►
From YouTube: Governance and Risk Meeting: Ep. 61
Description
# Governance
- Richard Brown: MCD, The next few weeks for Governance
- LongForWisdom: Governance at a Glance
# Risk
- Cyrus Younesssi: MCD, The next few weeks for Risk
- Vishesh Choudry: State of the Peg
We'll then spend the rest of the time in a general Q&A session.
Discussion Threads: https://forum.makerdao.com/t/782
Podcast version: https://feeds.soundcloud.com/playlists/soundcloud:playlists:623643288/sounds.rss
Summaries: https://community-development.makerdao.com/governance/governance-and-risk-meetings/summaries
Transcripts (Delayed by 48 hours): https://community-development.makerdao.com/governance/governance-and-risk-meetings/transcripts
Foundation Proposal: https://community-development.makerdao.com/governance/core-principles
Governance and Risk Framework - Part 1: https://community-development.makerdao.com/governance/governance-risk-framework/part-one
Website: https://makerdao.com
Twitter: https://twitter.com/makerdao
Forum: https://forum.makerdao.com/
Chat: https://chat.makerdao.com/home
Email: info@makerdao.com
A
Hello,
everyone
welcome
to
the
November
14th
edition
of
the
scientific
governance
and
risk
meeting.
My
name
is
Richard
Brown
I
am
the
head
Community
Development
at
maker
down.
This
is
going
to
be
an
interesting
call.
I
say
that
every
single
time
and
I'm
right
every
single
time
so
I
can
continue
to
do
it.
So
the
reason
why
this
is
interesting
is
we
have
a
launch
date
coming
up.
I
want
to
talk
about
that.
A
bit
I
want
to
talk
about
what
the
future
looks
like
as
well.
A
Anybody
has
been
a
long
time
listener
and
slash
watcher
if
these
calls
know
how
reticent
I
am
to
predict
the
future,
but
predicting
the
future
two
weeks.
That
might
not
be
that
bad.
So
this
this
trend
restrict
are
from
proselytizing.
Is
that
the
right
word
for
to
see
what
world's
gonna
look
like
post
on
TV,
presumably
posted
on
TV,
assuming
that
the
community
around,
if
I
sent
that's
entirely
up
to.
B
A
Guys
and
then
we'll
have
a
chat
with
Cyrus
and
he'll.
Tell
us
a
bit
about
his
thoughts
about
what's
reversed
world
is
going
to
look
like
post
on
TV
and
then
we'll
open
it
up
for
a
Q&A
session,
because
we
have
a
lot
potentially
a
lot
of
questions
that
could
be
asked,
and
it's
within
this
call
today,
I'm
going
to
do
the
preambles.
We
are
enormous
ly
interested
in
hearing
from
everybody
in
the
call
we
need
feedback
from
the
community.
A
So
if
you
have
a
question
or
comments,
please
type
it
in
the
chat
or
if
you
have
access
to
a
microphone,
please
jump
right
in
and
ask
what
we
don't
mind
being
interrupted.
If
you
think
of
a
question,
basically,
like
the
timing,
is
moved
against,
you
never
fear
we
stick
around
for
a
half
hour
at
the
top
of
the
hour
for
just
a
generalized
Q&A.
A
A
A
This
is
going
to
be
an
interesting
about,
not
only
because
it
could
potentially
change
the
face
of
be
fine,
but
it's
also
very
interesting
because
we
have
a
new
mechanism.
That's
been
built
into
this
boat.
That
mechanism
is
delay
a
timer
delay
on
a
timer
time
delay.
Obviously,
it's
been
a
long
write
to
this
column.
You
have
to
forgive
me
if
I
might
speak
a
few
times.
A
This
is
great,
whispering
news
when
I
heard
about
this
feature
sent,
because
we
have
I've
touched
on
this
and
some
earlier
calls-
and
this
is
just
to
present
some
additional
context
about-
what's
actually
happening
with
this
launch.
It
is
it's
hard
for
me
to
understand
how
complicated
is
the
this
thing
that
we
have
going
on
here
maker
down
the
maker
protocol
is,
we
can
come
up
with
all
kinds
of
analogies,
but
it's
it's
basically,
the
API
that
drives
defy
and
anybody
who's
managed.
A
A
large
API
in
the
past
knows
what
upgrading
an
API
that
third-party
vendors
rely
on
can
look
like
it's
a
complicated
thing.
Not
only
do
we
have
this,
this
underlying
protocol
that
other
the
ecosystem
actors
rely
on.
We
also
have
a
series
of
properties,
websites
and
tooling
that
rely
on
that
protocol,
and
there
are
numerous,
some
would
say,
countless
is
very
wrong,
but
there's
a
lot
of
other
third
parties
out
there
from
from
analytics
tools
to
defy
insurance
tools.
It
just
goes
on
and
on
that
require
that
are
sorry
that
are
built
on
our
protocol.
A
So
coordinating
a
launch
like
that
is
fantastically
complicated
and
we
can't
have
a
launch
date.
That's
indeterminate!
That's
a
little
too
much
to
ask.
So
what
we
did
is
we
were
leveraging
this
tool
where
there's
a
delay
built
into
the
executive,
and
this
is
the
way
it's
going
to
work
tomorrow.
At
5:00
p.m.
the
executive
goes
out,
the
community
can
votes
should
that
vote
be
ratified
before
Monday,
the
18th
at
4:00
p.m.
UTC,
the
it
will
be
held
in
a
queue,
and
then
it
will
go
live
in
exactly
Monday
at
4:00
p.m.
A
UTC
which
this
this
will
give
maker
demo
or
to
make
a
foundation
the
DAO
protocol
and
the
wider
ecosystem,
an
opportunity
to
plan
ahead,
and
so
I.
We
can
get
ready
and
they
can
oppress
their
big
red
buttons.
At
the
same
time,
this
is
as
close
to
a
definitive
date
as
we're
capable
of
providing,
because
this
is
what
the
community
this
little
devil
looks
like
you
know.
You
can't
control
everything,
and
so
this
is
the
middle
ground
between
having
a
hard
launch
date
and
having
a
very
predictable
window
for
a
launch
thing.
C
A
A
What
I
want
to
do
is
encourage
the
community
and
I'm
going
to
post
all
these
thoughts
into
a
forum
thread
this
afternoon,
and
then
we
can.
We
can
iterate
together,
but
what
I
want
to
ask
the
community
to
do,
and
particularly
the
stakeholders
in
this
call,
is
to
help
the
foundation
evaluate
exactly
how
governance
works
and
what
this
is
going
to
be
a
great
opportunity
to
sort
of
after
the
dust
settles
to
have
a
look
at
our
tooling.
Have
a
look
at
well.
A
I
should
listen
to
this
one
through
a
bit
of
a
list
here,
so
post
MCD
we're
potentially
going
to
be
dealing
with
more
parameters
than
we
have
in
the
past.
Now,
let's
speak
specifically
to
what
those
parameters
are,
but
we're
coming
from
a
world
where
we
have
this
weak
least
stability,
febo
and
we've
sort
of
evolved
around
the
requirements
of
that
vote.
A
We're
moving
into
a
two
universe:
world
where
we
have
that's
size,
stability
for
you
potentially
have
a
thigh
stability
for
you
yet
give
us
I
was
great
of
destiny's,
know
the
Reston,
and
we
need
to
start
thinking
about
some
of
the
things
that
we've
internalized
and
making
sure
that
those
those
processes
were
pressed
in
the
future.
What
I
would
like
for
us
to
do
as
a
group
is
started,
asking
re
asking
some
of
these
long-standing
questions
or
bring
them
to
the
floor
again.
A
We
have
some
empirical
data
now
that
we
can
start
looking
back
towards
and
then
we
can
start
thinking
about
back
testing
or
creating
some
models,
but
we
can
start
really
digging
to
whether
weekly
cadence
is
work
for
us.
If
it
does,
let's
keep
on
doing
it
if
it
doesn't,
let's
figure
out
what
a
more
appropriate
cadence
looks
like,
let's
think
about
how
we
validate
the
decisions
that
we've
made
in
the
past-
and
this
is
I-
think
that
this
is
going
to
be
a
great
opportunity
for
us
to
be
able
to
look
all
right.
A
This
is
how
much
we
would
disability
fee
this
week.
This
is
the
impact
on
the
peg
the
week
after
start,
drawing
some
correlations,
I
think
between
the
activity
of
the
people
in
this
this
group
and
in
the
forums
and
in
the
portal
to
determine
whether
we
are
moving
the
needle
in
the
way
that
we
excite
it
sounds
like
a
bit
of
a
leading
question:
I'm
not
entirely.
Maybe
it's
just
me
I'm
not
entirely
sure.
A
We've
seen
AB
direct
cause
and
effect
when
we've
been
altering
these
parameters,
so
it'd
be
a
good
opportunity
to
do
some
some
analytics
and
trying
to
figure
out
whether
that's
wrong
or
whether
I'm
right.
We
can
start
thinking
about
how
whether
rate
stepping
worked
for
us
or
not.
We
can
start
thinking
about
whether
we
want
to
bring
some
of
the
contingency
plans
that
we
came
up
with
in
the
sad
world
into
the
dive
world
and
one
was
contingency
plans.
A
The
best
example
is
the
raising
of
the
debt
ceiling,
so
the
community
empowered
me
to
keep
an
eye
on
the
debt
ceiling
and
if
we
get
within
10
mil
of
the
current
ceiling,
then
the
ceiling
gets
raised.
The
language
in
that
poll
was
very
specific
to
having
it
only
be
valid.
Pre
MCD
we're
now
almost
in
a
post
into
you
will
be
world
if
everything
goes
according
to
plan.
So
we
need
to
think
about
whether
those
mechanisms
are
carried
over
or
whether
we
want
to
think
of
something
a
bit
more
targeted.
A
It
served
us
well,
I
believe
up
until
this
point,
but
I
think
that
we
need
to
make
sure
that
we're
not
being
constrained
by
the
tools
and
that
the
tools
provide
us
the
flexibility
to
do
it
is
this
community
wants
to
do
so.
There's
maybe
an
opportunity
for
us
to
start
thinking
clearer,
deeply
about
what
the
requirements
are
for
this
system.
What
are
the
requirements
for
polling?
A
How
does
continuous
approval
voting
work
in
a
world
where
we
have
a
lot
of
things
happening
at
once,
and
that's
that's
going
to
be
a
bit
of
a
wake
up,
call
I,
believe
I
will
make
this
prediction.
People
are
going
to
begin
to
evaluate
what
that
means,
and
we
need
to
figure
out
if
there's
adjustments
that
needs
to
be
made
there.
I
also
want
to.
A
A
How
do
we
explain
what
all
this
is
about
to
the
wider
world
in
a
way
that
reduces
on-boarding
friction
some
of
the
cognitive
dissonance
that
we've
seen
in
the
community,
where
every
once
in
a
while,
a
group
of
people
on
Twitter
will
discover
that
they
were
wrong
about
maker
to
maker
protocol,
the
entire
time
and
now
they're
mad.
So
we
need
to
start
hitting
that
stuff
off
at
the
past
and
how
do
we
provide
enough
information
to
the
wider
world
that
allows
stakeholders
to
get
engaged
with
the
system
and
cast
their
votes
in
a
way?
A
That's
has
less
friction
than
it
does
right
now,
which
leads
us
into
another
major
issue
that
we
should
start
dealing
with
head-on,
and
that
is
around
apathy,
and
so
this
is
a
very
interesting
conversation
that
I
want
us
to
devote
some
time
to
you,
because
this
is
a
text
centered
world
and
technologists
are
highly
critical,
basically
of
anything
that
moves
so
I
want
to
reevaluate
with
the.
What
we
Cathy
is,
how
we
measure
apathy,
whether
we
just
whether
we
actually
are
experiencing
voter
apathy,
and
if
we
are,
how
do
we
fix
it?
A
I,
don't
think
it's
just
cut
and
dry,
there's
yeah.
Why
she'd
know
you
know
what
I
got
spare
everybody
this
this
discussion
about,
how
we
measure
apathy,
but
we
should
do
it
very
soon.
Another
major
topic
that
we
should
discuss
in
earnest
very
soon
is
I
mean
this.
This
falls
back
into
that
bucket
of
recurring
thoughts
and
ideas
that
haven't
quite
taken
hold.
Yet,
though,
figuring
out
whether
incentivization
is
possible
in
this
world,
there's
been
piles
of
conversations
about
if
they
fall
into
generally
two
different
buckets.
A
The
first
bucket
is,
let's
just
straight-up,
pay
people
somehow
to
vote.
My
thoughts
on
that
are
not
ambiguous,
I
hope
and
then
there's
a
second
method
where
we
start
thinking
very
clearly
about
skin
in
the
game
and
what
that
actually
means
and
real
skin
in
the
game
means
that
you
lose
some
if
you
make
a
bad
decision.
So
that's
that's
another
discussion
we
can
have
so
do
we
actually
reward
and
or
punish
people
for
their
decision-making
capabilities
in
this
ecosystem.
A
That's
a
very
interesting
discussion
and
that
leads
to
the
third
sort
of
prong
of
apathy
discussions
delegation
is
its
delegation.
To
think.
Is
it
reasonable
to
implements?
Is
it
something
that
people
will
actually
engage
with?
If
so,
how
do
we
design
it?
Do
we
design
it
some
of
just
somebody
else
design
it,
and
do
we
encourage
people
to
organize,
or
should
they
self
organizers,
there's
a
lot
of
really
interesting
questions,
so
I
am
NOT
going
to
keep
on
throwing
questions
out.
There
I
think
that
what
I
would
like
to
do
is
get
MCD.
A
It's
a
main
that
deal
with
to
discover
this
new
world
that
we
live
in,
or
this
double
universe
world
that
we
live
in
next
week
and
as
we
do
that,
let's
be
very
cognizant
to
vote,
keeping
track
of
what
works
and
what
doesn't
work
and
not
just
grandfathering
in
system
that
might
not
be
the
one
that
we
need
and
figure
out
how
to
improve
governance
going
forward.
I.
A
Think,
oh,
you
know
there's
another
thing
that
I
want
to
touch
on,
so
we're
gonna
own.
This
vaguely
sauernotes
there's
been
a
lot
of
really
really
good
initiatives
coming
out
of
the
community
and
it's
fantastically
exciting
to
see
these
things
happen.
Where
groups
of
people
in
this
call
forums
and
chats
have
determined
that
there's
a
hole
in
the
ecosystem
that
they
can
fill
and
self-organizing
and
then
interacting
with
the
foundation
in
order
to
solve
those
problems,
and
that
is
a
model
that
makes
me
super
happy
because
that's
what
this
is
all
about
its
own.
A
The
interim
roles
that
maker
has
presented
to
the
community
and
is
currently
funding
our
breach
trafficking
of
prototypes
of
rules
that
we're
hoping
the
community
will
eventually
take
over
for
us,
and
we
can
continue
that
model
with
improving
governance
in
general.
So
I
have
to
put
this
forum
thread
in
the
side
of
the
forums.
Can
we
have
all
these
thoughts,
a
discussion
going
and
maybe
get
a
working
group
from
the
community
together
and
then
we
can
build
some
direct
bridges
between
the
community
and
the.
A
A
Chant
when
it
matters
people
of
votes,
yeah,
that's
another
great
conversation
that
we've
had
a
couple
of
times.
It's
it's
my
everybody.
In
the
absence
of
empirical
data,
it's
my
belief
that
people
have
been
voting
that
often
because
voting
is
born
and
planks
basis
points.
I'm,
a
stability
is
dull
as
dirt.
A
A
A
So
you're
saying
that's
you
lead
by
an
example.
I
think
that
we
could
probably
just
reverse
engineer
that
off
of
the
contract
kitchen-
oh,
that
makes
this
a
little
risky.
So
this
this
is
where
it's
an
interesting
conversation
picked
out
for
delegation
is:
are
we
going
to
as
a
community?
Are
we
going
to
be
moving
that
much
farther
ahead
and
t-bone
I'm,
not
calling
you
X
I've,
seen
your
chats
and
I
know
you're
super
super
serious,
but
assume,
let's
assume
for
a
second,
you
weren't.
A
The
intention
was:
do
we
just
copy
somebody
else
and
just
hope
that
they're
thinking
about
stuff
and
that
that's
one
of
the
risks
with
with
delegation,
and
that's
one
of
the
reasons
why
it's
an
experiment
and
experiments
make
me
nervous
with
governments
in
this?
Why
I
don't
believe
that
we
need
to
bake
this
into
the
core
protocol?
A
That
I
am
very
very
encouraged
about
trying
to
encourage
the
third
party
ecosystem
to
build
on
this,
and
we
have
grants
programs
and
funding
available
for
teams
that
are
willing
to
do
delegation
proof
of
concepts
and
then
experiment
in
their
little
sand
boxes.
And
then
we
can
learn
from
that,
but
I
don't
think
messing
with
our
core
protocol
anytime
soon,
all.
A
D
D
There
are
actual
methods
to
calculating
debt
ceilings
and
stability,
fees
and
all
sorts
of
things,
and
we
will
be
getting
into
that
just
starting
the
next
few
weeks,
maybe
even
as
early
as
next
week,
depending
on
how
launching
the
migration
goes,
we
have
a
lot
of
cool
stuff
coming
down
the
pipeline.
A
lot
of
interesting
research
to
share
I
think
it'll.
D
It
not
only
is
it
kind
of
difficult
and
more
overhead
to
govern
in
this
new
style,
but
I
think
it'll
take
a
while
to
even
just
be
absorbed
by
the
community,
so
I
think
we're
we're
just
gonna.
Try
to
release
some
materials
on
a
weekly
basis
and
just
try
to
engage
the
community
and
open
things
up
for
discussion,
certainly
not
forcing
anything
down
on
governance
but
yeah.
That
should
be
interesting
to
explore,
and
one
of
the
consequences,
though,
is:
how
do
we
fit
all
this
stuff
in?
Do
we
need
a
separate
governance?
D
Probably
aligned
voters:
how
do
we
keep
track
of
all
the
different
parameters?
These
are
all
pretty
much
unsolved
problems
at
this
point
so
we'll
see,
but
for
today
have
a
super
short
presentation
on
the
remaining
system.
Parameters
that
were
part
of
the
initial
launch
parameter
set
so
I'm
going
to
go
through
those
hopefully
shouldn't.
Take
too
long.
D
D
D
This
can
be
useful
if
you
want
to
increase
every
collateral
stability
fee
up
or
down.
The
use
case
might
be
to
offset
that
cost
of
the
DSR.
So
if
you
increase
the
DSR
by
1%-
and
you
wanted
to
make
an
equivalent
offsetting
adjustment
to
the
collateral
types
you
wouldn't
have
to
go,
you
wouldn't
have
to
go
in
and
do
that
on
a
case-by-case
basis.
You
could
just
change
the
DSR
and
the
global
stability
fee
for
launch.
D
D
The
global
debt
ceiling
similar
concept,
so
each
individual
collateral
has
a
debt
ceiling,
but
there's
also
a
global
global
cap
for
the
protocol.
So
if
you
have
three
individual
collateral
types
with
50
million
each,
but
the
global
is
set
to
a
hundred,
then
I
you
can't
have,
then
certain
collateral
types
might
be
disallowed
because
the
global
ceiling
has
been
hit.
So
this
might
be
interesting
in
the
case
where
you're
the
the
community
is
indifferent
between
two
collateral
types.
D
They
just
don't
want
to
have
both
in
there,
so
you
may
be
able
to
tinker
with
these
numbers
a
little
bit
similar
to
the
stability
I
think
this
one
may
be
a
little
bit
confusing
if
you
have
too
many
different
collaterals.
So
this
is
another
one
that
we're
we
should
kind
of
approach
slowly
for
now,
though,
we're
just
the
global
debt
ceiling
is
just
the
sum
of
the
individual
collateral
types.
D
The
lot
size,
I
believe
for
those
auctions
was
ten
thousand,
which
means
that
every
ten
thousand
accrual
of
stability
piece
there
is
a
surplus
auction,
but
you
can
set
a
buffer
in
there,
where
the
auctions
only
trigger
above
that
buffer,
so
community
might
can
say
we
want.
We
want
to
have
a
certain
amount
of
reserves
for
the
system,
and
this
could
be
for
various
reasons.
One
is
the
stability
fee,
funding
that
gets
talked
about
from
time
to
time.
D
There's
there's
a
few
there's
a
particular
game
theory
issue
when
it
comes
to
emergency
shutdown
and
what
happens
to
this
buffer.
That
is
a
little
bit
confusing,
and
so
it's
not
entirely
clear,
at
least
to
me
that
it's
it's
safe
to
have
this
number
be
enormously
high,
but
I
think
having
it
set
to
a
couple
hundred
thousand,
where
you
can
absorb.
Maybe
smaller
losses
and.
D
There
there
there
is
a
buffer,
it's
yeah,
it's
kind
of
almost
like
a
discretionary
account
for
em
care
holders,
but
I
think
there's
a
lot
of
discussion
and
work
that
needs
to
be
had
before
it
starts
getting
used
as
such,
because
it
is
multi-purpose
and
there
are
some
risk
considerations-
are
we
will
and
and
of
course
the
other
issue
damages
so
yeah.
We
can
get
more
into
that
later.
D
D
Mm
does
immediately
get
sent
to
that
account
for
where
bad
debt
gets
collected
for
em
care
inflation
and
then
host
auction.
Once
the
die
is
collected,
it
would
then
get
sent
to
that
same
bucket
and
erase
the
and
and
wipe
out
the
debt
that
was
just
put
there.
So
there's
an
issue
where,
if
the
died
debt
is
higher
than
the
threshold
for
a
debt
auction,
then
it
would
trigger
immediately
before
the
auction
has
even
had
a
chance
to
complete.
D
So
we
dropped
this
by
a
day
to
two
days.
Additionally,
with
extremely
large
CDP's,
they
could
take
a
few
day.
They
may
take
a
day
or
more
to
to
completely
finish
their
auctions.
Last
week
we
discussed
that
large
CDP's
are
broken
down
into
smaller
chunks,
and
only
a
certain
number
of
these
auctions
can
occur
per
block,
so
yeah
just
write
considerations
going
into
the
debt
auction
delay.
I
think
we
went
with
two
days.
D
D
This
is
set
to
24
watch,
which
may
be
a
little
bit
annoying
or
confusing.
So
it
will
not
be
possible
to
draw
less
than
20
dye
from
from
a
vault
at
launch
the
reason
here
being
a
potential
spam
atack,
where
someone
can
kind
of
create
a
very
large
number
of
CDP's
with
extremely
low
dye
dye
amounts,
and
they
would
become
too
expensive
to
write
and
sent
an
auction
and
and
go
through
that
whole
process.
And
it
would
result
in
the
system
effectively
being
under
collateralized.
D
Number
certainly
can
be
lower
a
lot
lower
than
20
and
it
will
be
I
would
say
just
in
a
few
weeks
after
launch,
we
should.
We
should
probably
reduce
this
to
something
more
reasonable.
This
is
just
kind
of
like
an
extra
safe
buffer
at
launch
to
kind
of
reduce
the
scope
of
things
that
I
need
to
be
monitor.
D
D
E
C
D
Okay,
just
a
few
more
so
governance
security
module
is
the
delay
between
governance,
approving
a
approving
the
vote
and
when
the
changes
are
actually
applied.
Our
head
of
engineering
louder
spoke
about
this
at
length
I
think
two
weeks
ago.
So,
if
you're
interested
more
in
about
the
details
here
and
check
out
that
that
recording
this
is
a
really
useful
parameter
for
preventing
against
governance
attacks.
D
But
this
is
actually
worse
for
technical
risks,
anything
that
had
any
sort
of
protocol,
bug
or
error
that
hopefully
doesn't
happen.
But
if
anything
were
to
happen
that
needs
to
be
fixed
in
the
next
few
weeks.
This
security
module
would
only
hamper
those
efforts,
so
the
delay
will
be
set
should
be
set
to
zero
at
launch
and
then
kind
of
once
we're
out
of
the
weeds
we
can.
We
can
start
to
increase
this
to
you
know.
This
could
be
something
like
a
few
hours
and
then
I
think,
as
the
system
gets
more
and
more
ossified.
D
Emergency
shutdown
delay
is
the
amount
of
time
after
emergency
shutdown
is
called
from,
after
which
you
can
actually
redeem
your
third
for
collateral,
I.
Think
the
only
consideration
here
is
that
we
make
sure
that
all
the
auctions
that
are
in
process
have
ended.
So
that's
why
it's
set
to
three
days
plus
an
hour
there's
the
max
auction
duration
is
three
days.
D
Mkr
I
think
this
number
is
probably
going
to
I
think
it
should
start
a
lot
of
discussion
and
it's
something
that
we
should
reflect
on
frequently
and
from
time
to
time,
because
I
think
this
number
would
necessarily
need
to
change
over
time,
and
it
is
an
important
parameter.
I
think
intuitively
people's
first
reaction
is
this
should
set.
This
should
be
set
extremely
high,
such
that
no
attacker,
no
one
can
kind
of
arbitrarily
trigger
emergency
shutdown.
Right
50,000
seems
may
seem
a
bit
low
at
first,
it's
only
5%
of
the
supply.
D
The
the
argument.
The
counter-argument,
though,
is
that
this
number
isn't
designed
to
make
it
difficult
to
achieve.
It's
actually
designed
to
be
small
enough
where,
if
we
need
to
trigger
it
for
any
reason,
it
can
be
fairly
easily
attained.
I
think
there's
certainly
a
lot
of
confusion
over
the
mechanics
of
emergency
shutdown
and
I.
Don't
blame
them.
D
D
Given
that
governance
can
kind
of
come
to
agreement
that
this
was
like
a
good
use
of
emergency
shutdown
and
wasn't
some
sort
of
speculative
attack.
So
anyone
trying
to
borrow
MKR
or
somehow
trying
to
maliciously
trigger
emergency
shutdown,
because
the
threshold
may
seem
arbitrarily
low
I
would
get
extremely
penalized.
D
No,
they
can't-
or
in
fact
this
this
threshold
just
builds
up,
so
you
yeah,
so
this
is
kind
of
like
a
running
running
tally.
Anyone
can
start
to
trigger
this
whenever
and
then
just
whenever
the
threshold
is
reached,
it
can
be
called,
but
as
far
as
I
know,
there's
no
way
to
kind
of
counter
over
time.
I'm.
E
Yeah
yes,
similarly
in
that
vein,
if
somebody
accidentally
deposits
like
a
bit
of
MKR
they're
like
over
time,
maybe
over,
like
let's
say
over
ten
years,
emergency
shutdown
never
actually
happens
and
like
just
accidentally,
you
see
like
slowly
but
surely
approaching.
We
could
always
raise
the
threshold
right
to
counteract
that,
instead
of
actually
like
removing
it
or
anything
like
that,
so
we
could
change
the
that
threshold
yeah.
E
So,
just
for
a
couple
of
points
of
clarification
there,
this
is
actually
not
a
tunable
parameter
in
the
sense
that
we
can
change
it
in
the
emergency
shutdown
module,
it
has
to
be
deauthorized
and
then
a
new
module
gets
launched
with
a
different
threshold,
which
is
the
way
that
two
chasers
question
governance
could
come
back.
The
accumulation
of
NPR
in
the
emergency
shutdown
module
is
by
D
authorizing
the
emergency
shutdown
module
before
it
gets
fired.
E
D
B
Know,
I
I
didn't
have
any
slides
ready
and
you
already
have
everything
you
know
here:
yeah
if
you
look
at
Cyrus's
a
beautiful,
slides,
there's
kind
of
two
parameters
that
we
have
to
focus
on
for
for
launch
better
that
irrelevant
here,
the
first
being
the
Oracle
30
module
delay.
So
for
those
of
you
who
don't
know
we,
the
Oracle
kind
of
component
has
two
small
contracts.
B
This
is
kind
of
as
a
defense
mechanism
in
case
that
there
is
an
Oracle
attack
right,
so
it
gives
us
the
delay
period
to
react
to
you
know.
Perhaps
you
know,
freeze
that
particular
OSM
to
make
sure
that
the
the
malicious
price
doesn't
affect
the
system.
At
the
same
time,
right,
the
the
Oracle
security
module
delay
is
kind
of
latency
that
we're
adding
to
the
sensitivity
of
the
the
price
right.
B
The
next
thing
is
the
Oracle
feed
quorum
right.
So
how
many
feeds
do
you
need
to
come
and
come
to
agreement
come
to
consensus
on
on
what
the
prices
so
there's
a
couple
things
to
play
here
right?
So
if,
if
you
have
a
quorum
right,
which
is
some
subset
of
the
total
number
of
feeds,
then
the
median
value
of
that
quorum
is
the
price
that
is
going
to
be
used
by
by
the
median
Iser
right.
So
if
you
can
effectively.
B
Manipulate
that
median
value
in
that
quorum
set,
then
you
can
launch
write
a
price
attack
right.
So,
let's
start
putting
real
numbers
to
this
thing,
so
we
have
a
total
of
20
feet
and
multi
collateral
by
it's.
The
15
original
feeds
that
we
had
in
Oracle's
B
1
in
single
collateral
by
plus
the
5
defy
projects
that.
A
B
B
Now,
on
the
flip
side,
there's
also
something
called
a
censorship
attack
in
this
case,
kind
of
an
Oracle
is
unable
to
reach
quorum.
So
if
you
take
the
total
number
of
feeds,
20
minus
the
quorum
13,
that
means
it
costs
7
feeds
to
launch
a
censorship
attack
and
effectively
what
that
means
is
that
whatever
price
the
the
Oracle
is
currently
reporting,
that's
kind
of
the
price
the
system
is
stuck
on
now
someone
may
be
asking
you
know:
why
is
the
cost
to
launch
a
censorship
attacking
a
lower?
B
B
So
let's
say
that
even
one
feed
kind
of
goes
offline
right.
So
now,
instead
of
twenty
feeds,
you
only
really
have
19
feet
submitting
crisis.
All
of
a
sudden.
The
censorship
attack
drops
requirement
drops
by
one
right,
because
now
you
don't
need
eight
malicious
feeds
to
stop
posting
prices
anymore.
You
only
need
seven
right
and
and
so
on
and
so
forth
right,
and
so
until
we
see
and
have
some
empirical
data
that
the
the
feeds
are
rock-solid
stable.
B
B
B
A
Nick
we
had
a
question
for
you:
I'm
gonna,
read
it
out
with
the
volatility
of
East
and
the
future
MCD
approved
assets
like
bats.
What's
the
response
to
the
argument
that
the
one
our
Oracle
security
delay
lag
time
may
not
be
appropriate
comparable
to
single
collateral
Oracle
is
out
getting
every
time.
The
price
of
these
fluctuates
I.
B
Mean
there's,
there's
different
risks
here
right,
so
there's
two
things:
one
is
I
know
what
the
current
value
of
the
OSM
is
and
I
know
what
the
cued
up
value
of
the
OSM
is
right,
so
people
can
actually
see
that
they're
going
to
be
liquidated
in
one
hour
right
and
then
close
their
positions
right.
So
in
that
way,
it's
actually
a
risk.
Mitigating
type
of
deal
right
instead
of
adding
risk
right,
because
we
actually
have
less
liquidations
right.
B
B
Adding
an
extra
hour
really
should
not
have
that
they
give
an
impact,
given
that
the
risk
parameter
should
already
have
a
healthy
buffer
right
in
terms
of
collateralization
ratio.
So
essentially
the
Oracles
security
module
delay
has
already
been
kind
of
priced
well,
not
priced
in,
but
baked
into
the
risk
parameters
that
that
we
assign.
So
it's
not
that
the
delay
is
kind
of
screwing
everything
up
right.
B
D
Sure
and
I
think
there
is
a
there's,
a
question
from
Matthew
about
the
threshold
being
a
percentage
of
outstanding
em
care,
I'm,
not
sure
what
the
what
the
right,
what
the
right
metric
is
going
forward.
I
do
know
that,
if
it's,
if
we're
trying
to
keep
it
small
enough
so
that
it
can
be
reached
when
needed
as
the
MKR
ecosystem
grows
and
the
supply
becomes
less
concentrated,
fifty
thousand
becomes
a
harder
and
harder
number
to
reach
the
more
distributed.
D
The
token
is
so
that's
one
thing
to
keep
in
that's
kind
of
the
consideration
that
I
have
in
my
mind
about
how
that
number
should
evolve
over
time.
We
don't
want
to
end
up
in
a
situation
where-
and
you
know
for
in
the
future.
Everything's
really
successful
are
so
successful
that
it
becomes
almost
impossible
to
raise
5%
of
the
of
the
em
care
supply
in
the
time
needed.
D
A
All
right
we're
approaching
the
top
of
the
hour,
so
I'm
going
to
hand
it
off
to
our
two
special
regular
guests,
long
for
wisdom
to
talk
about
government's
at-a-glance
and
then
we'll
hear
about
the
state
of
the
peg
from
the
shesh
we're
also
gonna.
Hang
will
be
hanging
around
for
another
30
minutes.
So
if
you
asked
a
question,
and/or
thought
of
a
question
and
get
a
chance
to
ask
it,
please
hang
around
and
we'll
have
a
generalized
Q&A
right
after
these
next
two
segments,
bunk
or
elbows
long
one
for
wisdom.
E
Yeah,
okay,
so
speaking,
the
forum
we've
had
sort
of
some
scattered
discussion
over
a
couple
of
threads,
but
a
couple
sort
of
stick
outs,
and
so
we
have
an
official
welcome
thread
now
and
after
make
a
man
joins.
So
if
anybody
wants
to
like
introduce
or
talk
about
themselves,
if
you've
been
here
for
a
while,
there's
a
good
place
to
do
it.
E
D
Jump
in
for
a
quick
second
yeah,
so
after
last
week's
call
posted
kind
of
a
written
version
of
the
auction
parameters
and
there
will
be
a
written
version
of
today's
system
parameters
as
well
today
or
tomorrow,
already
gotten
some
good
feedback
on
the
auction
parameters,
challenging
a
lot
of
assumptions
and
I
expect
this
same
for
these
I.
So
I
just
wanted
to
reiterate
that
these
are
obviously
just
early,
safe
defaults
and
as
more
data
becomes
available,
we
will
certainly
be
fine-tuning
all
these
parameters.
F
F
E
There's
no
discussions
see
what
else
oh
yeah,
so
as
contra
pronouncement
working
on
a
new
forum
theme
am
I
supposed
to
thread
in
the
forum
showing
awesome,
screenshots
and
looking
for
feedback.
So
if
anybody
cares
deeply
about
what
the
form
looks
like,
they
can
go
in.
That's
that
throat
and
talk
to
him.
F
Yeah
we're
doing
two
responding
stuff.
Okay,
so
just
gonna
touch
on
the
debt
ceiling
was
raised.
Post
raise
of
the
debt
ceiling
nervous,
a
few
more
drawers
bringing
us
up
over
to
about
a
hundred
and
two
million,
which
is
good
that
there
wasn't
just
a
sudden
explosion
in
supply
post
raise
because
then
that
really
would
have
been
evident
of
a
lot
of
hoarding
behavior
and
potentially
would
have
damaged
the
bank.
F
Overall,
the
peg
for
the
last
seven
days
has
been
pretty
smack
dab
on
a
dollar
with
relatively
small
trading
volumes
and
then
for
the
last
24
hours
very
similar,
but
again
small
trading
volumes,
so
not
much
dyes
moving
around
on
the
decentralized
exchanges,
despite
the
fact
that
there
was
a
pretty
significant
increase
in
supply
not
even
two
weeks
ago.
So
that's
gonna,
be
something
interesting
to
watch
is
what
happens
to
that
dye?
It
seems
people
are
still
just
holding
on
to
it.
F
So,
looking
at
the
volumes,
the
compound
borrow
volume
has
come
down
a
bit
more
since
the
five
percent
drop
and
the
supply
volumes
have
also
come
down
a
little
bit
since
that
drop.
So
what's
interesting
is,
though,
the
total
supply
has
come
up
quite
a
bit.
There
is
also
been
some
refinancing
of
those
positions
net
over
from
secondary
lending
platforms
back
over
to
maker
additionally
like
in
outside
of
just
the
amount
of
refinancing.
F
So
it'll
be
interesting
to
watch
because,
as
maker
migration
shakes
out
the
rate
differential
between
the
stability
fee,
DSR
and
borrowin
supply
rates,
some
secretary
lending
platforms
should
theoretically
be
pretty
impactful
in
terms
of
how
quickly
that
migration
occurs.
So
watching
these
rates
I
think
will
continue
to
be
an
important
data.
Point
I'll
pause
for
any
questions.
F
A
A
A
We
closed
lost
all
right,
so
there's
an
interesting
question
here.
That's
I
like
hard
question,
so
Chris
suggests
that
a
big
part
of
the
apathy
problem
is
around
which
decisions
the
foundation
makes
on
its
own,
versus
which
decisions
are
finally
put
to
a
vote
and
that's
a
great
conversation.
We
should
have
it
and
I
shot.
My
pilot
questions
at
the
top
of
the
call
that
we
need
to
address
as
an
eco
or
as
a
community
I
should
get
to
Mandy
asking
as
protocol
shepherds.
A
Maintainer
is
whatever
we're,
gonna
call
ourselves
which,
which
leavers
get
exposed
to
the
community
and
went
and
in
what
order,
and
how
they're
managed
is
basically
essentially
what
this
is
all
about.
That's
what
gradual
the
centralization
is.
It's
baked
into
our
DNA.
It's
in
our
foundation
principles.
A
That's
the
thing
that
opens
the
floodgates
to
more
leavers
or
an
understanding
that
yes,
this
is
where
I
get
philosophical
or
anything
the
concept
of
leavers,
because
it
could
be
that
we
find
out
that
exposing
50
different
parameters
to
the
ecosystem
and
asking
them
to
work
that
the
board
might
be
unrealistic
and
maybe
just
the
community
doesn't
want
to.
So
who
knows
what
the
future
is?
Gonna
hold.
It's
a
it's
a
great
discussion,
though,
and
I'm
looking
forward
to
having.
E
Now
we're
having
a
conversation
about
the
Oracle
price
delay
and
if
that
has
any
kind
of
risky
scenarios
or
attack
vectors,
and
only
real
one
is
as
Cyrus
and
I
both
pointed
out
the
idea.
If
the
collateral
is
crashing,
you
know-
and
you
have
that
hour-
delay
where
you
could
still
draw
die
based
on
like
the
higher.
You
know,
the
the
could
just
draw
more
died
before
it
actually
crashes
right,
you're
in
sense
of
us
to
draw
that
die.
A
B
D
That's
just
a
trade-off
of
having
the
security
of
the
OSM
and
really
the
only
solution.
There
is
to
keep
a
tight
debt
ceiling
on
the
collateral
type
and
then
basically
increase
it
as
needed,
using
instant
access
modules,
for
example,
if
there's
50
million
of
empty
buffer
in
the
debt
ceiling,
then
that's
just
unnecessary
risk
to
the
protocol
from
from
the
Oracle
security
module.
That's.
A
Interesting
it
Considered.
That's
I,
have
a
fairly
naive
perspective
on
what
the
debt
ceilings
for
I
think
it's
mitigating
liquidity
risk,
but
I
suppose
that
this
it's
potentially
another
reason
why
empowering
the
government's
facilitator
to
just
simply
bump
the
debt
ceiling.
Every
time
we
get
close
might
not
be
a
great
idea,
or
at
least
not
as
sophisticated
as
it
could
be
well.
D
No,
this
is
this
is
more
like.
If,
if
we
know
we
want
our
total
max
debt
ceiling
to
be
say,
200
million
for
each,
we
wouldn't
just
hard
code
that
into
the
protocol
today,
because
we
wouldn't
be
anywhere
near
there
and
then,
if
eath
were
to
crash,
then
users
would
draw
up
to
the
maximum
200
million
died,
which
is
completely
unnecessary
risk
for
the
protocol.
It.
E
F
Wouldn't
those
use
cases
only
like
make
sense
when
you
know
that
the
price
is
crashing
by
more
than
whatever
your
buffer
is
above
100%
collateralization
right.
So
if
you
have
CDP's
that
are
two
hundred
percent
collateralized,
it
would
only
make
sense
to
engage
in
that
kind
of
behavior.
If
you
know
that
eath
is
going
to
drop
more
than
50
percent
in
price
breaks,.
D
A
A
D
A
E
A
How
do
we
accelerate
the
pace
of
governance
without
increasing
the
overhead,
maintaining
our
Jillybean
velocity,
but
also
do
we
move
towards
this
glorious
future
where
machines
are
running
the
world
for
us
that
this
is
sort
of
like
this
overlap?
We're
setting
up
a
sandbox
assumably
created
by
some
some
models
and
some
empirical
data,
which
allows
rapid
reactions
that
shouldn't
there
at
least
roped
off
and
then
it's
an
interesting
discussion
have
we
talked
at
all
Sarge?
Are
you
aware
of
what
timelines
look
like
that
when.
D
Any
engineers
on
the
call
have
any
insight
into
that
I
mean
it
was
I
know
it
was
put
on
hold
for
launch
I
mean
it
wasn't
in
scope
for
launch.
But
if
I
remember
correctly,
it
was
pretty
much
done
it
just
wasn't
kind
of
wasn't
finalized
tested
all
that
stuff.
So,
oh,
please,
presumably
very
soon
right,
I
think,
especially
because
for
the
DSR
right,
yeah
I
think
it's
a
high
priority
for
DSR
changes,
so
I
I
don't
expect
it
to
be
long
before
we
have
these
modules.
D
A
D
A
I'm,
hoping
that's,
will
have
a
bit
of
a
cooling-off
period
as
far
as
releases
go
post,
launch,
there's
gonna
be
I,
didn't
like
the
thing
I
touched
on
this
as
well,
but
we
have
to
wrap
our
heads
around
what
dual
universe
looks
like
and
what
the
requirements
are
on
us
and
the
ecosystem
and
we're
largely
driven
by
external
forces.
We
need
to
start
allow
for
some
reactivity
to
whatever
NCD
is
a.
B
C
C
One
thing
that
we
should
consider
is
when
we
make
changes
how
many
of
those
spills
we
really
want
to
like
bundle
into
one
giant
lump
we've
had
this
sort
of
discussion
casually
but
I
think
there's
you
know
possibly
governance
risk
if
we
split
them
all
out
and
at
the
same
time
you
know
might
change
the
cadence.
We
should
kind
of
really
start
to
flesh
out
what
it
is
we
want
to
do
with.
A
C
A
We're
running
it,
so
that's
one
of
the
things
that
we
need
to
I
think
this.
This
potentially
could
be
one
of
the
primary
flash
points,
I
guess
I
about
when
we
talk
about
what
governance
is
and
how
it
works.
Post-Launch
is
that
we've
been
lucky
enough
to
deal
with
a
single
parameter
world
for
a
long
time
and
the
system
that
we
have.
That
works
very
well
with
a
single
parameter
world,
but
people
are
just
there.
There
are
gonna
start
developing
mental
models
that
don't
mesh
up
with
continuous
approval
voting.
So
hey.
A
B
C
And
then,
obviously
we
don't
want
to
split
governance
on
a
bunch
of
different
votes
in
our
current
system
right
now,
because
that'll
put
a
week
and
what
we
need
to
look
to
get
the
Hat.
So
we
always
want
to
make
sure
whatever
we're
doing
with.
Regarding
the
Hat
really
well
so
yeah,
it's
gonna
require
some
thought.
Yeah.
A
And
which
comes
back
to
why
I'm
so
enormous
they
hesitant
to
predict
the
future
and/or
build
for
a
world
that
we
haven't
seen
yet.
So
that's
that's.
Why
I
really
want
to
take
this
opportunity,
post,
launch
to
audits
and
evaluate
all
the
processes
and
tooling
that
we
have
as
we
go
and
then
keep
the
secondary
track
of?
How
do
we
improve
this
then
and
build
up
the
initial?
The
additional.
A
A
C
A
Sure
the
I
hope
parameters
of
the
votes
are
included
in
the
two
holes
that
are
in
the
voting
portal
right
now.
There's
the
system
parameter
poll
and
then
there's
the
amendments,
Kol
idealist
parameters
where
and
what
one
was
incorrect
and
one
need
to
be
modified.
So
taking
those
two
things
taken
in
conjunction
were
represents
the
end
user
adjustable
configuration
parameters
of
the
software
package.
We
call
the
protocol,
so
that's
what
people
are
gonna
be
voting
on
how
much
maker
requires.
A
This
is
continuous
approval
vote,
like
any
other,
so
it
just
needs
to
exceed
whatever
the
current
hat
is,
and
if
somebody
has
a
portal
open
before
I
get
there,
they
can
feel
free
to
chime,
in
so
I
mean
so
that's
to
beat
60
okay
maker
in
order
to
ratify
but
I'm
sure,
okay
I
will
be
moving
from.
That's
never
done
exactly
to
the
new
170
memberships.
So.
A
E
Also
the
caveat
that
I'm
kr
moving
from
the
from
the
current
standing
proposal
as
it
moves
out
of
that
proposal,
the
threshold
for
what
a
new
proposal
needs
to
pass
as
lawyers.
So,
even
though
it's
60
km
care
right
now.
If,
if
on
the
vote
tomorrow,
20,000
mkr
moves,
you
only
need
40,000
m
care
to
actually
be
able
to
pass
a
new
proposal.
So
it's
kind
of
like
a
sliding
scale,
as
one
proposal
surpasses
the
other.
A
A
No
I'm
not
aware
of
any
specific
initiatives
like
you
so
you're,
presupposing
that
the
reluctance
here
is
technical
or
custodial.
What?
Because
that
my
impression,
the
the
main
challenges
of
getting
BTC
as
a
collateral
type,
is
that
nobody's
figured
out
completely
decentralized,
custodial
solution
and
the
custodian
once
they
do
have
are
complicated
and
essentially
have
some
have
some
technical
flaws.
So
I'm
not
an
expert
on.
So
somebody
once
wanted
to
speak
to
a
WPT
CEO.
C
I
I
don't
know
as
much
about
wbg
C's
mechanism,
but
I
know
that
there's
the
significant
amount
of
a
sort
of
counterparty
risk
or
centralization
and
their
approach.
There
is
another
approach
right
now,
I
think
it's
T
V
T
C,
which
has
sort
of
figured
out
the
decentralized
bridge
model,
but
they
require
like
a
hundred
and
fifty
percent
collateralization
on
the
etherium
network,
which
in
a
way
means
that
there's
collusion
on
the
Bitcoin
side
of
the
network
that,
whatever
we
have
you
know
as
collateral,
would
really
just
reduce
down
to
more
etherium.
C
So
yeah
I
mean
I
think
both
both
approaches,
that
I'm
aware
of
and
maybe
there's
more
approaches
out
there,
but
both
would
require
some
kind
of
insurance,
because
you
would
really
want
to
make
sure
that
whatever
we
had
as
collateral
was
a
sort
of
decentralized
guarantee
on
actual
Bitcoin.
At
the
end
of
the
day,
yeah.
A
So
it's
a
whole
world
of
complexity
there
that
I,
don't
know,
has
been
proven
yet
it'd,
be
an
interesting
question
of
all
Nexus
I.
Think
there's
a
couple
other
major
players
now
that
are
doing
the
same
thing
I've
been
surprised
that
it
hasn't
been
more
widely
adopted,
or
maybe
I.
Maybe
it
has
been
I
just
haven't
been
watching
very
carefully,
but
it
seems
like
insuring
some
of
these
smaller
projects.
Contracts
would
be
a
compelling
option
for
a
lot
of
people,
but
I
don't
have
a
lot
of
insight
to
them
over
how.
E
C
D
D
D
D
It's
just
much
less
important,
but
if
you
have
other
illiquid
assets
less
frequent
bids,
maybe
it
takes
a
couple
days
for
a
team
of
sophisticated
investors
even
price
an
asset
or
come
to
consensus
on
what
they
want
to
submit
as
a
competitive
bid,
certainly
having
a
few
days
time,
but
having
more
than
a
day
to
do.
That
is
useful.
I,
just
think
it's
much
less
relevant
or
uh-huh
and.
E
Yeah
cuz
I
was
yeah.
Cuz
I
was
trying
to
understand
that
I
thought
that
basically,
the
one
thing
is
is
perfect,
like
preventing,
basically
profit
profiting
out
of
auctions
from
things
like
network
congestion
attacks.
That
would
like
prevent
people
from
bidding
at
all,
or
something
like
that.
Yeah
I
was
just
trying
to
wrap
my
head
around
it
thanks.
D
A
Okay,
I
think
we're
gonna
wrap
it
up
here.
Big
things
are
afoot.
We
have
a
significant
shift
in
the
ecosystems
appearing
tomorrow,
essentially
assuming
that
people
come
out
to
vote.
So
please
do
come
out
to
vote.
It's
gonna
be
a
big
one.
You
can
tell
your
grandchildren
and
throne
to
show
them
your
Easter
stand
that
you
were
there
heard
that
voted
in
Antigua,
so
get
your
keys.
Have
a
keep
your
eyes
out
for
that
executive.
Join
us
in
the
forum
share
your
thoughts.
Help
us
focus,
build
up
this
down.