►
Description
Topics covered
----------------------------
1:37 - Guest's bio
2:55 - PBS Presentation
3:29 - Protocol-agent Problem
5:25 - How to become a validator?
6:15 - Proposer-agent Problem
7:15 - How to make a good block?
7:39 - Execution Payload
10:30 - Ordering matters
11:45 - How to allocate first slot?
16:05 - Block construction in PoW
17:55 - Block construction in PoS
19:24 - MEV trickle down
20:05 - Present PBS (mev-boost for PoS Ethereum)
22:40 - Block building (today in PoS)
24:26 - Proposer Builder Separation
24:36 - Out of protocol markets
25:35 - Block building tomorrow (In-protocol PBS)
26:56 - Why delegate?
29:50 - Opportunities of outsourcing : Danksharding
31:43 - Risk of outsourcing - Censorship resistance
34:19 - Why in-protocol PBS?
36:15 - Why not in-protocol PBS?
39:34 - Wen mainnet?
40:57 - Recap
43:50 - End of the presentation
44:24 - Qn - Why change the bid post sign?
45:17 - Is this enforced at protocol layer or just in mev-boost?
46:00 - Importance of first slot?
47:20 - Friendly Fee Market design
50:54 - Any progress on Friendly Fee Market design
52: 50 - Roadmap & timeline
54:40 - Biggest blocker to have it integrated with Ethereum protocol?
56:28 - Message to community
Resources:
-----------------
Slides - https://docs.google.com/presentation/d/1Tp8E88CMSCjsQ6G9FCHo1hkWLlQjEl66xI-8zHRCtFw/edit?usp=sharing
https://barnabe.substack.com/p/pbs
https://notes.ethereum.org/@vbuterin/pbs_censorship_resistance
https://ethresear.ch/t/proposer-block-builder-separation-friendly-fee-market-designs/9725
Shanghai Playlist - https://www.youtube.com/playlist?list=PL4cwHXAawZxpok0smGmq-dFGVHQzW84a2
EIP-1559 - https://www.youtube.com/watch?v=AC1FS3LmoT4&list=PL4cwHXAawZxqu0PKKyMzG_3BJV_xZTi1F&index=55&t=15s
PEEPanEIP - https://www.youtube.com/playlist?list=PL4cwHXAawZxqu0PKKyMzG_3BJV_xZTi1F
Questions for Guests - https://docs.google.com/forms/d/e/1FAIpQLSdsmujnH_ncIlSEJ23s9jzXAKwfpTGNeJqjREWPloBbkJVQTA/viewform
Check out upcoming EIPs in Peep an EIP series at https://github.com/ethereum-cat-herders/PM/projects/2
Follow at Twitter
-----------------------------------------------
Barnabé Monnot @barnabemonnot | Pooja Ranjan @poojaranjan19
Check out upcoming EIPs in Peep an EIP series at https://github.com/ethereum-cat-herders/PM/projects/2
Contact Ethereum Cat Herders
---------------------------------------------------
Discord: https://discord.io/ethereumcatherders
Twitter: https://twitter.com/EthCatHerders
Medium: https://medium.com/ethereum-cat-herders
Website: https://www.ethereumcatherders.com/
A
A
Welcome
to
p
penny
episode,
93
a
series
dedicated
to
sharing
information
on
upcoming
proposals
and
protocols
for
ethereum
blockchain
and
to
provide
an
overview
on
proposal,
Builder
separation
or
PPS.
We
are
joined
by
barnabay
mannat
Barnaby
is
a
researcher
and
a
member
of
robust
incentive
group
at
ethereum
Foundation.
We
have
him
on
the
show
earlier
to
talk
about
his
analysis
to
support
eip1559
check
out
episode,
37
on
T20
playlist
at
ethereum,
Caterers
YouTube,
coming
back
to
the
post
merge
world
thanks
for
joining
us
by
the
way.
A
B
Sure
so
I
started
getting
into
a
theory
while
I
was
doing
my
PhD
thesis.
My
physics
was
on
Game
Theory,
but
not
applied
to
blockchain,
applied
to
something
else
and
I
started
getting
into
ethereum
realizing.
Oh
there's
a
lot
of
like
this
nice
Game
Theory
type
problems
of
mechanism
design
in
ethereum
and
the
community
is
like
really
awesome
and
the
project
was
so
interesting.
So
after
my
PhD
physics,
I
joined
the
ethereum
foundation
and
we
started
this
robust
incentive
group
that
was
in
January
2020.
B
So
it's
been
almost
trivial
that
we
that
we
are
doing
research
and
yeah
we've
been
working
on
I
would
say
most
of
the
big
upgrades
proof
of
stake.
Eap1559
now
proposal,
Builder
separation-
some
of
us-
are
also
working
on
roll
ups
or
different
topics.
So
I
would
say:
we've
really
expanded
our
range
and
we're
trying
to
still
deliver
high
quality
research
for
the
community.
A
B
A
B
Is
to
try
and
widen
the
frame
a
bit
to
give
a
sense
of
why
it's
happening
and
why
it's
kind
of
already
happening
today
and
then
talk
a
bit
about
the
outstanding
research
questions
and
and
give
you
like
more
background
on
what
we're
currently
thinking
about
relative
to
proposal.
Business
separation,
so
yeah.
A
B
You
have
this
idea
of
principal
agent
problem.
You
have
a
principle
like
an
employer,
and
you
have
an
agent
that
you
delegate
some
function
to
like
running
your
shop
or
something
and
you're
trying
to
ensure
that
the
person
you're
delegating
to
is
is
doing
as
good
a
job
as
you
would
do
yourself,
and
so
the
first
thing
that
we
see
like
the
first
kind
of
instance
of
this
principle
agent
problem
that
we
see
and
it
pops
up
everywhere
in
the
centralized
system,
because
we're
constantly
like
dealing
with
other
parties.
A
B
That
fuzzy
set
of
rules
and
specification
is
somewhat
delegated
to
an
actual
setup,
is
equal,
full
nodes
that
are
running
that
protocol,
and
this
is
the
the
network
of
validators,
so
the
ones
that
are
bonded
state
in
the
system
and
and
produce
the
blocks
make
the
votes
and
everything,
and
so
for
these
validators
you
to
align
their
incentives
with
the
incentives
of
the
ethereum
protocol.
You
have
rewards
penalties
that
try
to
say:
okay,
if
you
do
a
good
job,
you're
rewarded.
B
If
you
do
a
budget,
you're
penalized
and
so
yeah,
so
you
really
have
this
first,
let's
say
original
delegation
to
to
run
your
your
network.
I
am
picturing.
It
like
this
and
I'll
come
back
on
this
graph
throughout
the
token
and
I
like
more
things
and
so
yeah.
So
you
have
like
this
boundary
of
a
protocol
like
what
the
ethereum
protocol
sees
and
what
it
tries
to
have
control
over
it
tries
to
have
control
over
listening
to
the
validators.
It
doesn't
have
full
control.
B
Validators,
maybe
have
a
bit
of
space
to
make
their
own
decisions,
but
the
protocol
exerts
a
certain
amount
of
control.
We
have
the
rewards
and
the
penalties
rapidly
like.
Let's
talk
a
bit
about
how
you
become
a
validator,
so
this
is
the
equivalent
of
being
a
manual
in
proof
of
work,
but
it
now
in
proof
of
stake.
You
have
to
stake,
so
you
look
at
32
if,
after
some
time
you
get
activated
and
then
as
a
validator,
you
pull
the
bond
to
perform
some
consensus
duties.
B
So
once
in
a
while
you're
sampled
from
that
very
large
set
of
validations
and
when
you
are
sample,
you
have
to
to
make
a
block
that
contains
both
consensus
data,
which
is
the
votes
and
execution
data
which
is
the
transactions
and
then
relatively
frequently.
Every
six
minutes.
Also,
you
are
called
upon
to
produce
what
we
call
in
that
test
station,
which
is
your
your
viewable
consensus
and
which
the
protocol
uses
as
a
way
to
to
finalize
blocks
and
coordinate
everyone
over
a
single
Ledger
of
you
and
so
okay.
B
So
I
said
earlier
that
you
know
the
protocol
sort
of
delegates,
its
execution
to
this
network
of
validators,
but
now
what's
surprising
today,
and
what
we
are
talking
about
in
this
talk
is
that
we
observe
that
daily
details
start
delegating
their
crawl
functions,
which
is
the
core
function
of
making
a
block
and,
most
specifically
a
core
function
of
packing.
These
execution
payloads
with
transactions.
B
The
validators
are
delegating
that
to
another
set
of
party
that
we
now
know
as
the
builders
and
and
the
object
of
PBS
is
to
understand
that
relationship
that
creates
between
the
validators
and
the
builders
and
and
understand
like
why
it
came
about,
and
what
is
the
the
best
way
to
to
organize
it.
So
that's
really
what
I'm
planning
to
dig
into
here
so
yeah.
A
B
Understand
like
why
this
delegation
happen
in
the
first
place,
I
think
it's
really
nice
to
take
a
step
back,
or
rather
like
get
into
the
nitty-gritty
of.
How
do
you
make
actually
a
good
block
like
if
it
was
so
easy
to
make
a
block
like
there
would
be
no
need
to
to
delegate
your
functions
as
a
as
a
validator.
You
would
just
make
it
yourself,
so
what
makes
it
so
difficult
to
to
make
a
good
block,
and
why
do
we
see
the
emergence
of
this
network
of
Builders?
B
So
I
said
earlier
that
the
protocol
boundary
doesn't
extend
all
the
way
to
recover
like
everything
that
the
validator
does,
and
so
one
thing
that
the
protocol
doesn't
cover
is
exactly
how
the
validator
is
expected
to
to
make
the
block
they're,
not
really
inclusion
or
ordering
constraints,
there's
really
one
or
two
constraints.
So
the
the
first
constraint
that
validators
have
to
abide
to
is
they
can't
put
transactions
in
their
block
that
don't
pay
the
eip1559?
Basically,
so
you
have
this
fee
Market.
The
protocol
requires
a
at
least
a
reserve
price
for
the
transaction.
B
If
the
transaction
doesn't
pay
that
you
can't
put
it
in
your
blog,
that's
one
constraint:
another
constraint:
is
you
can't
go
above
the
gas
limit
of
your
block,
so
you
contact
the
block
with
more
than
30
million
worth
of
gas.
So
these
are
two
constraints,
but
they
are
fairly
weak
right.
So,
when
you're
a
block
proposal,
you
have
your
main
pool,
you
can
decide
to
insert
all
the
transactions
that
pay
the
fee
as
long
as
there's
enough
space,
you
can
decide
which
way
to
order
them.
B
B
So
you
really
have
like
a
very
large
space
of
how
you
how
you
make
your
blog
and
so
what
we
observed
early
to
ethereum
and
maybe
until
2018
2019,
is
it's
somewhat
incentive
compatible
for
the
block
proposal
to
order
transactions
in
decreasing
order
of
gas
price,
so
packing
the
the
most
fee,
paying
transaction
at
the
top
of
the
block
and
then
going
down
from
them?
And.
B
Is
the
first
spot
in
a
block?
It
actually
has
a
lot
of
value,
because
when
you
have
the
first
transaction
to
to
be
executed
in
a
block,
you
can
do
a
lot
of
Arbitrage.
So
you
you,
you
you're
acting
upon
the
state,
Left
To
You
by
the
previous
block.
But
now
the
markets
have
changed,
prices
have
been
updated
and-
and
you
really
want
to
be
like-
have
that
first
slot
be
the
first
one
in
the
block
and
because
this
has
a
lot
of
value.
A
B
B
And
the
value
of
your
transaction
really
depends
on
the
state.
You're
executing
it
on
is
valuable
or
not
it's.
For
instance,
V
Arbitrage
still
exists
at
the
moment.
You're
sending
your
transaction.
You
may
tend
to
make
a
million
dollars,
but
if
the
Arbitrage
is
gone,
if
that
state
is
okay,
vrb
trash
has
already
been
wiped
out.
B
The
blog
proposal
is
really
the
person
who
has
a
final
word
on
what
goes
into
their
blog
in
what
order
and
we
can
even
insert
their
own
transactions.
So
if
nobody
is
speaking
at
that
Arbitrage,
they
can
be
the
ones
doing
it
by
just
inserting
their
their
transaction
in
the
blog,
and
so,
if
we
think
in
terms
of
mechanism
design
like
if
we
say
okay,
this
first
plot
is
very
valuable.
What
is
like
a
good
mechanism
to
to
allocate
it
to
to
the
person
who
cares
about
it.
B
And
we
can
go
through,
Alternatives
I
think
it's
kind
of
helpful,
because
at
the
end
of
the
day
we
will
see
sort
of
the
seeds
of
how
this
proposal
Builder
separation
has
been
operationalized
from
the
start,
and
so,
let's,
let's
think
of
a
a
very
simple
allocation,
is
you
don't
have
any
rule
for
allocating
this
first
load?
What
you
see,
then,
is
what
we
just
described.
You
have
priority
gas
auctions,
which
is
an
ascending
price
option,
but
it's
very
wasteful.
You
have
Bots
competing
in
the
main
pool.
It
creates
a
lot
of
congestion.
B
We
have
many
trades
in
the
block
that
are
failed
because
we're
all
trying
to
get
that
first
load,
but
when
they
don't,
they
use
a
bit
of
space.
We
revert
to
transaction.
So
that's
not
so
great
another
ID.
How
would
you
allocate
this
this
first
slot
is?
Is
you
let's
say
as
a
proposal
I
could
say
whatever
I
see
in
the
main
pool
I'll
just
pick
one
at
random
and
I
will
put
that
as
my
first
transaction
in
the
Block.
It's
not
incentive
compatible
as
a
proposal.
B
I
I
would
want
to
use
the
decreasing
gas
price
order,
because
that
leads
to
these
priority
gas
options
that
pay
me
more
fees,
but
let's
assume
that
for
some
reason
I
commit
to
that
rule,
it's
very
wasteful
if
I'm
trying
to
get
the
Arbitrage
as
a
user.
What
I
will
do
is
we
spray
and
pray
where
I
just
like
send
millions
of
transactions
in
the
hope
that
one
of
them
makes
it
to
the
top
of
the
block
like
is
chosen
by
the
library?
B
So
you
have
a
lot
of
spam
and
it's
it's
not
really
a
nice
way
to
to
deal
with
that.
Another
one
that
people
are
discussing
quite
often
is
the
idea
of
first
come
first
cell,
so
you
say:
well
it's
fair.
If
the
first
person
who
realizes
that
this
Arbitrage
is,
there
gets
the
right
to
to
extract
that
I'll
be
trash,
but
operationalizing
this
rule
in
decentralized
system
is
very
difficult,
because
you
have
many
blog
proposals.
You
have
many
people
who
are
trying
to
come
to
consensus
over
well.
B
Actually,
who
was
first,
like
you
have
to
come
to
consensus
over
it
because
the
blog
proposal
they
can
cheat.
They
can
say:
oh
you
know,
my
transaction
came
first,
and
so,
if
you
have
to
come
to
consensus,
it
induces
what
we
call
latency
games
where
people
are
just
trying
to
rush
so
that
they
are
very
fast
to
to
get
their
transactions
into
the
first
come
first
cell
queue
and
it
it
wastes
a
lot
of
value.
B
So
if
you
have
a
million
dollar
on
the
table,
that's
on
on
the
let's
say
on
chain:
these
people
are
going
to
try
and
spend
a
million
dollars
to
improve
their
Network
infrastructure
like
build
very
fast
fiber
optic
connections
like
high
frequency
trading
and
and
get
that
arbitration,
so
yeah,
it's
a
bit
of
a
waste,
because
then
that
value
just
flows
out
of
your
system.
Another
thing
that
we
started
seeing
appearing
I
would
say:
2018
and
2019
is
private
channel.
B
So
you
will
just
have
mining
pool
saying
you
know
if
you
care
so
much
about
this
first
slot
just
send
us
like
under
the
table
and
offer
for
how
much
you
care
about
it
and
we'll
put
whoever
is
giving
us
the
best
offer
at
the
top
of
the
block
that
type
of
thing
it.
It
tends
to
favor,
I
would
say,
dark
pools.
So
the
idea
that
now
you
have
exclusive
order
flow.
You
have
very
close
integration
between
the
blog
proposals
and
the
people
who
are
making
with
Arbitrage.
B
So
it's
it
can
lead
to
to
centralization
vectors
that
are
not
super
desirable,
then,
in
other
ways
auction.
So
you
you
try
to
have
some
kind
of
open
option
where
you
say
this.
First
slot
is
valuable
to
most
people,
let's
just
auction
it
up
and
try
to
see
who
can
give
the
most
value
to
to
capture
investing,
and
so
now
you
move
to
the
system
where
you're
thinking
of
bundles
and
bundles
are
just
packages
of
transactions.
B
The
block
ordering
is
is
not
at
all
in
the
controller
of
the
appearance
protocol
and
it's
something
that
the
validators
are
can
decide
by
against
so
okay,
and
so
now
we
have
these
options.
So
we
have
the
system
where
you
have
entities
that
we
call
Searchers
that
are
trying
to
bid
to
the
blog
proposals
for
the
right
to
get
their
bundles
into
the
blog.
So
if
I'm,
a
Searcher
I'm
trying
to
capture
that
million
dollar
Arbitrage
I'm
going
to
try
and
say
to
the
blog
proposal.
B
B
Instead,
so
the
auction
reveals
that
this
bundle
as
a
certain
value
and
Searchers
compete,
and
so
that
value
is,
is
coming
back
to
the
to
the
blog
proposal
and
that
option
was
known
as
the
flashbots
auction
there's
been
others
that
were
deployed
in
in
proof
of
work
and
they
relied
on
a
somewhat
permissioned.
Let's
say
Marketplace
between
the
block
proposals
and
and
researchers
who
were
contributing
these
vendors.
That
worked
because
you
had
mining
pools
which
were
reputable.
You
had
cell
trails
that
were
comfortable
sharing
their
bundles.
B
The
vendors
contain
a
lot
of
very
valuable
information.
It's
basically
how
to
extract
that
value
from
the
from
the
bundle
and
so
yeah,
so
it
relied
on
on
some
trust
between
the
the
mining
pools
and
researchers
who
were
publishing
these
these
bundles.
And
now
we
move
to
proof
of
six,
so
September
15,
20
22..
B
So
they're
saying
that
block
is
worth
this
much
to
me,
and
so
I
am
willing
to
give
it
to
you
the
proposal,
and
so
now
we
move
to
this
system
where
yeah,
it's
no
longer
bundles
given
to
the
proposal.
But
it's
the
whole
block
itself
made
on
behalf
of
the
of
the
proposal.
I
like
this
picture.
A
lot
I
think
it
gives
you
like
a
lot
of
insight
in
a
very
concise
way,
and
it's
the
idea
that
yeah
most
of
the
value
even
Market,
is
competitive
enough.
Okay,
the
social
Market
is
competitive.
B
The
Builder
Market
is
competitive,
like
we
are
going
to
try
and
outbeat
one
another
until
all
the
value,
somehow
trickles
down
to
the
to
the
proposal,
who's
the
again,
the
ultimate
beta
of
what
gets
into
the
block
yeah,
and
so
what
this
value
really.
What
it
is
is
med
the
value
that
can
be
extracted
from
The
Proposal
being
this
all-powerful
entity
at
the
time
of
of
making
the
blocks
okay
and
so
really
that's
the
presence
of
PBS.
B
So
the
system
managers
described
with
the
builders
and
the
proposals
we
you
can
really
think
of
it
as
PBS
today
live
on
the
ethereum
protocol,
it's
known
as
MacBooks.
It
was
deployed
by
flashbots.
Well,
it's
a
it's
a
neutral
Marketplace
in
which
flashbots
does
interact
as
a
relay
and
as
a
builder,
but
it's
open
to
much
more
Builders
and
relays
than
and
just
flatbots,
and
so
very
briefly
like
how
it
works
to
give
a
sense
of
of
the
the
mechanics.
B
If
you
have
three
phases
in
the
first
phase,
the
builders
send
their
full
block
and
their
bid
to
what
we
call
the
relay
the
relay
it
seats
between
the
proposal
and
the
Builder.
So
the
Builder
is
not
going
to
send
the
food
block
directly
to
the
to
the
proposal,
because
they
still
want
to
make
sure
that
the
proposal
makes
a
strong
commitment
for
it
before
they
release
the
data
that
the
block
contains.
That
data
again
is
is
very
valuable.
So
the
relay
acts
as
this
kind
of
middle
man
or
broker.
B
A
B
Perspective,
it
validates
that
the
Builder
is
indeed
paying
the
bid
that
they
announced.
That's
the
first
phase
in
the
second
phase
of
the
beat
selection,
The
Proposal
gets
all
of
his
Beats
without
the
blocks,
just
the
bits
from
the
relays
and
by
default
math
boost
today
selects
the
highest
bid.
So
you
have
some
options
now,
where
you
can
modulate
the
behavioral
math
books,
but
you
can
think
of
it
as
taking
in
a
stream
of
Beats
and
just
saying
okay
I,
like
that
one
which
is
the
highest
and
I'll
commit
to
it.
B
The
proposal
says:
okay,
I'll
sign
that
B.
That
bit
is
actually
a
block
header.
So
once
the
proposal
signs
it,
they
can't
sign
something
else
like
they
can't
see
the
contents
of
the
block
once
it's
released
and
try
to
double,
spend
or
to
steal
the
content
of
the
block
when
we
assign
the
bid,
if
they
make
any
of
our
block,
then
the
builders
block
they
can
get
slashed.
B
B
Where
the
Builder
is
trying
to
get
a
commitment
from
The
Proposal
and
then
the
proposal
is
able
to
say,
okay
I
committed
to
it
now
release
the
block,
and
let's
share
in
that
that
value,
just
to
recap,
with
the
I
think
a
plot.
That's
helpful
to
to
have
in
mind
is
the
different
ways
that
you
can
build
a
block
today
in
proof
of
stake.
So
you.
A
B
Always
the
option
to
use
your
local
blog
building,
so
your
execution
client
is
going
to
look
at
the
main
pool.
Backup
block
itself
propose
that
to
the
consensus,
client
and
proposition
to
the
network.
But
if
you
want
to,
you
can
also
connect
to
that
math
boost
software,
so
net
boost
Marketplace.
We
have
a
builder
API
and
and
receive
these
blocks
from
this
external
blog
building
Network.
So
yeah,
it's
it's
very
much
like
more
complex.
B
If
they
leave
out
certain
transactions
like
either
essential
or
not,
so
the
proposal
is
able
to
to
choose
which
relays
to
connect
to
based
on
on
their
own
preferences,
okay,
so
that
was
kind
of
like
how
DBS
arrived
on
the
scene
and
the
economic
reasons.
Why
you?
You
see
the
separation
now
I
want
to
get
back
into
maybe
a
bit
more
Theory
and
and
address
like
some
of
the
questions
that
I
often
see
people
asking
when
they,
when
they
think
about
PBS.
So
first
thing
to
notice
is
we
actually
do
have
PBS
today?
B
So
many
people
say
oh,
when
do
when
this
PBS
hitting
main
net
in
a
sense
it
already
is
because
PBS
is
not
a
specific
in
protocol
mechanism,
but
it's
really
this
ID
or
this
architecture.
That
tells
you.
Okay.
Proposals
are
Outsourcing.
Part
of
their
blood
construction
to
to
someone
else,
and
so
that
is
expressed
today,
but
it's
out
of
protocol.
They
have
this
math.
A
B
Software
to
guarantee
that
out
of
particular
Market
you
have
a
relays
which
have
addressed
to
parties,
they
validate
the
good,
that's
being
sold,
they
guarantee
the
payment.
So
you
have
a
system
that
works
and
and
the
evidence
that
it
works
is
many
proposals
are
opted
into
it.
But
what
you
can
ask
as
a
protocol
designer
is,
is
this
enough
in
a
sense?
B
Should
the
protocol
recognize
that
okay,
most
of
my
network
is
delegating
the
block
construction
to
builders,
like
should
I,
be
doing
something
about
it
and
and
start
concerning
myself
with
that
separation
that
could
be
block
building
tomorrow.
So
what
I
think
most
people
refer
to
when
we
talk
about
PBS?
Is
the
IDE
or
in
protocol
PBS,
so
removing
the
relays
and
having
the
protocol
itself
become
the
broker
of
the
middleman
between
the
proposals
and
the
builders,
and
so
we
remove
that
relay
row
that
was
in
the
in
the
middle
here
again
picturely.
B
B
You
have
this
Builder
API,
but
that's
not,
let's,
say
a
protocol
rule
it's
more
a
facility
that
the
clients
have
built
into
the
into
their
software,
and
so
the
ethereum
protocol
itself
is
still
in
this
Paradigm,
where
okay
validators
are
making
block
I
reward
them
I
penalize
them,
they
get
fees,
they
get
stuff,
but
it
just
has
no
idea
that
validators
have
been
delegating
these
block
construction
to
the
building.
And
so
the
question
is:
do
we
make
protocol?
Has
such
an
ID?
B
So
we
move
forward
the
boundary
of
a
protocol
so
that
we
recognize
that
okay
Builders
exist
in
our
Network
and
we
entrench
them
as
protocol
actors,
I
think
it's
a
very
important
question
and
it's
not
something
that
should
be
done
and
without
like
careful
thought,
because
whenever
you
try
to
move
it
under
user
protocol,
like
you,
you
say
when
and
I
doing
it
right.
It's
it's
costly
to
move
that
down
the
rates
because
lead
to
do
anything
at
the
political
level.
B
So
you
really
want
to
to
make
sure
that
it's
it's
worth
it,
and
so
some
elements
of
response
here
and
again
I
I,
don't
came
back
to
have
a
full
picture.
It's
it's
something
that
there's
a
lot
of
open
research
on,
but
I
just
try
like
to
give
some
ways
to
to
think
about
it,
and
the
first
thing
is
why
delegate
in
the
first
place,
so
when
metal
boost
started
and
when
there
was
Wisconsin
around
censorship.
I
think
a
lot
of
people
were
naturally
and
and
almost
positively
careful
to
say.
B
So
the
role
of
a
network
as
a
whole
is
to
to
check
that
the
protocol
is
executed
correctly,
but
at
the
time
of
making
a
block
at
the
time
of
attending
something
to
the
to
the
blockchain.
This
might
require
a
lot
of
resources
or
a
lot
of
knowledge
and,
and
then
the
Builder
role
might
be
very
helpful
so,
for
instance,
with
block
construction
so
packing
the
transaction
in
the
Block.
B
It's
a
One-Shot
exercise
like
it
might
require
a
very
sophisticated
algorithm
to
pack
to
do
RB
Treasures
everything
the
once
that's
done,
will
work
to
verify
that
the
block
rules
is
correct
and
the
block
is
valid
from
the
Indian
perspective.
This
is
fairly
cheap.
You
just
have
to
execute
the
transactions,
and
so
what
you
want,
since
making
a
good
block
is
hard
or
making.
Let's
say
the
optimal
block
is
hard.
Is
you
don't
want
to
shut
out
proposals
who
are
interested
or
unsophisticated
from
accessing
these
very
good
books,
because.
B
Case,
if
you
say,
when
there's
no
mechanism
for
you
as
an
interested
proposal
to
to
access
that
Network,
that
can
make
very
good
blocks.
You
start
having
an
economic
centralization
effect,
where
a
rewards
that
these
solo
validators
are
getting
they're,
just
not
as
much
as
the
rewards
that
very
sophisticated
pools
or
very
reputable
pools
can
get
by
having
access
to
this
Builder
Network
and
so
really.
The
idea
of
proposal.
B
Builder
separation
is
making
sure
that
these
superpowers
that
the
builders
have
are
somewhat
fairly
distributed
over
the
whole
network
and
that
the
network
stays
decentralized
by
this
effect.
Some
opportunities
of
Outsourcing,
for
instance,
something
that
we
can't
really
do
without
a
builder
role
is
then
shouting
so
Deng
sharding
is
his
proposal
to
to
greatly
expand
the
data
availability
that
the
ethereum
protocol
can
provide
and
and
to
to
get
that
huge
increase
in
in
bandwidth
of
data.
You
really
need
to
have
very
powerful
entities
in
the
network,
so
so
this
Builder
works.
B
So
the
Builder
in
the
enchildren
is
responsible
for
making
a
let's
say
a
media
block,
so
something
that
is
16
megabytes
per
block.
It
doesn't
sound
that
big,
but
but
if
you
are
the
person
who's
making
the
block
at
the
time
you're
making
it,
you
have
to
send
it
to
gossip
it
over
many
of
your
peers,
so
the
bandwidth
requirements
are
very
high
for
for
you.
B
So
once
you
make
that
make
a
block,
and
once
you
disseminate
parts
of
it
on
the
network,
you
can
you
can
basically
shout
the
contents
of
that
block
and
and
then
for
the
rest
of
the
network.
Let's
say
solo
validators
with
just
regular
consumer
internet
connections.
They
don't
need
to
see
the
mega
bloke.
B
They
can
just
play
these
games
of
data
availability,
sampling
and
verify
for
themselves
and
for
the
rest
of
the
network
that
the
data
is
indeed
available,
and
so
you
have
again
this
asymmetry
that
a
very
good
Builder
that
can
summon
very
high
resources
can
do
something
very
nice
like
just
improve
the
capacity
of
the
network
without
imposing
that
cost
to
the
to
the
network
as
a
home.
So.
A
B
You
can
have
big
blocks
and
also
you
can
make
sure
that
the
Big
Blocks
are
are
still
verifiable
by
your
network
as
well,
and
the
Max
Capacity
just
can't
be
provided
without
this
Builder,
and
so
that's
really
an
opportunity
of
PBS,
but
they
are
restore.
So
every
time
you
delegate
something
to
someone
else,
there's
always
the
risk
that
that
person
doesn't
really
do
what
you
want
them
to
do.
So
this
principal
agent
problem-
and
one
of
them
is
very
clear
today-
is
censorship
resistance.
B
So
if
we
go
back
to
these
interlocked
sets
of
delegations,
the
first
thing
was
okay.
If
your
protocol
kind
of
tells
you
you
should
be
neutral
and
you
should
include
any
and
all
transactions
which
which
pay
the
fee
right,
like
your
blog
proposal,
you
see
a
transaction
that
pays
basically
and
maybe
pays
you
a
little
tip.
It's
rational
for
you
to
to
include
it.
That's
the
protocol
now
do
proposals
always
include
every
transaction.
They
see.
B
Some
proposal
might
decide
that
they
want
to
to
censor
a
transaction,
but
when
you
really
rely
on
the
fact
that
your
proposal
set
or
your
validator
set
is
very
decentralized,
and
in
that
case,
if
someone
doesn't
include
the
transaction
you
you
know
that
fairly
soon,
another
person
is
going
to
to
include
it.
But
now
you
have
this
phenomenon
where
proposals
are
Outsourcing
block
production
to
agents
that
by
Design
are
much
less
decentralized.
B
So
if
we
say
Okay
Builders
have
to
be
sophisticated,
they
have
to
have
all
this
knowledge
to
make
very
good
blocks,
so
very
high
resource
requirements
like
you
should
expect
that
the
set
of
builders
in
your
network
is
going
to
be
much
smaller
than
the
set
of
validators
but
bonded
in
in
proof
of
state,
and
so
the
builders
we
might
have
their
own
preferences
like.
If
they
are
legal
entities
in
some
jurisdiction,
they
might
say
you
know
the
Assumption
transactions
I
I,
don't
want
to
put
them
in
my
block
despite
the
proposal's
preference.
B
So
the
proposal
might
have
no
problem
with
the
transaction
being
included,
but
the
Builder
may
say
you
know
not
in
my
blog
and
then
the
proposal
can't
really
do
anything
about
it.
Or
can
we
a
solution
that
came
up
a
year
ago?
Is
the
idea
of
getting
the
proposals
to
require
the
inclusion
of
some
transaction
list
by
the
billions
when
they
tell
the
Builder?
You
know
make
a
block
for
me.
They
require
the
Builder
to
to
include
some
transactions
and
we
call
that
a
inclusion
list.
It's
a
proposal
by
Francesco.
B
You
can
do
inclusion
lists
out
of
protocol
if
you
could
build
some
kind
of
inclusion
list
in
net
boost,
for
instance,
but
the
fact
is
that
it
works
better
if
a
protocol
is
involved
and
if
you
can
build
these
inclusion
lists
in
a
way
that
one
proposal
builds
it
for
another
proposal.
I
won't
get
into
the
details
here,
but
I'll
give
a
link
at
the
end
where.
A
B
Can
read
more
on
this
discussion,
which
is
very
briefly
last
couple
of
slides?
Okay,
we've
seen
like
why
people
delegate,
we
said.
Okay,
there
are
opportunities,
there
are
risks.
Now.
The
question
is:
why
should
we
move
that
in
protocol?
So
we
we
know
that
delegation
happens
and
we
have
ways
of
organizing
this
out
of
protocol
markets.
Why
do
we
as
protocol
designer,
maybe
care
that
we
need
to
move
PBS
in
in
the
protocol?
So
the
first
thing
that
I
was
pointing
out
earlier.
A
A
B
With
respect
to
the
protocol
in
math
boost,
you
have
some
failure,
cases
which
are
now
fairly
rare,
which
were
maybe
more
problematic
at
the
beginning,
but
but
as
the
market
matured,
it
seems
like
it's
fairly
stable,
but
we
could
have
cases.
We
have
a
relays
tail
to
validate
the
bids
and
the
proposal
thought
they
were
getting
a
million
dollars
and
then
they
just
end
up
getting
10.
B
So
if
you
have
in
protocol
PBS,
you
have
ways
to
make
any
bid
that
is
made
by
the
Builder
as
fully
binding
and
the
Builder
says
you
know:
I'll
pay
you
a
thousand
if
they
have
to
pay
a
thousand.
Another
thing
that
might
be
nice
is,
and
maybe
I
won't
organize
so
much,
but
the
idea
of
consensus
bids.
So
the
protocol
now
can
see
the
auction
happen
and
it
can
get
a
sense
of
what's
the
actual,
correct
value
of
the
block.
B
In
a
sense,
it
elicits
the
the
true
value
of
the
block
via
the
auction,
and
it's
able
to
to
detect
that
into
action
upon
this.
This
knowledge
and
this
can
be
used,
for
instance,
for
any
of
this
moving
or
Med
burning
proposals.
There
was
some
reasons
to
do
in
protocol
PBS.
Some
reasons
why
we
might
not
want
to
do
in
protocol
PBS
the
first
one
is
always
it's
just
more
protocol
complexity.
B
B
Usually
things
are
we
get
it
right
like
it's
tested
a
lot,
but
it
still
means
that
yeah,
you
just
have
like
more
complexity
and
more
overhead.
Another
question
that
I
think
is
maybe
more
important
is
if
we
do
in
protocol
PBS.
In
a
sense,
we
are
enshrining
a
very
specific
Market
structure
and
the
market
structure
is
proposals
are
delegating
to
builders
to
make
their
blocks
now.
The
question
is:
do
we
still
have
builders
in
10
or
50
years?
So
sometimes
we
think
you
know,
like
you,
want
to
build
a
protocol
so
that.
B
It
every
five
years,
or
so
it
has
to
osify,
and
if
you
build
in
the
protocol
beside
you
that
it
has
a
separation
to
proposals
between
Builders,
maybe
we
can't
get
rid
of
it,
and
maybe
it's
irrelevant
in
10
or
50
years,
where
we
have
super
amazing,
cryptography
or
proposals
with
like
really
high
bandwidth
yeah.
Then
we
just
like
Stoke
with
that
mechanism,
and
another
thing
is.
We
also
need
to
make
sure
that
whatever
we
enshrine,
we
get
like
the
exact
shape
of
the
mechanism.
B
Right
and
here
you
really
have
like
a
very
large
design
space
to
to
do
PBS
in
protocol
and
the
design
Space
is
really
has
to
do
with
the
way
you're
doing
the
the
auctions.
For
instance,
you
could
decide
to
do
potential
block
option
where
the
proposal
makes
one
path
and
the
Builder
makes
another
path
versus
a
whole
block
auction.
You
also
have
like
some
design
space
around
when
the
builder
needs
to
commit
to
the
block
contents.
B
So
there's
beside
your
blog
auction
and
Slot
auction,
the
Block
action,
the
Builder
commits
to
block
content
before
the
bid
in
the
slot
auction
they
commit
after
the
Beats.
It's
a
bit
more
flexible
as
opposed
by
Julian
who's,
a
researcher
at
the
week
that
will
come
out
very
soon
on
this,
and
then
you
also
have
this
idea
of
timing
games,
which
is
this
incentive
for
people
to
be
as
late
as
possible
to
do
their
duties.
B
So
you
want
to
be
the
last
one
to
place
your
bid
because
you
have
the
most
information,
the
most
Mev.
If
you're
a
proposal,
you
want
to
be
as
late
as
possible
to
propose
your
blog
so
that
you
capture,
like
the
most
any
possible
and
this
time
in
games
like
they
tend
to
appear
in
very
nasty
ways
in
these
in
these
scenarios.
B
As
a
note,
if
we
feel
okay,
maybe
we
are
not
compatible
within
postcard
PBS,
we
can
still
do
these
inclusion
lists
in
protocol
without
having
in
protocol
begins.
We
have.
We
still
have
ways
to
deal
with
censorship.
If
it's
really
a
problem,
add
the
protocol
level
without
going
through,
like
the
whole
trouble
of
enshrining
PBS
in
in
the
protocol.
So
I
think
this
is
something
that
maybe
is
overlooked
in
the
public
discussion
around.
A
B
B
It's
just
not
in
protocol
mainnet,
but
math
boost
is,
is
part
of
the
ethereum
network,
not
part
of
a
protocol,
but
part
of
what
Network
agents
are
doing,
and
so
yes,
when
the
question
is,
is
the
value
of
PBS
in
protocol
like
stronger
than
the
cost
of
doing
it
all?
Do
we
feel
confident
that
it's
like
a
good
way
forward
to
to
enshrine
tbs
in
the
in
the
protocol
and,
in
my
opinion,
there's
more
research
that
is
needed
to
to
evaluate
both
options?
B
But
it's
very
interesting
like
even
if
we
decide
not
to
go
with
it
or
if
we
decide
to
go
with
it,
I
feel
like
research
questions
that
this
has
prompted
they've,
been
very
productive
on
many
other
fronts,
such
as
understanding,
Mev,
understanding,
the
space
of
mechanisms
that
the
protocol
can
do.
So
it's
been
like
a
yeah,
a
really
productive
research
trend
and
I'm,
mentioning
here
that
I
released
the
post
a
month
ago.
B
A
A
B
Validator
is
PBS
and
EIP
no,
but
it's
an
architecture
that
can
be
instantiated
in
many
ways
and
when
we
do,
if
we
do
PBS
in
protocol,
there
will
definitely
be
an
EIP
that
will
explain
how
to
how
to
do
that.
Is
it
currently
in
protocol?
It's
not,
but
it
does
exist
today
in
the
in
the
form
of
Med
boost.
Does
it
serve
censorship?
B
No,
and
to
some
extent,
maybe
it
makes
it
worse,
because
you
have
these
builders
in
the
network
that
are
more
centralized
and
maybe
are
more
likely
to
to
censor
Builders
or
relays
in
in
net
boost,
but
but
really
what
it
tells
us
is.
It's
kind
of
revealed
that
maybe
the
person
who's
the
best
able
to
to
make
the
block
is
also
the
person
who
who
wants
to
to
censor
it
and
and
it's
a
bit
of
a
unfortunate
outcome.
B
But
it's
one
that
seemingly
is
not
unresolvable
and
and
seems
to
improve
over
time,
and
especially
we've
seen
like
more
and
more
blogs
that
are
built
by
Builders
and
relays
that
are
declare
themselves
as
non-censoring,
and
so
it
does
feel
like
the
market
is
trending
in
the
right
direction.
For
this,
if
we
move
PBS
to
the
protocol,
would
we
immediately
sort
censorship
in
a
sense
not
really
that
we
would
remove
a
relays?
B
B
Then
we
might
have
some
use
for
for
these
Builders
to
provide
State,
Witnesses,
provisional
state
to
stateless
validators
or
make
this
very
easy
proofs
based
on
the
entrance,
EK
VM,
so
yeah
there's
plenty
of
ways
that
PBS
can
be
helpful
in
the
in
the
future
and
yeah.
That
was
it
and
thank
you
for
listening.
I
hope
this
was
interesting
and
that
it
gave
like
a
bit
more
insight
into
PBS
and
now
I.
Think
I
will
take
questions
from
Puja.
A
Awesome
presentation,
I
think
it
explained
difficult
terms
in
a
simple
way:
slashing
conditions
will
keep
validators
and
check,
which
is
helpful
for
the
security
of
the
network.
Last
slide,
I
think
was
the
best.
The
recap
was
great
well
obviously,
this
has
helped
get
answers
to
many
questions
that
our
listeners
and
viewers
may
have
in
mind.
We,
however,
will
take
a
few
more
questions
to
maybe
get
more
clarification
for
people
trying
to
get
involved.
B
I
see
a
question
on
the
chat,
should
I
answer
it
now
the
professor
has
chosen
the
highest
beads.
Why
would
we
want
to
change
that?
Both
sign
yeah?
That's
a
good
question,
so
we
don't
want
to
change
a
bit,
but
they
might
want
to
try
to
get
the
Builder
to
reveal
their
blog
copy
paste
that
block,
but
replace
everywhere
that
the
Builder
is
making
money
with
their
own
address.
So
they
might
try
to
steal
the
alpha.
A
B
Novel
block,
once
you
see
the
Builder
block
being
released,
but
what
is
end
first,
that
protocol
layer
is,
if
you
do
make
two
blocks,
one
that
you
got
from
a
builder
and
one
that
you
made
yourself
and
you
sign
both
of
them.
Someone
is
going
to
slash
in
and
that's
going
to
be
and
cost
at
protocol
level.
So
we
it's
always
in
your
power
to
make
two
blocks
in
the
same
slot,
but
you're
going
to
be
slash
for
it.
So
you
probably
shouldn't.
A
B
A
B
It
just
happens
to
be
that
yeah,
if
you're,
trying
to
Arbitrage
the
silica
unicever
pool
like
you,
want
to
be
the
first
one
in
the
block
who
touches
that.
So
maybe
what
you
care
about
is
not
the
first
slot
exactly,
but
it's
it's
to
be
first
in
relation
to
everyone
else
who
is
trying
to
access
the
same,
let's
say,
pool
or
address
in
the
in
the
state
and
so
yeah.
B
What
what
it
looks
like
is
just
that
people
are
trying
to
outbeat
one
another
so
that
they
land
above
their
competition,
so
I
call
it
first
slot
I
think
more
as
a
as
a
mental
model,
but
you
can
really
think
of
it
as
yeah
you're.
Just
trying
to
raise
ahead
to
be
the
first
and
the
race
is
not
based
on
time.
It's
based
on
how
much
the
beat
you're
making
like
puts
you
ahead
of
your
competition.
A
B
A
B
A
B
B
Neither
The
Proposal
nor
the
Builder
can
sort
of
grief
one
another.
That's
the
first
post
that
was
there.
The
second
post
I
think
is
probably
what
I
would
say,
the
most
credible
option
for
protocol
PBS.
It's
called
the
two
slot
design.
So
in
that
one
you,
you
think
of
like
the
slots
that
we
have
today
in
the
proof
of
stake
consensus
as
you
divide
them
into
first
The
Proposal
listens
to
all
the
bits
from
the
builders
and
then
we
make
like
a
consensus
block
that
that
says.
B
Testers
that
are
voting
on
that
block
so
via
testers.
They
they
make
safe.
The
fact
that
the
proposal
has
committed
and
when
the
Builder
sees
okay,
the
proposal
has
made
the
block
the
bead
is
made
safe,
like
the
proposal
can't
go
back
on
their
words,
so,
as
a
builder
I
feel
confident
now
to
release
my
blog,
so
the
second
slot
of
the
second
partial
slot.
The
Builder
really
sees
the
execution
payload.
A
B
Have
this
game
of
commit
reveal
commit
reveal
that
is,
there's
another
proposal
that
the
Builder
would
encrypt
the
contents
of
their
blog
to
to
a
public
key
that
is
used
for
threshold
decryption
and
the
the
set
of
the
testers
is
is
the
committee
that
tries
to
decrypt
the
payload
of
the
of
the
building?
So
if
you
have,
let's
say
two
thirds
of
via
testers
that
are
online,
then
you
can
decrypt
the
the
Builder
block.
B
I.
Think
it's
a
it's
a
nice
solution,
because
it
avoids
like
that
one
step,
two
stepping,
but
it
does
mean
that
you
need
much
more
likeness
requirement
in
your
system
like
you
need
to
have
sufficiently
enough
adjusters
online
in
the
network
and
in
reality
it's
the
case.
When
you
see
the
statistics,
99.6
percent
of
your
testers
are
online
at
any
time,
but
in
many
ways
POS
protocol
was
designed
so
that
you
you
stay
live
even
you
are
below
that
critical
to
short
threshold
and
so
I.
A
B
Yeah
I
would
say,
there's
been
like
some
puts
on
actually
maybe
less
vote
on
the
in
protocol.
Pbs
designs,
mostly
I,
feel
because
with
two
slot
design
yeah
seem
like
it
could
be
specified
fairly.
Well,
what
probably
prevents
people
from
specifically
is
yeah
understanding.
Do
we
need
it
at
all?
Do
we
we?
How
do
we
do
it
like?
How
do
we
do
that
auction?
B
A
B
Different
ways
that
you
can
use
and
maybe
even
like
some
of
these
ideas
that
I
discussed
like
threshold
decryption,
so
there's
been
ideas
on
how
you
would
potentially
remove
relays
by
having
Builders
encrypt
their
blocks
to
to
some
kind
of
committee
or
using
things
like
eigen
layer
to
to
secure
the
Builder
proposal
relationship
without
relying
again
on
these
relays.
So
yeah
there's
been
kind
of
like
a
creative
explosion
of
ways
to
to
to
make
these
made
boost.
Market
more
more
resilient,
very.
A
Interesting
so
for
people
who
may
have
missed
the
concept
of
Mev
in
the
first
place,
we
have
recorded
an
episode
with
Thomas
tanza
on
Mev
and
Flash
bonds.
So
it
is
highly
recommended
to
give
a
watch
about
Mev
and
then
probably
learning
about
MAV
boost
concept
would
be
a
bit
easier.
Only
for
people
who
are
relatively
new.
A
That's
right
all
right.
My
next
question
is
about
the
roadmap
and
the
timeline,
though
you
have
already
mentioned
that
it
is
on
ethereum
and
mainnet
in
a
way
with
the
help
of
Mev
boast
yet
I'm
curious
to
hear
if
there
are
any
chances
of
having
an
EIP
for
PBS
in
future.
You
did
mention
about
like
the
positive
and
the
like
negative
side
of
having
it
on
the
protocol.
A
B
How
can
we
attack
these
questions,
so
it
does
feel
like
we're
making
progress
in
terms
of
closing
out
these
open
questions
that
we
have
around
PBS
I,
wouldn't
be
surprised
that
if,
at
some
point
there
is
an
EIP
that
tries
to
specify
a
bit
better
okay,
what
would
it
look
like
if
we
had
this
thing
in
the
protocol
and
then
onto
the
basis
of
these
EIP
people
would
probably
discuss
and
say:
okay?
Can
we
do
it
better?
Can
we
do
it
another
way,
or
should
we
do
it
at
all?
B
A
B
Right,
that's
a
good
question
too
I
I
would
say:
there's
a
relaxer
momentum
behind
having
it
like
it's
on
the
roadmap
of
Italy,
for
instance,
and
for
good
reason,
I
mean
it
does
make
a
lot
of
things
simpler,
like
in
terms
of
organizing
that
relationship
I
would
say.
Probably
the
biggest
obstacle
is
convincing
ourselves
that
we're
doing
that
for
a
good
reason
and
that
we're
enshrining
something
that
fundamentally
benefits
the
ethereum
protocol
and
durably
so
something
that's
helpful
in
the
in
the
future
as
well
yeah.
A
With
this
conversation,
I
can
say
we
definitely
cannot
see
it
in
Shanghai.
Obviously,
Shanghai
proposals.
Considerations
are
closed.
Probably
we
can
Target
it
for
Cancun,
considering
that
developers
are
convinced
by
that
time.
Maybe
if
some
kind
of
research,
work
or
feedback
from
community
may
be
needed
for
for
having
that
things
affirmed
that
it
is
needed
with
the
community
support.
Probably
we
can
see
that
in
future.
B
Yeah
I
do
think.
Actually
the
community
support,
and
just
generally
like
Community
agreement
with
with
the
ideas,
is
very
important,
because
I
I
feel
it's
a
very
big
change
as
to
how
we
think
of
the
ethereum
protocol
so
yeah.
There
definitely
needs
to
be
quite
a
lot
of
dying
to
go
through
that
change.
B
Feel
free
to
contribute
I'm,
always
like
very
happy
to
to
talk
to
anyone
who's
thinking
about
these
problems,
maybe
in
the
links
then
I
will
share.
You
can
find
we
have
some
open
questions
related
not
only
to
PBS,
but
also
to
things
that
are
maybe
more
tangentially
related,
so
yeah,
if
you
find
that
interesting.
A
Thanks
for
joining
us
today
to
talk
about
PBS,
Mev
boost
and
its
future
plan,
I
suppose
that's
helpful
for
new
and
existing
validators
to
understand
the
concept
and
follow
why
ethereum
POS
shows
Mev
boost
I,
hope
aspiring
depth
will
continue
following
ethereum
research
work
with
robust
incentive
group,
and
on
this
note,
thanks
to
everyone
watching
or
listening
to
this
non-eit
special
episode.
Should
you
have
any
questions?