►
Description
Topics covered
------------------------
01:25 - Guest’s introduction
04:49 - About EIP 4895
05:35 - Potuz's Presentation
05:35 - I want my ETH, what do i do?
08:15 - Validators' Keys
12:24 - Two types of withdrawal credential(Examples)
13:08 - Partial Withdrawals
15:18 - Tooling
18:25 - Full Withdrawals
20:00 - Inside Withdrawals
23:01 - Withdrawal Messages
25:57 - Withdrawal, EL's perspective
27:04 - Some interesting cases
30:10 - Ongoing devnets
35:07 - Hsiao-Wei's Presentation
35:11 - Staked ETH Withdrawals & Testing Agenda
35:35 - Validator status
39:03 - Full Validator Lifecycle
42:58 - Validator and Staker FAQs by Hsiao-Wei
43:13 - What kind of address I can use in the BLSToExecutionChange message?
43:54 - How long it takes to change my withdrawal credentials to execution address format?
45:07 - Can I change my execution address AGAIN after my BLSToExecutionChange was applied?
46:04 - I'm an active validator. How long it takes to be "withdrawable" status?
50:11 - How long it takes to get my partial/full withdrawal if I have set up my withdrawal execution address?
52:06 - When can I withdraw on the mainnet?
53:03 - Progress and Testing
53:18 - Specs of Withdrawal feature
54:03 - Software Components
55:18 - Withdrawals testing approaches
59:45 - CL executable pyspec
1:00:55 - Pyspec as the test vector generator
1:04:00 - Test case discussions
1:07:45 - Q&A
1:07:59 - Is there any client way to schedule a broadcast a pre-signed BLS change withdrawal credential message?
1:11:13 - If a validator is kicked out or forced exited, how will he claim the remaining ETH which I assume is less than 32 ETH will get pushed along with the exit?
1:13:15 - Is there is a provision to validate the withdrawal address before executing?
1:15:40 - Changes done on the engine Epi side?
1:22:33 - How far we are from the launch of public testnets?
Resources:
-----------------
Slides - https://docs.google.com/presentation/d/1ojzvCMVKaXIedNdTk1zWnPC_HmGVCjwgr126m7PY3AQ/edit?usp=sharing
EIP-4895 with Alex S. - https://www.youtube.com/watch?v=CcL9RJBljUs
EIP-4895: Beacon chain push withdrawals as operations - https://eips.ethereum.org/EIPS/eip-4895
https://notes.ethereum.org/@ralexstokes/validator-withdrawals-meta-spec
https://notes.ethereum.org/@marioevz/Withdrawals-hive-testing
https://github.com/marioevz/hive/tree/engine-withdrawals-2
https://github.com/wealdtech/ethdo
Shanghai Playlist - https://www.youtube.com/playlist?list=PL4cwHXAawZxpok0smGmq-dFGVHQzW84a2
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
-----------------------------------------------
Hsiao-Wei Wang @icebearhww | 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
C
A
B
There
aren't
that
many
people
within
our
team
working
on
capella,
because
most
of
the
heavy
work
is
now
moving
towards
other
areas,
but
anyways
so
capella,
even
though
it's
a
light,
Fork
is
a
fork
that
is
being
waited
for
a
long
time
in
this
taking
community
and
we're
excited
to
be
delivering
it
soon.
We
are
at
a
stage
that
in
just
a
few
weeks
we're
gonna
be
opening
large
chestnuts
for
the
public
to
to
test
withdrawals,
and
it's
very
exciting
that
this
is
just
really
the
conclusion
of
the
merch.
A
Definitely
it
makes
a
lot
of
sense.
Merge
was
just
the
beginning,
I
hope
to
see
ethereum
Community
growing
like
this
and
more
validators
and
staking
coming
on
the
network.
Well,
thank
you
both
for
sharing
about
yourself
before
we
dive
into
today's
presentation.
Let
me
provide
some
more
context:
EIP
4895
Beacon
chain
push
withdrawal
as
operations
was
selected
in
April
this
year
to
be
the
way
forward
for
withdrawing
staking
rewards.
A
After
a
series
of
discussion
on
different
proposals
on
withdrawals
in
client
devs
meeting
early
this
year,
it
is
recommended
to
watch
Peep
and
Eve
episode,
68
an
overview
of
EIP
4895
with
Alex
Stokes
to
have
a
better
understanding
of
the
proposal
and
definitely
following
what's
going
on
with
it.
With
that
said,
I
don't
want
people
to
be
waiting
anymore
for
withdrawal,
testing
updates
so
without
further
Ado.
Let's
peep
in.
B
So,
okay,
so,
let's
start
with
this
I
am
an
operator
I've
been
staking
for
over
two
years,
since
the
beacon
change,
launch
and
I
have
I
mean
most
most
takers
now
will
have
over
like
seven
e
extra
that
are
not
being
used
for
anything
they're,
not
securing
the
network,
they're
just
sitting
there
in
the
beacon
chain,
and
they
would
want
at
least
that
back.
So
what
do
you
need
to
do
to
get
your
teeth
back
either?
B
If
you
want
just
the
excess
over
your
32
deep
or
if
you
want
everything
that
you've
got
in
well,
you
may
need
to
do
two
things.
You
may
need
to
do
nothing,
so
there
are
two
different
areas
to
work
with
Raul.
One
of
them
is
to
say,
I
want
my
money
to
this
address.
This
has
to
be
an
execution
address
outside
of
the
vehicle
chain
and
the
other
one
is
I.
Want
all
my
money
or
I
just
want
my
excess
money.
B
B
So
the
first
message
that
the
one
that
you
need
to
tell
you
may
need
to
send
to
tell
your
Deacon
I
want
to
have
my
money
in
this
address
is
to
send
a
message
that
is
called
a
BLS
to
execution
change,
that
BLS
to
execution
change
is
a
message
that
is
signed
with
your
withdrawal
address
and
I'll.
Tell
you
in
a
second.
B
What
is
that,
and
if
you
want
all
of
your
money,
then
you
need
to
exit
your
validator
first,
and
this
is
a
message
that
is
called
a
voluntary
exit
and
is
signed
by
the
validating
key,
which
is
a
different
key
than
the
withdrawal.
Key
I
say
that
you
may
need
to
send
these
messages
and
you
may
need
to
to
to
exit
your
validator,
but
you
may
not.
For
example,
you
may
already
have
an
execution
address
in
your
validator
as
a
withdrawal
address,
and
you
may
already
have
exited
your
validator.
B
B
This
is
the
lucky
first
validator
that
actually
proposed
the
first
block
in
the
beacon
chain
two
years
ago
and
here's
one
key.
You
can
just
look
on
Beacon
chain
on
beaconchat.in
and
you're.
Gonna
see
this.
This
is
the
key
that
most
people
would
look
at
in
a
validator.
This
is
the
public
validating
key,
but
today
we're
gonna
be
mostly
interested
in
a
different
game,
which
is
the
withdrawal
key
and
the
withdrawal
key.
Not
even
the
public
key
is
available
now
to
anyone.
B
You
can
see,
for
example,
when
you
go
to
the
deposit
that
made
this
validator,
so
you
can
see
here
in
etherscam,
for
example,
this
is
the
Public's
key
and
it's
the
same
public
key.
That
I
showed
it
for
the
other
validator
up
there
for
the
same
validator
up
there
and
then
the
withdrawal
credentials
is
this
hex
string
of
32
bytes.
B
So
this
withdrawal
credential
in
this
case
is
a
hash
of
the
actual
public
key,
the
public
withdrawal
credential,
except
that
the
first
byte
here
and
that's
the
important
one.
The
first
byte
is
changed
to
be
a
zero
instead
of
whatever
the
hash
would
have.
So
this
zero
that
I
see
here
is
a
prefix
and
this
prefix
being
zero
means
that
this
address
is
a
BLS
address
and
it's
not
an
ethereum
address
a
withdrawal
credential
for
this
valuable.
B
B
Okay,
so
each
validator
has
two
pairs
of
keys,
two
pair
of
publics
and
and
private
Keys.
One
of
them
is
in
your
validator
client
and
it's
hot.
The
validator
client
has
that
part
that
private
key
to
sign
messages.
You
sign
every
ebook,
you
sign
an
attestation
and
from
time
to
time
you
sign
a
proposal
and
that's
that
signature
is
being
done
with
the
validating
key.
This
is
a
pair
of
BLS
Keys.
B
B
If
you
have
the
public
key
for
that
one
and
you
hash
it,
and
you
change
that's
going
to
give
you
32
bytes
instead
of
48,
and
you
change
the
first
byte
for
a
zero
and
that's
what
your
withdrawal
credentials
are.
So
the
withdrawal
credentials
that
are
public
facing
feature
object
of
your
validator
consists
of
the
hash
of
the
withdrawal
public
key,
and
it's
just
changed
the
first
byte
for
a
zero
to
set
the
signal
that
it
is
a
BL
SQ.
B
But
there
are
other
validators,
for
example,
rocket
pool
validators
would
have
an
an
ethereum
address
already
as
a
withdrawal
credential
since
with
reward
credentials
were
supposed
to
be
these
32
bytes
that
were
given
by
a
hash,
but
ethereum
addresses,
consists
of
20
bytes.
Then
the
way
we
signal
that
with
our
credential
actually
mean
it's
actually,
an
ethereum
address
is
by
putting
the
first
byte
to
be
a
one.
So
this
is
going
to
be
a
call
on
execution
prefix.
B
Then
we're
going
to
choose
filling
with
blank
zeros,
so
this
is
11
zeros
so
that
when
you
add
the
two,
the
remaining
20
bytes
of
your
return
address.
This
is
the
full
withdrawal
credentials,
so
there
are
some
withdrawal
credentials
of
some
validators
that
already
are
pointing
towards
an
ethereum
address.
Here's
an
example:
this
is
the
first
validator
that
is
on
top
is
the
body
data
that
I
showed
you
before.
It
has
a
BLS
with
rubber
credential,
but
this
other
validator.
B
This
is
a
validator,
and
that
is
the
first
validator
after
the
400
000
mark,
and
this
validator
already
has
an
execution
with
our
credentials.
You
can
see
it
here
in
the
withdrawal
condition
it
starts
with
a
one.
It
has
11
zero
bytes,
and
then
we
set
the
address
that
consists
of
20
bytes,
so
this
validator,
for
example,
if
you
were
to
exit
to
exit
today,
then
when
capella
hits,
all
of
its
balance
is
going
to
go
to
this
address
here
in
the
ethereum
execution
chain,
all
right.
So
let's
talk
about
now.
B
What
type
of
withdrawals
we're
gonna
see?
Well
all
validators
that
already
have
an
execution
address
like
the
previous
one,
that
I've
shown
you
or
all
those
that
already
changed
their
address.
When
capella
after
capella
you're
going
to
be
able
to
change
your
address,
then
you
don't
need
to
do
anything
else
every
once
in
a
while.
Currently,
this
is
going
to
be
every
five
days
more
or
less
you're
gonna
get
everything
that
you
have
on
excess
of
32
eat
to
your
address,
and
this
is
going
to
be
deposited
automatically
in
your
execution
address.
B
B
If
you
do
not
have
an
execution
address,
an
addiction,
prefect
and
withdrawal
prevention
you're
going
to
have
to
change
your
address.
If
you
want
to
have
your
access
balance,
you
can
do
this
whenever
you
want.
You
can
send
this
message
whenever
you
want,
and
you
need
to
send
this
message
that
is
called
a
BLS
efficient
change.
So
here's
such
a
message
we're
using
this
to
test
on
devnets.
This
is
changing
the
the
withdrawal
credentials
for
this
validator
index
number
three
that
has
a
given
public
BLS
keys.
B
That
I
just
chose
them
to
be
one
and
then
in
one-
and
it's
just
a
signed
message-
the
important
thing
about
this.
This
message
is
that
it's
signed
with
your
withdrawal
key,
since
you
can
only
change
addresses
from
a
validator
that
has
a
BLS
address,
then
that
means
that
that
validator
has
appear
as
a
BLS
pair,
so
it
has
to
have
access
to
the
withdrawal
private
key
and
that
private
key
has
never
been
online.
Hopefully,
so
you
probably
want
to
keep
it
that
way.
B
You
really
want
to
sign
this
message
still
offline
in
an
air
gaps
computer.
So
let's
go
to
that.
We'll
need
tooling
to
sign
these
messages.
Most
of
people
that
work,
careful
enough
did
not
create
their
mnemonic
in
a
computer
that
was
connected
to
the
internet,
so
hopefully
their
mnemonic
was
never
exposed
to
the
internet
and
we
want
to
keep
it
that
way.
So
no
client
that
I
know
of
no
none
of
the
consensus
clients
that
I
know
of
will
actually
Implement
any
sign-in
picture.
B
So
we're
gonna
rely
on
external,
tooling,
currently
Etho,
which
is
there
is
the
JT
Hub,
the
GitHub.
You
can
just
check
it.
Can
actually
produce
currently
now
these
messages
and
it
can
actually
broadcast
them.
That's
a
feature
that
I
don't
want
you
to
use
I
want
you
to
use
the
feature
of
signing
your
messages
online
offline
in
an
air
gap
computer
so
that
your
mnemonic
is
never
exposed.
B
B
Once
you
have
this
message
signed
in
an
air
gap
computer,
you
can
actually
just
transfer
this
message
to
your
online
validating
computer
and
broadcast
it
for
the
word
to
see.
So
that's
a
possible
workflow.
You
just
boot
up
your
laptop
with
a
pen
drive
on
Tails,
Os
or
whatever
depends
on
your
level
of
paranoia.
B
You
run
your
tooling
to
generate
this
message
and
sign
it
with
your
private
key.
If
you
have
access
to
your
mnemonic,
the
tooling
would
actually
you're
gonna,
say:
I
want
to
withdraw
validator
number
three,
that
I
created
with
this
mnemonic,
then
tooling,
would
know
which
private
key
to
use
and
which
is
generated
with
the
the
BLS
change
for
that
Val
data.
B
Once
you
have
that
changed,
you
can
just
go
to
your
client,
and
here
client
might
choose
to
implement
a
particular
command
to
assist
you
most.
All
clients
would
just
take
a
simple
curve,
which
is
here
that's
exactly
a
common
that
my
command,
that
we're
using
prison,
for
example,
already
has
a
PR
open
to
actually
have
an
actual
command
that
would
take
your
file
and
assist
you
like
making
sure
that
you
want
to
exit
this
validator
tell
you.
This
is
the
address
that
you're
sending.
B
This
is
a
checks
on
address,
or
not
so
just
add.
Some
s
just
take
some
extra
safeguards
to
prevent
you
from
saying
in
the
wrong
to
sending
the
wrong
message.
I
think
Lighthouse
and
techu
were
thinking
about
doing
the
same.
I
haven't
heard
back
from
other
clients
to
whether
or
not
they're
gonna
have
this
this
functionality.
But
it's
not
really
necessarily
a
simple
car
would
just
send
you
the
file.
If
you
were
careful
enough
offline
signing
this
message.
B
All
right,
so
the
other
kind
of
withdrawals
that
are
there
not
the
ones
that
are
partial
everything
that
is
in
excess
of
32,
is,
if
you
want
everything
back,
if
you
want
all
of
your
money
that
you
have
on
the
beacon
chain.
Well,
then
you
need
to
exit
your
validate,
and
the
funny
thing
about
exiting
your
validator
is
that
this
is
signed
with
a
different
key.
This
is
signed
with
the
key
that
is
already
hot
in
your
validator
client,
so
all
clients
already
can
sign
and
submit
this
for
you.
B
So
this
is
the
command
that
you
would
use
if
you're
using
prism
and
similar
commands,
if
you're
using
a
different
consensus,
layer,
client
it'll,
let
you
choose
which
validator
you
want
to
exit
and
you
just
exit
your
validator.
If
your
validator
already
has
an
execution
address,
then
your
validator,
when,
whenever
its
turn,
is
it's
going
to
completely
withdraw
and
it's
going
to
give
you
everything
that
you
have
in
your
BLS
address
in
your
I'm?
Sorry,
your
execution
address.
B
B
B
Here
this
is,
this
is
the
point
of
view
of
The
Proposal.
Okay,
so
withdrawals
are
messages
that
are
going
to
be
included
in
blocks
and
the
way
they
are
going
to
happen
is
as
follows:
the
proposer
of
a
slot,
so
a
proposal
in
slot
n
plus
one
knows
that
it's
going
to
propose
in
the
next
slot.
So
during
slot
n
is
going
to
ask
its
execution
layer
to
prepare
a
block,
and
in
that
message,
when
the
consensus,
let
your
client
using
the
engine
app
is
sends
the
engine.
B
Api
sends
the
message
to
the
execution
layer
to
prepare
the
block.
It
sends
them
the
full
list
of
withdrawals
that
should
go
in
that
block.
So
the
proposal
just
queries,
it's
vcom.
It
gets
a
list
of
withdrawals.
It
grabs
this
list
of
withdrawal,
sends
it
to
the
execution
layer.
One
slot
before
is
actually
proposing
the
execution
layer
will
grab
these
proposals
pack
transactions
from
users
into
an
execution
block.
We
will
prepare
that
block
and
send
it
back
to
the
proposal.
B
So
when
the
proposer
is
in
his
slot
in
N
plus
one
The
Proposal,
which
is
query
for
the
a
full
payload
from
the
execution
layer,
the
execution
layer
would
send
the
payload
back
with
the
exact
withdrawals.
The
same
withdrawals
that
the
proposal
sent
before
and
this
proposal
just
signs
and
submit
the
block
forever.
B
So
now
what
happens
on
the
validation
side,
this
block
went
through
the
network
and
then
validators
grabbed
this
block
and
they
need
to
attest
to
it
and
check
how
withdrawals
are
actually
enforced.
So
now,
validators
would
get
this
block
with
the
withdrawals
and
the
first
thing
a
validator
does
is
send
the
execution
payload.
What
we
used
to
call
block
in
the
Pro
area
to
the
execution
layer
and
the
execution
layer
grabs
the
list
of
withdrawals
within
that
block.
B
I'll
show
you
in
a
bit
how
a
withdrawal
message
looks
like,
but
it
contains
an
address
to
give
it
and
it
contains
an
amount
on
how
much
it
needs
to
go
to
that
address.
So
this
is
minted
in
the
exection
layer
and
then
the
execution
layer
checks
the
validity
of
the
block.
It
sends
back
to
the
consistency
that
you're
saying
this
block
was
valid
from
my
perspective
and
only
then
the
consensus,
layer,
client,
validator
processes,
the
block
and
says
okay.
B
B
This
is
a
withdrawal
message:
it
contains
an
index.
This
is
just
the
withdrawal
indexes
just
to
keep
an
account
an
accountant.
This
is
a
monotonic
monotonically
increasing
index,
it's
going
to
start
at
zero
and
it's
going
to
increase
by
one
and
each
withdrawal
that
we
see,
so
our
clients
can
check
that
is
sequential.
B
It's
going
to
have
a
validator
index
indicating
the
validator
that
is
actually
withdrawing.
It's
going
to
have
an
address
which
is
an
execution
address.
This
is
a
20
byte
address
that
is
going
to
be
credited
an
amount
and
an
amount
that
is
going
to
be
Credit
in
that
address
again.
I
am
a
conscious
layer.
Developer
so
amounts
for
me
anyway.
This
is
how
the
message
looks
for
a
consistent
layer,
client.
B
We're
gonna
find
the
first
16
of
them
and
just
give
them
back.
We're
Not
Gonna
sweep
forever.
If
we
sweep
for
16
384
validators
and
we
haven't
found
any
well,
we
haven't
found
the
16
withdrawal
validators.
That
is
the
maximum
that
can
be
included
in
the
block.
Then
we're
just
gonna
stop
there
and
just
send
back
whatever
we
have
now.
What
what
does
it
mean
that
a
validator
is
withdrawable
here?
B
Well,
if
you
have
more
than
32
each
and
you
have
an
execution
address,
that
I
can
send
you
money,
then
you
are
withdrawal
and
you're
gonna
execute
a
partial
withdrawal,
except
in
one
case,
which
is
the
case
that
you
can
actually
perform
a
full
withdrawal,
and
this
happens
when
you
have
any
balance
in
your
validator.
That
is
not
zero
and
you
have
already
exited
your
validator
and
already
went
through
the
queue
to
actually
be
withdrawal,
so
validators
that
have
exited
for
some
time
and
have
any
balance
in
them.
B
B
B
Withdrawals
are
not
usual
transactions,
they
are
system
level
operations,
they
do
not
carry
any
gas
cost
and
they
can.
They
are
only
performed
after
when
so
the
execution
layer
gets
a
block,
it
performs
it,
it
sinks
all
transactions
from
users,
it
adjusts
the
state
and
then,
after
doing
all
of
this
it
it
just
carries
the
withdrawals.
It
means
it
to
each
one
of
the
withdrawals
in
the
list.
B
B
Here
are
some
interesting
cases.
Xiao
is
going
to
be
talking
certainly
about
this
I
suppose,
but
this
is
just
to
give
you
an
idea
of,
like
all
of
the
possible
consensus,
breaks
all
of
the
possible
consensus
problems
that
we
may
have
because
of
these
edge
cases,
Slash
validators
can
change
their
messages
and
exited
validators
can
check,
can
send
this
message
to
change,
addresses
fully
withdrawn
validators,
so.
B
Your
validator
you're
fully
withdrawn
you're,
fully
withdrawable
the
suite
passes
by
you,
and
it
gives
you
all
of
your
balance
and
then
circumstances
is
that
you
can
still
get
money
in
the
beacon
chain.
For
example,
you
can
exit
your
validator
whilst
being
in
a
sync
committee
and
if
you're
still
in
the
same
committee
and
your
validator
is
still
active,
even
though
he's
succeeded,
he
can't
sign
messages
for
the
same
committee.
It
can
actually
get
some
eth
on
the
beacon
chain,
even
though
it's
already
withdrawn
everything.
B
So
in
the
next
sweep
it's
going
to
give
you
a
few
way
one
and
it's
going
to
be
an
extra
full
withdrawal.
Another
way
of
Performing
several
full
degree
travels
on
the
same.
Validator
is
just
by
depositing
more
if
you
have
a
validator
that
is
already
withdrawn,
and
you
decide
to
deposit
more
money
in
that.
That's
fully
allowed
we're
gonna
just
give
you
your
deposits
in
the
beacon
chain
and
the
next
week,
you're
gonna
we
throw
a
day
again.
B
There
are
there's
some
use
case
into
signing
messages
before
the
capella
Fork
happens.
If
your
key
was
stolen.
For
example,
there
are
several
cases
of
people
that
are
advocating
to
have
some
kyc
kind
of
system
to
try
to
prove
that
your
key
was
hacked
that
your
we
throw
a
key
that
your
mnemonic
was
exposed.
Then
you're
gonna
be
racing
against
hackers
to
have
your
change
address,
included
in
a
blog
before
them,
and
clients
are
going
to
implement
I'm
hoping
that
clients
are
going
to
implement
this
prism
already.
B
Has
this
and
I
think
I
talked
to
Lighthouse?
They
were
going
to
do
this
I
suspect
all
clients
are
going
to
have
one
way
or
another
of
doing
this,
of
allowing
your
message
to
be
in
their
pulley.
The
operation
pool
before
the
capella
Fork
happens
so
that
you
can
have
large
staking
pools
having
white
listed
changes
of
addresses
of
people
that
they
already
know
or
people
that
they
think
they
know
when
they
they
have
proven
somehow
that
their
referral
credentials
were
stolen
so
that
they
are
included
in
blogs
earlier.
B
B
Currently,
if
you
want
to
I,
can
tell
you
later
how
to
sign
them
for
withito,
for
example,
we
did
to
themselves,
for
example,
but
you're
gonna
need
some
extra
information
like
the
fork
version,
The,
Genesis,
validator
roots
and
a
bunch
of
fields
that
are
needed
to
sign
these
messages
before
the
capella
Fork
happens
all
right.
So
let
me
just
quickly
tell
you
what
is
going
on
on
devnets,
and
this
is
my
last
slide-
I'm
just
going
to
pass
this
to
xiaoway.
B
We
tested
very
early
about
a
month
ago,
already
most
of
the
of
the
features
on
the
consensus.
Like
your
side,
we
tested
a
month
ago,
four
causes
later
clients
all
of
them,
but
Nimbus.
If
I'm
not
wrong
and
two
execution
layer,
clients
get
nevermind
and
we
checked
everything
except
the
P2P
stack,
which
is
an
untrivial
part
of
it.
So
we
didn't
check
that
you
could
actually
broadcast
BLS
execution
changes.
B
We
could
include
them
in
blocks
change
your
address
and
people
would
take
the
blog
and
process
the
BLS
execution,
change,
change
and
and
change
the
address
and
withdrawals
were
working.
Fine.
We
found
some
interesting
consensus.
Bugs
prison
had
a
bug
about
what
I
just
told
you
in
the
previous
slide
that
a
validator
that
could
withdraw
several
times
by
being
in
the
same
committee,
and
there
were
some
some
free
case.
If
you
were
in
a
single
meeting
multiple
times,
person
will
Fork
off
there.
B
After
those
early
death
Nets,
we
were
in
trouble
because
Britain
could
not
start
on
a
debt
net
on
a
post,
merge
estate
and
bezel
could
not
Fork,
so
he
could
not
start
from
phase
zero
and
Fork
into
capella
capella.
The
capella
Fork
is
different
for
an
execution
layer
perspective
because
it's
going
to
be
a
fork
on
a
time
stamp
base
instead
of
on
a
Block
Base,
and
this
is
something
that
wasn't
implemented
in
bezel
at
the
time.
B
So
we
couldn't
test
base
one
person
at
the
same
time,
so
we
had
Barnabas
a
huge
shout
out
to
Barnabas
this
new
hire
of
the
term
Foundation
that
has
been
taking
care
of
the
debt
and
it's
for
all
of
us
and
he
was
running
two
sets
of
web
Nets,
one
with
bezel
included,
one
with
prism
included.
One
was
starting
after
the
merge.
The
album
was
started
before
the
merge.
Now
Casey,
one
of
our
teammates
implemented
a
PR
in
person
to
actually
be
able
to
start
Prison
from
a
merged
State.
B
Now
we
can
test
all
clients
together,
Nimbus
already
joined
the
devnets
and
we're
all
working
I.
Believe
Aragon
is
the
only
client
that
is
missing
on
those
lab
Nets
and
were
fully
featured
or
almost
fully
featured
completely
fully
featured
on
the
consensus
layer
side
we
can
submit
and
broadcast
BLS
changes.
We
we
are
gossiping
them.
We
are
we
throwing
food
withdrawals,
partial
withdrawals
and
also
on
the
execution
side.
There
are
other
changes.
B
Besides,
withdrawals
on
capella
or
Shanghai,
Shanghai
is
going
to
have
eof
on
well,
as
presume
other
people
were
more
capable
than
me
are
going
to
talk
about
these
changes,
but
these
changes
are
already
being
tested
and
I
think
most
of
them
are
already
included.
There
are
some
things
that
are
extra
features.
B
C
So
purpose
has
covered
the
first
part
that
would
draw
functionality
introductions
so
now
I'm
going
to
introduce
to
talk
about
what
it
is
for
from
the
perspective
of
the
data
and
so
to
describe
the
status
of
that
later.
There
are
mainly
four
stages
here.
First
is
that
if
the
waiter
made
a
deposit
message-
and
this
message
has
been
included
to
the
weekend
chip,
but
now
it
will
create
a
new
entry
in
the
radiator
registry,
but
at
the
first
stage
the
radiator
hasn't
been
activated
yet,
and
the
next
stage
is
the
activated.
C
So
if
a
radiator
is
activated,
then
we
say
that
this
spectator
is
has
the
rated
duties
and
they
have
the
responsibility
to
validate
and
send
message,
send
their
attestation
message
to
the
application
Network
and
they
have
to
propose
the
block
Etc
then
so
I
think
one
thing
that
I
want
to
note
to
you
is
that
there
is
a
rate
limited
when
we
are
handled
the
validator
set
update.
So
this
is
for
both
the
security
and
the
performance
considerations
that
there's
a
red
limited
here.
C
The
Chun
rate
is
based
on
how
many
active
radiators
are
in
the
big.
You
can
check
at
the
moment
so
yeah,
so
once
it's
activated,
they
have
to
perform
like
all
the
active
radiators
are
doing
right
now,
and
the
next
stage
here
is:
is
the
excited
so
a
letter,
my
initiate
the
button
Theory
exit
or
they
might
they
may
get
forced
to
exceed
so
the
active
activated
and
exceed
the
both
stages
are
that
they
have
the
same
rate,
which
is
calculated
by
the
beacon,
State
Beacon
State
transition
in
protocol.
C
Yes,
it
is
also
in
protocol
in
the
preset
configuration
and
we'll
also
talk
about
it
later
and
then
next,
let's
see
the
full
Vader
life
cycle.
I
mean
all
this.
This,
the
on
the
right
hand
side
if
there's
a
very
detailed,
a
transition
lower
here
and
think
we
can
just
mentioned.
That
main
point
is
that
about
the
exceed.
There
are
two
approach
one
is
like
I
said
before
the
voluntary
exceed
so
the
weather.
C
So
after
they
get
slash
because
they
had
sent
the
malicious
attestations
like
the
double
Vote
or
the
double
progress
block,
then
they
will
get
stashed.
In
this
case,
the
Predator
also
will
get
close
to
exceed
in
protocol,
and
the
third
possibility
is
that
my
due
to
the
leaking
like
they
didn't
perform,
wait
their
a
test,
a
tester
Duty
for
a
while.
Then
they
will
be
leaking
their
balances.
C
C
Stake
in
pool
to
include
this
message
as
soon
as
possible,
but
if
even
if
it's
included,
it's
all,
there's
also
the
red
limit
here.
Benjamin
athletes
in
the
chat
like
each
of
us
become
black,
can
include
at
most
about
16
POs
to
Extrusion
change
messages,
but
as
I
said,
but
we
are,
we
will
do
the
definite
stress
test
later
so
based
on
that
result,
then
we
might
change
this
preset.
C
Third
question
is:
can
I
change
my
social
address
again
after
I
have
applied
my
POs
to
Extrusion
change
message
so,
unfortunately,
there's
no
interface
to
allow
users
to
do
that.
So
please
make
sure
that
your
pos-2
shouldn't
change
message
is
correct
and
double
check
the
address
again
before
you
send
it
and
okay
one
more
question:
if
not
encoding
the
first
step.
Well,
it
be
required
to
keeps.
C
The
first
question
is
like
so,
if
I
active
by
data,
how
long
it
takes
from
active
weighted
to
withdrawable
status,
so
there
are
like
three
different
kind
that
we
have
to
combine
together.
The
first
factor
is
the
the
time
that
one,
the
editor
to
include
your
voluntary,
exceed
message
in
the
beacon
chain.
C
So
even
if
there's
no
other
queuing
relatives
that
in
front
of
you,
there's
also
a
delay
of
like
a
full
epochs,
so
you
have
to
if
I
mean,
if
there's
no
way
other
way,
data
behind
you
there's
still
this
these
delay,
because
we
don't
want
to
it.
I
mean
List
delays,
because
if
the
values
get
a
c
slashed,
I
mean
if,
like
one
layer
can
exit
inventory
exit
and
then
during
the
time
after
once,
they
do
it
I
mean
after
their
exchange
message,
got
included.
C
C
So
you
can
calculate
that
from
these
parameters
to
see
how
long
it
takes
for
you
to
become
the
withdrawable
radiator
and
the
last
part
is
the
delay
from
a
stated
to
withdrawable.
This
is
also
the
oh
sorry.
This
is
for
the
slash
validator
account
consideration,
I
mistook
to
a
stage
and
yeah.
So
so
think.
C
Next
question
is
like
how
long
it
takes
to
get
my
partial
for
withdrawal
that
I
have
set
after
I
have
set
up
my
withdrawal,
Extrusion
stress,
so
there's
another
preset
called
the
max
withdrawals
per
payload,
so
the
current
value
is
also
set
to
16
withdrawals
per
block.
So
with
the
current
size
of
the
radiator
State,
we
might
get
we'll
get
like
100
and
two
hour
per
cycle,
so
it
will.
If
you
are
waiting
painting
on
your,
we
are
waiting
for
getting
the
partial
withdrawal.
C
It
might
take
that
a
long
time
to
get
your
turn
to
be
paid
in
unchain,
but
also
on
this
number
is,
could
be
changed
because
the
The
Sweep
that
I'll
just
mentioned
before
so
there's
a
case.
If
it's
possible
that
there's
no
no
previous
radiators
are
eligible
to
perform
the
partial
withdrawal,
then
this
number
will
change
in
that
case
and
also
as
soon
as
before
that
this
had
this
kind
of
presets
might
be
changed.
When
we
see
the
deafness
stress
test
later.
C
C
And
the
last
question
is
that,
like
people
want
to
ask
like,
when
can
I
withdraw
out
a
minute,
so
the
previous
code,
desktop
score,
has
decided
that
this
feature
will
must.
We
must
be
included
in
the
next
protocol
upgrades
for
sure
and
the
spec
size
that
is
is
nearing
competition.
So
now
everything
is
just
how
to
make
sure
how
to
make
sure
that
the
software
is
fully.
C
Okay,
then
any
question
before
we
talk
about
the
progress
and
the
testing,
you
know,
okay,
thank
you
and
there's
the
the
the
list
of
the
specs.
That
of
the
that
will
be
changed
for
the
withdrawal
features.
We
have
the
consensus
layer
and
the
execution
layer
by
the
way,
the
consensus
layer
folk
is
called
capella
and
the
expression
layer
book
is
called
the
Shanghai
evoke
and
also
beside
that
we
have
the.
We
have
three
main
API
interfaces,
which
is
the
engine
API
and
Beacon
API
and
Builder
apis.
C
So,
on
the
left
hand
side,
it's
the
CL,
the
consensus,
layer,
Mark
components
and
the
right
hand,
side
the
Extrusion
layer
components,
so
the
Econo
and
you
editor
type
communicated
through
the
Bitcoin
apis
and
we
can
know
the
Extrusion
engine
are
communicated
through
the
engine
apis
so
and
both
the
beacon
you
know
as
its
own
P2P,
Network
and
engine
has.
It
also
has
their
own
P2P
Network.
So
things
we
are
going
to
add
a
view
to
the
beacon
State
and
you
will
also
add
the
feel
to
the
equation
payload.
C
C
So
the
first
one
is
the
static
test
vectors.
So
the
client
is,
they
have
their
own
unit
tests
and
for
the
spec
side
we
have
the
fake
test,
vectors
and
I,
and
while
the
maintenance
of
the
CEO
Pi
space,
so
I
will
talk
about
it
deep
later
and
the
second
price
the
fasting
test.
So
we
have
the
fasting
test
is
to
to
test
the
components
against
the
random
data,
so
Mario's
from
get
gold
student
guest
team.
Okay,
he
maintain
some
process
to
test
the
EVN
and
engine
API.
C
C
For
example,
if
you
can
simulate
the
engine
API,
so
you
can
simulate
a
rendering
API
to
talk
with
the
CEO
clients
and
to
or
to
talk
to
the
El
clients
in
the
test,
and
this
is
very
interesting
testing
suit
that
we
had
before
before
the
merge.
So
the
testing
kudos
to
the
Eastern
Foundation
testing
team.
C
They
provide
this
very
good
testing
platform
for
us
and
now,
after
the
merge,
we
have
like
five
CL
times
and
sticks
at
least
six
yield
kinds,
so
that
leads
to
like
13
cents,
30
different
combinations
only
for
the
kind
only
for
the
clients.
So
the
internet
test
is
very
important
to
both,
not
not
just
the
internal
program
upgrade,
but
also
things.
We
want
the
client
diversity
to
be
improved,
then
these
tests
are
required
for
us
and
the
first
component,
which
is
what
the
plantings
are
currently
working
on,
is
the
definite.
C
C
Okay,
we
are
a
bit
a
tough
time.
I
will
speed
up
so
I'd
like
to
talk
more
about
the
CL
tuition.
It's
suitable
specific
features,
that's
what
I'm
working
on
mainly,
and
so
the
price
back
is
like
it's
a
repository
of
space
and
it's
a
collection
of
the
course
consensus
qualifications.
So
it
all
the
core
consensus.
C
C
To
provide
a
static
test
for
the
kind
things
to
use,
and
this
is
an
example
of
how
we
write
a
pi-spec
test
here,
so
you
can
see
that
we
have
the
basic
fake.
If
you
see
my
mouse
I
think
so,
oh
you
can.
Oh
am
I
display
is
like
it's
a
it's
a
python
module
here,
so
it
could
be
like
base
zero
or
alter
Pediatrics
or
capella
specs,
and
also
we
have
this
pre-state
gear.
C
C
So
we
say
that
okay,
we
want
to
apply
this
side
block
to
the
pre-state
here
and
what
we
will
get
is
the
post
date,
because
let's
stay
transition
function
here
so
yeah,
that's
basically
how
and
what
the
I
think
we'll
get.
They
will
get
the
pre-state
the
post
date
and
the
block
that
we
apply
to
the
state
transition
functions.
C
C
C
Okay,
so
during
the
withdrawal
test,
writings
I
think
so
compared
to
the
the
merge
capitalized
quests,
it's
much
more
singular
than
the
merge,
and
this
just
I
would
say.
There
are
three
different
three
men
consider
points
when
we
are
writing
the
test
against.
Why
is
the
if
we
can
validate
the
basic
verification
of
the
new
message?
C
So
in
this
case,
is
the
site
POs
to
Extrusion
change
message,
and
the
second
is
if,
if
actually
recent
rise
by
others
and
parents
from
presenting
that
it
is
also
important
to
test
the
order
of
the
balance
operating
calculations,
because
we
have
like
different
different
State
transitive
functions,
I
mean
the
different
sub
functions
in
the
state
transition
that
could
change
the
balance
updates,
for
example,
the
from
the
deposit
to
withdrawal
and
all
the
rewards
and
penalty
updates.
So
this
is
important
to
make
sure
that
the
client
teams
are
the
other.
C
C
A
Great
work
done
on
testing
and
for
the
wonderful
presentation,
I
know,
testing
anything
before
putting
into
the
main
net
is
important
for
software
development,
but
I'm
always
carried
away
by
the
extent
ethereum
team
put
into
a
testing
part
I,
think
adding
technical
aspects
from
photo,
slides
and
FAQ
from
shivay
slide
where
A
Plus.
Thank
you
for
sharing
with
us.
Icb
already
have
received
a
few
questions
in
chats,
so
yeah
opening
the
floor
for
questions
from
audience.
If
the
author
can
maybe
read
the
question
and
point
towards
the
guess,
hey.
D
Thanks
for
a
great
presentation
to
both
my
my
main
question
was
around
the
rate
limiting
of
messages
specifically
the
BLS
change.
Credential
operation
I
see
that
there's
a
rate
limit
of
only
16
operations
per
block.
But
if,
if
there's
many
different
gospel
gossip
messages,
is
there
any
rate
limit
to
the
actual
gossip
messages,
thanks
again
for
the
presentations?
D
B
I
can
take
that
one,
because
we
were
discussing
this
not
long
ago,
there's
no
limiting
currently
on
the
gossip
Network,
so
you
can
just
send
and
we
we
can
be
flooded,
but
with
these
messages
I'm,
not
we
don't
really
know.
We
don't
believe
it's
a
worry
that
we
should
have.
These
are
signed
messages,
so
it's
not
that
you
can
actually
float
us
with
with
invalid
messages,
because
they're
not
going
to
continue
on
the
on
the
network.
However,
we
will
test
against
the
Sasa
dos
Vector
on
devnets.
B
D
Yes,
that
was
another
question,
so
if,
if
you
did
mention
in
your
presentation
that
it
is
some
clients
at
least
allow
the
commission
committing
a
BLS
change,
credential
cooperation
early,
if
you,
if
that
client
has
already
committed
it,
will
they
proceed
to
re-gosip
it?
If
you
submit
it
again
and
how
do
you
ensure
that
you're
still
transmitting
that
gossip
yeah.
B
So
we're
not
going
to
broadcast
those
messages,
because
you
cannot
broadcast
those
messages
until
the
fork
happens.
So
there's
not
going
to
be
any
gossip
network
before
the
fork.
But
the
question
is
whether
or
not
those
messages
are
going
to
be
already
in
the
pool
and
by
having
this
message
already
in
the
pool,
then
you
know
that
these
messages
are
going
to
be
starting
to
being
broadcasted
immediately
at
the
fork.
A
All
right,
I
have
a
few
more
questions
that
we
have
collected
earlier,
or
these
are
the
very
basic
questions
for
General
understanding
of
people.
So
my
first
question
is
related
to
withdrawn,
obviously
Pula
partial.
We
know
that
there
are
over
490
000
validators
and
we
have
over
16.6
million
each
state
on
ethereum
blockchain.
My
question
is
from
shiva's
slide.
It
was
mentioned
the
way
of
validated
exiting
condition
in
the
validators
life
cycle.
C
So
after
like
seeds
and
then
they
will
be
with
durable,
then
in
that
stage
they
can
initiate
the
POs
to
Institution
change
if
they
happen
to,
they
might
already
did
before
and
since
then
then,
the
the
process
withdrawal
is
will
be
automatically
performed
in
the
through
in
the
protocol.
So
each
Epoch
will
perform
sorry,
not
apart
which
each
block
then
we
will
perform
the
process
withdrawal.
Then
it
will
go
in
that
function.
They
will
go
through
by
data
registry
field
so
to
check
each
radiators,
not
each.
C
A
That
makes
sense.
Thank
you
for
that.
My
next
question
is
about
change.
The
withdrawal
credential
address.
If
I
understand
correctly,
it
has
to
be
done
by
an
individual
validator.
So
from
Portal.
Slides
withdrawal
is
a
system
level
operation
and
cannot
be
reverted
as
you
mentioned,
and
definitely
we
would
like
to
minimize
errors
on
creating
transaction
or
signing
the
withdrawal
message.
I
wonder
if
there
is
any
provision
to
validate
the
withdrawal
address
before
executing
I
am
taking
it
from
the
context
of
some
facilities
provided
by
safe
life.
A
B
I
I
don't
think
the
protocol
that
we
can
actually
do.
These
are
the
protocol
level
there's
no
way
of
validating
the
mattresses
filed
or
not.
We
can
check
for
balance,
for
example
or
not,
but
but
this
can
be
done
at
the
client
level.
For
example,
some
mitigations
are.
Is
it
check
summed,
for
example,
is
it
is
it
the
right?
B
Is
it
you
have
addresses
that
you
can
type
certain
characters,
capitalize
and
certain
characters
not,
and
this
gives
you
a
way
of
a
check
summing
for
like
mistakes
in
just
copy
pasting,
for
example
and
clients
before
they
submit.
The
message
can
actually
check
this.
They
can
actually
ask
you
a
question
like
wait
with
a
big
sign.
Are
you
sure
you
want
to
send
this
address
these
address?
Is
it
yours
or
not?
We
have
already
implemented
this
within
prism
and
I.
Think
other
clients
are
going
to
follow
through
with
similar
designs.
B
I
am
not
really
sure
if
there's
I
mean
I'm
hoping
if
anyone
opens
an
issue
on
the
prison
side,
we're
gonna
be
happy
to
take
it
I'm,
not
really
sure
we
can
actually
make
more
of
a
change
like
this.
This
is
one
of
the
problems
of
of
blockchain.
That
one
mistake
is
your
key.
Is
your
money,
but
one
mistake
and
you
no
longer
have
the
keys
and
you
no
longer
have
your
money.
A
A
My
next
question
is
related
to
Hive
testing
a
roadmap,
so
the
withdrawal
flow
was
one
of
the
main
point
of
discussion
for
developers
before
48.95.
So
when
we
reached
consensus
that
it
will
be
48.95,
that
means
we'll
be
using
the
push
withdrawal
which
introduces
the
system
level
operation.
I
understand
that
it
is.
It
has
become
more
of
a
CL
Centric
proposal,
but
I'll
be
curious
to
understand
about
the
changes
done
on
the
engine
API
side,
because
48.95
requires
some
The
Hive
testing
document
mentioned
engine
API
execution
client
as
Suite.
A
C
C
I
I
Nash,
surely
because
please
correct
me
if
you
guys
so
like
I,
said
that
there's
the
the
new
field
is
added
to
the
instruction
payload,
so
the
exclusion
there
will
be
like
once
they
get
that
message.
They
will
have
to
do
something
else,
but
I
think
this
is
the
push
approach,
so
the
engine
apis
and
things
pretty
clear,
cling
to
me.
I
mean
not
much
changes
there.
B
So
the
the
endpoints
are
changed
in
such
a
way
that
now
well.
They
would
respect,
for
example,
the
list
of
withdrawals
and
there's
a
new
version
for
the
endpoint.
So
if
I,
if
one
of
them
is,
for
example,
the
notepad
for
Choice
update.
So
in
that
message,
when
you're
sending
payload
attributes
because
you're
proposing
on
the
next
Vlog
as
I
just
play
in
one
of
my
slides,
you
need
to
send
the
list
of
withdrawals
and
there's
a
new
version
for
that.
Endpoint.
B
That
would
take
the
list
of
withdrawals
and
the
old
version,
of
course,
doesn't
accept
it.
But
then
there's
a
failure
mode
in
which,
if
you're
before
the
capella
fork-
and
you
send
with
troubles,
then
that's
an
invalid
payload.
That's
an
invalid
code
and
if
you're
after
the
capella,
the
capella
fork,
and
even
if
there
are
no
withdrawals,
you
need
to
send
an
empty
list
instead
of
like
no
no
field
with
roads.
B
So
there's
these
two
endpoints
they're
they're
backwards
compatible
in
the
sense
that
I
can
start
calling
the
new
endpoint
with
a
with
without
any
withdrawal
list
before
the
fork
and
the
execution
that
consists
that
your
client
can
actually
start
doing
that.
If
you
already
updated
your
execution,
client
I
can
start
calling
the
new
endpoint
before
the
fork.
Without
any
we
throw
a
list
and
not
the
fork
I'm
going
to
start
sending
the
list.
That's
one
of
the
things
that
we
could
do
prison
decided
to
do
something
different
I.
B
So
we're
gonna
just
use
the
previous
endpoints
right
up
to
the
moments
of
the
fork
and
then
start
the
forward
start
using
the
new
endpoint.
This
has
a
lot
of
advantages
that
you
can
just
change,
update
your
execution
client
few
minutes
before
the
the
fork,
but
it
has
a
big
disadvantage
that
you're
not
going
to
know
that
you're
not
updated
right
up
before
to
the
fork.
B
So
if
you
forgot
to
update
your
client
and
you
might
have
updated
prison
but
not
say
nevermind,
then
you're
gonna
find
out
that
you're
not
working
right
off
capella
instead
of
before
these
are
just
design
decisions.
But
this
is
one
of
the
issues
with
the
with
the
execution
engine
changes
the
engine
API
changes.
There
are
some
other
failure
modes,
as
I
mentioned.
One
of
them
is
the
fact
that
clients
now
return
the
block,
value
and
present,
which
is
for,
for
example,
fail.
B
B
A
B
Just
started
to
interrupt
that,
one
of
you,
one
part
of
your
question-
was
an
answer
which
is
whether
or
not
it
was
already
implemented.
Yes
right
so
so.
This
is
mostly
for
the
execution
plan,
not
the
consensus,
client,
of
course,
and
as
far
as
I
know,
I've
only
been
in
touch
with
Netherland
and
on
gas,
and
both
of
them
have
implemented
everything
that
we
needed
from
the
engine
ABI.
So
both
of
them
have
been
implemented.
B
A
block
value
both
of
them
are
implemented,
the
right
Behavior
and
the
withdrawals-
and
this
is
already
implemented.
I
think
bezu
already
has
everything,
but
there
was
one
of
the
Shanghai
Robert
in
one
of
the
not
not
related
to
the
engine.
Abi,
so
basil,
I
think
is,
is
feature
complete
from
our
perspective
and
I.
Think
ethereum
GS
JS,
which
I
didn't
mention
in
my
slides
but
I,
think
JS
also
was
working
fine
with
with
the
engine.
Api
I
do
not
know
about
other
planets.
A
That
is
good
to
know,
because
I
could
not
find
information
about
the
implementation
in
executions,
but
at
El
level.
Maybe
we
need
to
update
the
chart
very
well.
I
have
a
couple
of
more
concluding
questions.
So
do
we
expect
more
testing?
Obviously
I?
Remember
you
mentioned
about
a
few
weeks
of
testing
left
there.
How
far
we
are
from
the
launch
of
a
public
test
net.
B
There
was
a
small
impromptu
poll
in
the
Discord
r
d
and
client
devs
voted
not
to
have
a
public,
a
public
testing
now,
mostly
because
there
are
many
people
going
on
holidays
and
things
will
break
for
sure.
There
are
some
other
reasons.
Why
not
to
have
a
public
test
net
right
now.
One
of
them
is
that
tooling
is
not
ready.
We
have
one
tool
which
is
it
do,
but
the
state
CLI
is
not
working
yet
to
sign
messages,
so
I
would
suspect
that
we
would
want
to
have
that
ready
before
going
public.
B
So
I
can
give
you
a
hard
deadline,
which
is
by
the
there's,
a
there's,
a
meeting
by
the
end
of
January,
where
the
both
consensus
and
and
execution
clients
that
are
going
to
meet,
and
even
some
people
from
other
areas
and
research
as
well
are
going
to
meet,
and
the
idea
is
that
have
everything
feature
complete
by
then
and
by
then
have
at
least
a
big
Shadow
fork
and
a
big
public
Testament
ready.
So
I
would
say
that
by
the
end
of
January
is
sort
of
like
a
hard.
B
C
C
C
B
No,
that's
absolutely
perfect,
so
the
D
canonical
place
to
look
is
the
interrupt
channel
on
the
ethereum
RNG
Discord
I
would
say
that
when,
when
the
time
comes
for
the
public
debt,
Nets
clients
should
have
documentation
ready
as
to
how
to
test
each
consensus.
Client
and
each
execution,
client
should
have
documentation,
ready
and
I
would
ask
people
to
start
looking
into
those
documentation.
Person
should
be
ready
soon
to
pinpoint
mistakes
or
things
that
are
not
clear.
That's
that's!
That's
really
useful.
B
A
That's
great
and
I'm
also
hoping
that
there
would
be
an
EF
blog
post,
explaining
the
process
of
participation
or
maybe
pointing
towards
the
repository
in
right
places
to
follow
all
right.
It's
time
to
wrap
up.
This
is
the
launch
of
ethereum.
We
can
change.
The
community
is
curious
to
know
when
one
will
be
able
to
withdraw
rewards
from
staking
on
ethereum
blockchain.
A
Hopefully,
the
release
of
this
episode
will
provide
answers
to
people
waiting
for
withdrawal,
and
we
see
a
decent
number
of
community
participation
when
withdrawal
is
available
for
public
test
net
shining
and
portals
appreciate
you
taking
out
time
sharing
the
test
net
progress
thanks
to
all
other
participants
of
the
call
for
joining
I
hope.
You
received
answers
to
your
questions
as
well.
On
this
note,
thanks
to
everyone
watching
or
listening
to
the
Shanghai
special
episode,
should
you
have
any
questions?
Let
us
know
in
the
form
available
in
description
or
leave
a
comment
reach
us
at
eatcatter.