
remember
right.
They
are
supposed
to
evaluate
the
feedback
they
get
from
the
broader
Community
they're
not
like
making
this
election
based
on
their
own
knowledge.
D
Only
right,
they're
really
supposed
to
look
at
what
comes
in
right,
so
and
and
that
I
think
they
can
do
fine,
even
if
they're,
mostly
only
remotely
participating,
I,
think
the
threat
that
Ronald
talked
about
is
like
Outsiders
stuffing
this
thing
right
and
and
without
an
intent
to
participate
in
good
faith.
But
that's
that's
not
a
participant
either
remote
or
in
person.
That
is
just
a
disruptive
element,
and
if
we
have
too
many
of
those,
we
have
problems.
D
So
maybe,
if
we
sort
of
try
and
focus
more
on
the
you
know,
how
do
we
exclude
that
risk?
And
rather
like
not
worry
so
much
about
whether
who's
remote
is
actually
participating
to
what
degree,
because
I
think
they
are,
and
they
I
agree
that
we
will
continue
to
have
people
that
are
going
to
be
remote
all
the
time
or
more
frequently
about.
J
Yeah
so
I
just
we're
talking
a
lot
about
costs
and
benefits
of
bigger
and
smaller
pools.
The
other
one
I
call
an
attention,
if
not
if
the
volunteer
pool
is
too
small,
then
the
assumptions
in
the
mathematical
analysis
and
the
security
concentration
breakdown.
If
they're,
like
20
volunteers,
then
like
it's
actually
quite
easy
to
catch
the
thumb
com
with
with
like
a
small
number
of
people,
so
that
yeah
I
mean
I,
don't
think
we're
in
any
case
going
to
drive
to
20
but,
like
smaller,
you
know,
mixed
probabilities,
work
out
quite
differently.
J
J
Oh
I'm,
sorry
well,
I'm,
sorry,
just
like
so
I
could
I've
got
a
bunch
of
notes
on
like
stuff
to
add
insecure
to
considerations,
which
is
great,
so
I
will
like
file
some
issues
for
that
and
try
to
come
up
with
some
text
regarding
like
this
like
regarding
this
particular
path.
One
do.
Is
there
like
a
measurable
and
desire
a
instrumentable
and
desirable
criteria?
N
N
A
A
Okay,
thanks
so
I'm
next,
in
the
queue
as
a
participant.
My
comment
is
that
I
personally
think
we're
getting
too
wrapped
around
the
idea
that
we
might
get
one
or
two
people
on
the
nomcom
who
don't
know
the
community
very
well.
As
Lars
put
pointed
out,
the
nomcom
uses
a
lot
of
years
ago.
The
nomcom
didn't
even
tell
people
who
was
standing
for
positions
and
solicited
input
selectively,
but
now
everybody
knows
who's
standing
for
positions.
Everybody
is
welcome
to
make
comments
and
the
nomcom
isn't
expected
to
know
everybody.
A
I
Yeah
I
I
I
just
want
to
try
to
draw
several
different
things
together:
I
I'm,
at
least
as
concerned
about
about
ways
to
game
and
attack
the
system
as
as
Donnie
is
and
I'd
be
concerned
about
it.
Since
long
before
we
started
having
having
all
remote
meetings
but
an
equally
or
more
concerned
that
whatever
we
do,
we
keep
the
option
and
and
encourage
a
non-com
which
is
really
representative
of
the
community.
I
It
includes
some
people
who
are
remote,
for
whatever
reason,
probably
doesn't
include
anybody
who's
dead,
but
other
than
that.
We
probably
want
to
do
that
as
possible
in
terms
of
representation
of
the
community.
It's
only
because
those
different
perspectives
may
lead
to
more
useful
nom-com
discussions
about
the
appropriateness,
particular
candidates,
so
so
I
I'm
worried
that
we're
that
you
know
they
can
get
this
balance
wrong.
If
we
continue
to
focus
on
trying
to
protect
against
attacks
much
as
I'm
concerned
about
the
attacks,
thanks.
I
I
J
Okay,
so
to
try
to
put
a
bore
on
this
issue.
I
have
not
heard
like
a
strong
consensus
to
change
path.
One,
although
I
certainly
recognize
that
we
are
nowhere
near
convergence
on
this.
Nor
is
there
a
concrete
proposal
for
a
different
path.
One
so
I,
don't
know
what
this
bodes
for
working
last
call,
probably
not
well,
but
I,
don't
know
chairs.
What
do
you
think.
A
J
J
A
B
J
I'm
happy
I
mean
I
I,
I'm
gonna
file.
Some
issues
to
to
add
some
some
things.
I
think
Ted's
threat
model
is
an
interesting
one
that
we
should
write
up
and
a
few
others,
and
all
this
like
stuff
that
people
should
do
that
leadership
should
do
Etc,
but
yeah
I'll
welcome
text.
I.
Don't
need
to
write
all
my
stuff
by
all.
A
J
A
J
Right
all
right,
so
the
other
issue
is
has
two
and
three
which
are
working
group.
So
there
was
right
so
path
to
his
working
group
chairs
and
we
were
talking
about
actually
what's
the
addressing
I've
gotten
confused,
yes,
chair
or
secretary,
and
the
question
is:
should
we
have
secretaries
and
I?
Don't
know
if
that's
actually
I,
don't
know
what
the
Venn
diagram
is
for
that,
whether
that
actually
adds
anyone
I
mean.
In
my
personal
opinion,
this
is
a
pretty
small
matter.
B
N
So,
based
on
my
experience,
a
secretary,
would
be
really
helpful
to
the
chair
because
during
my
Nom
come
the
chair
was
overworked
and
he
you
know
he
had.
He
had
a
lot
of
things
on
his
plate,
plus
his
current
work,
so
you're,
jumping
in
and
out
to
help
him,
but
there
were
stuff
that
we
could
not
ashamed.
Member
help
him.
So
he
has
a
potential
backup
on
some
of
the
issues.
I
think
it
could
be
useful.
We're.
G
Saul's
Akamai
nam-com
working
group
chairs
is
sort
of
like
Bigfoot.
You
rarely
see
them
I'm.
Sorry,
secretaries
are
rarely
like
Bigfoot
and
you
fairly
really
see
them,
and
in
order
for
them
to
be
effective,
they
really
have
to
be
at
the
meetings,
the
virtual
or
remote.
So
I
don't
think
it's
worth
adding
a
special
category
for
them.
Apart
with
any
more
than
a
working
group
chair,
you
can't
be
a
working
group
chair
and
not
have
attended
three
out
of
the
last
five.
A
G
B
So
they
can't
sit
at
the
front
easily
and
all
this
kind
of
stuff,
especially
before
remote
attendance,
but
they
often
have
very
long
long
views
of
what's
happening
and
that's
why
they
wind
up
as
secretaries
I,
think
it's
more
common
in
routing,
for
instance,
to
have
secretaries
that
have
that
situation
and
I,
don't
I,
don't
know,
we'd
have
to
go
the
numbers,
but
I
suspect
that
there's
five
or
six
secretaries
that
would
not
qualify
in
half
one
but
would
qualify,
and
we
probably
they
probably
know
their
names.
Well
right.
A
Do
we
consider
that
somebody
who
would
be
appointed
as
a
working
group
secretary
is
likely
to
be
somebody
we
think,
would
be
reasonable
on
the
nomcom
or
are
we
likely
to
see
people
put
in
as
working
group?
Secretaries?
Who
would
be,
who
don't
know
anything
about
how
the
ietf
works?
Who
don't
know
the
people
in
the
ietf
and
who
would
not
be
someone
we'd
want
on
the
Namco
and.
E
I
Oh
I've
had
people
was
working
group
secretaries
who
have
been
appointed
precisely
to
guide
them
into
the
iitf
and
make
them
more
involved
than
they
have
otherwise
been
and,
and
those
people
are
first
of
all,
those
experiments
often
fail
and,
and
secondly,
those
people
are
often
not
good
non-com
candidates
unless
they're
attending
a
meetings,
for
other
reasons,
so
I'm
a
little
bit
at
best
a
little
bit.
What's
about
this.
A
D
Thanks
Lars,
so
large
I
go
there's
an
ad
I
would
actually
disagree
with
John.
So
sometimes
we
appoint
secretaries
because
we
can't
fire
the
chairs
yet,
and
so
so
sometimes
it's
sort
of
it's
actually
that
the
role
of
the
secretary
to
do
the
actual
work
and-
and
they
are
maybe
more
qualified
than
the
chair.
So
it's
I
think
there's
a
spectrum
here,
but
I
I
don't
see
any
reason.
D
K
K
M
Hey
this
is
Pete,
and
this
came
up
in
my
head
earlier,
but
I
think
it
applies
to
this
discussion.
I
I
worry
about
us
over
analyzing
the
what
makes
a
good
quality
non-com
member,
because
there
are
loads
of
people
who
qualify
under
the
old
criteria
who
make
lousy,
non-com
members.
What
I
think
we
should
be
focusing
on
is
to
a
certain
extent
what
would
allow
reasonably
qualified
people
who
we
know
exist
in
reasonable
numbers
to
get
onto
a
non-com
where
they
currently
can't
and
so
I'm
I'm
ambivalent,
about
secretary
I.
J
B
So
so
the
biggest
the
biggest
problem
I
have
with
this
discussion
is
the
would
qualify
otherwise
discuss
part
of
it
right
and
so
a
lot
of
weight.
We
do
like
it
sort
of
assumes
that
we
know
what
the
other
paths
are
when
we
evaluate
a
certain
path
that
we
say
would
qualify
otherwise
I'm
like
well.
We
just
went
back
and
forth
for
an
hour
as
to
how
we
think
about
remote
I
think
we
got
very
close
to
removing
remote
from
the
list,
but
we
didn't
do
it.
B
So
that's
why
I
have
a
problem
when
whenever
he
says
it
would
qualify
otherwise,
qualify
that
to
me
is
like
well
wait
a
minute
so
you're
telling
me
they
are
qualified,
but
you
just
don't
want
to
qualify
them
under
this
path,
and
so
that
I
have
a
problem
with
with,
without
anything,
I
think
we
should
simply
evaluate
Things
based
on
their
direct
thing,
not
that
it's
Superfluous.
If
it's
really
hard
to
evaluate
and
Robert
says
I
can't
do
this,
then
that
will
be
different
in
my
mind,.
O
A
D
Like
individual
I'd
like
to
keep
it
simple
right,
because
what
if
somebody
is
like
a
work,
a
working
group
secretary
and
five
working
groups
does
that
count,
5x,
right
or
sort
of
it
gets
us
down
a
very
detailed
process.
That
I
think
we
actually
don't
really
need
here,
because
the
goal
is
to
prevent
abuse
of
the
system
and
not
like
optimal
pool
qualification,
right
Elliot.
O
A
J
A
J
J
A
C
B
C
J
J
J
J
A
H
J
I
I
will
prioritize
this
work
when
I
get
back
to
home,
but
I
would
encourage
anyone
who
is
like
very
unsettled
about
path,
one
to
actually
like
propose
a
PR
just
in
the
interest
of
paralyzing
this
work
and
like
we
can
withdraw
it.
If
the
security
considerations
address
everything,
but
if
not
like,
we
can
start
discussing
it,
because
maybe
it's
just
better,
even
if
we
have
security
considerations.