►
From YouTube: CasperLabs Community Call
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
A
How
do
%
all
right?
Let's
kick
it
off
happy
new
year?
Everyone
very
excited
to
be
here.
Welcome
to
our
newly
branded
community
call
where
we
provide
status,
update
on
the
project
and
then
talk
about
the
economics
of
the
blockchain.
We
share
some
of
our
research
and
thinking
and
our
progress
on
the
project,
so
people
can
engage
I'm.
A
Those
of
you
who
know
me
love
me,
I'm,
Etta
and
I'm
joined
here
with
Ashok
and
owned
or
whose
owner
is
one
of
our
economists
on
the
project
and
Ashok
is
our
program
manager,
and
so
what
I'm
going
to
start
off
with
is
a
status
update,
and
then
we
will
dive
into
some
of
the
economics
research
and
our
current
thinking
on
the
project.
So
you
know,
is
quite
holiday
season.
We
had
some
folks
that
were
pushing
really
hard
working
on
consensus
over
the
holiday
season.
I
myself
was
overseas
and
I'm.
A
Now,
back
with
a
lot
of
vigor,
we
have
started
planning
out
some
of
the
key
events
that
we'd
like
to
have
a
presence
at
in
2020
as
well
as
working
on.
You
know
a
developer
engagement,
validator,
engagement
and,
generally,
you
know,
adoption
and
training
on
how
to
use
the
program
with
respects
to
releases.
I
am
currently
focused
on
no
12,
no
13
and
our
March
test
net.
That
is
planned
our
alpha
test
net,
which
will
have
an
implementation
of
highway.
A
We
intend
to
do
this,
so
we
can
learn
how
the
protocol
will
behave
in
real
life.
You
know
taking
a
math
paper
and
translating
to
code.
What
looks
great
on
math
in
math
may
not
necessary
work
properly.
Once
you
have
to
engineer
solution,
so
that's
our
first
milestone,
so
we
released
note
10
on
December
19th
and
we've
updated
the
dev
net
with
the
new
software,
no
10.
If
you
look
at
the
medium
post,
it
will
talk
a
lot
about
what
is
in
node
10.
A
We
didn't
have
any
huge
earth-shattering
features
in
no
10
a
lot
of
it
was
around
hardness.
We
spent
a
lot
of
time
implementing
big
features
that
we'll
be
dropping
in
node
11
they're,
not
externally,
facing
big
features,
but
primarily
it's
the
Kasper
lab
type
system.
So
you
won't
see
this
necessarily
in
contracts,
but
it
basically,
we
found
about
30%
more
performance
in
the
execution
engine,
so
we're
really
excited
about
some
future
work
that
we're
planning
in
the
execution
engine
that
we
suspect
will
give
us
another.
A
Almost
100%
performance
improvement
like
essentially
doubling
the
performance
of
the
platform.
So
that's
really
cool
I'm
really
excited.
You
know
to
to
report
that
finding
so,
let's
talk
about
our
current
focus.
Current
focus
is
production
of
the
implementation
of
Kasper
labs
highway
with
Arras
without
rounds
and
switch
blocks,
and
what
that
means
is
the
March
network
will
have
a
single
large
era.
A
There
won't
be
any
validator
set
rotation
so
anytime
we
upgrade
the
system
and
we
have
to
bounce
it
with
you
state,
because
we
found
in
you
bug
or
we
have
a
significant
update
to
deploy,
because
we've
learned
something
that
will
be
the
opportunity
for
new
validators
to
join.
So
basically,
you
have
to
bond
in
at
Genesis
and
then
you're
locked
in
for
that
era,
with
the
beta
test
net
is
when
we
will
have
validator
sent
rotation
and
economics.
A
We're
going
to
also
we're
going
to
extend
the
protocol
to
include
bonding
auctions
and
what
this
means
is
basically
validators.
Will
the
Genesis
validators
will
have
slots
that
are
likely
designated
for
them
because
they
have
requirements
to
stay
in
the
network
for
a
year,
but
new
validators
can
join
in
by
base
we
bidding
to
join
for
a
validator
slot.
We're
gonna
have
a
large
number
of
validator
slots.
The
goal
is
to
have
a
thousand
validator
slots,
so
it's
not
like
it's
gonna,
be
you
know.
A
Twenty
one
or
fifty
or
some
really
small
number
it'll
be
a
sufficiently
decentralized
set.
We
believe
it'll
be
a
large
enough
set
to
keep
this
the
protocol
sufficiently
decentralized,
but
our
goal
is
that
over
time,
as
a
protocol
proves
itself
out,
there'll
be
more
and
more
demand
to
get
a
validator
slot.
So
we
are
going
to
have
bonding
auctions
for
validators
to
get
a
slot
in
in
the
next
era,
we're
also
initiating
Ashok
that's
assembly
script.
We
want
to
be
really
crisp.
A
We
don't
want
to
tell
people
it's
type
script
support
because
type
script
in
the
assembly
script
do
have
significant
differences,
so
assembly
script
is
a
type
script
like
library
that
supports
compilation.
Web
assembly,
we're
implementing
our
contracts
ABI
for
typescript,
and
we
expect
to
release
that
I'm
sorry
assembly,
stripped
I'm
doing
it
to
assembly
script.
We
expect
to
release
that
in
the
March
timeframe.
A
Performance
and
stabilization
we've
got
a
lot
of
work,
we're
doing
on
automation,
of
infrastructure,
set
up
and
speeding
up
of
our
test
harnesses.
So
we
believe
strongly
in
automated
testing,
and
we
have
you
know.
Presently
it
takes
about
five
to
15
minutes
to
run
all
the
automated
tests
that
we
want
to
speed
that
up
more,
it's
a
good
amount
of
chunk
of
spend
in
AWS.
A
Every
pull
request
goes
through
the
full,
automated
harness,
so
we
use
Bors
for
that,
but
we
want
to
speed
it
up
more,
so
Marc
Greenslade
is
doing
some
good
work
on
that
got
some
more
functionality,
we're
adding
to
clarity
and
then
we're
also
specifying
the
requirements
for
interaction
between
consensus
and
execution
engine.
So
this
really
talks
about
proof
of
stake
requirements,
mint
and
token
requirements,
and
what
consensus
is
going
to
need
from
the
execution
engine
when
we
drop
in
the
highway
protocol
right
now,
we've
got
a
coach
doing
a
lot
of
the
work
on
highway.
A
If
you
look
in
github,
you
will
see
two
pull
requests.
Actually,
if
you
pull
requests,
I,
don't
know
if
he's
already
merged
him.
They
were
already,
but
here
you'll
see
that
they
are
eras.
It's
a
highway
work-in-progress,
the
other
one
was
actually
merged
and
then
era
aware,
dag
I
believe
he
merged.
One
pull
request
yesterday
for
highway
and
this
was
rounds.
So
the
work
is
progressing.
A
We've
chosen
to
have
a
single
author
work
across
andreas
and
Michael,
birch,
Andreas
and
Michael
birch
are
doing
prototypes
and
then
a
coach
has
taken
those
prototypes
and
he's
doing
production
implementations
working
across
them.
We
felt
that
one
person
needed
to
do
it
because
we
weren't
sure
if
we
have
to
turn
the
protocol
on
its
side
to
get
an
implemented.
A
Let's
see,
let's
talk
a
little
bit
about
node.
We
are
looking
closely
at
the
node
architecture.
We
want
to
build
something
that
allows
validators
to
scale
out
sub
sections
of
the
system.
So
what
we
want
to
do
is
build
like
a
micro
service
based
system
where,
if
you
want
to
scale
out
the
consensus
layer,
you
can
do
so
it
basically
the
node
architecture
or
if
you
want
to
scale
out
the
execution
engine
layer
separately,
you
can
do
so
or
if
you
want
to
scale
out
the
graph
QL
layer
and
you
can
do
so
separately.
A
So
we
see
that
there
are
three
core
features
or
core
functionalities
that
the
node
provides.
One
is
execution
of
smart
contracts,
the
second
one
is
computation
around
consensus
and
the
third
one
is
servicing
a
data
request
from
their
users,
and
so
we
believe
that
we
need
to
come
up
with
like
a
micro
services
architecture
that
enables
each
of
these
components
to
scale
out
independently
the
execution
engine
team
and
the
SRE
team
are
likely
going
to
take
point
on
this
and
collaborate
with
the
node
team,
we're
having
an
off-site
in
February.
A
Where
we'll
be
discussing
this
extensively
to
figure
out,
you
know
what
that
micro
services
architecture
looks
like,
but
I'm
sure
if
I
was
to
take
stuff
on
aside,
he
would
say
yes,
yes,
yes,
this
is
exactly
what
we
need
right.
We
need
to
have
an
architecture
that
enables
the
system
to
be
upgraded
without
necessarily
having
liveness
faults
or
downtime.
This
is
something
that
you
know
service-oriented.
Architectures
do
very
well
I'm
using
a
big
IP
architecture,
so
we
need
to
provide
something
similar
for
validators
as
well.
A
So
this
is
something
that's
really
top
of
mind,
we're
also
thinking
of
the
node
really
as
an
enterprise
business
bus
right.
We
believe
that
even
we
want
to
build
a
piece
of
software
that
is
robust
and
and
useful
enough
that,
even
if
enterprises
want
to
use
the
software
by
itself
to
you
know
to
fuel
enterprise
business
applications
that
it
can
do
so
so
we're
not
just
limiting
ourselves
to
this
public
blockchain.
We
also
see
opportunities
if
private
implementations
as
well,
and
we
want
to
build
something
that
is
robust
and
can
really
cross.
A
You
know
kind
of
straddle
both
both
use
cases
we're
looking
at
tests
an
SRE.
We
don't
want
to
be
hit
up
by
ethical
hackers
again,
so
we're
looking
at
definite
security,
making
sure
that
we've
got
the
latest
patches
installed
on
all
the
systems
and
we're
figuring
out
name
policies
for
logs
folders
and
metrics.
This
will
be
really
important
when
we
start
rolling
this
out
to
more
validators
that
we
need
them
to
ship
logs
and
in
you
know,
basically
ship
metrics.
A
A
Fantastic
we're
also
going
to
implemented
app
developer
guide.
We
got
our
rustic
piece
done
and
next
sprint
hope
is
going
to
be
I'm
consuming
that
so
we'll
have
a
very
nice
tap
developer
guide.
I
would
like
to
follow
the
rust
dock.
Rust
has
great
documentation
and
because
we're
supporting
rust,
we
should
have
a
seamless
experience
between
rust
stock
and
adapt
developer
documentation
I'd
like
to
have
that
coming
in
no
twelve,
we
actually
will
offer
an
IDE
integration.
A
A
You
would
see
all
of
our
examples
there
and
you
would
have
a
stub
that
would
allow
you
to
run
the
contracts
and
use
break
points
within
the
execution
engine,
so
you
can
actually
debug
your
contracts
right
against
ie
and
hopefully
see
the
effects
in
the
global
state
or
how
what
would
be
mirrored
in
the
global
state
locally
within
within
the
IDE
itself.
It's
not
quite
breakpoints
in
the
node,
but
it
is
a
nice
stepping
stone
towards
that.
There
is
an
ambitious
goal
to
offer
break
pointing
in
the
node
against
a
global
state.
A
C
C
A
All
right
sounds
good
economics,
research,
we're
gonna
talk
about
some
of
our
economics
research
today,
we're
designing
the
open,
bonding
auctions
and
then
we're
talking.
We
thinking
a
lot
about
rewards
distribution,
slashing
and
transaction
fee
distributions
and
then
investigating
the
impact
of
equivocations.
The
team's
really
been
focused
on
you
know,
equivocation
catastrophe,
and
how
do
we
integrate
social
consensus?
Basically,
what
that
means
is
the
network.
You
know
the
validators
in
the
network
realized
that
we're
having
a
catastrophe
on
the
network.
A
What
is
the
protocol
for
them
to
basically
get
the
network
back
on
track
and
that's
something
that
we
have
been
thinking
a
lot
about
and
with
that
and
also
I
want
to
do
a
weekly
workshop
session.
This
Thursday
look
for
announcements
in
our
discourse
chain
or
a
telegram
channel.
I.
What
I'd
like
to
do
is
demonstrate
the
weighted
keys
functionality
with
multi
signature
and
demonstrate
how
it's
more
secure
than
the
current
implementations
that
you
see
in
a
theorem
and
a
lot
of
other
projects
or
any
project
that
uses
the
EVM.
A
We
have
a
unique
way
of
doing
multi
signature
that
keeps
signatures
secure
and
gives
controls
to
the
contract.
Author
to
control
the
authority
of
who
can
sign
deployments
so
that
will
be
happening.
This
Thursday
so
do
join
us
and
I'll.
Be
posting
posting
the
plan
and
announcements
by
tomorrow
with
that
I'm
gonna
turn
it
over
to
own.
Nor,
let
me
stop
by.
B
B
So
there
have
been
many
things
that
I've
been
working
on
lately:
I
incorporated
the
feedback
from
andreas
to
reward,
distribution
and
I.
Think
it's
ready
to
specify,
but
it's
too
technical
I
thought
centralization
would
be
a
like
a
broader,
more
general
topic
which
people
might
be
interested
in,
and
it's
how
it
relates
to
delegation
and
by
delegation
here,
I,
don't
mean,
provoke
state
delegate
approve
of
stake,
I
mean,
like
you,
know,
broiler
sense
users.
B
So
so
decentralization
is
what
we
want
in
blockchains.
It
means
there
is
more
economic
freedom
for
those
who
run
it
for
do
for
people
who
want
to
participate
when
a
miner
or
something
or
for
the
users
in
general
because
like
if
you
think
about
it,
you
can
freely
join
the
network
can
grow.
Bitcoin
grew
to
the
size.
It
is
today
because
it
was
permissionless
and
users
don't
have
to
ask
anyone
to
transact
like
don't
have
to
get
get
permission
from
anyone
they
to
send.
B
Bitcoins
and
decentralized
networks
are
hard
harder
to
kill
and
we,
as
human
beings,
tend
to
favor
systems
that
are
more
likely
to
survive.
Think
about
big
BitTorrent,
for
example.
It's
a
good
example,
so
it
hasn't
been
stopped
since
many
years
and
there
the
reason
is
because
it's
decentralized
there
is
no
single
Authority
responsible
for
it,
and
I
want
to
make
a
distinction.
So
there
are
different
types
of
decentralization
I.
Don't
want
to
I
want
to
go
into
the
pitfalls
that
many
people
do
so
in
this
context,
I'm
using
political.
B
How
many
individuals
or
organizations
ultimately
control
the
computers
that
the
system
is
made
up
made
up
of
so
I?
Don't
care
about
the
other
two?
So
how
so,
let's?
Let's
look
at
how
delegation
affects
centralization?
We
know
in
proof
of
work.
We
have
miners
in
proof
of
stake.
We
have
validators
and
delegation
is
a
recurring
concept
delegation
in
delegation
like
one
one
one
agent
in
the
system
assigns
authority
to
another
agent
right
and
I
will
give
examples.
So
when
you
think
of
delegation,
you
only
people
generally
think
only
about
proof
of
work.
B
Oh
sorry,
proof
of
stake
because
it's
been
it's
been
only
mentioned
there,
but
if
you
think
about
Bitcoin
mining
pools
are
again
a
form
of
delegation,
so
you
have
this
pool
operators
they're
responsible
for
by
operating
hardware.
They
buy
through
software
that
the
full
members
install
they
track
contributions.
They
they
they're
responsible
for
distributing
block,
rewards
fairly,
and
so
that
is
why
itself,
if
you
look
at
the
definition,
is
delegation,
miners
are
assigning
authority
to
a
pool
operator.
B
Like
an
entity
that
didn't
exist
before
hand
right
and
the
proof
of
State
delegation
is
like
I,
said,
users
staked
their
token
speak
without
becoming
evaluator
themselves,
so
users
just
hand
out
just
delegate
their
tokens.
They
give
the
right
to
stake
their
tokens,
but
they
don't
just
hand
out.
They
can
take
them
away.
They
can
under
legate
sambong
their
tokens
at
any
time.
So
it's
free.
It's
not
like
the
validators,
all
the
tokens.
B
Now
they
as
as
long
as
they
have
the
delegate
tokens
they
are
and
more
more
profit
from
them,
and
then
they
they
take.
It
take
a
cut
out
of
the
commission
and
then
they
return.
The
surplus
for
the
delegating
user
and
the
thing
I
want
to
emphasize
this
by
its
nature
delegation
causes
these
networks
to
become
centralized.
If
you
think
about
it
like
before,
before
mining
poles,
there
are
no
such
distinction
like
you,
then
you
assign
new
new
type
of
roles
to
the
system
like
agents.
B
If,
like
the
efficiency
of
the
whole
network,
the
growth
or
throughput
or
whatever
becomes
higher,
as
as
the
number
of
delegates
decreases,
then
you
can
expect
all
the
30
to
be
concentrated
in
the
hands
of
a
few
like,
like
the
variants
of
minor
revenue,
is
minimized
when
only
a
single
mining
pole
serves
a
whole
market
and
you
don't
just
have
a
single
miner
pole
I
brought
up
the
chart.
We
have
this
many
right,
so
1,
2,
3,
4,
5
mining,
pools
and,
like
I,
don't
know
how.
B
How
much
are
the
bit
main
operated
mining
pros
amount
to
now,
but
it
was
once
50
50
percent.
So
if
you
think
about
it,
there
was
a
time.
50
percent
of
the
mind,
total
hash
rate
was
operated
by
a
single
company
and
I
don't
mean
controlled.
I
mean
operate,
so
it's
it's.
The
I
will
talk
talk
about
the
difference
now.
B
So
central,
so
this
is
centralization
right.
You
expect
like
thousands
of
nodes,
but
now
you
have
entities
handful
entities
controlling
like
some
aspects
of
the
net
network
and
I
argue-
and
this
is
a
something
like
debatable-
that
centralization
doesn't
have
to
hurt
freedom
in
block
chains
as
long
as
delegation
is
permissionless
and
then
doesn't
have
any
barriers
to
entry
or
exit
and
the
services
provided
by
delegates
are
the
same,
very
producible,
identical
in
nature.
B
And
and
the
argument
I
present
is
it's
not
centralization
with
church
freedom,
but
the
ability
to
monetize
oligopolies
Authority.
So
when
a
minor
or
a
validator
or
a
pool
with
their
power
granted
power,
try
to
block
others
tries
to
try
to
increase
prices
whatever
that
means
like
could
be,
they
could
just
increase
transaction
fees
right.
If
you
have
just
a
polygon
poly
of
miners,
they
could
just
not
mind
blocks.
B
Sorry,
not
include
transactions
with
to
of
transaction
fees,
so
they
could
just
set
a
floor
so
that
that
is
what
I
mean
by
monopolizing
Authority
and
as
long
as
you
don't
have
it.
As
long
as
you
have
this
permissionless
nature,
I
argue
that
centralization
is
not
a
bad
thing,
so
you
can
have
centralization
and
economic
freedom.
At
the
same
time,
it's
an
important
distinction.
It's
it
is
act
like
this
is
not
something
and
making
up.
This
is
how
Bitcoin
is
operating,
and
this
is
like
this
is
a
if
you,
if
you
think,
bitcoin
is
decentralized.
B
If
you
have
heard
about
it,
then
you
see
that
bit
name
this
company
huge
company
operating
it
then
you're
like
confused.
How
can
how
can
a
network
be?
You
know,
decentralize
if
it's
half
of
it
is
operated
by
a
single
company?
So
it's,
but
the
answer
is
this:
so
it's
not
it's
not
about
centralization
is
about
deterrence,
so
centralization
allows
those
in
power
to
restrict
freedom,
but
if
you
have
permissionless
permissionless
nests,
you
have
deterrence.
B
If
a
mining
pool
just
wants
to
you
know,
do
something
malicious,
not
hand
out.
Rewards
miners
can
just
leave,
because
it's
not
that
hard
to
create
a
mining
pool
right.
So
this
deterrence
is
is
is
what's
important
for
for
economic
freedom,
a
delegate
will
be
deterred
from
acting
and
monopolists
because
they
can
be
easily
placed
there
in
there.
Another
always
solid
ground
this.
This
deterrence
is
the
key
to
making
centralized
blockchains
work
and
yeah
I
can
take
any
questions.
If,
if
there
are,
you
know
chucked.
A
Yeah
I
think
one
of
the
most
important
things
to
note
is
that
it
with
permissionless
blockchains.
The
key
is
that
the
validators
decide
whether
or
not
they
want
to
support
either
an
upgrade
right,
so
their
ability
to
control
the
version
of
the
blockchain
and
control
how
the
blockchain
is
operated
is
really
I.
Think,
what's
key
and
central
to
this,
how
the
validator
set
participation
that
can
change
right
and
similarly,
you
know,
what's
with
staking
delegation
I
think
you
know.
A
B
A
A
And
I
think
I
I
believe
that
the
integration
of
some
kind
of
reputation
system
is
is
eventually
going
to
come
into
blockchain,
where
people
will
know
those
valve
ala
daters
that
are,
you
know,
giving
the
best
like
say,
for
example,
the
best
return
on
their
investment,
providing
the
best
service
providing
the
best
right.
So
you
know
have
the
best
Pro.
You
know:
management
protocol
for
the
nodes,
etc,
etc.
Uptime
I
think
those,
though
that
will
eventually
become
clear
over
time.