►
From YouTube: PEEPanEIP #96: KZG Ceremony with Carl Beekhuizen
Description
Topics covered
------------------------
01:21 - Guest’s introduction
04:27 - Scaling
05:34 - (Proto) Danksharding / EIP-4844
06:18 - Checking availability
07:03 - Serving the data
07:37 - Throw it all away!
08:18 - Why KZG?
09:58 - Minimising downsides
11:10 - Quick and Easy
11:53 - More Participants than Powers!
13:48 - Specs first design
15:14 - How does the ceremony work?
17:43 - What could a malicious sequencer do?
19:15 - What does an implementation do?
21:01 - How does the math really work?
24:41 - Anti-sibyl
23:56 - Write your own client
31:52 - Crazy entropy generation
31:10 - Timelines
34:32 - Why choose VLS?
35:27 - What is the idea behind the contribution signature?
38:41 - What is the next sequences and who will design?
41:37 - Who else are invited in the ceremony?
44:42 - attack report
46:40 - Talk on Main Audit
Resources:
-----------------
Slides - https://docs.google.com/presentation/d/1DSo_xY7Nt1WYUMWaQOC1D-0FkJ77L-Hm7S_x6S9X4NU/edit#slide=id.p
Website - https://ceremony.ethereum.org/
GitHub - https://github.com/ethereum/kzg-ceremony
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
carlbeek.eth @CarlBeek | 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
A
A
B
Yeah
sure
that
sounds
like
a
good
plan.
I
got
started
in
the
ethereum
community.
About
four
years
ago,
a
friend
of
mine
had
started
making
a
bunch
of
money
doing
some
weird
crypto
things
which
I
had
no
idea
what
was
happening
there,
but
a
bunch
of
money
sounded
cool
to
me.
So
I
had
to
look
up
what
was
what
this
was
all
about:
anyways,
the
the
more
I
looked
into
this,
the
more
I
was
like
hey.
B
This
is
kind
of
interesting
like
the
the
the
the
there's
some
interesting
problems,
and
this
is
some
cool
protocol
stuff
happening,
but
it
didn't
really
grab
me
and
then
I
saw
all
the
proof
of
stake
work
happening
at
the
time.
There
was
a
lot
of
work
on
EIP
1011,
which
was
the
the
Casper
proposal
for
for
proof
of
mistake
as
a
smart
contract.
Anyway,
the
requirement
was
1500
each
which
I
did
not
have
so.
B
My
I
came
up
with
complicated
solutions
for
basically
what
we
call
today
liquid
staking,
but
the
idea
being
basically
I
could
split
up
those
1500
each
between
myself
and
my
friends
and
as
a
result
of
all
of
this
I
started
solving
a
lot
of
interesting
problems
around
consensus
and
networking,
and
look
looking
at
how
the
the
database
side
of
these
things
work
and
hashing
systems
or
whatever
and
then
very
quickly.
I
realized
I
was
solving
all
the
same
problems
that
were
like
problems
that
needed
to
be
solved
at
the
protocol
there.
B
I
basically
ended
up
making
the
search
into
full-time
research
to
work
on
the
same
things,
but
just
sort
of
one
level
deeper
where
it
was
at
the
actual
protocol
level
and
solving
those
problems.
And
since
then
I've
been
doing
research
with
ethereum
Foundation.
My
specialty
I
guess,
if
I
have
such
a
thing,
is
generally
on
the
like
projects
that
happen
slightly
to
the
side
of
the
protocol.
So
those
kcg
ceremony
is
a
good
example.
A
Getting
into
a
ethereum
blockchain
with
the
interesting
way
like
getting
to
know
more
about
crypto
and
that
gets
to
know
about
protocol
is
really
interesting
and
I.
Remember
the
time
of
the
1500
each
proposal
when
it
was
there
at
the
time
of
when
we
were
having
the
ghost
proposal,
it
was
really
scary,
but
thanks
to
the
development
team,
it's
not
that
big
for
staking.
As
of
now
yeah
right,
so
I'm
super
excited
to
learn
more
about
it.
Maybe
we
can
start
with
screen,
sharing
and
presentation.
B
It
could
be
jpegs
of
your
cat
if
you
so
wished.
The
protocol
doesn't
really
care,
but
the
idea
is
basically
you
take
these
these
blobs
and
we
commit
to
those
blobs
so
much
like
a
we
use
local
trees
or
mocha
Patricia
trees
or
whatever
to
to
commit
to
the
data
on
ethereum
right
now.
A
kcg
commitment
is
just
another
commitment
device
where
we
like
bind
ourselves
to
these
these
data
blobs
and
so
after
a
block
proposer.
B
Does
this
the
validators
who
are
assigned
to
to
look
at
that
block
check
whether
this
data
is
actually
available,
whether
this
data
exists
and
that
this
block
proposer
has
or
the
the
stock
block
proposals
put
the
data
out
there
such
that
everyone
can
download
it?
And
then
the
idea
is
that
if
they,
if
they
see
that
this
is
the
case,
then
only
do
they
attest
to
the
data
which
is
part
of
their
current
roles.
It's
just
like
an
additional
roles
are
validated
and
are
you
going
to
be
like?
B
Do
I
see
these
data
blobs
coming
in
these
data?
Blobs?
Don't
aren't
actually
a
part
of
the
chain
they're
only
committed
to
on
the
chain
and
the
ties
we
have
between
them
come
via
these
the
commitments
and
then
the
validators
saying
that
this
data
is
actually
around
and
once
they've
done.
This
validators
store
this
data
and
make
this
like
provide
this
data
to
whoever
needs
it
to
the
network.
B
So
if
you
are
using
an
L2
and
you
want
to
check
that
your
transactions
on
chain,
you
should
be
able
to
find
your
transaction
in
these
blocks.
The
L2
sequences
are
going
to
verify
that
all
the
state
is
there
and
the
validators
are
just
going
to
provide
this
data
out
of
their
own
local
storage.
Much
like
it
can
serve
blocks
now
to
those
who
who
ask
for
it
and
then
a
little
while
later
the
validators
take
that
data
and
they
throw
it
away.
B
So
they
exp
there's
like
a
sliding
window
of
the
data
that
validators
are
storing
at
any
one
time
and
after
this
sliding
Windows
passed.
They're
no
longer
responsible
for
providing
this
data,
and
the
idea
here
is
to
prevent
the
storage
requirements
for
validators
from
blowing
up
massively
to
handle
this
extra
data
that
come
in
the
the
form
of
these
blobs.
B
B
So
why
do
we
have
like
the
we
have
this
commitment
mechanism
what's
wrong
with
local
trees
or
any
of
your
other
favorite
commitment
things,
and
it
comes
down
to
this
chart
which
metallic
posts
on
each
research
a
while
ago,
plus
like
get
some
modifications
for
myself.
But
the
idea
here
is
basically
that
we
need
a
commitment
scheme
that
is
very
easy
to
calculate
very
easy
to
verify.
B
If
you
that
have
these
other
properties,
if
you
use
hash
based
commitments
that
just
become
very
expensive,
the
downside
for
kcg
here
is
that
it
requires
a
trusted
setup,
so
trusted
setups,
they're,
never
good.
If,
ultimately,
if
we
could
avoid
them,
that
would
be.
That
would
be
ideal.
They
they
add
a
text
surface
to
your
protocol
and
they
change
some
of
the
trust
assumptions
you've
been
thinking
about.
B
B
So
we,
the
the
goal
here,
is
that
we
want
to
change
this.
Like
trust
assumption
from
like
some
of
the
early
trusted
setups,
where
it
was
six
people
all
together
in
a
room
who
did
some
fancy
computation
and
there's
some
some
interesting
stories
to
hear
about
some
of
the
early
Z
cash
setups,
but
basically
they're,
where
there
were
some
few
people
who,
in
essence,
you
had
to
trust
that
trust
them
to
try
to
scale
this
out
to
reduce
these
factors
so
like
what
are
the
major
concerns
here.
B
B
B
So
this
interface
is
at
ceremony.etherian.org,
and
from
here
you
can
basically
just
in
about
15
minutes
or
so
contribute
yourself
to
the
ceremony
with
like
no
prerequisites
in
terms
of
signing
up
or
needing
to
basically
think
through
any
of
this
process
and
the
idea
here
being
that
we
want
to
make
it
as
easy
as
possible.
We'd
love
to
have
like
anyone,
no
matter
their
technical
background,
if
they're
a
part
of
the
ethereum
community
or
not
just
part
of
a
wider
Tech
Community.
B
B
B
The
transcript
is
going
to
be
mostly
like
here's,
the
list
of
people
who
contributed
and
then
like
a
tiny
like
adults
like,
oh
by
the
way,
here's
the
app
that
you
actually
cared
about
because
then
it's
not
like
you
have
to
trust
a
single
people
or
a
few
people.
It's
that,
in
order
for
something
to
have
gone
wrong
in
the
ceremony,
we
needed
to
have
basically
everyone
collectively
coming
together,
recombining
their
secrets
and
reconstructing
this
like
Master
secret
together
and
with
this
many
people.
B
It
becomes
very
hard
for
this
actually
to
be
a
thing
that
happens.
There's
just
you
need
a
single
person
to
be
to
be
honest
amongst
this,
this
group
and
then,
as
long
as
they
have
installed
the
secret.
Then
we
know
that
this
entire
ceremony
is
safe,
but
it
goes
one
further
and
that's
because
like
for
the
average
person
participating
in
this
website,
you
don't
just
see
the
secret,
the
browser's
sampling,
some
Randomness.
In
the
background,
it's
it
it's
doing
a
bunch
of
extra
calculations
for
you.
B
The
second
thing
is
that,
like
code
is
never
perfect,
so
although
the
that
implementation
I
refer
to
earlier,
has
been
audited
and
has
had
a
lot
of
eyes
on
it,
it
you
you
never
have
guarantees.
So,
just
like
it,
the
ethereum,
El
and
cl
are
written
from
like
a
Spec's
first
standpoint
like
first,
there
are
some
specifications
and
then
ultimately,
the
multiple
clients
that
implement
it.
B
So
we
like
eliminate,
like
I,
don't
know,
there's
a
Swiss
Cheese
model
of
security
where
it's
like.
If
you
have
a
failure
at
one
level,
one
hole
in
the
Swiss
cheese
and
a
fail
and
another
level
where
they're
like
different
holes,
they
don't
necessarily
line
up
so
by
stacking,
multiple
different
ways
that
it
could
fail.
If
they're
like
uncorrelated
or
low
correlation,
then
you
make
the
chance
of
like
all
these
cascading
failures
happening
very
low.
B
So
the
idea
here
is
that
each
individual,
ultimately
the
ceremony-
wants
to
generate
a
secret
and
then
some
like
mathematical
variations
on
the
secret
and
the
secret
is
just
a
number.
So
the
way
this
works
is
we
get
multiple
people
to
each
contribute,
their
their
own
bit
of
Randomness
to
the
number.
B
B
Oh,
the
secret
is
30.,
so
these
secrets
aren't
like
we
don't
actually
have
the
a
secret
as
a
plain
text
number,
it's
a
a
an
elliptic
curve,
Point
on
the
the
the
BLS
12
381
curve-
and
this
is
this-
is
the
fundamental
to
how
kcg
commitments
work.
Is
they
require
these
committed
collection
of
points?
And
then
this
is
like
what
the
output
of
the
ceremony
is
so
by
having
everyone
mix
in
their
their
little
bits
of
the
secret.
B
So
there's
one
other
step
here
and
that's
that
in
this
naive
or
in
this
previous
thing,
is
that
the
final
person
or
some
of
the
earlier
people
could
do
the
math
incorrectly
or
try
cheat
or
delete
someone
else's
contribution,
or
something
like
that.
So
we
have
this
role
called
the
sequencer
who's.
This
centralized
entity
who
checks
that
all
the
files
and
contributions
are
correct.
B
Foreign,
so
I
mean
I
mentioned
the
scary
centralized
word
and
like
now
it
sounds
like
the
sequencer's
got
all
sorts
of
powers.
So
what
can
the
sequencer
actually
do
and
in
essence
the
like
the
primary
malicious
things
I
could
do
outside
of
just
crashing
or
censoring
contributions,
so
you
could
try
you,
you
could
contribute
and
everything's
going
fine
and
then
you
you,
try
to
send
it
to
the
secret
Zone
and
when
it
comes
to
the
verification
step,
the
sequences
like.
Oh
no,
there
was
a
mistaking
a
contribution.
B
It's
not
valid
when,
when
an
actuality,
it
was,
it
was
valid
and
the
second
thing
the
sequencer
could
do
is
it
could
just
simply
not
give
you
a
turn
when
you
try
contribute
the
the
the
the
the
the
sequence.
I
could
just
never
allow
you
to
to
do
so,
but
they
are
like
two
easy
mitigations
to
this,
and
one
is
that
this,
whenever
you
hand
over
some
information
to
the
sequencer,
the
sequencer
like
returns
and
acknowledgments,
which
signs
this.
B
So
if
the
sequencer
were
to
like
say,
oh
what
you
sent
me
is
invalid.
You
can
give
them
this.
Well,
you
have
a
you,
have
a
proof
that
the
sequencer
was
lying
and
in
the
second
case,
which
is
that
the
sequence
is
just
not
giving
you
a
turn
and
they're
trying
to
censor
you
from
from
starting
this
whole
process.
We
can
so
can
navigate
this
by
just
starting
with
a
you.
B
So
I
mentioned
that
they're.
These
implementations
that
are
like
speaking
to
the
sequencer,
to
try
like
contribute.
So
what
exactly
is
like
happening
through
this
whole
process?
So
the
first
step
is
to
sign
in
and
that's
just
because
we
need
like
a
sort
of
anti-sible
thing.
We
don't
want
the
the
entire
ceremony
to
just
be
basically
tossed
by
everyone
trying
to
contribute
to
fake
files,
so
there
are
two
options
available
for
this.
B
One
is
sign
in
with
ethereum,
where
you
can
join
and
contribute
with
with
your
normal
ether
account,
which
is
I,
think
how
most
of
us
have
established
identities
within
the
ethereum
space
and
then
the
second
is
GitHub.
If
for
like
those
community
members
who
maybe
don't
have
an
ethereum
account
or
don't
meet
the
minimum
nonce
requirements
which
I'll
touch
on
in
a
sec.
B
So
what
does
this
actually
look
like?
We
have
this
notion
of
a
Lobby,
so
after
you've
signed
in
you'll,
get
put
in
a
Lobby,
and
then
everyone
in
the
lobby
just
keeps
asking
the
sequencer
if
they
can
contribute.
So
here,
as
you
can
see,
the
the
yellow
person
is
actually
it's
their
turn
to
contribute
to
their.
B
So
this
is
sort
of
how
we
handle
lots
of
people
wanting
to
contribute
when
those
those
process
was
like,
fundamentally
sequential
the
the
contributing
to
the
ceremony,
so
I
mean
earlier
on.
I
had
the
this
nice
example
with
with
mixing
of
colors
and
people,
but
like
how
does
the
math
really
work?
That
seemed
a
little
bit,
overly
simple
and
and
what
are
the
actual
requirements
here?
B
So
what
we
actually
the
the
the
ceremony
and
many
ceremonies
established,
what
are
called
piles
of
Tau
the
the
piles
of
Tau
here
are
some
secret.
This
is
the
number
we're
referring
to
earlier
and
then
the
secret
Square,
the
secret
squared
and
then
cubed,
etc,
etc,
and-
and
we
have
many
like
a
series
of
of
powers
of
this
all
encoded
as
all
of
the
code
points,
and
so
this
is
like
the
ultimate,
the
ultimate
thing
that
we'd
like
to
get
out
of
the
transcript.
B
So
how
do
we
go
from
this
requirement
to
what
we
actually
have?
Which
is
you
so
the
sequencer
will
send
you
the
files
when
it's
your
turn
to
contribute,
and
then
you
or
more
specifically,
your
computer,
will
come
up
with
a
secret.
This
is
some
some
Randomness
that
you're
gonna
hopefully
destroy
afterwards,
and
then
you
perform
some
mathematical
operations,
so
you
here
calculate
powers
of
your
secret,
just
normal
integer
multiplication
that
you're
doing
here
well
find
that
filter
anyways
and
then
you
do
element
wise
multiplication.
B
So
you
take
your
the
first
part
of
your
secret
and
multiply
it
by
the
first
Power
in
the
powers
of
Tau
and
continue
this
on
until
you've
multiplied
each
of
these
out
and
this
basically
because
elliptical
math
is
linear,
multiplies
your
secret
and
mixes
your
secret
with
the
the
secrets
that
came
before
you,
the
existing
powers
of
top-
and
this
is
how
I
can
sort
of
build
on
these
Powers
that
that
already
exist,
and
then
you
send
you'll.
You
send
these
piles
of
tile
back
to
the
sequencer,
which
verifies
things.
B
This
looks
awfully
complicated
and
there
are
a
few
pairing
equations
all
over
the
place
here,
but
the
the
top
is
just
information
on
what
what
you
basically
send
back
the
sequencer
and
then
the
sequencer.
Using
this
information
checks.
Two
things
there's
one
is
the
update
check.
So
like
did
you
take
or
honestly
take
the
the
powers
that
were
given
to
you
and
mix
in
your
secret
correctly,
and
then
the
second
Checker
does
is
like?
Are
these
actually
powers
like?
Is
this
consecutive
x
x,
squared
x,
cubed,
moving
off
from
there.
B
One
further
complication
is
that
does
the
ceremony
is
not
actually
one
ceremony,
but
it
consists
of
four
sub-ceremonies,
so
it's
four
ceremonies
parading
as
one
as
a
user
of
the
ceremony.
This
is
totally
opaque
to
you,
but
under
the
hood,
this
is
because
one
day
in
the
future,
we
may
wish
to
further
scale
the
data
that
that
that
we
make
available
via
these
via
4844
or
when
you
scale
up
to
full
dark
shotting,
and
so
we
may
just
simply
need
more
points.
B
More
elliptical
points
to
perform
this
math
and
this
this
ceremony
is
so
cheap
and
easy
that
we
can
just
have
more
than
we
need.
So
ultimately,
in
the
the
4844
specifications
right
now
you
only
need
4096
points,
but
no
matter
how
what
our
requirements
are.
We
should
be
good
because
we
have
another.
He
orders
of
magnitude
here
available
to
us.
B
So
like
what
what
what's
up
with
the
the
sign
in
earlier
and
that's
that
we
need
some
kind
of
anti-civil
mechanism
as
I
mentioned.
If
we
just
if,
if
there
were
no
requirements
and
anyone
could
just
walk
up
and
join,
then
we
could
have
like
Bots,
just
spamming,
the
sequencer
being
contributing
and
preventing
legitimate
users
from
Black
uoi
from
getting
involved
in
the
ceremony.
B
But
basically
the
requirement
here
is
that
you
have
at
the
launch
of
the
ceremony
on
Friday
the
13th.
You
have
a
month
of
at
least
four,
and
if
you're
going
to
use
GitHub,
we
just
check
that
your
account
was
created
before
Friday.
The
13th
for
for
similar
reasons
and
the
goal
here
is
not
to
be
perfect.
So
we
could
have
someone
contributing
five
times
or
ten
times
if
they
had
many
accounts
lying
around,
but
it
just
needs
to
be
good
enough
that
it
isn't
entirely
full
of
these
bam.
B
B
B
B
So
when
we
put
this
in
the
the
final
output
of
the
ceremony,
which
is
a
this,
what
we
call
the
transcript,
we
can
see
that
Puja
did
participate
and
we
can
prove
that.
It
really
was
her
because
she
signed
this
contribution
with
her
ethereum
address
and
this
if
she
she's
the
only
one
who
could
have
made
this
claim,
because
she
has
she.
She
has
this
account
and
the
the
the
the
counter
to
that
is.
B
B
This
the
ceremony
so
to
touch
on
some
other
aspects.
We
are
really
trying
to
encourage
people
to
write
their
own
implementations.
So,
as
mentioned
this
came
from
specifications
first,
it
really
is
like
we've
stripped
out
all
of
the
complexity.
There's
no
weird
funny
ZK
math,
it's
a
few
elliptical
multiplications
and
that's
it.
B
So
it
really
is
a
simple
thing
to
implement
to
highly
encourage
anyone.
Who's
who's
got
some
coding
capability
to
just
write
one
for
themselves,
because
it's
easier
to
trust
then-
and
the
second
thing
is:
we
offer
grants
to
people
who'd
like
really
like
to
to
dig
in
further
and
maybe
maybe
write
a
more
robust
implementation
with
better
documentation
to
try
offer
alternative
ways
for
other
people
to
contribute,
or
maybe
in
some
specialized
language,
etc,
etc.
B
I've
even
been
encouraging
people
to
to
roll
their
own
crypto
in
this
case,
which
is
not
something
you
should
ever
do.
But
what
I
mean
here
is
writing
your
your
own
crypto
implementation,
because
a
it
would
be
a
fun,
a
fun
thing
to
do:
I
think
for
many
people
and
B.
It
can't
really
go
wrong
because,
if
you're,
if
there
is
a
bug
in
this
Beyond
leaking
your
local
secret.
But
if
there
is
a
bug
in
your
comp,
your
contribution,
then
the
sequence
will
just
be
like
hey.
B
B
From
Mario
Hubble,
who
is,
has
got
these
accelerometers
inside
of
a
little
ball
that
he's
going
to
give
to
his
cat
Who's
Gonna
Knock
it
around
a
bit
and
he's
gonna
use
his
cat
to
to
create
some
entropy,
so
all
sorts
of
weird
and
fun
ideas.
So,
if
you'd
like
to
do
that
would
also
be
happy
to
to
fund
that
those
those
kinds
of
projects
and
help
people
Workshop
ideas
for
that
as
well.
B
B
So
if
you
want
to
get
a
grant
in,
please
apply
before
then
and
then
after
two
months,
so
on
the
13th
of
March,
two
months
off
the
start
of
the
ceremony,
we're
going
to
close
the
ceremony
for
the
most
contributions
and
then
we're
going
to
allow
for
people
doing
these
weird
and
wacky
contributions
to
do
their
thing
in
case.
We
need
to
allocate
them.
B
So
the
idea
here
would
be
that
as
soon
as
the
client
devs
indicate
that
they're
ready
for
fight
for
Photoshop
with
one
month
or
so
lead
time,
we
can.
We
can
shut
down
the
the
ceremony
and
give
this
final
output
to
the
client
there,
so
they
can
make
it
into
the
clients
and
yeah
I.
Think
that's
that
there's
all
the
slides
I
have.
A
In
the
beginning
of
the
presentation
for
the
concept,
because
I
think
that
the
concept
behind
is
anyway
is
way
more
encrypted
for
novice
users
and
yeah.
Also
thanks
for
sharing
the
grand
information,
I
hope.
The
QR
code
that
you
have
shared
in
the
slides
will
provide
more
information
about
how
to
apply
for
the
grant.
A
B
That
is
a
good
question.
I
think
it's
because
we
have
a
very
good
implementations
of
it.
Already,
it's
already
baked
into
the
consensus
layer,
clients
I
mean
there
are
some
some
alternative
curves
which
which
could
work
really
nicely
here
as
well,
some
of
them
potentially
a
little
bit
faster,
but
I,
don't
think
we
have
like
the
robust
infrastructure
and
it
would
potentially
delay
things
to
do
that.
D
Did
I
have
a
question,
of
course:
okay.
Thank
you
for
the
beautiful
presentation.
First
of
all,
I
wonder
what
is
the
idea
behind
the
contribution
signatures
I
mean
why
the
only
material
measures
generate
ecdc
signatures
on
contributions.
Why
do
not
the
other
users
with
GitHub
credentials,
sign
the
their
contributions.
B
D
That
the
contribution
I
mean
I,
I,
checked
the
contribution
file
and
I
saw
that
the
signatures
are
stored.
Taken
in
the
in
contribution
file
and,
however,
it
has
noted
the
spec
in
the
spec
that
the
signatures
are
recommended.
It's
not
an
obligation
as
far
as
I
understand,
so
why
it
is.
Is
it
so?
Why.
B
B
So
it's
just
supposed
to
encourage
people
to
be
able
to
write
their
own
implementations
and
also
that
the
some
of
the
the
the
crypto
libraries
that
exist
out
there,
particularly
in
the
academic
space,
don't
necessarily
support
these
signatures.
So
it
just
made
it
easier
for
everyone
to
not
require
it.
And
then,
if
you
decide
to
write
your
own
implementation
and
go
down
that
route,
then,
like
you,
you
know
you're
doing
this
at
a
slightly
reduced
level
of
being
able
to
prove
to
other
people
and
that
this
was
your
contribution.
D
B
That's
enough
time
that
if
there
are
problems
with
a
sequencer
that
we
should
be
able
to
debug
it
or
point
it
out,
at
least
as
a
community
and
then
handle
those
use
cases,
so
the
additional
complexity
of
passing
around
between
multiple
sequences
just
didn't
seem
to
be
worth
it
I
see.
Thank
you
very
much
sure.
Thanks
for
questions.
C
I
have
an
extra
question:
have
you
heard
about
any
other
project
that
is
planning
to
Leverage
the
powers
of
Tau
that
will
be
calculated
for
their
own
project
so
like
considering?
They
need
the
same
kind
of
amount
of
powers?
Maybe
they
can
if
they
trust
the
ethereum
ceremony,
which
makes
some
sense,
because
if
they
are
really
on
top
of
ethereum,
you
kind
of
should
do
that,
because
if
that
breaks,
everything
breaks
so
it
may
be
interesting.
B
It
becomes
hard
to
run,
and
so
the
the
approach
of
this
ceremony
is
like
to
have
the
minimum
or
a
fairly
minimal
set
of
points
that
are
only
really
useful
in
this.
This
kcg
context
or
4844
context.
Specifically.
So
it's
like
a
bespoke
set
of
points,
so
people
are
welcome
to
use
it.
I
think
it'd
be
a
very
sensible
thing
if,
if
it
matches
someone
else's
use
case
but
I
think
outside
of
these
kind
of
commitments,
I
think
it'd
be
fairly
hard
to
to
do.
A
All
right,
I
think
I
have
a
few
more
questions.
Those
are
not
very
complicated,
I
mean
like
not
very
technical,
but
on
a
high
level.
I'm
just
curious
to
understand
about
a
few
details
of
it.
You
mentioned
about
a
ceremony.
Participants,
ethereum
client
developers
are
obviously
the
invitees
of
that
who
else
is
invited
to
participate,
but
are
you
looking
for
infrastructure
provided
from
ethereum,
or
do
you
think
that
participations
of
other
L2
blockchain
developers
could
add
value
here.
B
So
when
I
was
using
the
word
participant
I
mean
someone,
who's
gonna
contribute
towards
the
ceremony.
So
that's,
hopefully
anyone
within
the
ethereum
community,
but
in
terms
of
like
writing
your
own
implementation,
then
that
really
is
anyone
who
thinks
they
could
extract
some
use
out
of
the
the
ethereum
ecosystem
or
out
of
what
eip4844
could
provide
and
has
an
afternoon
or
so
spare.
B
Nothing
like
this
is
all
as
per
I.
Guess
normal
theorem
protocols
or
standards
is
all
written
in
the
open
all
with
public
specs,
and
there
are
no
special
gating
that's
happening
here,
so
it
really
is
up
to
whoever
would
like
to
to
participate
or
contribute
in
whichever
way
they
want
to
to
get
involved.
That.
A
B
So
I
guess
for
most
people,
that'll
be
the
the
the
the
website
I
mentioned.
So
if
you
go
there,
the
the
process
is
first
to
to
collect
some
entropy
you're
gonna
move
your
mouse
around
a
bunch,
you're
gonna
type
in
some
secret,
and
then
it's
going
to
ask
you
to
to
sign
in
and
there
you're
going
to
connect
one
of
your
bulb.
You
connect
a
wallet
that
meets
the
the
minimum
launch
requirements
and
then
the
the
sequence
is
going
to
do
check
that
you.
B
You
meet
these
requirements,
so
it
is
using
that
that
EIP,
under
the
hood,
as
like
the
the
communication
standard
and
if
you're
writing
your
own,
your
own
implementation,
you're
gonna
you're,
going
to
be
interested
in
something
like
this,
but
signing
with
a
theorem
is
a
very
easy
thing
to
use.
And
it's
part
of
why
it's
nice
to
use
in
this
case
is
that
it's
it's
a
very
simple
auth
system
which,
if
you're
using
pretty
much
any
common
web
tooling,
should
be
able
to
interface
with
that.
A
That's
very
interesting
to
know
like
we
are
trying
to
make
use
of
all
the
available
standards
and
protocol
in
the
ethereum
ecosystem
to
build
up
something
new.
On
top
of
that
for
people
who
aren't
aware
of
4361
sign
in
with
ethereum
EAP,
we
do
have
an
episode
on
a
peep
and
ebook
with
the
authors.
The
episode
number
is
61.,
please
check
out
and
have
more
information
about
The
Proposal.
A
My
next
question
is
about
threats
and
attack.
We
when
we
are
talking
about
trusted,
that
means
there
is
existence
of
threat,
and
you
did
talk
about
some
of
the
breaking
points
for
the
trusted
setup.
I
vaguely
remember
a
recent
tweet
about
an
attack
on
kg's
setup
on
Twitter.
If
you
are
by
any
chance
aware
of
it,
would
you
like
to
share
what
that
attack
report
was.
B
A
A
I
know,
that's
really
good
to
know.
My
next
question
is
about
audit
I
really
remember
about
the
mention
of
art.
It's
like
a
cryptographic,
audit
and
sequencer
audit
I.
Think
I
found
something
on
the
road
map
of
it.
Would
you
like
to
maybe
talk
about
this
scope
or
the
key
highlights
of
these
Audits
and
if
there
were
any
fixes
that
were
done
based
on
their
recommendations,.
B
Yeah
sure
so
we
did
two
two
audits
and
the
first
of
the
second
idea
being
there
that
was
a
crypto
audit
and
to
dig
into
the
specifications
we
didn't
really
have.
We
had
some
early
code
written
of
implementations,
but
the
the
idea
here
was
to
to
to
to
find
the
bugs
in
the
specs,
because
if
those
exist
then
that's
going
to
translate
into
to
all
high
levels
of
the
stack
and
they've
audited
a
bunch
of
Ceremonies
in
the
past.
So
it
seemed
like
a
good
team
to
do
so.
B
They
didn't
find
any
any
major
issues
there,
some
some
some
minor
typos
and
some
some
bad
references
and
that
kind
of
thing.
But
in
terms
of
the
actual
core
logic
there
wasn't
anything
that
needed
to
be
addressed
and
then
the
second.
The
second
audit
which
finished
up
at
the
end
of
last
year,
was
by
Sigma
Prime,
who
have
a
bunch
of
experience
or
auditing,
crypto,
stuff
and
and
rust
code,
which
is
what
the
sequence
is
written
in.
B
B
So,
as
I
said,
the
the
smallest
ceremony
size.
We
have
is
4096
points.
But
if
you,
if
you
were
to
do
a
hypothetical
ceremony,
we're
using
only
one
point,
then
there
was
like
an
off
by
one
error
that
could
cause
the
ceremony
to
crash
and
cases
like
that
which
are
like
good
to
address
for
a
robot
robust
code
base.
But
none
that
caused
us
to
worry
about
the
the
security
of
this
or
the
safety
of
the
ceremony.
A
B
I
think
mostly
that
this
is
this
is
something
to
to
seriously
consider,
because
this
this
is
an
important
part
of
ethereum's
security
model
moving
forward,
so
I
think
it'd
be
cool
if
everyone
thinks
about
like
how
this,
how
this
plays
into
that
and
whether
they
feel
comfortable
with
the
way
this.
This
is
approaching.
B
I
hope
that
this
is
happening,
I,
think,
on
top
of
that,
the
making
sure,
as
many
people
as
possible,
contribute
so
contribute
yourself,
encourage
those
around
you
to
do
so
so
that
we
can
have
many
people
contributing
to
like
help
distribute
these
this.
This
trust
amongst
as
many
people
as
possible
and
that
that
should
hopefully
lead
us
to
have
the
most
secure
ceremony
possible,
which
I
think
is
something
that
everyone
wants
here
and
yeah
I.
Think
that's
that's
the
summary
of
everything.
A
Like
many
others
out
in
the
world,
I
have
my
eyes
peeled
on
blockchain
scalability
solution.
Best
wishes
for
the
KCC
ceremony,
starting
on
13th
January
Carl,
thanks
for
joining
us
today
to
talk
about
kgz
ceremony
and
its
roadmap.
I
suppose
that's
helpful
for
many
developers
and
hobbies
who
would
like
to
make
a
contribution,
and
on
this
note,
thanks
to
everyone
watching
or
listening
to
this
non-vip
special
episode.
Should
you
have
any
question?