►
From YouTube: Governance and Risk Meeting #125
Description
## Introduction
* @LongForWisdom: Hosting, Introduction, Agenda.
## Weekly Updates
* Governance - @Gov-Facilitators
* Smart Contracts - @Smart-Contracts
* Oracles - @Oracles
* Risk - @Risk
* RWA - @SebVentures
* Community Development + Operational Support - @amyjung
* Collateral Calls - @juanjuan
* MIPs - @MIP-Editors
* PSM Update - @hexonaut
## Procedural
* @LongForWisdom: MIPs Governance Poll Review
## Segments
* @Andy_McCall: VaR Calculation
## Links
Website: https://makerdao.com
Twitter: https://twitter.com/makerdao
Forum: https://forum.makerdao.com/
Chat: https://chat.makerdao.com/home
Email: info@makerdao.com
Governance and Community Events Calendar:
https://calendar.google.com/calendar?cid=bWFrZXJkYW8uY29tXzNlZmhtMmdoaXBrc2VnbDAwOWt0bmlvbWRrQGdyb3VwLmNhbGVuZGFyLmdvb2dsZS5jb20
A
My
name
is
length
wisdom,
I'm
the
government
facilitator
at
maker
dao.
I
host
these
meetings
and
yeah.
We've
got
an
agenda
today
as
usual
through
some
general
updates
from
a
number
of
people.
Let
me
maybe
go
to
segment
later
regarding
risk
value,
calculating
value
or
risk.
I
believe
correct,
so
yeah
general
kind
of
reminders
feel
free
to
interrupt
people
as
they're
talking,
and
we
want
to
hear
from
you.
Please
don't
be
shy
and
yeah.
A
A
Uni
lp
volt
and
the
usdc
eth
unilp
volt,
along
with
a
bunch
of
other
changes,
so
yeah
look
out
for
that
tomorrow
in
terms
of
governance
domain
generally,
I
don't
think
I
have
any
any
substantial
updates
from
last
week
continuing
to
onboard
new
contributors,
which
seems
to
be
going
well
and
yeah
I'll.
Give
you
a
more
substantial
update
next
week
or
the
week
after.
B
B
B
For
those
that
don't
know
about
it,
makerburn.com
is
a
great
way
to
kind
of
check
out,
what's
happening
there
and
they
even
got
a
live
update
on
how
much
is
going
to
system
surplus,
and
if
you
click
on
stability
fee
zone,
you
can
see
about
how
much
our
growth
rate
per
second
is
in
stability
fees.
Right
now,
that's
about
a
dollar
and
10
or
1.1
die
per
second,
which
is
pretty
exciting.
B
B
The
core
unit
operating
model
brought
to
us
by
juan,
and
I
believe,
he's
going
to
speak
to
that
a
little
further
today.
So
I
won't
dive
too
much
into
it
other
than
say
that
it
looks
a
little
intimidating
as
a
bit
set.
But
it's
really
really
really
well
packaged
and
kind
of
lays
out
the
groundwork
for
how
we
can
move
to
paying
our
own
teams
and
operating
entirely
separate
of
the
foundation.
B
So
that's
a
pretty
neat
feature
and
they
asked
for
some
feedback
as
well
in
terms
of
how
we
see
that
maybe
being
used
with
maker,
so
check
out
that
on
the
forums
pretty
light
week
in
terms
of
signaling,
probably
thanks
to
all
the
other
signals
that
expired
this
week
or
a
few
days
before,
but
we
do
still
have
an
active
one
going.
That
is
new
from
the
last
time
we
gave
an
update,
and
that
is
for
an
eth
fault
with
a
higher
liquidation
ratio.
C
The
one
thing
that
got
bumped
for
the
liquidations
2-0
work
was
the
mip
21
work
and,
while
we've
been
working
on
liquidations2o
lev's
made
a
number
of
changes
to
the
21
stuff,
some
stuff
we
asked
for
and
we
we
had
pushed
back
last
week.
I
think
because
we
wanted
unit
tests
and-
and
that
was
maybe
a
good
thing.
Lev
ended
up
finding
a
a
bug
that
none
of
us
had
seen
just
in
the
manual
review.
C
C
So
they're
they're
gonna
go
at
it
at
that
point,
at
which
point
we
will
flip
our
focus
back
over
to
mip
21
and
getting
rip
21
in
covan
and
possibly
even
mainnet
for
next
week
and
and
then
as
soon
as
that
is
finished,
we'll
we'll
flip
gears
over
to
mip22
so
which
I
think
which
needs
to
go
through
that
a
bunch
of
the
code
is
written
for
that.
But
we
need
to
do
the
sort
of
like
full
review
process.
C
So
that's
our
our
kind
of
timeline
right
now
and
and
then
obviously
wherever
other
like
lp
collaterals
and
what
not
fit
in.
I
do
want
to
hand
off
to
sam,
though,
because
this
week
is,
you
know,
the
community
executive
as
many
people
know
and
sam's
got
a
bunch
of
stuff
that
he's
been
working
on
in
there
and
I
think
he
and
will
have
been
working
on
that.
So
maybe
sammy
could
like
give
us
an
idea
of
what
is
going
in
this
week.
D
D
C
A
E
Yeah
sure
so
we
yesterday,
we
deployed
the
the
new
lp
oracles
for
the
two
lp
tokens
that
are
going
live
tomorrow,
so
that's
wrapped
btc,
eth
and
usdc
eth
on
the
audit
front.
So
we
we
finally
did
talk
to
the
auditor
that
that
lev
had
connected
us
with
abdk
kind
of
scoped
them
out
a
bit.
They
seem
to
be
pretty
legit,
so
they
they
had
forwarded
us,
an
audit
that
they
did
for
tornado
cache.
E
So
after
looking
through
that
it
looked,
it
looked
pretty
thorough,
so
the
the
timetable
they
gave
us
is
that
they
should
be
able
to
do
it
kind
of
beginning
of
february
and
then
finish
by
by
february
15th,
so
that
that
timetable
looks
looks
very
looks
very
nice
for
my
opinion.
So,
even
if,
even
if
it's
a
bit
optimistic,
you
know
at
the
very
latest
end
of
february.
You
know
we
should
have
this
audit
back
so
turn
turnaround
time
on.
That
is
going
to
be
fairly
quick.
E
So
after
we
incorporate
those
changes,
you
know
we
should
be
able
to
increase
the
debt
ceilings
relatively
quickly.
So,
given
that
kind
of
short
interval
between
between
now
and
and
then
I
think
the
the
priority
for
collateral
onboarding,
you
know,
apart
from
the
the
real
world
assets
from
from
centrifuge
and
success,
should
be
to
kind
of
just
on
board,
lp
tokens
and
mass,
because
right
when
we
get
the
audit
results
back,
we
can
just
you
know,
crank
up
the
debt
ceilings,
and
you
know
we're
off
to
the
races.
E
E
Yep,
so
so
that's
thanks
are
moving
forward
on
that
end.
So
that's
good
on
the
stark
warefront
development
is
pretty
much
done,
so
we're
going
to
be
a
couple
things.
Well,
one
we're
going
to
be
putting
it
into
into
testing
in
our
staging
environment
and
two
mark
andre
has,
you
know,
compiled
the
the
mip
10
for
that,
so
that
should
be
going
live
before
the
end
of
this
week
as
well.
E
E
E
So
so
they
can
start
doing
some
of
the
stuff
that
people
like
dfisaver
are
doing
right
where
they
can,
they
can
save
you
from
from
liquidation
or
kind
of
manage
your
position
and
a
little
bit
of
foreshadowing
in
this
direction.
So
I
think,
in
terms
of
whitelisting
the
kind
of
governance
overhead
in
terms
of
you
know
the
quantity
that
has
been
coming
through
lately
and
the
time
you
know
it
takes
me
and
the
rest
of
the
oracle
domain
team.
E
You
know
to
write
up
the
the
polling,
votes
and
stuff.
I
I
was
gonna
propose,
say
a
mip
that
adding
someone
to
the
white
list
probably
shouldn't
need
a
a
governance
proposal
right.
The
the
only
thing
that
there's
really
not
any
reason
we
would
ever
reject
someone
right,
and
it's
really
just
you
know
more
money
for
the
dow.
E
Once
we
end
up
turning
on
the
the
oracle
fees,
the
only
thing
that
really
needs
to
be
checked
right
is
that
the
smart
contract,
that's
reading,
the
the
oracle
value
is,
is
kind
of
closed
out
that
it's
like
private.
So
no
one
can
just
kind
of
steal
the
oracle
value
from
there.
You
know
they
do
have
to
come
to
us
to
get
the
value
and
the
oracle
domain
team
already
does
those
evaluations
anyway.
E
Important
to
note
is
the
oracle
domain
team
should
not
get
the
authority
to
remove
anyone
from
the
white
list
right,
because
that
would
give
them
right
leverage
to
disrupt
protocols
that
that
are
using
the
oracle.
So
so
that
would
not
be
good
right,
so
so
government
governance
would
need
to
step
in
were
we
to
ever
want
to
to
remove
anyone.
E
F
Just
to
quickly
add
to
that,
I
had
yield
protocol
allen
from
yield
protocol
come
on
the
community
call
on
tuesday
and-
and
one
of
the
questions
we
talked
about
was
exactly
how
they
rely
on
our
oracles
without
being
whitelisted,
and
he
he
basically
like
described
like
a
a
bit
of
a
workaround
way
that
they're
reading
the
the
price
feeds
so
having
something
like
a
free
eat.
Feed
for
projects
that
are
still
in
like
super
early
stage
is
a
great
way
to
cement.
F
F
A
G
Yes,
so
yesterday
we
published
a
risk
evaluation
on
a
farmable
cosdc
crop
joint
adapter,
which
is
mip
30..
We
also
propose
parameters
for
this
wall
type.
I
suggest
you
read
the
risk
assessment,
since
this
kind
of
voltaic
is,
is
a
bit
more
complicated.
It
has
a
bit
of
an
unusual
risk
profile
compared
to
all
other
worlds.
We
onboarded
so
far.
G
Also,
this
this
adapter
could
be
very
useful
right
for
any
other
lp
tokens
that
have
liquidity
rewards.
So
I
think
this
is
important,
because
maker
can
position
itself
quite
uniquely
here
we
are
also.
We
also
started
evaluating
the
next
batch
of
uni
lp
tokens,
for
which
we
already
know
there's
a
high
demand,
so
in
the
upcoming
days
we'll
be
evaluating
usdt
unilpa
pair.
G
This
one
was
already
greenlit
recently
and
potentially
also
die
usdc
unilp
per,
but
this
one
is
still
pending,
so
if
it
gets
greenlit
we'll
work
on
it,
it
also
looks
like
don
quinn,
paul
for
depth
ceiling,
instant
access,
module
implementation
for
other
wall
types
test
today,
so
we'll
need
to
propose
parameters
for
each
wall
type.
In
order
to
have
this
implemented,
we're
probably
going
to
roll
them
out
in
few
batches
each
week,
or
maybe
you
know
all
at
once-
it
really
depends
if
you
can
find
some.
G
You
know
easy,
comma
formula
to
determine
all
these
parameters
needed
for
each
wall
type.
If
you
want
to
have
this
implemented,
I
think-
and
I
think
I
mentioned
this,
and
that
we
don't
believe
that
we
should
implement
this
module
for
stablecon,
vaults
or
psm,
because
this
might
limit
that
supply
when
it's
named
mostly
right
in
some
liquidity
crunch
events-
and
this
is
something
we
may
want
to
avoid
to
so
we'll-
probably
not
propose,
including
instant
access
module
for
something
like
psm.
G
I
think
it
even
doesn't
make
sense
if
you're
using
curb
module
anyway.
I
also
mentioned
last
week
we
started
preparing
budget
for
risk
team.
This
team
now
consists
quite
a
decent
amount
of
people.
I
think
it's
already
like
six,
maybe
even
eight,
if
we
count
part-time
for
doing
working
on
collateral
evaluations,
but
we
first
need
to
properly
read
this
new
me
proposal
that
I
think
juan
will
talk
about
so
the
pure
core
units,
one
to
understand
how
exactly
this
is
meant.
G
So
we
can
propose
budget
aligned
with
this
and
then
finally,
andy
has
serendi's
new
on
our
team.
He's
prepared
something
on
value
calculations
of
maker
of
maker
portfolio,
and
he
should
give
a
presentation
at
the
end
of
this
call
today.
So
stay
tuned
for
for
that,
and
that's
pretty
much
it
for
for
past
week
from
our
site.
H
Cool,
hey
primos,
I
wanna
circle
back
to
the
crop,
join
collateral,
risk
evaluation
that
you
put
out
yesterday,
you
mentioned
in
that
piece
that
compounds
governance
might
change.
The
speed
of
the
comp
rewards
that
are
distributed
to
particular
markets
in
this
case
would
be
cusdc,
and
you
said
that
it
creates
a
problem
for
makers
governance.
H
Do
you
think
that
the
maker
community
should
have
a
team
of
quote-unquote
ambassadors
that
communicate
with
compound
governance
and
try
to
work
on
building
a
strategy
that
is
beneficial
to
both
protocols
and
same
thing
goes
with
keepers
right
I
was
reading.
I
forgot
what
post
it
was,
but
something
to
do
with
keepers.
H
J
G
G
No,
no
bro.
So
in
regards
to
compound
this
speed
change,
this
is
already
effective
right.
So
this
this
is
not
really
an
issue.
The
yield
on
usdc
is
still
fine,
but
what
I
wrote
about
in
the
risk
assessment
is
that
there
are
other
upcoming
governance
changes
that
you
know
are
being
proposed
or
are
going
to
be
potentially
proposed,
and
we
need
to
keep
an
eye
on
so
one
of
them
is,
I
think
there
there's
going
to
be
a
cooldown
period
for
comps
claimed.
G
So
this
is
something
we
need
to
keep
an
eye
on
if
they
implement
it.
This
might
affect
adapter,
but
probably
left
should
know
better
and
there
could
be
other
things
as
well.
So
definitely
we'll
need
to
you
know,
keep
an
eye
on
their
governance
and
need
to
react.
You
know
as
soon
as
possible.
They
also
have
a
similar
procedure.
So
before
something
gets
implemented,
they
discuss
it
on
forums,
there's
an
active
signal,
I
think,
and
then
before
it
gets
executed.
G
I
think
it's
two
days
and
I
think
monet
supply
who's
on
our
team
is
quite
active
there.
So
he
knows
what's
going
on
so
this
is
all
always
good,
so
yeah,
that's
the
thing
about
compound
and
then
about
keepers
yeah.
So
we
know
that
the
maker,
you
know
makers
risk
is
struggling
a
bit
and
I've
talked
about
mitigating
risk.
We'll
talk
about
it.
G
Actually
the
end
of
the
call
as
well
but
yeah,
one
of
the
things
that
that's
kind
of
limiting
currently
is
50
million
bucks
and
of
course
we
would
primarily
want
to
increase
it.
If
we
were
confident
enough
that
keepers
had
enough
capital,
you
know
my
speculation
was
that
yeah,
perhaps
the
keepers
ecosystem
evolved
in
the
past
six
months
and
they
should
have
more
capital
on
their
hands.
But
you
know
this
is
only
speculation.
I
can't
really.
You
know
we
don't
have
any
evidence.
G
We've
tried
evaluating
unchanged
reserves
of
keepers,
but
it's
kind
of
hard
because
many
of
them
they
actually
mean
to
die
when
when
needed
right,
they
don't
necessarily
hold
it,
they
mint
it
or
they
already
minted
it.
So
it's
hard
to
just
simply
look
at
their
wallets
and
say
you
know
this
is
the
amount
of
money.
So
the
last
decision
we've
made
on
50
million
box
was
yeah
more
or
less
based
on
the
feedback.
G
We
we
we
got
from
keepers
right
and
it
sounded
a
bit
small,
but
this
was
still
good
right
because
if
they
would
say
a
high
number,
you
know
it
could
you
know
this
could
be?
Could
not?
We
couldn't
be
really
confident.
This
would
be
a
problem
like
if
this,
if
they
say
a
small
number,
it's
always
great,
because
we
know
you
know
for
them.
I
Yeah,
I
can
jump
in
really
quick,
so
the
problem
we
have
is
that
it's
very
cumbersome
infrastructure
to
run
right
now
and
there's
very
little
incentive
to
run
because
there's
just
not
that
many
auctions
you're
waiting
for
a
one-off
event
that
may
happen
twice
a
year
and
as
a
desk
deploying
capital
they
just
they
don't
want
to
do
that.
And
the
other
issue
is
that
a
lot
of
them
have
compliance
departments
that
have
to
approve
all
these
things.
I
And
if
this
infrastructure
is
made
redundant
by
liquidations
2.0,
which
it
seems
like
it
will
be,
then
they
don't
want
to
put
in
the
effort
to
go
to
their
compliance
departments
and
get
a
whole
setup
approved
and
put
in
all
these
resources
for
something
that's
going
to
be
redundant
in
a
few
months.
So
yeah
I
mean
it's
just
with
the
current
infrastructure.
H
Old
school
billionaires
talking
about
that,
there's
a
euphoria
right
now
in
the
financial
markets,
traditional
markets,
etc.
So
how
do
you
sleep
at
night?
Do
you
think
that
the
first
wall
that
market
relayer
has
tim
black,
has
put
out
a
market
relay
letter
this
week
and
it
shows
that
there's
about
35
million
died
to
be
liquidated
at
1071
I
mean.
H
G
Well,
the
the
good
part
is
that
overall,
the
collateralization
ratios
are
much
higher
than
they
were
in
the
past
right.
So
it's
you
know,
the
portfolio
looks
quite
it
looks
more
healthier
than
it
was
ever
before.
Actually,
but
if
you
look
at
the
nominal
exposures,
they're
still
big,
so
this
is
a
worry.
That's
why
you
know
the
last
few
weeks.
I've
been
kind
of
you
know
shouting
a
bit.
You
know
we
need.
We
may
need
to
change
some
things.
We
may
need
to
think
about
high
liquidation
ratio,
volts
yeah.
G
This
is
exactly
the
point
of
the
presentation
towards
the
ends
of
call
so
yeah.
I
am
a
bit
worried.
On
the
other
hand,
I
can
see
that
you
know
world
users
are
a
bit
more
sophisticated.
You
could
see
every
time
price
falls.
You
could
see
people
unwind
in
time,
there's
more
automated.
You
know
unwinding
going
on.
Edb,
though,
looks
a
bit
worse
right.
If
you
just
look
at
auctions
today,
currently
they're
running,
you
know
they
just
started
kicking
off.
So
maybe
that's
not
the
best
insight.
G
D
Sure
so
we
are
working
a
lot
still
on
the
legal
documents
provided
by
centrifuge
to
see
if
everything
will
work
as
mostly
as
expected.
Currently
we
don't
have
a
lot
of
issues.
There
will
be
a
past
in
the
following
days
and
we
hope
that
there
will
be
at
some
point,
a
legal
team
at
maker
as
well,
to
provide
another
view
and
more
more
insight.
D
D
Some
are
very
good,
not
only
the
fun
crazy
projects,
but
at
the
at
the
time
being,
it's
always
difficult
to
put
the
propositions
inside
the
current
structure,
and
so
that's
why
I'm
turning
to
the
community
to
see
if
a
long-term
lending
is
an
option
or
not,
and
obviously
every
time
there
is
someone
wanting
to
borrow
from
us.
I
told
them
that
we
it's
either
meep
21
or
mip
22.
D
D
Think,
I
guess,
oh
sorry,
guys
yeah
if
the
call
is
going
to
be.
As
short
as
last
week's
I
just
wrote
in
the
chat
as
well,
then
I
think
I
would
be
super
curious
to
have
a
bit
of
a
debate
discussion
around
interest
rates
and
like
long-term
versus
short-term
borrowing
yeah.
I
would
love
to
hear
what
the
community
says.
D
I
I
mean
I'm
obviously
like
share
a
lot
of
the
experiences
that
you
as
well
just
mentioned,
said
with
demand
for
longer
term
assets
and
different
kind
of
interest
rates
and
sort
of
as
it
as
maker
like.
How
do
we
deal
with
that?
How
do
you
compare
traditional,
traditional
banks
and
funders,
but
maybe
that's
towards
the
end
of
the
call,
or
I
don't
know
long
what
the
agenda's
like.
A
L
Hey
so
a
few
one
thing
that
I
like,
it
was
post
like
right
before
the
holidays,
so
I
would
just
want
to
remind
people.
There
is
a
beta
program
for
a
it's,
a
limited
edition,
limited
duration,
beta
program
that
the
data
services,
which
is
an
internal
team
in
the
foundation,
dropped.
So
they,
the
api,
will
be
giving
you'll
be
able
to
receive
historical
data
on
mcd,
vaults
and
they're,
going
to
it's
one
piece
of
a
whole
suite
of
products
that
are
coming.
L
A
J
J
So
if
you
haven't
seen
it
check
it
out,
it's
deployed
on
ipfs
so
really
excited
about
that.
Regarding
the
the
community
calls
in
yesterday,
we
had
a
call
with
the
with
harvard
trade
credits
on
real
world
assets.
Quite
interesting.
The
video
is
already
on
youtube.
So
if
you
want
to
check
it
out,
it's
it's
there
and
notes.
J
Coming
soon
and
next
week
we
will
have
lido,
which
is
I'm
probably
not
doing
justice
to
the
project
but
they're,
bringing
if
two
staked
back
to
each
one
for
better
or
worse
chris.
So
that's
that's
on
on
that.
On
that
front,
I
don't
know
if
you
want
me
to
to
show
a
brief
presentation
that
I
prepared
about
the
korean
operating
model
or
or
we
can
do
it
at
the
end
of
the
call
up
to
you
guys.
J
Screen
is
it
good,
yeah,
okay,
so
yeah
before
starting
a
very
brief
presentation,
we
will
have
a
specific
call
being
hosted
by
pros
on
monday.
So
know
your
mip
number
three.
I
guess
we
are
at
the
cadence
of
one
per
month,
so
it
I
guess
it's
good
and
good
timing,
but
yeah
a
lot
of
people
helped
doing
this.
So
by
means
it's
it's
my
my
own
work.
So
thanks
to
everyone
that
helped
and
also
everyone
that
provided
feedback
on
the
on
the
forum,
a
little
bit
of
context.
J
J
J
So
again
why
the
core
unit,
the
korean
operating
model,
is
the
first
step.
So
it's
a
it's
a
face
approach
and
the
idea
is
to
have
double
primitives
and
they
are
designed
to
be
easy
to
transition
to
capable
of
accommodating
the
current
structures
with
minimal
changes.
I
think
that's
important,
so
people
were
asking
about
the
main
themes
and
current
moves.
J
So
the
idea
is
that
these
things
continue
to
work
with
or
under
the
this,
this
new
framework-
and
we
don't
change
everything
that
we
can
receive
the
newly
delegated
operations
from
the
foundation
right,
so
anything
that
they
won't
be
able
to
do
anymore.
We
want
to
be
able
to
to
do
it,
and
the
important
part
is
that
we
want
a
coherent
framework
and
not
just
governance,
voting
on
things
that
are
hard
to
understand
and
keep
track
of
so
yeah,
as
I
was
mentioning
a
facebook
approach.
J
So
what
we're
focusing
on
right
now
is
the
initial
approach,
with
this
mipset,
adding
the
temporary
measures
that
we've
seen
and
hopefully
in
the
future.
We
can
reiterate
with
more
mips
and
mipsets
and
seeing
if
there's
any
need
in
the
future,
but
yeah.
The
the
plan
of
of
this
of
this
snip
set
was
flexibility
and
modularity.
J
J
So
the
dow
can
provide
the
recurring
cash
flows
and
expect
the
services
in
exchange.
Everything
is
clear
and
listed
in
in
in
a
mip
and
then
yeah
a
core
unit
is
very
simple
and
very,
very
flexible.
So,
according
has
a
name
and
a
mandate,
that's
it
then
there's
one
or
many
facilitators
with
their
name
under
commitment.
J
J
This
is
the
implementation
and,
and
the
breakdown
is
a
little
bit
the
rational
behind
it
and
then
the
facilitator.
With
this
budget,
they
will
be
able
to
hire
contributors
to
become
a
company
and
hire
employees
to
form
a
dow
allow
split
into
several
core
units
bringing
another
facilitator
to
help.
So
it's
quite
flexible
in
that
sense
and
that's
a
little
bit
what
we're?
What
we're
looking
for,
moving
to
the
next
slide
and
as
an
example
of
of
a
core
unit
and
any
resemblance
with
reality,
is
pure
coincidence,
but
yeah.
J
J
J
So
we
want
to
keep
it
very,
very
simple
and,
and
then
yeah
nick
would
be
able
to
hire
either
contributors
or
a
company
if
he
needs
an
audit,
for
example,
or
if
he
wants
he
can
organize
and
and
then
yeah
there
are
different
questions
like
can
nick
form
two
core
units.
Potentially,
yes,
can
a
domain
team
split
into
two
core
units
to
specialize?
Also,
yes,
so
so
we
want
to
keep
it
flexible
right.
J
We
think
that
the
current
structure
of
of
having
domains
is
not
enough
for
everything
that
we
want
to
achieve
in
the
in
the
next
year,
and
this
is
a
good
first
step.
I'm
almost
done
guys
or
sorry
modular
approach,
and
what
do
we
mean
with
this?
It
means
that
we
want
it
to
be
compatible
with
with
current
mips
and
also
future
mips.
So,
for
example,
in
the
budget,
you
can
see
that
someone
could
do
a
manual
suck
or
we
can
implement
a
keg.
So
some
no
pressure
on
that
one.
J
We
have
the
mip
42.
That
was,
I
think,
published
for
rfc
this
morning.
So
that's
potentially
something
else
that
could
work
with
this
budget
or
any
other
implementation
that
we
can
think
of
in
the
future.
Right
so
pretty
much
the
budget
and
it
could
be
several
budgets,
so
one
could
be
four.
I
know
salaries
and
the
other
one
could
be
a
bonus,
for
example,
but
a
core
unit
could
have
one
or
multiple
budget
assigned
to
it
with
different
rules.
I
need
different
implementations.
So
that's
that's
a
little
bit
plan
same
thing
with
facilitators.
J
J
We're
going
to
do
a
q,
a
know
your
mip
call
on
monday
january
25th,
so
this
coming
monday
at
8,
00
pm,
utc,
so
join
us
pros,
will
post
the
agenda
soon.
Then
we
have
also
created
a
core
unit
operating
model
on
boarding
a
rocket
channel.
I
know
that
it's
not
the
best
name
but
well,
at
least
it's
descriptive.
J
So
so
that's
that's
a
bit
on
the
on
the
pipeline.
Regarding
this
mipset
yeah,
I
need
to
incorporate
all
the
feedback
so
again,
thanks
for
for
all
the
feedback
that
I've
received,
the
second
one
I'm
amy
for
republishing
next
week.
The
big
change,
as
I
mentioned
that's
happening,
is
multiple
facilitators
per
core
unit
or
the
possibility
of
multiple.
It
could
be
one,
but
the
rest
is
almost
cosmetic
right,
so
it's
making
sure
that
it's
clearer,
because
long
for
wisdom
wasn't
particularly
happy
with
my
with
my
writing.
J
No
I'm
I'm
half
kidding,
but
but
yeah
potentially
we
can.
We
can
get
this
out
submitting
it
and
voting
so
call
your
whale
friend
for
support,
I'm
looking
at
you,
planet,
x
and
and
yeah
thanks
again
for
for
everything
I
don't
know
if
there
are
any
questions.
If
we
have
time
for
questions
or
you
can
join
us
on
monday,
there's
also
a
really
good
video
that
mio
did
and
posted
in
the
in
the
forum
where
he
goes
more
in
details
about
different
possibilities.
So
that's
also
helpful,
so
yeah
any
questions.
A
A
I
would
encourage
everyone
to
read
those
over
either
now
or
after
once
on
some
updates,
because
yeah,
it's
it's
a
pretty
major
step
in
in
the
dallas
development
and
it's
great
to
see
yes
good
job
to
everyone
who
is
involved.
I
know
it
was
the
not
just
one
all
right
that
me
brings
us
to
mips,
then
I
think
david.
I
believe,
you're
doing
something
this
week.
A
F
Oh,
okay,
that's
how
you
do
it
there.
It
is
all
right,
no
matter
how
many
times
I
screen
share.
I
somehow
always
manage
to
not
know
how
to
do
it.
Happy
thursday,
everybody
today
we
are
going
to
be
going
over
the
weekly
mips
update,
as
we
usually
do
on
this
call.
My
name
is
david
utrobin.
I
am
a
community
development
leader
at
the
foundation
and
resident
myth,
editor
in
training
actually
yeah.
So
let
me
not
share
geez.
F
My
brain
man
today
is
the
day
all
right
cool.
So
let's
go
for
it.
I
created
a
little
like
slide
for
a
visualization,
because
I
feel
that
visualizations
often
help
people
kind
of
place
themselves
and
understand
things
a
little
bit
better.
So
we
are
here
today
is
january
21st,
so
the
governance
poll
actually
just
ended
so
we're
coming
up
at
to
the
end
of
week,
three
in
our
january
governance
cycle
and
next
week
on
monday,
we're
gonna
have
a
bundled
executive
vote.
F
So
yeah,
let's
go
through
it.
So
there's
a
number
of
proposals
moving
ahead
that
are
part
of
this.
Also,
I
can't
see
the
comments
so
feel
free
to
to
shout
out
anything
long
that
I
should
pay
attention
to
so
yeah.
So
there's
a
number
of
proposals
moving
ahead
with
the
january
governance
cycle.
Long
is
actually
going
to
confirm
them
later
on
in
this
call,
but
we
have
the
peg
stability
module,
which
actually
has
already
been
implemented.
F
This
technical
map
being
passed
is
really
a
kind
of
formality:
mip
30,
farmable,
c
usdc
adapter
is
currently
there
and
also
a
declaration
of
intent
for
the
governance
communications
team.
Some
more
proposals.
There
are
six
in
total
there's.
Also
a
sub
proposal
for
operational
support
domain
facilitator
onboarding
this
on
actually
onboards
our
very
own,
juan
as
operational
support
domain
facilitator.
F
Likewise,
we
have
a
domain
team,
onboarding
mip
as
well,
which
on
board
sebastian
duro
as
a
risk
team
domain
facilitator
and
then
the
last
domain
team
onboarding
mip.
That's
in
this
month's
cycle
is
for
smart
contracts
which
on
board
sam
mcpherson
as
a
smart
contracts,
team
domain
facilitator,
and
so
this
is
really
cool
because
none
of
these
people
are
like
ex-foundation
employees.
So
it's
really
amazing
to
see
more
and
more
community
members
being
drafted
into
these
mandated
actor
roles.
F
That's
it
for
the
proposals
that
are
moving
moving
ahead
in
the
current
cycle.
Here
are
some
proposals
that
are
in
rfc.
We
have
mip
26
the
governance
contract
redesign.
We
have
nip
32
the
peg
stability
module
for
compound
mixed
exposure.
This
takes
the
psm
idea
and
and
applies
the
the
concept
of
a
farming
comp
and
some
yield
with
the
usdc.
That's
just
sitting
in
the
psm.
F
Creating
some
revenue
for
maker
for
the
protocol.
Mip
33
is
the
maker
stability
price
module.
This
is
a
proposal
that
allows
governance
to
defend
a
given
mkr
price
by
allocating
die
to
buy
it
out
of
the
surplus
buffer.
Basically,
it
like
throttles
the
the
idea
of
like
the
buy
and
burn
at
a
steady
rate.
Instead,
it's
like
yeah,
it's
it's.
An
interesting
idea,
definitely
request
for
comment.
There
nip
34
as
well.
We
have
the
keg
streaming
payments
module.
This
is
the
evolution
of
the
flapper
distributor.
F
It
basically
allows
streaming
of
dye
into
specified
addresses
out
of
the
surplus
buffer,
so
this
is
going
to
be
really
cool.
As
we
start
establishing
things
like
strategic
reserve
funds-
maybe
you
know
overhead
operating
budget
like
reserves
and
and
other
interesting
things
like
that,
a
few
more
additional
proposals
in
rfc
mip
35,
the
peg
stability
module,
which
is
the
compound
mixed
exposure
one
again,
except
this
one-
is
with
farming,
so
the
other
one
just
collects
the
yield
for
for
lending
out
the
usdc.
F
While
this
one
does
that
and
also
the
comp
farming
mip
36
is
the
peg
stability.
Compound
governance
die
leverage.
This
is
also
part
of
that
same
project.
Mip
42
is
for
the
proposed
compensation
addendum
for
maker,
dow
contributors.
This
this
mip
is
actually
really
cool.
It
proposes
to
pay
contributors
partially
in
time
locked
mkr
and
proposes
this
idea
of
like
v
mkr
similar
to
vcrv
and
so
yeah
definitely
throw
some
comments
in
there
and
just
note
on
that
mip.
F
It
is
a
bit
early
since
there
are
a
number
of
things
in
the
air,
for
example
like
the
keg,
for
example
like
the
governance
contract
redesign,
so
there's
a
lot
of
dependencies
there,
but
the
idea
is
really
great
to
have
formerly
an
rfc,
also
subproposal.
There
is
a
eight.
I
believe
this
isn't
yeah.
This
is
a
declaration
of
intent.
F
It's
onboarding,
a
new
collateral
type
backed
by
the
protocol
also
interesting
indeed,
b
protocol
is,
is
a
protocol
that
actually
aids
it
like
creates
almost
like
a
liquidation,
dao
type
of
construct
and
helps
yeah
helps
make
her
take
care
of
liquidations,
essentially
by
letting
users
migrate,
their
vaults
there
and
then
giving
them
first
priority
for
liquidations
so
yeah.
It
would
be
interesting
to
see
what
a
new
collateral
type
backed
by
them
would
be.
Would
look
like
only
a
couple
of
mips
in
conception.
F
At
the
moment
we
have
mip
31
the
active
reserve
amm.
This
is
a
technical
proposal.
It
basically
makes
a
case
for
having
a
mkr
die
reserve
that
also
acts
as
a
amm
within
the
maker
protocol
and
then
mip
37,
which
is
the
static
reserve
for
governance,
which
is
like
the
most
basic
version
of
a
reserve
that
we
can
have,
which
can
just
receive
any
tokens
and
can
send
any
tokens,
as
governance
would
vote
to
do
so
if
they
wanted
to
so
yeah.
F
Really
cool
ideas
definitely
jump
in
and
just
a
note
on
what
the
difference
between
conception
and
rfc
is
for
mips
whenever
a
mip
is
initially
posted.
It's
in
conception,
but
after
mip
editors
reach
out
to
the
author
to
fix
just
basic,
formatting
and
and
kind
of
making
sure
nothing's
going
going
too
crazy
there.
It
gets
moved
to
rfc
and
also
with
being
with
rfc
there's
a
pull
request.
F
That's
submitted
to
the
mips
repo
a
number
of
premium
discussions
continuing
to
go
on
in
the
forum,
thoughts
on
allowing
mkr
defy
liquidity
providers
to
vote
on
how
that
would
be
technically
possible
if
it's
even
worth
it
to
prioritize.
Something
like
that.
The
throttled
surplus
buffer
idea,
the
the
delegation
of
authority
idea,
the
real
estate
tokenization
protocol
idea
and
and
a
brainstorm
on
how
we
ensure
small
vaults,
have
a
place
to
maker
and
so
yeah
all
right
and
then.
Lastly,
just
some
collateral,
onboarding
updates
as
they
relate
to
the
mips.
F
I
believe
we
are
waiting
on
an
executive
for
three
of
them.
I
actually
would
love
some
confirmation
from
whoever's
putting
together
the
community
executive.
That's
going
up
tomorrow,
I'm
wondering
if
success
is
a
part
of
that.
I
know
I
have
it
on
pretty
good
authority
that
usdc
ethan
wbtc
ethp
tokens
are
going
in,
but
if
somebody
can
quickly
comment
on
that,
I
would
really
appreciate
it.
C
F
The
assessments
that
have
been
done
and
assigns
like
act
like
the
current
status
of
every
single
collateral
application
so
definitely
use
that
as
a
resource
collateral,
onboarding
updates
continued.
There
were
a
number
of
community
greenlight
polls
for
this
month
that
I
believe,
are
actually
still
active,
so
definitely
vote
on
those
I'm
not
going
to
read
through
all
of
them
here.
H
H
Much
for
the
work
you've
done
in
the
past,
and
you
know
I
learned
a
lot
from
you.
You
helped
me
when
I
didn't
understand
a
certain
myth,
or
even
something
so
basic.
So
thank
you
for
the
time
you
spent
working
for
the
foundation,
and
I
can't
wait
for
you
to
start
your
own
core
unit
and
continue
to
be
a
part
of
the
community
yep.
That's
pretty
much.
It.
A
G
G
So
this
kind
of
presentation
should
give
us
some
kind
of
insight.
You
know
about
how
well
protected
maker
truly
is,
and
I
also
think
the
presentation
should
serve
governance,
not
just
to
understand
this
risk
profile,
but
also
soon
start
thinking
about.
You
know
risk
mitigation
decisions,
because
the
debt
exposure
can
continue
to
grow
and
we
may
want
to
react
sooner
rather
than
later
so
I'll.
Let
actually
andy
present
his
work,
and
then
we
have
some
discussion
at
the
end
of
his
presentation.
K
You
should
be
able
to
see
this
correct,
correct,
yeah,
so
just
real
quick,
my
name
is
andy.
You've,
probably
seen
me
running
around
in
the
forums
I've
been
around
for
a
long
time,
yeah
and
I
have
been
helping
the
this
team
over
the
past.
You
know
month
or
two
to
refine
the
old
collateral
risk
model
and
run
some
simulations
for
them.
Why
didn't
that
work
here?.
D
K
I
won't
read
this
whole
disclaimer
here,
but,
like
this
is
sort
of
a
holdover
from
previous
presentations
that
the
risk
team
has
done
more
or
less.
It
says
this
isn't
financial
advice
and
you
can.
You
should
consult
your
own
advisors
to
those
matters
I'll
post
this
slide,
probably
after
this
is
done
so
this
set
of
slides.
So
you
you
can
probably
be
able
to
find
these
on
the
forum
within
the
day
yeah.
K
So
the
first
thing
that
we
need
to
talk
about
is
what
is
diode
risk.
Var
var,
you
may
hear
me
say
all
those
terms
here.
This
is
a
nice
little
definition
that
comes
from
wikipedia
feel
free
to
read
it.
You
know,
basically,
it
estimates
how
much
a
set
of
investments
might
lose.
Given
a
period
of
time,
you
know
here's
a
picture
that
kind
of
illustrates
what
we're
talking
about
here.
K
You
know
you
have
this
bell
curve
type
shape,
and
this
red
line
here
represents
the
value
at
risk
at
five
percent
on
this
particular
bell
curve.
So
the
area
under
this
that
shaded
in
red
here
that's
five
percent
of
the
total
area
underneath
this
curve-
and
you
know
in
this
example-
basically
the
that
number
is
8.2,
but
that's
not
really
terribly
important,
just
kind
of
to
give
you
a
sense
of
what
we're
talking
about
here.
K
So
why
do
you
care
about
this
particular
metric?
Like
pretty
much
said,
the
nominal
risk
at
maker
is
growing.
I
think
today
the
like
max
theoretical
debt
ceiling
is
something
like
2.2
billion
as
up
this
morning,
and
so
you
you'll
see
that,
like
you
know,
there's
even
in
this
call
previously
some
some
comments
about.
K
You
know
certain
media
figures
making
statements
like.
Oh
there's,
this
much
maker,
this
much
diet
risk
in
the
maker
portfolio.
Basically,
this
value
at
risk
metrics
helps
you
make
statements
like
I
have
quoted
here.
That
says,
you
know
reasonably
confident
that
make
her
like
sit.
You
know
make
her
losses
will
not
exceed
x.
You
know,
depending
on
your
confidence
in
a
given
year,
you
know.
Potentially
you
can.
You
can
use
that
sort
of
information
to
help,
inform
your
stability,
buffer
decisions,
potentially
rates
decisions
etc,
and
you
know
kind
of
just
overall.
K
It
gives
you
a
sort
of
easy
to
consume
format
to
talk
about
how
risk
is
growing
in
the
portfolio.
So,
as
you
raise
the
rates,
you
can
point
to
this
metric
and
say
look
where
this
number
is
going
up
I'll,
be
real,
quick
here,
pretty
much
as
far
as
covered
most
of
this,
but
here's
a
link
to
the
collateral
risk
model,
here's
the
source
code.
K
So
there's
a
bunch
of
inputs
that
you
need
to
come
up
with
in
order
to
actually
run
the
simulation
and
a
lot
of
these
inputs,
especially
a
few
ones
in
particular,
are
quite
complicated
to
derive
so
over
the
past
month,
we've
written
some
software
that,
like
helps
us
to
quickly
figure
this
out
for
a
given
collateral
type,
and
the
last
thing
is,
you
know
you
you
can
with
this.
With
this
model,
like
I
said,
you
can
run
a
simulation
for
a
hypothetical
vault
type,
but
you
need
to
account
for
that.
K
K
K
It's
basically
you
know,
like
I
said
here-
a
function
of
keepers,
profit
expectations,
plus
the
price
impact
of
liquidating
things.
I
won't
go
into
this
too
much
because
I'm
already
up
on
six
minutes,
but
I
highly
recommend
I
have
a
link
here
on
the
the
wi-fi
slippage
analysis,
which
is
where
I
got
this
graph.
I
highly
recommend
somebody
just
going
in
there
and
looking
at
that
thing,
you
don't
have
to
really
take
it
in,
but
it's
amazing.
This
sheet
is
amazing.
K
There's
like
five
or
six
pages
worth
of
random
math
that
the
wrist
team
put
together
just
to
you
know,
basically
make
these
four
lines
here.
I'm
dumbfounded
every
time
I
look
at
it.
It's
it's
great,
so
we
wrote.
I
wrote
a
program
that
kind
of
can
do
the
calculation
performed
in
this
spreadsheet
here
and
using
that
we
can
sort
of
determine
the
slippage
for
each
individual
collateral
type
and
having
those
slippage
values
basically
allows
us
to.
You
know:
look
at
the
cl
how
faults
are
being
collateralized.
K
K
K
K
K
K
The
unix
swap
liquidity,
tokens,
wptc
and
psn
reason,
for
that
is
basically
because
there's
some
debate
about
how
to
calculate
slippage,
and
I
thought
it
best
to
just
leave
them
out
and
ignore
that
they
exist
for
purposes
of
this.
You
know
these
numbers
are
sufficiently
high
that
you
know
I
don't
feel
like
it's
necessary
to
talk
about
them
too
much,
but
just
know
that
when
you
read
this
and
yeah
so
here's
some
numbers.
K
You
know
basically
here's
a
few
arbitrary
confidence
thresholds
I
picked,
but
you
know
kind
of
basically
how
you
can
read
this.
A
little
bit
is
basically
for
every
number
you
see
here,
bar
p
of
n
right
1,
minus
n
is
like
basically
what
the
model
predicts
the
probability
of
lost
single
day
losses
exceeding
this
number
on
the
in
the
right
column.
So
you
know
the
model
predicts
that
there's
a
20
chance
of
exceeding
86.4
million
in
the
next
one-year
period
of
time.
That's
one
way
you
can
read
this.
L
K
Is
the
ether
versus
e3
thing
here?
You
know
obviously
those
out
of
the
faults
that
were
simulated.
You
know
I
talked
about
all
the
ones
that
weren't
all
the
ones
that
were
simulated.
These
two
are
the
highest
amount
of
exposure
from
that
set.
They
okay,
they
account
for
almost
all
of
the
value
at
risk
there
and
you'll
notice.
In
these
forum
threads,
like
various
risk
team
members
commenting,
I
don't
feel
comfortable
raising
these
things.
This
sort
of
thing.
K
E
K
This
is
a
quote:
I've
linked
the
website
here,
where
you
can
find
this
quote,
but
basically
what
this
point
with
this
point
is:
this
quote
is
trying
to
tell
you
it's
not
like.
Okay,
I
gave
you
a
bunch
of
numbers
that
the
models
spit
out
for
diet
risk
at
different
confidence
levels,
those
numbers
dude,
there's
a
there's,
a
term
called
a
var
breach,
which
you
know
those,
basically,
that
the
model
splits
out
a
confidence
level
and
a
number
right.
That
says.
K
We
think
that
there's
only
20
chance
that
the
loss
will
exceed
x
right.
It
does
not
state
how
large
that
loss
will
be
if
it
exceeds
x.
So,
for
instance,
we
saw
on
sciatica
the
model
thinks
that
there's
a
one
percent
chance
of
it
exceeding
1
million,
if
there
is
say,
for
instance,
a
half
a
percent
chance
that
the
loss
could
be
1
billion.
K
It
doesn't
that
that
there's
nothing
wrong
with
that
statement,
that
there
is
a
one
percent
chance
that
it
will
only
it
will
exclude
only
one
100
million,
so
be
a
little
bit
careful
looking
at
those
numbers,
they're
they're
often
misinterpreted,
and
they
do
not
sort
of
speak
to
the
nominal
size
of
any
potential
bar
breach,
only
the
likelihood
that
they
would
occur,
and
you
know.
Lastly,
I'm
pretty
much
ready
for
questions.
C
K
K
K
G
Maybe
I
could
jump
in
so
I
think
I
think,
with
the
box
of
50.
It
really
depends
how
the
you
know.
What's
the
price
trajectory,
but
I
would
say
normally
you
know
it
protects
from
really
worse
events
like
zero
bits
and
something
like
that.
So
if
we
adjust
it
for
the
box,
we
would
get
more
optimistic
numbers.
I
think
better
numbers,
but
on
the
other
hand,
what
we
are
also
not
what
the
model
is
also
not
doing
it's
not
taking
into
account
people
unbinding
before
before,
getting
with
it
right
so
yeah.
G
You
know
we
can't
really
model
for
every
single
thing:
user,
behavior
and
this
kind
of
stuff,
but
like
on
average
we
may
be
under
you
know
we're
not
really
taking
into
account
one
part
with,
maybe
which
may
be
too
optimistic
on
the
other
part
too
pessimistic,
or
something
like
that.
So
on
average
I
would
say
the
slippage
curve,
which
is
kind
of
the
real.
K
K
I
G
No,
it's
it's!
It's
basically
simplifying
it.
It's
basically
what
the
model
is
doing.
It's
calculating
clauses
on
each
day,
so
we
have
thousands
of
simulations
and
it
there's
a
one
year
period
right
and
then
this
crash
occurs
and
then
you
know
you
have
random
walk.
We
have
thousands
of
simulations
and
you
get
this
distribution
of
losses
and
the
average
of
those
losses
is
the
expected
loss.
That's
the
risk
premium.
G
This
is
the
one
that
we
use
when
we
when
we
set
rates
right,
but
when
you
want
to
look
at
the
tail
losses
right,
the
worst
case,
events.
This
is
how
banks
do
it
as
well
right
they
calculate
value
freeze
because
they
want
to
be
cashed
against
worst
case
events,
and
this
is
the
number
right.
So
this
is
the
one
percentile
of
divorce
losses
that
we
simulated
and
thousands
of
them
and
daily
losses.
G
G
But
I
get
your
point
like
we
could
make
an
alternative
simulation
which
would
be
focused
just
on
a
short,
shorter
time
interval
just
simulate.
You
know
how
one
day
the
price
trajectory
during
the
day
try
to
you,
know
micro
estimate
that
loss
and
then
extrapolate
it
to
a
yearly
figure.
It's
also
doable,
but
it's
not
how
the
model
currently
works.
I
Yeah-
and
I
guess
I
you're
looking
at
it
from
a
little
bit
different
of
a
lens
than
I
do
like.
I
always
worry
about
the
an
osm
attack
where
you
could
see
a
vault
go
under
100
collateralization,
but
the
rule
of
thumb
I
do
for
that.
Personally
is
just
what
was
the
worst
hourly
candle
on
any
chart
in
march
and
then
overlay
that
into
the
vault
and
assume
that
that
will
happen
so
yeah.
I
guess
we're
kind
of
taking
it
from
two
different
angles
too.
G
But
thank
you
sure
I
just
want
to
jump
in
here.
So
what
why
I
wanted
to
have
this
discussion
today
is
really
to
start
thinking
about
you
know:
how
can
we,
you
know
catch
catchmaker?
What
are
the
tools
available?
I've
discussed
about
it
so
their
surplus
buffer?
You
know
we
may
need
to
think
about
increasing
it
even
further,
but
you
also
need
to
agree
that
you
know
it
takes
time
to.
G
You
know,
get
to
a
10
million
figures,
so
I
think
it's
gonna
take
what
about
two
months
or
three
months
to
get
to
10
million
from
now
right.
This
might
be
already
too
late.
Then
the
second
thing
is
box
parameter.
We
discussed
it
like
right,
it's
hard
to
justify
high
number,
but
would
be
really
beneficial.
G
Then,
of
course
liquidations
2.0.
I
think
it
was
in
the
slide.
You
know
this
would
totally
improve
the
slippage
function,
so
the
number
would
be
much
lower.
So
that's
why
I
think
liquidation
2.0
is
kind
of
a
priority
from
the
development
perspective
right
now
and
then
the
other
thing
right
about
higher
liquidation
ratio,
volts
something
I've
brought
up.
G
Can
change
instantly
like
if,
if
current
worlds
think
that
new
ethereum
price,
the
current
ethereum
price,
is
kind
of
a
stable
price,
they're
going
to
mean
more
die
and
they
will
push
their
ratios
down
there
towards
250..
Currently,
it's
pretty
high,
like
just
never
so
high.
If
we
get
to
some
equilibrium
state
that
we
you
know
we
observed
in
the
past,
then
we
might
have
like
50
of
the
portfolio
lowly
collateralized
and
then
those
numbers
become
even
worse
right,
so
yeah.
G
G
A
A
A
Monday,
all
right:
okay,
we're
about
half
past.
Anyone
has
any
final
comments.
They
want
to
make,
I
guess
or
anything,
and
we
should
maybe
think
about
wrapping
up
as
usual
I'll
leave
the
call.
Well,
as
usual,
I
will
leave
the
call
open
if,
in
case,
people
want
to
discuss
in
more
detail
after
we
stopping
the
recording.
So
if
anybody
does
want
to
discuss
more
with
dreamershaw
or
andy
and
they're
willing
feel
free
to
stick.