►
Description
Visit the https://archive.devcon.org/ to gain access to the entire library of Devcon talks with the ease of filtering, playlists, personalized suggestions, decentralized access on Swarm, IPFS and more.
https://archive.devcon.org/archive/watch/6/the-history-and-future-of-decentralized-operations-at-makerdao/
MakerDAO is one of the foundational projects in Ethereum that originally created DeFi, and the creator of the DAI stablecoin. The project has a rich history of gradual decentralization and, today, is one of the largest and most advanced DAOs in the space.
In this talk I will outline the evolution of decentralized operations and governance at MakerDAO. We will have a look at a few of the most recent scaling challenges, how they developed, and how MakerDAO is dealing with them.
Speaker(s): Wouter Kampmann
Skill level: Beginner
Track: Governance & Coordination
Keywords: MakerDAO,scalability,decentralization
Follow us: https://twitter.com/efdevcon, https://twitter.com/ethereum
Learn more about devcon: https://www.devcon.org/
Learn more about ethereum: https://ethereum.org/
Devcon is the Ethereum conference for developers, researchers, thinkers, and makers.
Devcon 6 was held in Bogotá, Colombia on Oct 11 - 14, 2022.
Devcon is organized and presented by the Ethereum Foundation, with the support of our sponsors. To find out more, please visit https://ethereum.foundation/
A
A
In
2019
there
was
a
launch
of
Multicultural
dye
which
also
added
other
collateral
types,
and
then
after
2019,
when
maker
was
still
functioning
as
a
classic
Foundation,
we
started
to
decentralize
and
that
simply
means
that
we
dissolved
that
foundation
and
we
rebuilt
everything
in
the
Dao.
So
we
have
had
almost
two
years
of
experience
with
that
now
and
yeah.
We
are
wondering
what
is
coming
for
2023.
Will
it
be
make
it
RV
next
so
very
quickly,
myself,
I
have
a
background.
A
In
software
engineering,
I
joined
makerdown
in
November,
2017
I
had
to
make
a
foundation.
I
was
the
head
of
engineering
and,
after
that,
when
I
also
joined
the
Dao
I
became
the
co-founders
of
one
of
the
20
core
units,
as
we
call
them
the
contributor
teams
at
maker
Dao.
My
core
unit
is
called
makerdale
SCS.
This
is
short
for
sustainable
ecosystem
scaling,
we've
removed
barriers
between
decentralized
Workforce
capital
and
work.
A
Lately
we
have
been
thinking
more
and
more
about
decentralized
operations,
and,
while
you
may
ask
why
don't
I
just
use
the
term
governance
as
any
proper
Dao
member?
Well
decentralized
operations,
I
Define
it
very
simply.
For
me,
it's
the
art
of
getting
things
done
in
an
open,
transparent
and
decentralized
organization,
and
by
things
we
like
to
have
work
that
produces
actual
value,
not
just
activity
but
actually
creating
value.
A
So
why
the
term
decentralized
operations?
Well,
Dow
governance
involves
mostly
discussion
and
decision
making.
It
is
not
enough
to
get
things
done.
This
decentralized
decision
making
needs
to
be
supported
by
effective
execution
and
decentralized
operations
includes
governance,
but
it
emphasizes
the
execution
part.
So
we
will
be
talking
a
lot
about
voting,
we'll
be
talking
a
lot
about
governance
stuff,
but
we
say
decentralized
operations
because
we
realize
we
need
to
execute
and
deliver.
A
So,
let's
return
to
make
it
out
and
what
does
operations
and
governance
looks
like
today
at
maker
Dev
there
are
a
number
of
well-known
Concepts
that
have
been
implemented,
basically
when,
when
the
Dow
first
started
for
real,
so
one
year
and
10
months
ago,
and
of
course,
the
foundation
of
it
all
is
token
token
voting.
So
maker
has
token
voting
by
mkr
holders,
and
it
also
works
with
delegates
so
mkr
holders
they
can
improve
changes
to
the
maker
protocol.
A
They
have
to
approve
these
changes
because
the
maker
protocol
is
permissionless.
So
without
a
majority
of
the
vote
you
will
not
be
able
to
make
any
changes
in
any
way.
These
changes
they
can
range
from
simple
parameters
such
as
stability
fees,
debt
ceilings
and
so
on,
but
also
bigger
things
such
as
collateral,
onboarding
and
off-boarding.
A
Today
about
15
of
the
total
mkr
Supply
is
delegated,
and
this
is
delegated
by
222
addresses.
We
have
to
say
addresses
because
we
don't
know
how
many
people
are
behind
those
addresses.
There
are
24
recognized
delegates,
92
Shadow
delegates,
and
these
recognized
delegates.
They
do
receive
compensation
proportional
to
the
voting
rate,
although
there
is
a
limit,
so,
for
example,
the
current
payout
of
the
delegates
is
around
120
000
in
cost
for
the
protocol
per
month.
If
you're
interested
in
voting
or
delegates,
then
definitely
have
a
look
at
our
voting
portal
vote
on
makeada.com.
A
A
This
lasts
three
weeks
and
one
week
where
the
proposal
is
frozen,
so
that
the
governance
participants
can
properly
form
an
opinion
about
what
is
in
there.
If
the
RFC
proceeds,
then
there
will
be
a
governance
poll.
Typically,
this
is
about
three
weeks
and
yeah.
This
poll
will
then
be
up
on
the
voting
portal
that
you
just
saw.
A
Mpr
holders
and
delegates
will
then
be
voting
on
the
individual
proposals
could
be
that
one
proposal
makes
it
another.
Doesn't
all
the
proposals
that
make
it
then
are
then
bundled
in
the
executive
vote?
So
this
is
one
one
smart
contract
that
bundles
all
the
changes
that
have
been
voted
in
by
governance
that
month
and
then
yeah.
If
mkr
holders
approve
this
executive
vote,
then
these
changes
are
applied
to
the
blockchain
protocol.
The
MIP
system
is
quite
Advanced.
For
example,
we
use
templates,
which
we
call
sub
proposal
Maps.
A
There
are
hundreds
of
proposals
that
have
been
voted
on
since
early
2022..
Do
you
want
to
know
about
all
the
mips
that
went
to
maker
governance
go
to
mipsommakerdire.com
and
the
last
piece
of
the
puzzle
are
the
contributors
to
core
units.
As
you
may
know,
a
maker
is
a
pretty
big
down,
so
we
have
20
contributor
teams,
100
full-time
equivalents
of
contributors
around
and
these
core
units,
the
way
that
they
voted
in
it's
with
three
proposal
types,
so
they
Define
or
they
propose
a
mandate.
They
propose
a
budget
and
they
propose
their
facilitator.
A
The
person
who
will
be
responsible
for
interfacing
between
governance
and
the
core
unit,
so
maker
Dao
grew
quite
spectacularly
in
2021
and
as
a
result,
the
total
expense
has
also
grown
to
a
total
of
about
30
million
a
year.
The
scalability
of
the
elements
that
we
saw
is
is
reached
now
in
2022,
with
the
bear
Market
that
has
kicked
in
maker
has
long
been
known
as
the
protocol
or
one
of
the
few
protocols
in
D5
that
is
actually
profitable
in
the
the
longer
run,
if
you
average
it
out,
it's
still
the
case.
A
But
if
you
look
at
the
latest
months,
that
is
no
longer
the
case,
because
the
bear
Market
has
taken
its
toll
and
you
can
see
that
the
dye
expenses
remain
around
30
million.
So
what
are
the
challenges
that
we
have
faced
in
maker
governance
and
what
are
some
ideas
that
we're
thinking
about
to
improve
on
the
current
design?
The
first
challenge
that
I
want
to
talk
about
is
stakeholder
alignment
and
political
Deadlock.
A
So,
in
a
truly
open
and
decentralized
organization,
it
is
very
difficult
to
agree
on
things,
even
the
smallest
things,
let
alone
long-term
vision
and
strategy
maker
has
been
experiencing
that
firsthand.
Throughout
the
last
two
years
there
has
been
a
lot
of
governance
drama
at
maker.
You
probably
read
about
us,
even
if
you're,
not
following
the
dial
up
close
and
at
times
core
units,
the
contributors-
they
really
got
sucked
into
this
drama
and
they
broke
down
entirely
in
certain
cases.
A
A
So
there
was
a
new
core
unit
that
was
up
for
vote
to
be
voted
in,
and
this
was
the
landing
oversight
core
unit,
and
then
there
was
a
maker
Hathaway
fund
that
was
up
for
vote.
I
won't
go
into
the
details
of
what
these
proposals
were
about,
but
they
came
to
an
ultimate
standoff
between
the
two
factions
and,
as
you
can
see,
in
the
screenshot,
there
was
a
record
MTR
voter
turnout
and
yeah.
This
attracted
quite
a
bit
of
attention
and
it
caused
a
lot
of
distraction
in
the
Dao
at
that
time.
A
Definitely,
political
rhetoric,
reigned,
Supreme,
Over,
the
rational
debate,
and
one
of
the
reasons
is
that
makerdale
has
very
few
consensus,
building
tools,
other
than
just
a
forum
to
have
a
discussion
and
then
majority
voting
which
really
isn't
a
consensus.
Building
Tool
you
do
that
when
you
can't
reach
a
consensus.
A
So
what
might
be
an
idea
to
improve
on
this
recently,
there
has
been
a
lot
of
talk
about
the
idea
of
decentralized
voter
committees.
In
fact,
this
is
a
main
element
of
The
Proposal
that
was
put
forward
called
The,
End
Game
Plan
by
the
one
of
the
original
Founders
roon
Christensen,
the
voter
committees.
They
recognize,
on
the
one
hand,
the
political
nature
of
decentralized
organizations,
but
at
the
same
time
they
try
to
apply
some
structure.
A
They
try
to
offer
some
mechanisms
to
form
consensus
and
to
reduce
the
conflicts
that
are
going
on
all
the
time.
What
it
does
is
a
voter
committee
is
quite
simple:
it
is
just
a
group
of
MPR
holders
and
delegates
and
the
important
thing
is
they
are
self-selecting,
so
you
have
people
in
a
group
that
are
of
the
same
mind.
They
think
the
same
thing
about
what
makers
should
do
next,
what
the
long-term
strategy
should
be
Etc
and,
of
course
not
everyone
might
agree
with
that
group,
so
there
might
be
multiple
voter
committees.
A
A
So
another
part
of
the
design
would
be
a
number
of
fixed
scopes.
A
number
of
business
activities
that
makers
should
focus
on
and
not
go
outside
of
that
maker
is
not
an
ice
cream
factory.
For
example,
the
Scopes
that
have
been
proposed
are
so
far
protocol
engineering,
real
world
collateral,
permissionless
collateral
and
growth.
A
So
when
these
voter
committees
they
put
forward
their
strategy,
they
will
Define
for
each
one
of
those
Scopes
how
they
see
the
future
for
maker.
But
by
doing
so
there
will
be.
There
will
be
supported
by
a
number
of
expert
councils
that
have
been
put
forward
by
the
workforce
by
the
core
units,
and
this
is
one
idea
to
deal
with
the
yeah,
the
the
difficulties
within
makers
to
reach
agreement
about
the
long-term
vision
and
strategy.
What
is
the
next
challenge?
A
An
important
lesson
that
we've
learned
is
that
transparency
in
itself
is
not
enough.
The
right
information
really
needs
to
be
served
to
the
right
audiences
at
the
right,
diffraction
level.
People
don't
have
time.
No,
everyone
has
the
time
to
read
through
long
documents
and
just
distill
the
information
that
they
may
need
so
standard
size.
Structured
data
needs
to
be
made
available
via
apis
if
it
is
to
be
analyzed,
summarized
and
leveraged
successfully.
Today.
This
is
the
case
for
some
types
of
data
in
maker,
but
not
all
so.
A
The
clearest
example
of
this
was
the
core
units
and
their
budgets.
All
information
was
always
available.
It
was
on
the
forums,
but
it
was
very
difficult
to
find
our
user
research
has
shown
that
stakeholders
were
unable
to
find
even
the
simplest,
relevant
information
which
core
units
exist.
What
are
they
doing?
What
is
their
budget
and
what
are
they?
A
Spending
their
money
on
a
result
was
a
breakdown
of
trust
between
mkr
holders
and
core
units
and
even
MPR
holders
and
delegates,
because
people
had
this
impression
that,
despite
all
the
so-called
transparency,
there
was
a
lot
of
obfuscation
going
on.
So
what
is
an
idea
how
to
deal
with
this
there's
the
ideal,
the
idea
of
decentralized
operations
platform
we
want
to
Leverage
The
Power
of
software
to
serve
the
right
information
to
the
right
audiences
and
automate
processes
and
embed
best
practices.
A
In
fact,
we
have
already
been
doing
this
for
the
core
units
and
the
budgets
this
piece
of
information
that
was
so
difficult
to
to
understand
for
stakeholders.
We
created
a
limited
prototype
that
focuses
specifically
on
that
information.
So
if
you
go
your
expenses
up,
make
it
down
the
network,
you
will
see
a
clear
overview
of
the
core
units
right
now.
This
platform
is
being
used
for
core
units
to
submit
their
budgets
and
Koreans
can
use
it
to
verify
that
they've
properly
reported
on
their
budget.
A
This
information
is
available
on
the
Forum,
but
if
you
would
want
to
collect
the
same
information
that
you
have
in
five
minutes
here,
you
would
probably
need
five
days
or
longer.
So
what
is
the
next
challenge?
Coordination?
Failure
between
core
units?
There
is
a
construction
error
at
the
heart
of
maker
data
that
we
were
not
really
aware
of
when
we
put
forward
the
first
proposals
for
the
Dao.
The
mechanism
goes
like
this,
so
core
units
are
voted
in
individually
by
mkr
holders
and
delegates.
A
For
example,
you
may
have
the
growth
core
unit
that
is
voted
in
separately,
then
from
the
smart
contracts
or
the
protocol.
Engineering
core
unit
is
work
done
separately
from
the
Oracle
score
unit,
for
example.
So
one
coordination
is
needed
between
these
core
units
and
if
you
want
to
be
successful-
and
if
you
want
to
deliver
value,
a
lot
of
coordination
is
needed.
These
core
units,
I,
can
only
propose
work
to
other
core
units,
but
they
can't
enforce
a
collaboration.
So
there
is
a
constant
negotiation
that
goes
on
between
the
core
units.
A
A
The
result
is
because
these
core
units
they
have
a
defined
mandate.
They
have
a
long-term
work
stream,
but
the
wrong
end-to-end
deliverables
that
are
defined
so
as
a
result,
koreanists
don't
feel
responsible.
If
coordination
fails,
so
you
might
have
a
core
unit
that
say,
builds
a
front
end
and
does
a
tremendous
job.
A
And
then
you
might
have
a
core
unit
who
needs
to
promote
the
front
end,
but
they
haven't
been
aware
that
it
it
was
built
or
that,
where
it
is
available
and
the
marketing
might
fail-
or
you
might
have
disagreements
between,
for
example,
technical
core
units
which
architectures
to
use
and
if
they
don't
agree
on
that
well,
the
end
result
will
simply
not
work.
So
a
lot
of
work
is
done.
It's
just
that.
Sometimes
not
a
lot
of
value
is
created
because
the
pieces
aren't
integrated
or
working
together.
A
Well
so,
which
idea
can
help
with
that?
We
have
been
talking
a
lot
about
project-based
budgeting
at
makerdale
and
I
believe
that
this
will
be
a
much
better
structure
to
deal
with
this.
It's
a
real
value
is
really
only
produced
when
delivering
an
integrated
solution,
delegates
and
mkr
holders.
They
should
approve
these
Integrated
Solutions
projects
more
so
than
mandates,
so
they
wouldn't
be
paying
for
work
like
for
coming
to
to
work
every
time
every
day,
but
they
would
be
paying
for
the
actual
results.
A
So
if
core
units
disagree,
how
to
implement
a
certain
piece
of
work,
then
the
already
proposing
the
project
should
have
the
authority
to
say
well,
we'll
go
with
your
solution
and
not
yours.
The
current
system
that
we
have
with
mid
40,
which
defines
the
budget
of
core
units
and
their
MIP
39
with
their
mandate.
A
A
Instead,
we
would
be
transitioning
to
a
budget
that
is
split
up
between
still
a
partially
retention
budget,
but
separate
from
that
clear
project
budgets,
where
the
cost
items
are
directly
related
to
the
integrated
and
valuable
work
that
is
delivered
then
for
the
last
challenge,
Talent
acquisition,
onboarding
and
the
compensation
question
maker
has
been
struggling
a
lot
consistently
with
hiring
onboarding
and
comp
challenges
maker,
dial,
SCS
micro
unit.
We
have
run
an
incubation
program
for
some
time,
but
now
we
are
winding
that
down
why?
A
A
But
if
you
really
want
to
propose
a
core
unit,
you
need
a
lot
more
work
than
that
at
least
two
months
of
preparation
adds
up
to
a
total
of
four
it's
difficult
for
people
to
just
do
that
out
of
either
their
own
effort
without
compensation.
However,
for
a
core
unit,
it's
also
very
difficult
to
make
these
calls.
A
So
we
have
been
doing
that
for
a
while,
but
we
felt
that
the
model
really
was
it
wasn't
suited
for
the
new
situation
that
we're
in
and,
of
course,
a
large
element
of
that
is
simply
the
bear
Market.
This
has
ended
the
onboarding
and
of
new
startup
teams,
and
it
also
has
created
a
lot
more
discussion
about
where
the
money
should
go.
A
It's
easy
to
get
a
budget
approved
if
there
is
there
is
about
enough
budget
to
do
everything.
Comp
questions
in
general,
they're,
extremely
difficult
in
a
global,
open
and
decentralized
organization.
With
such
diverse
areas
of
expertise,
even
if
you
know
how
much
an
engineer
is
worth
or
how
much
a
designer
a
web
designer
is
worth,
you
may
not
know
how
much
you
should
be
paying
for
a
risk
analyst
or
someone
working
in
banking,
for
example.
A
A
A
maker
Academy
would
be
a
platform
that
is
open
to
everyone,
and
that
is
also
permissionless,
meaning
that
it
is
out
there
in
the
open
in
the
Dow,
and
if
you
want
to
create
a
core
unit,
or
you
have
an
idea
for
a
project
that
you
want
to
see
funded,
you
need
to
know
how
can
I.
You
know
how
how
do
I
fulfill
the
role
of
being
a
good
core
unit
facilitator?
What
are
my
responsibilities?
Who?
A
How
do
I
need
to
interact
with
governance
and
all
these
things
that
can
be
made
available
in
open
platforms,
open
education
platforms
with
specific
courses,
but
also
General
topics,
especially
today,
when
maker
is
going
through
a
lot
of
transition.
There
is
a
lot
of
complexity
on
the
technical
side.
There
is
a
lot
of
complexity
on
the
Real
World
Finance
side.
There
is
a
lot
of
complexity
on
the
organizational
side
and
there
is
a
lot
of
complexity
in
the
new
proposals
that
are
up
for
vote.
A
So
yeah
an
open
and
available
platform
for
Education
and
Training
may
help
people
to
to
acquire
the
skills
they
need
and
then,
if
you
combine
it
with
the
possibility
of
Grants,
then
you
get
a
funding
mechanism
that
doesn't
need
to
be
inside
a
single
core
unit,
but
that
can
be
done
out
in
the
open
and
so
maker
can
continue
to
onboard
the
talent
that
it
needs.
So
these
were
my
four
challenges.
So
what
does
the
future
of
decentralized
operations
that
make
it
look
like?
A
No
one
knows
there
are
a
lot
of
ideas
that
are
floating
around.
We
do
believe
that
the
solutions
we're
thinking
about
the
solutions,
we're
building
they're
all
open
source
software
and
they
could
be
interesting
for
other
projects
as
well.
So
yeah,
thank
you
for
coming
to
my
talk
and
let's
continue
the
discussion.