►
From YouTube: Governance and Risk Meeting: Ep. 91
Description
# Agenda
## Governance Segment
- Richard Brown: The road from intention to implementation
- LongForWisdom: Governance at a Glance
## Community
- Maker_Man: Black Thursday Compensation Analysis presentation and discussion.
## Risk Segment
- Vishesh Choudry: State of the Peg
## General Q&A
We'll open the floor for any questions about Scientific Governance and Risk.
Please join us and help shape the future of the MakerDAO.
## Links
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
April
30th
edition
of
the
Scientific
government's
in
risk
meeting.
Today
we
will
be
talking
about
government's
at-a-glance,
we'll
be
hearing
from
vishesh
for
the
state
of
the
pegs
or
even
just
peg.
Can
we
go
back
to
the
singular
for
pretty
close
all
right
state
of
the
peg
possibly
can.
A
A
Surprise,
surprise
we'll
be
hearing
from
maker
man
who
has
done
a
fantastic
amount
of
work
for
the
ecosystem
in
generating
a
model
or
a
report
based
on
the
events
of
the
zero
food
option.
Episode
I
he'll
be
walking
us
through
some
of
the
numbers
there
and
hopefully
explaining
some
of
the
most
intimidating
graphs
I've
ever
seen
before
we
dig
into
too
much
I.
A
Bootstrapped
and/or,
internal
teams,
but
more
and
more
often
we
are
allocating
time
to
the
community.
So
we
have
community
led
teams.
The
community
is
beginning
to
self
organize
and
assume
responsibilities
get
mandated
by
other
members
of
the
community
and
basically
just
self
organized
in
ways
that
are
fantastically
healthy
and
we're
seeing
an
example.
C
A
Give
him
a
thumbs
up
in
the
chats
I,
just
I
think
that
it's
the
compensation
that
he
gets
from
the
community
at
this
points
and
I
pat
on
the
back
for
the
super
hard
work
that
this
was.
This
represents
all
right,
I
promise
to
keep
it
short
and
I
will
so
here's
the
rest
of
the
preambles.
If
you
have
questions,
please
ask
them
if
you
have
access
to
a
microphone,
if
you
don't
have
access
to
no
microphone
type
in.
A
D
All
right
cool,
so
yeah
not
to
still
make
a
man's
thunder,
but
like
the
biggest
discussion
going
on
the
forum,
currently
is
discussion
about
the
liquidation
report.
She's
going
to
present
later
so,
as
one
might
expect
that
throat.
That's
a
lot
of
vote
holders
in
its
discussing
possible
compensation
things
if
more
people
from
the
know
so
more
core
making
community
members
want
to
comment
in
that.
That
might
be
beneficial
and
yeah.
A
Like
to
follow
honor
that
actually,
because
there's
this
this
depending
on
how
the
community
Weston
before
does
maybe
one
of
the
most
complex
operations,
so
it's
over
organized
that
requires
alignment
in
communication
and
identifying
stakeholders
numerous
other
activities.
There
was
a
tremendous
amount
of
activity
in
the
outset
around
this
issue.
Lots
of
opinions
were
raised
and
it
feels
to
me,
like
the
conversation,
might
be
narrowing
down
a
bit,
and
that
feels
slightly
alarming
to
me.
A
So
if
you
consider
yourself
a
stakeholder
in
this
issue
or
even
a
trusted
third
party,
please
consider
hopping
into
this
thread
and
making
yourself
known
also
along
the
same
lines
as
my
previous
speech
about
maker
math.
If
you
have
skills
that
you
believe
you
can
bring
to
bear
I'm
sure
that
those
additional
resources
would
be
appreciated
by
the
group.
Please
have
a
look
at
the
form
to
back
to
you.
D
E
I
think
you
basically
covered
the
gist
of
it.
So
yeah
I'm
I
noticed
that
we
have
like
the
diced
ability
fee,
which
is
for
FAA
and
bad
day
and
sort
of
as
we
threw
us
DC
in
there.
We
just
kind
of
added
a
weekly
stability
fee
vote
for
it
as
well
so
I
kind
of
wanted
to
take
a
moment
and
generalize
this
structure
for
going
forward.
So
when
we
add
the
tenth
collateral,
we
don't
have
ten
weekly
stability
fees
being
voted
on.
E
E
Sure
so,
basically,
right
now,
since
we
had
a
single
collateral
dye
and
we
had
the
side
to
dye
migration
bridge,
we
basically
encoded
all
the
fees
into
the
duties
into
the
duty
parameter.
They
do
the
vault
specific
fees
so
I'm,
proposing
that
we
kind
of
split
the
stability
fee
into
two
portions,
the
first
portion
being
the
base
rate,
which
is
the
global
fee
that
is
sort
of
the
minimum
that
everything
gets
set
to.
And
on
top
of
that
we
have
a
vault
specific
fee.
That
would
mostly
be
the
risk
premium.
E
So
we're
thinking
about
separate
things
and
the
weekly
die
stability,
fee
vote
kind
of
remains
and
we
change
that
to
the
base
rate
vote
so
for
peg
stability,
week-to-week,
we
vote
on
the
base
rate
and
the
vault
specific
fees
are
voted
sort
of
on
well
like
on
Anna
collateralized,
chin
point
as
well
as
sort
of
like
if
we
need
to
change
it
based
on
risk
premium.
So
there,
in
the
end,
I'm
imagining
there
being
one
main
vote
for
PEG
stability
week-to-week,
and
that
is
the
base
rate.
A
All
right,
well
I'm
enormously
engaged
with
this
idea,
simply
because
it
reduces
my
administration
overhead
by
like
nine
ten,
so
I
love
it
miss
just
oh
no
I
scratch
all
that
just
do
it
cause
I,
say
so.
I
risked
have
you
had
a
chance
to
look
at
the
Stars
like
there's
a
too
early
to
presents
an
opinion
about
feasibility,
I.
A
E
A
Right
cool
and
maybe
we'll
dive
into
this
again
next
next
Thursday,
maybe
in
do
84
presentation
for
us
that'd
be
really
interesting.
Let's
get
that
also
touches
on
one
of
my
favorite
bullet
points
recently
is
yeah
and
regular
listeners
will
be
familiar
with
intention
versus
implementation
in
the
past,
from
one
to
the
other.
So
I
like
the
fact
that
there's
going
to
be
a
discussion
for
a
week
moving
into
a
poll
love
it
as
well,
we
need
to
have
a
discussion
in
the
ecosystem
of
boats.
A
D
Back
to
you
want
yep
sounds
good
all
right,
so
the
other
things
that's
gonna
cover
or
the
like.
The
polls,
the
consensus
seeking
threads
generally
I.
Think
most
all
the
polls
to
finish
now
that
you
actually
usually
have
this
time
right.
So
does
anybody
know
if
the
MIPS
timing
poll,
successful
I
didn't
look
want
to
stress
on
the
Ponce.
A
D
A
D
A
B
You
guys
can
still
hear
me:
yeah,
okay,
good,
alright,
so
I
just
said
basic,
so
we
kind
of
know
when
it
happened.
You
know
312,
Black,
Thursday
I
need
to
sort
of
reported
here.
What
was
the
zero
bits?
You
know
so
four
times,
1461
almost
63,000
minutes
into
320
volts
and
then,
basically,
when
you
look
at
the
numbers,
it
cost
the
protocol
six
point,
six
five
million
part
of
which
was
made
off
from
liquidation
fees
and
other
actions,
and
you
guys
we
did
stuff.
B
That
thought
would
be
make
this
better
for
us,
and
then
we
adjust
adjusted
and
Leslie
adjusted
risk
parameters
may
add
to
us
DC
that
made
more
Tyler
quiddity
so
and
afterwards
the
community
kind
of
responded,
something
like
so
I
think
the
one
guy
in
auctions
he
got
liquidated
and
think,
as
I
said,
his
bat
was
liquidated.
I,
don't
know
this
for
sure,
but
people
responded
by
putting
up
interfaces
so
that
people
could
participate
in
the
auction
system
and-
and
that
seems
to
actually
about
but
I'm,
precluding
myself
here.
B
So
you
guys
finally
signaled
you
wanted
to
do
something
and
I
was
in
this
from
the
start.
Just
because
it's
an
interesting
event.
It
really
is
your
black
swan
and
you
look
at
it
and
go
what
the
heck
happened
and
how
can
we
do
better?
So
it
was
it's
going
to
be
a
good
event
in
the
end,
I
think
for
us,
but
right
now
it's
very
painful
and
then
I
just
talked
about
I
got
data
I
was
going
to
refer
to
all
the
datasets
and
unfortunately
I
didn't
but
they're
already
in
the
forum.
B
You
can
find
them
the
first
two
like
torque
Twix
did
and
about
paste
it
and
thank
them
both
because
I
contributed
Twix.
He
did
most
of
the
hard
work.
I
wanted
to
look
at
the
whole
system
because
to
understand
what
happens
in
a
bad
system,
you
kind
of
got
to
see
when
it's
good.
Nobody
else
talked
about
that,
and
so
I
was
trying
to
lay
my
hands
on
the
data
and
that's
just
what
the
analysis
is
saying
here.
B
B
Okay,
here
are
the
two
figures,
the
first
ones
in
time,
and
what
you
can
see
is
the
113
percent
mark
maker
basically
got
its
hold
13%
and
recovered
a
styler
quiddity
and
with
the
red
dots
below
that's
the
percent
collateral
return
right
and
so,
and
then
you
can
look
at
the
markets
below
that.
That's
scatter
around
zero,
like
whether
the
keeper
made
or
lost
money
relative
to
the
market.
B
You
want
to
look
at
at
the
time
and
you
see
the
discontent
of
to
discontinuity
at
312,
313
and
then
another
one
in
3/16,
it's
like
when
we
kind
of
opened
up
and
change
stuff
and
I
don't
have
the
etherium
chart
on
here.
It's
unfortunate
I
really
wanted
to
categorize
this
in
the
events,
but
it's
difficult
to
present
complicated
data.
Well
with
poor
tools.
B
B
Well,
you
just
break
it
down
by
auction
number
and
just
lay
these
all
out,
and
you
could
have
the
time
sequences
on
here,
but
there
you're
really
looking
at
this
by
auction
number
in
mostly
312
to
313,
and
you
can
see
what
happened
from
like
76
760
is
the
first
point
where
we
started
getting
those
zero
pits
and
just
letting
the
collateral
go.
Some
keepers
come
back
on
up
online.
B
You
see
the
red
line,
come
up
in
performance
and
is
improving
then
just
BAM
again
another
chunk
of
a
whole
bunch
of
them
that
flew
out
when
archers
came
in.
When
keepers
came
in
around
2200,
it
started
off
pretty
well
we're
almost
getting
20%
again
and
then
just
kind
of
went
down
another
discontinuity,
BAM
and
then
something
else
happens
because
you
can
see
it's
like
auctions
are
giving
you
back
about
50
60
70.
B
A
B
Right
here
yeah
there-
and
so
it's
right
here-
basically
right
so
I'm
going
through
this
from
here-
is
it's
good.
Here's
our
first
break!
Here
we
come
back
here,
it's
bad.
We
come
up
really
good
and
then
you
just
let's
look
what
this
look
at
the
red
line.
It
really
tells
you
what's
going
on
above
and
then
BAM
down
again
with
all
of
this,
and
then
we
go
right
here
and
then
have
the
scatter
and
the
auction.
But
I
don't
really
understand.
I,
really
more
analysis
can
happen
here.
Guys,
there's
a
lot
of
more
information.
B
B
A
A
B
Blue
line
of
the
tab
and
I
said
it
to
one
hundred
thirteen
percent,
because
I
wanted
anything
under
a
hundred
means
the
protocols
losing
money
like
here
as
well
as
when
it
was
down
here
at
zero.
A
lot
of
these
things
are
at
zero
and
it's
just
you're
not
seeing
them
well
here,
yet
there's
a
zoom
in
you
can
see
a
lot
of
things
sort
of
end
up
at
zero
in
these,
but
you
can
see
the
discontinuities
right
and
so
that's
a
blue
line
of
the
tab.
The
red
line
is
the
collateral
return.
B
Here
we
in
good
times
we
get
20%
over
here
we
get.
We
get
north
of
20%
right.
At
the
end,
everything
that
happened
between
312
and
313
is
literally
like
up
into
316.
There's
this
chunk
right.
We
halted
auctions
and
then
we
started
the
system
different
parameters
right
and
it
was
doing
you
know
it
was
you
know
case.
Let
me
just
come
back
to
normal
afterwards,
right.
A
B
Don't
have
I
do
have
the
gas
fees
I
didn't
put
them
on
here
it.
Throughout
this
whole
time,
gas
views
were
like
hunger
quake.
It
was
really
hard
like
I,
said.
Look,
you
have
to
look
at
the
times
on.
We
could
do
what,
when
at
what
blah
I
didn't.
Do
that
analysis,
that's
a
whole
different
beast
in
here.
I
hope
somebody
does
it.
You
know,
go
ahead.
The
data.
Is
there
really,
you
can
see,
get
on
change,
you
want,
but
you
pull
it
from
the
data
sets.
B
B
What
is
it
under
severe
liquidation,
stress
and
compression
right
as
a
user
I'm
like
I,
read
the
stuff
I
understood
this
auction
performance
in
a
Black
Swan
event:
that's
going
to
be,
it
could
be
ten
dollars
instantly
and
if
you're
coming
down
fast
on
the
OSM,
it's
gonna
it's
going
to
liquidate
you
and
you
will
get
nothing
back
and
the
protocol
will
lose
money.
There's
no,
if
ands
or
buts
on
that,
I
want
to
be
clear
about
that.
B
You
look
at
the
ëthe
moves
on
here
and
they're
significant
and
strong
in
the
periods
that
they're
doing
it's
like
20,
almost
25
percent
each
hour
period
or
each
janna,
and
so
the
system
was
really
trying
to
hammer
out
liquidity,
hard
and
fast,
and
the
markets
just
couldn't
handle
it.
And
then
everyone
trying
to
do
everything
at
once.
Just
led
to
a
you
know:
gas
shortage
or
you
know
huge
gas
in
pre-crease,
and
so
everyone
was
hammering
the
system,
and
so
you
guys
are
all
trapping
each
other.
A
Interesting
so
there's
some
exogenous
well,
there's
other
factors
that
they
can
be
happy
to
hear
from
other
people
at
better
analysis
than
I
am
so
we
have
the
number
of
options
happening.
At
the
same
time
we
have
network
congestion
and
we
have
liquidity.
These
are
like
three
different
factors
that
would
have
impacted
the
successes
of
the
options.
Is
that
correct?
Oh.
B
That's
mine,
that's
I
mean
generally
that's
the
impression
right
and
and
to
look
at
the
data
more
carefully.
There
was
this
really
an
access
issue
you
can
see
when
we
come
up.
You
know
early
on
like
at
some
point.
We
start
to
see
these
happening
rights
like
what's
going
on.
Oh
hey,
let's
get
going,
and
so
they
did
and
so
stuff
happen
and
all
there
was
a
change
here.
B
You
know
you
have
to
look
at
this
in
time
and
kind
of
go
whoa,
because
the
way
that
I'm
timing,
these
auctions,
is
it
at
the
end
time
when
they
deal
is
that
when
they
initiate,
you
know,
there's
a
whole
time
period
for
and
even
though,
at
that
time
it
was
like
10
minutes
there.
It
is
a
period
of
time
that
stuff
happens,
and
so
what
do
you
call
the
time
stand?
B
Everybody
tends
to
look
at
the
deal
time
because
that's
when
you
can
actually
act
on
it
and
take
advantage
of
the
market,
even
though
it
will
take
you
time
and
so
I
think
I
think
this
was
the
deal
that
had
to
go
back
and
look
at
it.
But
you
pick
a
timestamp
for
some
of
that
and
then
you
would
line
it
with
that,
because
you
just
want
to
see
them.
Sequentially
and
I.
B
Don't
think
I
have
them
by
how
they
were
kicked,
or
you
know
in
the
in
the
tenth
in
the
first
phase
or
if
I
have
them
by
one
of
their
dealt
I.
Don't
remember:
I
have
to
look
I'd
have
to
look
at
that
carefully,
but
did
not
look
at
that
as
an
analysis.
Point
I
have
the
data
for
it
that
I
have
a
data
set.
That
has
that
data
and
you
can
pull
it.
Look
at
the
block
number.
You
know
you'll
never
see
when
it
goes
out.
You
know
but
you'll
see
when
it
finally
hit.
B
B
I
messed
it
up
originally
its
twenty
four,
instead
of
twenty
four
point,
six
of
instead
of
twenty
four
twenty
one
percent
that
I
had,
and
it's
up
here
like
we're
over
it
right
now
and
it's
mostly
because
of
the
bigger
you
know
improved
access.
Somebody
is
overpaying
in
auctions.
For
some
reason,
I'm
go
fat
Amit.
You
know,
someone
told
me
why
I
can
guess
so
it
was.
This
is
a
difference
what's
good
times
war.
Is
it
difficult?
What
did
we
do
when
it
got
difficult?
B
You
know
you
see
the
red
in
here
any
kind
of
average
all-out
it's
like
when
it
was
good.
It
was
like
around
10-ish
percent.
I,
don't
know
it's
just,
and
my
question
was
just
what
happened?
Can
I
get
a
look
at
it
and
so
that
I
was
interested
in
this
just
out
of
the
gate
and
I
knew
every
data
point
under
here
was
a
person.
You
know,
there's
320
volts
in
this
and
presume
they're
all
people,
so
there's
hundreds
of
people
involved
potentially
on
this
whole
onslaught.
B
No
one
here
is
just
the
totals
for
the
what
happened
to
the
system
and
each
period
that
I
picked.
You
know
in
the
beatty
period
for
Black
Thursday
is
kind
of
where
all
the
action
is
or
the
vast
amount
of
it.
You
still
see
some
losses,
even
in
the
pre
AGC
and
post
AGC
period
like
we
should
not
be
having
auctions
that
are
posting
zero
lot.
B
You
know
as
losses
to
the
system
at
all,
wasn't
that
bad,
but
it
was
something
in
periods
where
I
think
the
markets
were
relatively
stable
and
I
didn't
look
at
that
in
detail.
It's
just
noted
it
in
the
data
myself
and
then
presented
it
here.
The
biggest
issue
was
collateral
returned.
We
look
at
the
end,
you
know
and
pick
your
period
and
even
in
the
post
AGC,
where
we
should
be
good,
we're
we're
almost
as
bad
as
Black
Thursday.
It's
like.
What's
going
on
there?
B
That's
what
that
and
that's
where
I
broke
it
down
and
I
started
looking
into
bids
and
basically
just
breaking
them
out
by
bid
and
then
looking
at
kind
of
what
I
get
as
a
result
in
an
average,
and
these
should
be
weighted
averages
they're,
not
I'm,
just
taking
the
numbers
and
then
just
average
or
anyway,
as
a
number,
just
not
weighting
them
by
the
volume.
That's
going
underneath
the
law.
B
So
it's
just
kind
of
a
rough
idea
what's
going
on
and
what
I
saw
universally
was
the
one
bid
auctions
just
nailed
us
every
time
here
here.
It's
like
zero
percent
clatter
or
turn
look
at
this
tab
efficiency.
It's
like
37!
This
is
in
Black
Thursday.
So
of
course
it
was
horrible
here,
it's
almost
one
hundred
and
thirteen
percent.
This
is,
of
course,
a
good
time,
but
still
the
collateral
returns
is
0%
on
these
61.
Really
what
happened
there?
That's
when
it
was
good,
and
then
you
go
down
here.
C
B
Total,
when
you
go
down
to
where
it
was
supposed
to
be
good
again
and
it
kind
of,
is
we
get
no
one
bid
auctions
like
awesome,
one
to
bid
also
awesome,
but
when
we
got
it
it
kind
of
sucked,
you
know
tab
efficiency,
104
lower
than
all
the
rest.
Of
course,
not
that
bad.
You
know
off
the
OSM
discounted,
30
percent
looks
like
one
what
hoop-dee-doo,
but
we
get
three
or
more,
and
even
then
out
of
these
we
are
only
getting
12
13.
You
know
up
to
17
here
you
can
average
them
all
out
again.
B
You
know
it's
like
there's
some
stuff
going
on
here
and
in
the
post
AGC.
That
needs
to
be
looked
at
a
little
more
carefully
right
and
then
I
tried
to
break
this
down
by
lot
size
because
I
could
see
just
by
looking
at
the
auctions
that
the
smaller
Lots,
typically
we're
almost
always
doing
as
good
or
better.
You
know
this
is
skewed
to
be
good
for
a
smaller
lot
size.
It
means
people.
You
can
more
easily
bid
up
small
auctions
if
you're
already
in
there
for
a
big
auction.
B
A
B
Legislation
is
that
you
want
two
different
bidders,
at
least
that
and
then
ideally
three
bids,
and
if
you
at
least
have
that
as
a
minimum
requirement,
it
probably
will
get
you.
It
looks
like
it'll.
Give
you
a
better
result,
but
I
didn't
didn't
I,
don't
have
didn't,
have
the
unique
number
of
bidders
at
the
time
at
some
point,
I
will
get
my
that
data
will
come.
You
know
that
we
can
add
it
to
this
data
set
in
the
analytics
and
add
more
even
beyond
that
that
we
want
I
just
didn't
happen
to
have
it.
A
B
C
If
I
got
anything
wrong,
what
I
did
notice
is
that
you
kind
of
keep
bringing
up
the
the
OSM
with
regards
to
the
price,
but
I'm
not
sure,
if
that's
the
right
approach
here
right,
so
the
OSM
exists
not
to
say
that
this
is
the
price
of
something
right
now,
I
think
the
OSM,
in
my
opinion,
kind
of
just
exists
as
like
a
buffer
right
in
order
to
protect
the
system
from
Oracle
attacks.
It's
not
necessarily
indicative
of
what
the
price
of
something
currently
is
or
or
should
be
to
to
our
best
estimate.
C
B
Okay,
so
I'll
go
back
to
the
original
right,
so
I
happen
to
add
by
I.
Do
this
by
hand?
If
you
get
me
a
data
set,
it
can
be
added
to
the
data
and
no
problem.
I
need
a
median
Iser
data
set
marginalized
to
the
right
time.
Points
for
the
for
the
actual
are
the
points
in
the
auctions
themselves.
I
mean
and
I
only
have
one
set
two
sets
that
follow.
That
criteria,
which
was
it
happened
to
be
the
OSM.
B
This
is
all
on
chain
right
and
then
you
know
swap
which
is
effectively
also
on
chain.
So
you
can
use
those
you
I
have
both
of
those
here.
It
was
arbitrary.
What
datasets
it
was.
A
sport
did
I
have
at
the
time
that
I
could
use,
and
the
first
thing
I
put
in
here
was
the
by
hand
CVP
high
and
low
our
parks.
You
know
so
it
is
its
own.
B
But
if
you
look
on
here
the
this
stuff,
the
light
blue
and
the
purple
was
kind
of
me
all,
adding
by
hand
the
our
marks
and
they're
our
marks
a
real
fine
time,
scale
right
and
so,
ideally
for
the
Black
Thursday.
This
data
is
kind
of
like
aunt.
Whatever
you
know
is
just
here
for
its
own
complete
sake,
because
I
used
it
before
here
and
then
I
kinda
used
it
here
after
where
the
time
grain
starts
to
expand
again,
that
it
was
reasonable
to
have
some
it.
Just
that
was
one
of
the
metrics.
B
You
I
have
a
whole
bunch
of
them.
Take
your
pick
and
give
me
the
data
for
it
and
and
if
I
add
it
was
kind
of
my
reaction
to
you
know
the
OSM
I
completely
agree
with
you.
It's
not
the
greatest
point
price
to
use.
You
really
want
a
good
market
price
and
median
Iser
would
probably
be
ideal.
I
don't
understand
that
well,
but
if
it's
an
average
and
a
collective
price
volume
weighted
yeah
great
give
me
whatever
you
got.
I'll
add
it.
If
you
really
want
to
see
this
better.
Okay.
C
B
F
A
A
F
I
have
a
couple
comments
about
the
lot
size.
One
is
that
potentially
one
other
additional
factor
for
the
decreased
efficiency
could
be
the
6-hour
TTL
change,
so
the
lock
size
and
the
TTL
were
both
increased.
At
the
same
time,
it's
hard
to
say
how
much
each
one
is
contributing
to
the
decrease
in
auction
efficiency.
I
think
it's
both
but
I
think
it's
hard
to
say,
which
one
is
having
more
of
an
impact
comment.
B
That's
telling
you
if
you're
already
doing
better
at
the
lot
smaller
lot
size,
so
data
speaking
they're,
pretty
well
to
that
the
lock
size
I
have
a
recommendation
on
this,
too,
is
that
we
should
just
throw
out
different
kinds
of
lock
sizes
all
the
time,
and
that
way
we
can
get
data
and
gauge
market.
You
know
in
liquidation
conditions
and
pricing
conditions
over
the
liquidity
spread
right,
and
you
do
that
simultaneously
when
you
kick
out
an
auction,
and
so
it's
a
recommendation,
I
had
later
I
didn't
get
to
it.
So
I'll
pause
for
questions,
though.
B
F
Not
saying
the
lot
size
isn't
having
an
impact
I'm
just
saying
that
there
are
potentially
other
factors
as
well
and
then
another
another
comment
is
regarding
a
governance.
Trade-Off
is
that
if
we're
trying
to
optimize
for
the
average
case
or
the
worst
case,
because
one
of
the
issues
that
was
popping
up
during
Black
Thursday
was
just
the
insane
amount
of
transactions
and
network
congestion
and
having
to
having
to
bid
on
say
10
times
as
many
auctions
creates
an
additional
burden
on
keepers,
which
was
I,
think
one
of
the
main
reasons
for
increasing
the
lot
size.
F
So
that
would
definitely
result
in
subsequent
auctions
on
average
behaving
or
performing
much
more
poorly.
But
then
it
would
potentially
be
more
beneficial
during
times
of
extremely
high
network
congestion.
So
at
the
very
least,
it's
a
trade-off
if
we
want
better
performing
actions
on
the
day-to-day
or
if
we
want
to
have
better
keeper
activity
during
times
of
high
network
suggestion.
F
B
Ideally
I
going
to
sample
this
and
spread
right,
so
you'd
want
to
do
some
you'd
want
to
probe
the
network
with
some
large
size
network
auctions
right,
but
do
you
keep
consistently
pound
them
out
at
500?
I,
don't
know
if
I
have
this
as
a
suggestion,
but
one
of
the
things
I
was
talk
about
it's
just
a
sheer
amount
of
more
liquidity
that
tried
to
hit
the
market.
You
need
to
throttle
that,
basically,
you
can't
just
throw
out
like
5
million
or
3
million
or
whatever
the
real
numbers
were.
B
You
know
up
to
a
couple
thousand
and
then
other
keepers
that
will
go
from
that
couple
thousand
to
ten
and
then
another
chunk
of
keepers
that
are
going
to
do
the
tens
to
five
hundreds.
Maybe
you
know
500k
in
one
lot,
but
you
need
to
see
if
those
keepers
are
even
remotely
can
handle
that
kind
of
liquidity.
In
some
way,
that's
discussing
anything
else,
I
mean
I,
don't
mind
decreasing
decreasing
for
$500
to
50.
So
what.
B
Too
sorry
overrun
me,
but
I
agree
with
your
first
point
that
the
timing
part
of
this
is
an
issue
in
terms
of
how
long
you
have
to
keep
your
liquidity
there
and
how
long
you
basically
have
to
wait
through
the
market
fluctuations
to
actually
get
a
bid.
The
auction
process
itself
could
use
some
rethinking,
and
but
all
of
these
issues
are
in
clay,
I,
don't
I
think
we
should
keep
at
least
three
to
four
hours,
just
in
case
for
people
to
cycle
through,
and
so
we
have
a
better
mechanism.
B
B
I
think
longer,
auctions
are
better.
It
gives,
like
I've,
been
messing
them
myself,
honestly
as
a
person
some
of
them.
You
know
they
just
run
right
by
me
and
I'm
like
oh
well
whatever
and
move
on
to
the
next
one,
not
that
many
and
good
times
really
people
fight
over
I,
don't
know
whatever
anyway,
I.
D
D
B
You're
right,
it's
a
contract
would
require
changing
to
do
that,
I
think
to
spread
out
them.
There
is
a
multi
function,
so
you
can
just
pick
a
small
size
and
then
just
change
how
that
mall
is
amol
whatever
how
its
fanning
out
the
the
larger
and
you
just
make
that
smaller
and
then
have
it
do
its
fan-out
differently.
That's
all.
D
B
B
Gonna
do
anything
within
the
system.
Now
you
definitely
bring
it
down
and
it
looked
like
font.
50
was
pretty
good,
but
if
we
think
we
need
additional
liquidity,
I
don't
think
hundred
will
be
enough.
If
we
run
into
another
time,
like
a
you
know
a
Black
Swan
event
or
20
20,
25
%.
Isn't
that
bad?
It
was
the
back-to-back
25%
that
was
the
real
nailer,
the
second
one,
coming
literally
within
24
hours,
Bam
Bam,
and
that
was
after
already
going
down
I
think
10%
from
like
220
or
210
into
the
180
range.
B
If
I
remember,
my
price
is
right
here,
as
I
was
looking
at
all
this,
that
it
started
with
a
weekend
that
was
like
about
a
10%
down
and
then
did
a
25%
down.
I
was
like
BAM
and
then
it
caused
for
a
bit
and
then
within
24
hours.
It
did
another
25%
that
took
the
OSM
right
down
to
a
hundred,
and
then
it
really
came
back
up
basically
pretty
quickly
to
like
120
or
something
like
that.
No
people.
E
E
B
That
as
a
recommendation,
basically
that
before
you
even
deal
it,
you
can
do
it
as
a
kick
as
well
that
there
should
be
a
require,
a
minimum
required
bid.
That's
I
picked
it
as
a
percentage
of
the
OSM
just
because
the
OSM
is
kind
of
a
metric
that
maker
uses
to
determine
its
brain
of
action
like
when
it's
going
to
act.
It's
going
to
act
when
it's
over
some
changes
and
so
I'm
like
well.
Maybe
we
should
have
a
system
other
things
in
the
system,
change
and
that's
one
of
them.
E
B
I
would
encourage
all
the
noodling
on
this.
Just
put
it
in
the
forum's,
because
I
already
knew
this
would
spur
it's
gonna
tons
of
discussions
already
forming
around
the
number
of
the
things
in
the
report.
So
all
the
stuff
is
good
stuff.
We
need
it
put
it
there
more
data,
I,
think
on
the
data
and
analytics.
B
So
I
had
the
points
here.
You
guys
can
read
them.
You
know
I
have
conclusions,
you
know,
so
it
was
definitely
a
production
performance
was
impaired.
We
can
see
it
in
the
stuff
when
they
did
come
up.
We
got
back
ten
eleven
percent.
You
can
look
at
these
full
averages
and
you
can
look
at
the
figure
and
try
and
guess
what
you
think
the
number
is,
what
what
what
you
would
expect
there
right.
So
basically,
this
is
the
performance
was
impaired
when
they
had
more
keepers.
We
got
this
pre
GC
period.
B
B
We
need
to
look
at
the
bidders
and
a
lot
size
increases
by
about
you
know
a
lot
five
hundred
percent,
but
the
total
collateral
is
down
so
basically
for
our
vault
holders,
we're
kind
of
sucking
at
the
auction
system
right
now,
I
mean
three
point:
five
percent
now
and
it
should
be
kind
of
normal-ish
versus
almost
eighteen
percent
before
and
that's
even
short
of
the
twenty
five
percent
theoretical.
But
you
kind
of
I
guess
you
could
expect
that.
But
you
can
also
pin
this.
You
know
you
could
you?
B
Could
data
average
this
in
such
a
way
as
to
make
a
graph
kind
of
how
that's
happening
in
the
ranges
and
so
on
in
the
recommendations?
I
just
say,
you
know,
consider
rejecting
deals
and
retaking
auctions.
If
you
don't
have
three
bids
and
two
unique
bidders,
if
we
can
do
that
that
what
I
think
it
would
help
you
can
debate
whether
it
would
or
whether
it's
game
a
bowl,
it's
just
that's
a
separate
issue.
I
just
have
it
as
a
recommendation.
B
B
That's
a
that's
a
huge
unless
you
need
to
flag
governance
to
say,
am
I
going
to
let
this
one
go
or
not,
because
these
should
be
accumulated
and
then
reissued
with
some
thinking
behind
them,
because
if
we're
looking
at
like
a
third
of
market,
there's
a
certain
point,
I
have
ideas
on
how
to
liquefy
the
system.
But
you
know
there's
a
certain
point
where
price
drop
just
kind
of
nails
you
and
there's
nothing
you
can
do
about
it,
then
not
much.
B
You
can
do
about
it
and,
and
then
I
just
say,
consider
staggering
staggering
and
standardizing
lot
sizes,
but
you
just
kind
of
want
to
throw
out
lots
of
different
size
stuff
you
know
and
how
you
do
that
and
I
just
give
some
idea
an
idea.
You
want
to
get
a
feeling
for
what
markets
will
do
for
you
at
all
times
by
staggering
these
lot
sizes.
So
you
can
do
that
if
you
can
do
that,
you
can
sample
your
market
by
the
size
of
the
liquidity
and
it'll.
Tell
you.
B
It
immediately
give
you
statistics
on
system
operation.
So
just
a
suggestion.
I
know
it
requires
contract
changes.
I
suggested
we
shouldn't
go
back,
we
should
go
back
to
50,
like
it
shouldn't
be
more
than
50.
It
looks
like
you
know,
given
what
we're
seeing,
but
it's
all
debatable.
This
could
be
a
time
thing.
It
could
be
other
stuff
going
on.
We
don't
have
enough
data.
I
would
completely
agree
on
that.
It
looks
like
longer.
Auctions
do
get
more
bids,
though,
and
if
you
think
you
know,
you're
getting
more
bids
makes
it
better.
It
should.
B
But
then
you
can
go
back
and
look
we're
not
getting
better
bids
in
those
bigger
auction
sizes.
You
know,
and
that
kind
of
that
data
stands
out
itself,
that
whatever
is
going
on
in
the
bidding
system.
It's
definitely
related
to
a
lot
size,
whether
it's
related
to
the
new
time
element
or
not.
It's
unclear,
you
just
didn't,
have
big
auctions
back
then
you
know
and
then
I,
basically
just
I
try
to
run
this
in
a
compensation
analysis,
I'm
going
to
leave
this
we
can
go
in.
B
Just
given
the
number
of
scenarios,
I
was
already
looking
at
just
looking
at
this
from
the
collateral
return
deficit.
Basically-
and
so
you
know
the
one
a
lot
of
these-
what
you
see
is
that
the
number
of
vault
is
fixed,
of
course,
because
this
is
114
line-
is
pretty
much
all
the
auctions
that
fit
this
criteria,
the
nominal
collateral
return,
anything
that's
113
or
under
my
definition,
should
not
get
any
collateral
back,
and
these
should
all
stay
the
same.
B
This
was
kind
of
a
sanity
check
of
like
my
database
and
what
I
doing
the
right
thing,
and
hence
all
these
numbers
scale.
You
can
pick
a
percent
in
here
and
just
multiply
it
by
whatever
take
this
and
divide
it
by
25
and
then
multiply
by
whatever
percent
you
want
you'll
get.
The
number
in
here
same
is
true
all
the
way
up
to
here.
It's
only
here
that
you
start
to
get
it
get
different
auctions
that
fulfill
the
I
got
my
full
tab.
B
B
They
pay
the
system,
the
full
tab,
they
just
didn't,
get
all
the
collateral
back
and
so
and
that's
it-
you
go
through
here
and
you
decide
what
you
think
you
can
debate
what
you
think
the
value
of
the
apt
was
at
the
time
and
what
you
want
to
do
about
that
and
I
did
this
only
for
us
because
it
was
already
pretty
consuming.
I
was
looking
at
the
data
before
I
even
had
an
idea.
It
was
going
to
have
to
do
a
report
like
this.
B
I
was
just
curious
about
it,
and
this
finally
formally
evolved
into
this
report
and
trying
to
get
as
much
data
that
as
I
could
pack
into
it
and
then
do
what
I
cared
with
it
with
my
spreadsheet
versus
what
I
really
need
is
a
database
that
can
then
I
can
do
queries
on
and
there's
polish
stuff
I'll
do
graphs
and
everything's
just
clean
and
simple
I,
just
don't
have
those
tools.
So
so
this
was
just
the
compensation
report
and
then
you
know
so
you
guys
can
go.
B
B
B
I
was
like
you
can
do
different
things
with
this,
but
it
was
my
suggestion
for
if
we're
gonna
pull
and
even
compensating
you
kind
of
want
to
look
at
no
and
yes,
and
if
it's
yes
kind
of
how
much
and
how
you
figure
out
that
as
a
percentage
that
you
can
then
put
up
to
the
full
group
as
I
like
okay,
let's
approve
thirteen
point,
four,
eight
percent,
whatever
the
number
is.
However,
you
arrive
at
it
yes
or
no
kind
of.
If
we're
gonna
do
it
whatever
I
don't
know,
I,
don't
the
polling.
B
Someone
else
will
handle,
but
it
was
I
put
that
in
the
in
the
forum
as
a
suggestion
and
then
the
rest
of
it
is
just
you
know
thank
yous,
and
you
know
what
what
you
know
that
there's
this
could
have
errors.
Please
audit
it,
like
eventually
I'll
release
my
data
side,
I'm,
just
I'm,
pairing
stuff
away.
That
doesn't
need
to
be
there
because
I
have
a
lot
of
excessive
stuff
in
my
spreadsheet.
For
other
reasons,
and.
B
A
Gonna
use
that
as
a
segue,
to
thank
you
for
the
work
that
you
put
into
this.
Obviously,
the
main
takeaway
for
me
is
that
there's
a
lot
of
data
in
here.
This
needs
to
be
parsed,
but
this
is
the
starting
point.
I
feel
for
the
community
to
begin
to
solidify
around
a
solution
and
around
a
target
for
ourselves,
and
so
this
is
Spencer.
B
B
It
just
kind
of
finalizes
it
for
me,
but
I
want
to
put
that
four
thousand
died
up
against
like
ten,
for
one
I
want
to
raise
some
money
to
make
an
independent
analyst
group
for
maker,
and
so,
if
anybody's
interested,
like
I'm
willing
to
do
that,
I
can
give
an
address
whatever
it's.
My
only
pitch
I
think
if
we
can
do
something
independent
raise
some
funds,
get
people
working
on
something
and
do
it
independently,
somehow
better
I
think
it'd
be
great
and
I.
B
A
Amazing
and
super
generous
appreciate
that
there's
we'll
figure
something
the
the
foundation
is
enormous
ly
interested
in
bootstrapping
initiatives
in
the
ecosystem.
So
obviously
we
take
arms
lengths.
People
are
doing
some
cool
stuff.
Then
we
want
to
help
them
to
do
that
cool
stuff.
So
maybe
we'll
figure
something
out
grants
programs
about
these
programs
and
lots
of
other
programs.
We
can
do
some
resources
we
can
bring
to
bear
to
the
process,
but
that
was
a
great
recap
I.
A
What's
again,
thanks
for
all
the
work
they
put
into
this
I
encourage
the
community
to
like
I,
said,
jump
into
that
forum
thread
and
internalized
this
information.
What
a
part
of
its
take
a
look
at
those
recommendations
and
not
only
talk
about
the
compensation
issue,
but
let's
continue
the
discussion
on
the
improvements
to
the
auction
mechanisms,
because
in
a
sidebar
there's
less.
A
A
C
I
just
had
a
quick
question:
it's
a
little
bit
off-topic,
but
how
familiar
is
the
whole
community
with
you
know,
like
auctions
are
supposed
to
be
kind
of
a
last
resort
right.
There
should
be
a
way
like
divisive,
favor.
I
think
has
this,
where
they're,
basically
using
flash
loans
on
your
card
fault
to
kind
of
unwind
you
before
you
get
to
the
auction,
so
that
would
save
the
user
13%
liquidation
fees
right,
like
is
everyone
I'm
kind
of
familiar
with
this.
C
A
With
that
there's
a
larger
discussion
here
and
I
think
that
here
let
me
frame
before
you
why
options
in
the
first
place
it's
an
interesting
question.
There's
what
my
understanding
is.
This
were
the
the
protocol
is
looking
for
the
most
efficient
way
to
handle
price
discovery,
and
so
the
intention
is
that,
like
obviously,
auctions
are
an
enormous
ly,
efficient
way
of
discovering
price.
A
The
problem
here
is
people
have
to
use
the
option
for
that
to
happen,
and
so
those
questions
around
increasing
the
agility
of
those
auctions,
the
accessibility
of
them,
efficiency
of
them
versus
additional
source
place
to
source
Luke,
leading
against
and
I
know
that
Lev
is
in
the
call
and
I.
Don't
necessarily
once
you
push
him
under
this
boss.
Yes,.
C
C
To
my
two
cents
are:
if
you
have
a
small
vault,
you
should
be
using
those
mechanisms
to
avoid
the
13%
penalty.
But
if
you
have
a
large
vault,
then
you
cannot
use
that
bat
mechanism,
because
you
would
have
gigantic
price
impact.
If
you
just
kind
of
go
like
automatically
interacting
Lina
swap
and
just
pushing
the
prices
around,
so
you
would
end
up
who's
who's.
C
C
Opening
up
a
vault
so
that
user,
if
they're,
opening
up
a
large
vault,
should
not
use
that
automated
mechanism,
because
you
will
end
up
probably
losing
more,
and
you
should
kind
of
rely
on
the
auction
mechanism
that
the
maker
has.
But
if
you're
a
small
user,
then
you
can
avoid
those
13
percent
penalty
fees
by
just
using
something
where
it.
C
A
G
I
generally
agree
with
the
thing
that
Akash
said:
first,
that
I'm
in
general,
this
doesn't
seem
to
be
discussed.
Enough
is
actually
the
CDP
operator,
the
structure
and
it's
just
taken
as
a
given,
but
essentially
the
these
auctions
are
going
to
happen
and
that
people
are
going
to
subject
to
the
to
the
prices
that
they
get
on
the
auctions
when,
in
reality,
I
think.
G
As
it
says,
you
could
even
make
the
case
that
really
no
auctions
should
really
ever
happen,
because
you
know
a
user's
gonna
hit
with
behavior
this
very
high
liquidation
penalty
already,
even
if
assuming
the
auction
goes
there
very
well
I
mean.
Obviously
the
auction
might
might
not
go
that
well,
and
even
it
was
even
the
largest
CDP's.
You
know
on
the
market.
You
know
they
shouldn't
really
be
able
to
get.
You
know
much
less
than
a
thirteen
thirteen
percent
off
market
for
for
their
collateral.
So
there's
there's.
G
Basically
you
know
you
couldn't
even
you
could
even
say
that
really
no
one
should
just
be
going.
You
know,
holding
CDP's
naked
without
any
infrastructure,
that's
going
to
going
to
avoid
liquidation.
So
this
is
it
really.
An
auction
happening
is
really
some
kind
of
disaster
scenario
where
some
technology
has
already
failed
before
the
auction
is
even
there's
even
started,
and
the
auctions
are
not
really
intended
to
get
good
prices
for
users
they're,
really
a
backstop
to
protect
the
system.
G
There's
an
argument:
the
blue
asian
penalty
should
actually
be
a
hundred
percent
in
the
sense
that
you
never
get
any
collateral
back,
because
you've
really
really
screwed
up
the
people
of
it.
If
you
love
that
to
happen,
I'm
not
actually
proposing.
This
is
a
policy
decision
right
now.
I'm
just
saying
this
is
a
valid
perspective
on
the
system
that
the
auction
is.
You
know
it
says
essentially
a
you
know,
a
piece
of
last
resort
infrastructure
that
at
this
point
you
know
the
user
is
the
user.
G
The
situation
is
out
of
the
users
hands
and
yet
there's
no
obligation
to
them
anymore.
I
I
think
for
that
reason
it's
really
important
to
develop
and
promote
the
development
of
CDP
management
infrastructure,
whether
that's
these
third-party
services
or
or
smart
contracts
that
you
can
delegate
to
that,
can
basically
rescue
CDP
before
it
gets
liquidated
or
maybe
even
you
know
basically
services
where
you
basically
hire
someone
to
to
manage
a
CV
for
you
or
software.
If
you're
adventurous
software
that
you
run,
that
will
execute
some
trading
strategies
itself
liquidate
your
CDP.
G
Under
certain
circumstances,
all
of
those
things
is
like,
in
my
opinion,
everyone
should
be
doing
it.
You
know
at
least
one
of
the
things
if
you
have
a
CDP,
otherwise
you're,
really
just
like
you're
going
to
sleep
and
you're,
saying
I'm
the
wake
up
tomorrow
and
all
of
my
stuff
will
be
gone
because
I,
basically
technically
screwed
up
and
I,
don't
think.
That's
I.
Don't
think
that
that
is.
A
A
G
The
reason
there's
this
very
high,
so
the
message
should
really
be
set
by
this
very
high
liquidation
penalty
of
thirteen
percent,
which
is
saying
essentially
that
this
is
not.
This
is
not
just
a
handy
way
of
liquidating
position
like
13
percent,
crazy
right.
It's
really
saying
that
you've
really
screwed
up
now
and
we're
trying
to
was
trying
to
punitive
ly
discourage
this,
which
is
why
I
would
say
to
tosha's
last
point
is
that
we
need
to
make
sure
that
there
are
solutions
for
large
CDP's
that
can
manage
many
of
CBP's
before
liquidation
and
I.
G
Don't
agree
that,
basically,
that
the
right
thing
for
large
CDP
is
just
to
let
it
go
to
auction.
You
know,
I
mean
you:
could
you
could
get
a
much
better
price
than
that?
You
know
going
to
over-the-counter
to
a
market
maker
or
either
just
going
on
going
on
many
exchanges,
I
mean,
of
course
you
know
that
liquidity
might
not
be
atomically
available
on
decentralized
exchanges,
but
there's
no.
A
G
You,
wouldn't
you
would
have
definitely
been
able
to
liquidate
at
any
point
in
marks
at
12,
probably
any
of
these
positions.
It's
certainly
probably
pretty
pretty
pretty
nasty
slippage
for
the
in
in
you
know,
in
the
grand
scheme
of
things,
but
certainly
you
know,
certainly
probably
less
than
13
percent,
and
also
definitely
definitely
much
much
less
than
what
they
actually
got,
which
was
zero.
C
C
In
my
opinion,
kind
of
the
correct
sequence
of
events
is,
if
you
do
have
a
large
CDP,
what
you
could
do
is
go
to
the
UST,
see
fault
mint,
the
die
that
you
need,
close
off
your
aetherium
fault
and
essentially
now
you've
switched
yourself
to
a
US,
DC,
volt
and
then
slowly
by
the
die
and
then
close
that
off
instead
of
aggressively
pushing
prices
around
I.
Don't.
G
I
think
this
is
a
great
point.
I
definitely
read
about
the
general
point
of
not
using
the
not
using
the
tools
available.
On
the
specific
example,
I
would
say
that
actually
compared
to
CDP
users
that
have
gotten
liquidated,
that
CDP
user,
who
did
this
in
this
event,
that
you
described
was
actually
did
you
know,
was
actually
much
more.
What
much
more
effective
right,
because.
C
G
Until
you
managed
to
clean
up
dish
now,
it
would
probably
be
a
little
bit
more
efficient,
like
maybe
you
know,
maybe
would
save
a
couple
of
cents
on
the
dye,
but
basically
every
who's
getting
liquidate.
It
is
literally
like
ten
times
worse
than
that,
so
I
would
kind
of
the
low-hanging
fruit
to
me
is
basically
like.
No
one
should
be
getting
liquidated.
Every
liquidation
should
be
examined.
Is
some
kind
of
there's
some
kind
of
failure.
C
B
Yeah
I
wanted
to
echo
that
as
well
that
the
idea
that
we
could
have
some
sort
of
form
of
automated
vault
management.
You
know
liquidation
protection
system
that
people
can
buy
and
maybe
even
third
parties
can
provide
through
the
system
or
independently
I
like
all,
but
whatever
you
know
the
more
the
merrier
on
this,
but
its
redefinition
offer
it
itself
and,
however,
I
think
in
all
solutions
should
be
encouraged
on
that
that
I
didn't
get
to
look
at
that
data.
B
But
what
was
clear
was
and
get
to
say
this
at
the
end,
it's
like
out
of
the
forty
five
million
that
was
liquidated.
You
know,
roughly
at
thirteen
million,
went
through
the
maker
auction
system.
It
solved
the
rest
of
it
kind
of
self
liquidated
out
there
in
the
markets,
I
mean
a
lot
of
the
stuff
went
through
other
places.
B
It
had
to
offer,
and
so
I
didn't
even
begin
to
look
at
that,
and
so
you
know
that's
a
that
would
be
an
interesting
thing-
data
pole
and
then
analysis
on
as
well,
because
we
have
vault
holders
that
are
here
kind
of
like
hey,
I
self
liquidated.
All
two
two,
eight
eight
you're,
no
Felix
you're
off
I
gotta.
You
know
that
he
was
one
who
kind
of
nudged
us
a
little
bit,
but
he
was
one
guy
himself
liquidated
and
got
about
13
percent
of
his
collateral
back
if
I
have
them
right.
B
Fifty
five
hundred
turning
in
the
own
five
sixty
five.
Ninety
he
goes
five
ninety
and
he
did
that
himself.
2201
at
seventy
five
I
can
speak
to
some
of
these
real
trades
in
here,
and
so
you
know,
yeah
there
was
stuff
flying
through
there
and
yeah.
They
save
the
13%
liquidation
fee
that
this
is
a
I.
Even
have
it
in
that.
You
know
when
you
calculate
the
theoretical.
If
we
look
every
percent,
we
lower
the
collateral,
our
fee,
they
get
more
collateral
back
potentially,
and
so
it's
an
important
issue
to
discuss.
B
I
think
a
fee
is
important.
You
have
to
have
the
right
mechanics
in
the
system.
What
that
fee
is
what
you
do
with
it,
how
you
manage
him?
It's
a
whole
different
thing.
You
know
that
I
also
agree
with
love
in
particular
that
he
is
absolutely
right.
Every
auction
is
a
failure
and
the
system
for
some
participant
that
is
bad
and
needs
to
be
avoided.
We
chewed
through,
like
what
I
have
it
at
120,000
acts
out
of
what
was
like
around
2
million
I'm
guessing
I.
B
Don't
have
that
number
in
my
head
at
the
time,
but
it
was
like
10
percent
10
to
15
percent
of
all.
You
know
it
was.
It
was
pretty
good
for
a
50
percent
drop
that
we
only
ate
through
that
much
collateral
and
I
think
it
was
due
to
these
these
secondary
agencies,
kind
of
saving,
our
ass,
really
and
I
think
we
can
do
better
and
so
I
agree
across
the
board.
H
Okay,
all
right
so
just
share
my
screen,
yeah,
so
to
kind
of
touch
on
what's
been
happening,
I
mean
first
things.
First,
so
I
know,
we've
got
the
the
vote
going
through
about
the
dice
fire,
but
I
mean
this.
The
growth
of
the
dye
supply
has
continued
the
rate
that
we
identified
around
the
19th,
so
that
uptick
did
continue
exactly
the
same
pattern
so
as
called
out.
H
This
is
gonna,
hit
the
or
is
very
likely
to
hit
the
100
million
mark,
and
that
you
know
appears
to
have
been
addressed
now
with
increase
in
the
debt
ceiling.
So
as
far
as
like
what
impact
is
that
having
and
why?
Why
not
so
I
know,
a
lot
of
people
had
a
question
of
like
okay,
with
crypto
with
eath
pumping
so
well.
The
the
question
is:
what's
the
impact
on
the
dye
price?
H
We
saw
this
price
like
start
to
kind
of
level
out,
come
down
a
little
bit
and
then
sort
of
in
that
exact,
same
timeframe,
eath
continued
its
run-up
now.
The
interesting
thing
here
is
when
you
know
we
we've
seen
this
sort
of
inverse
relationship
track
between
the
eath
and
die,
particularly
when
eath
is
volatile
or
crashing
when
eath
is
rising.
This
and
people
have
sort
of
confidence
in
the
continued
rise
of
that
eath
price.
We
generally
tend
to
see
a
reversion
of
that
behavior.
H
That
makes
sense
right,
because
when
you
know
people
are
bullish
on
eath,
when
it's
starting
to
jump,
then
you
potentially
see
some
some
selling
of
dye
and
some
purchasing
of
eath
a
case
of
leveraging
behavior
and
the
dye
supply
has
been
increasing.
So
all
of
that
kind
of
adds
up.
What's
interesting
to
note,
there
is,
as
the
east
price
continued
to
rise,
then
the
dye
trades
started
to
happen
a
little
bit
higher
price.
H
So
we
can
sort
of
break
that
down
as
to
why
but
I'm
also
guessing
that
there
were
a
fair
amount
of
market
makers
that
maybe
picked
up
a
little
inventory
as
soon
as
it
dipped.
That
would
make
a
lot
of
sense,
and
so,
given
all
of
the
events
and
all
the
volatility
that
happened
prior,
it
makes
sense
for
a
lot
of
market
makers
to
have
been
cleared
out
of
inventory
and
for
this
sort
of
dip
to
have
been
an
opportunity
for
some
of
them
to
repurchase
dye
at
a
reasonable
price.
H
H
The
average
is
kind
of
still
sitting
at
you
know
a
dollar
1/5,
but
if
we
see
some
of
the
the
trades
continue
to
trend
like
lured,
a
dollar
you
know
0:05
or
at
a
dollar,
then
this
average
would
naturally
come
down
over
the
next
24-48
hours.
That
kind
of
behavior
would
make
a
hell
of
a
lot
of
sense
with
what's
going
on
with
eath,
but
there
are
certainly
certain
marked
combinations
of
market
conditions
where
it
wouldn't
necessarily
need
to
come
down.
So
it's
a
bit
of
a
wait-and-see
moment.
H
You
know
at
a
dollar
ones,
it's
not
coming
down,
it
might
go
back
up,
etc.
Like
again,
I
would
just
urge
a
little
bit
of
patience
to
wait
and
see
what
happens
next,
because
eath
did
just
you
know,
do
a
bit
of
a
180
and
I
think
that's
that's
gonna
be
throwing
off
the
market
a
little
bit
now.
The
other
point
to
be
made
here
is
if
a
lot
of
that
trading
activity
when
dye
dipped
below
was
purchasing
activity.
H
That
would
be
very
interesting
with
the
the
dye
supply
increasing,
which
would
mean
that
there's
a
lot
of
people
that
are
sitting
on
die
waiting
to
lever
up
at
the
right
moment
now,
given
the
eath
price
just
increased.
That
waiting
to
lever
up
at
the
right
moment
would
actually
mean
that
they'd
be
waiting
for
East
to
come
back
down
and
I.
Think
that's
a
fair
point
to
consider
here
is
I
sort
of
usually
mentioned
this
whole
easy.
H
So
you
know
it
doesn't
necessarily
make
a
lot
of
sense
that
you
know
the
dye
supply
has
been
continually
increasing
over
the
last
week
or
two,
but
that
the
diet
price
has
not
been
continually
decreasing,
except
if
you
consider
that
people
have
been
minting
that
diet,
but
not
necessarily
selling
it
and
the
reason
that
they
would
not
necessarily
have
been
selling.
It
is,
if
they're
sort
of
waiting
for
something
to
happen.
H
They're
stocking
up
on
inventories
to
take
some
potential
action
or
they're
they're,
just
uncertain,
given
the
speed
and
frequency
of
each
price
changes,
and
they
want
things
to
level
out
a
little
bit
before
they
take
those
actions.
So
these
are
all
potential
explanations
for
what's
going
on,
but
again
as
always,
we
can
only
ever
guess
at
people's
reasoning,
okay
and
then
just
to
touch
on
so
total
sigh
is
down
to
ten
point.
Two
million
total
die
is
up,
so
that
is
sort
of
resuming
progress
as
normal.
It
seems
in
terms
of
shut
down
migration,
etc.
A
Yeah,
it
doesn't
feel
like
if,
if
this
is
last
chance
to
ask
Mike
otherwise
I'm
going
to
wrap
things
up
a
bit
Thanks
making
around
once
again
heard
that
awesome
presentation
what's
to
digest.
Please
join
us
in
the
forums
s
what
governance
happens.
These
calls
are
broadcast
information
and
consensus
gathering,
but
not
creating,
so
they
join
us
in
the
in
the
forums
where
governments
actually
takes
place.
Keep
an
eye
on
the
portal.
There's
some
weighty
executives
coming
our
way
tomorrow
and
thanks
everybody
for
joining
us.