►
From YouTube: Baseline Summit 2020 Track7 Blockchain Clients - Edited
Description
The complete footage of this track from the Baseline Protocol Summit, from November 12-13, 2020, edited to remove blank sections from the livestream.
For more information on the Baseline Protocol, go to https://baseline-protocol.org
To compete in the Baseline Gitcoin Hackathon (Dec 9 2020 through January 6, 2021), sign up here: https://gitcoin.co/hackathon/baseline/onboard
A
B
B
B
B
C
A
So
I
think
we've
maybe
given
another
couple
of
minutes
before
we
kick
things
off
just
to
make
sure
everyone's
here
who
wants
to
be
here.
So
they
just
said
that.
C
I
I
tried
I
tried
to
find
a
comparable
picture
for
my
my
about.
A
E
Okay,
if
you
look
at
the
at
the
the
text,
underneath
the
stream
there's
actually
a
link
to
the
playlist
and
you
can
go
down
and
you'll
see
the
clients.
Breakout
is
streaming,
I'm
watching
it
as
we
speak,
so
take
it
away.
A
A
Cool
all
right
so
I'll
I'll
I'll
I'll,
try
and
make
these
slides
just
a
bit
bigger
for
the.
Hopefully
these
will
come
through.
Okay
right,
sorry,
I'm
gonna,
stop
this
screen
share
and
then
start
another
one.
Just
so
we
can
see
everything
all
right,
okay,
cool!
So
hopefully
everyone
can
see
the
slides
there
hope
you
get
a
thumbs
up
just
from
yeah
we're
all
good,
okay
cool,
all
right,
so
we'll
we'll
run
through
this
deck
and
then
we
could.
A
I
think
I
think,
hopefully,
as
we
as
we
go
through
it
we'll
have
some
opportunities
to
kind
of
you
know,
stop
and
discuss
and
so
on,
and
then
I
think
that
you
know
we
once
we've
kind
of
gone
through
the
motions
and
everyone's
kind
of
got
some
context.
We
can
then
start
thinking
about.
You
know
how
how
we
actually
want
to
proceed
and
start
capturing
some
more
specific
details
about
yeah.
You
know
what
we're
trying
to
achieve
here.
A
So,
of
course,
we'll
we'll
start
with
the
the
intros,
so
I'm
connor
svenson,
I'm
the
founder
and
ceo
of
web3
labs
web3labs
is
a
blockchain
technology.
Company
we've
been
in
the
space
for
a
number
of
years,
we're
responsible
for
the
web3j
java
and
android
integration,
library
for
ethereum,
also
the
epis
blockchain
explorer,
and
we've
done
a
number
of
you
know
commercial
engagements
as
well.
A
With
you
know,
various
companies
across
the
space
yeah
also
been
very
involved
in
a
number
of
industry
organizations
as
well
for
the
past
three
years.
So
at
the
enterprise
ethereum
alliance,
I
chaired
the
technical
specification
working
group
which
was
really
responsible
for
churning
out
the
enterprise
ethereum
client
spec.
That
was
first
released
a
couple
of
years
ago.
Then
this
year
I
got
involved
with
the
baseline
tsc
because
you
know
it
as
as
is
shown
by
the
number
of
attendants
we
have
here
like
it's.
A
It's
it's
a
very
awesome
premise
and
I
think
there's
a
lot
of
people
who
believe
in
the
potential
of
you
know,
baselining,
basically
on
top
of
mainnet.
So
you
know
it's
kind
of
a
no-brainer
when
it
launched
to
you
know
kind
of
get
involved
there.
I'm
also
involved
in
an
organization
called
the
interwork
alliance,
which
is
looking
at
kind
of
standards
for
really
business
processes.
In
the
context
of
blockchain,
so
you
know
creating
standard
ways
to
represent.
You
know.
Well,
you
know.
A
Processes
are
well
suited
to
dlt
in
in
a
manner,
that's
ledger,
agnostic
and
then
finally,
as
well
within
hyperledger,
I'm
involved
with
their
climate
special
interest
group
there
as
well.
I
will
now
pass
over
to
kyle
who's
going
to
be
running
the
session
too.
C
Hey
guys
thanks,
connor
kyle,
thomas
founder
of
provide,
provide,
was
created
to
to
blockchain
enable
enterprises
that
didn't
have
a
clue.
You
know
about
blockchain
yet
and
boy
did
they?
Do
they
not
have
a
clue
when
provide
was
created?
C
It
was
early
enough
to
you
know,
to
block
we
blockchained
enable
enabled
developers,
but
we,
you
know,
we
we
needed
to
make
sure
that
we
could
blockchain,
enable
the
enterprises
as
well,
because
obviously
they're
the
ones
that
employ
the
developers
and
make
them
make
the
decisions.
So
you
know
through
the
my
work
with
with
on
provide
over
over
the
last
you
know.
C
Few
years
we
found
ourselves
in
a
unique
position
to
contribute
to
the
baseline
protocol
to
the
the
enterprise,
the
eea
enterprise
ethereum
alliance.
We
were
able
to
you,
know
to
to
get
involved
with
the
main
networking
group
and
then
the
imminent.
The
imminent
task
force
which,
which
you
know
made.
C
C
Today
we're
we
have
a
commercial
baseline
as
a
service
offering
as
a
result
of
that,
working
with
with
customers
like
conan
cook,
one
north
america
and
some
other
international
large
international
organizations
on
on
creating
baseline
offerings
and
things
are
going
quite
well.
There.
G
C
I
I
the
eats
atlanta,
which
is
just
an
interesting
fact:
east
atlanta's
kind
of
we've
had
two
events:
one
one
hackathon
modeled
off
of
east
denver
that
was
back
in
2018
and
then
just
recently,
we
have
a
virtual
event
that
sort
of
featured
the
baseline,
the
the
the
usual
suspects
in
the
baseline
community.
Thanks
connor
back
to
you,
oh
hey,.
H
By
the
way
connor
I
I
noticed
that
we
don't
have
chat
in
here
as
my
fault
and
I'm
not
sure
why
so
something
in
the
configuration
of
the
eighth
concurrent
session
seems
to
have
gone
wrong.
Nobody
else
sees
a
chat.
Do
they
inside
the
slack
or
inside
the
zoom.
D
H
Yeah,
sorry
about
that
I'd
advise
you
know,
use
the
summit
chat
and
and
slack
while
I
try
to
work
that
out.
Probably
I'd
have
to
kill
the
session
to
set
that
up.
So
that's
probably
not
worth
it
so,
but
sorry
about
that.
A
That's
cool
so
just
in
turn
in
terms
of
the.
Where
are
we
the
for
for
for
the
actual
chat?
Where
was
it?
Wasn't
there
a
tab
across
the.
B
H
B
A
A
So
I
can
all.
I
can
also
see
there's
a
lot
of
good
good
people.
Who've
joined
the
session
here
as
well.
So
I
think
what
will
certainly
be
helpful
is
as
people
kind
of
pitch
in
and
ask
questions.
It
might
be
good
just
to
kind
of
introduce
yourselves
and
what
it
is
you
do
too
just
for
the
context
of
those
people
who
aren't
aware
of
you.
A
I
know
we
just
just
just
because
I
think
that
it'll
be
it'll
be
certainly
be
helpful
for
those
people
who
you
know
aren't
familiar
with,
say,
nether,
mind
or
you
know,
an
ethereum
or
you
know,
involved
in
the
baseline
in
some
shape
or
form.
It
will
definitely
be
good
for
those
people.
You
know
to
you
know,
make
make
them
aware
of
these
projects
and
what
you
do
on
them,
because
I
think
yeah
we've
really
got
some
some
some
key
people
here
for
this
discussion.
A
So,
okay,
so
fundamentally
the
you
know
the
goal
of
this.
A
This
working
group
is
basically
to
enhance
the
kind
of
client
and
language
diversity
we
have
for
baseline,
and
you
know
the
the
reason
for
this,
of
course,
is
that
you
know
there's
there's
multiple
ethereum
clients
out
there
and
there's
multiple
languages
that
people
you
know
developing
as
as
it
currently
stands,
the
sort
of
baseline
stack
so
to
speak,
it's
predominantly
written
in
javascript,
and
then
it
uses
the
clients
and
there's
good
reasons
for
that,
which
I'm
sure
that
carl
and
tomas
can
you
know
we'll
elaborate
on
shortly,
but
really
what
what
we
want
to
you
know
the
the
baseline
protocol
is
is
focused
on
having
you
know,
being
standards
backed,
and
so
you
know
with
any
sort
of
technical
standard.
A
A
It
just
gives
that
you
know
that
greater
choice
as
well,
so
that
you
know
if
an
organization
is
very
you
know,
they've
got
a
lot
of
technical
expertise
in
one
specific
you
know
stack,
then
it's
good
to
be
able
to
you
know,
focus
on
that
stack
rather
than
you
know.
Forced
to
you
know,
bring
someone
in
with
with
something
else,
and
so
really
you
know
what's
what's
what's
guiding
this?
A
Of
course
you
know
I've
got
a
lot
of
interest
in
jvm
development
on
ethereum,
but
you
know
also
the
jvm
and
the
net
framework
they're
very
widely
used
in
in
enterprise
and
and
also
you
know,
we
have
a
number
of
ethereum
clients
that
are
kind
of
you
know
enterprise.
You
know
suitable
for
enterprise
as
well,
because
of
course,
you
know
nether
mind
we
see
has
got
great
traction
in
the
public
network
community
as
well
as
in
you
know.
A
The
enterprise
community
kind
of
I
think
thomas
is
fair
to
say
kind
of
spans,
both
so
to
speak.
But
then
you
know
hyperledger
basu
is
certainly,
although
it
is,
you
know,
compatible
it's
being
pushed
more
as
an
enterprise
one
and
it's
likely
to
become
you
know,
kind
of
the
core
part
of
quorum
in
in
the
future
as
well.
So
there's
kind
of
you
know
in
order
to
help
grow
the
overall
baseline
ecosystem.
A
Really
the
you
know,
I
think
the
the
thought
is
that
if
people
have
the
ability
to
plug
other
clients
into
the
baseline
and
that
kind
of
suit
their
own
internal
architectural
stacks,
then
that's
going
to
be
a
better
thing
for
the
overall
baseline
ecosystem,
and
so
the
goal
here
is
really
to
talk
about.
You
know
how
we
think
we
would
need
to
go
about
it
in
order
to
get
there
yeah.
A
So
if
we,
what
we'll
do
is
we'll
we'll
go
we'll
touch
on
the
current
architecture?
And
again
you
know
these
slides
are
just
kind
of
to.
I
think
you
know
frame
everyone's
thinking
before
we
actually
go
into
the
specifics.
But
do
you
know
if
you
have
any
questions
or
want
to
speak
up
at
any
point?
A
Please
please
feel
free
to,
but
we're
going
to
run
through
the
architecture
and
then
just
talk
about
you
know
how
conceptually
you
could
potentially
break
up
the
baseline
packages
and
then
and
then
we'll
yes,
then
at
that
point
we
can
start
really.
You
know
get
getting
into
the
specifics.
A
So
the
this
diagram
here
reflects
it's
really
the
you
know
the
great
work
that
kyle
and
provide
did
over
the
last.
Would
it
be
fair
to
say
sort
of
six
months
of
their
about
kyle.
D
C
A
So
so
so
for
for
context,
you
know
the
the
the
baseline
protocol
was,
you
know
initially
using
death
as
its
clients,
and
there
was
just
kind
of
well
like
I
guess
it
wasn't
as
clean
and
whisper
as
well.
Oh
yeah
yeah,
it's
yeah,
it
was.
A
It
was
more
kind
of
created
as
a
as
a
demo
thing
to
prove
well
prove
a
concept,
as
opposed
to
being
something
that
was
like
a
robust
foundation
and
kyle,
and
his
team
basically
went
all
out
for
months
to
just
completely
turn
it
around
and
did
you
know
an
amazing
job
here
and
the
end
result
is
basically
you
know
what
what
we
have
in
front
of
us
now.
I
think
I
think,
for
the
context
of
you
know
this.
This
talk
anyway.
A
You
know
what
we're
looking
at
really
is
is
the
fact
that,
with
the
baseline,
you
can
kind
of
think
of
it.
In
terms
of
you
know
three
sort
of
groupings
of
components
so
to
speak
in
terms
of
you
have
you
know,
messages
coming
in
via
the
the
nats
messaging
protocol,
and
then
you
have.
You
know
your
your
actual
pieces.
A
Where
your
you
know
you
you,
you
you're,
generating
your
proofs
and
you're,
basically
processing
the
messages
and,
ultimately,
then
you
know
spitting
something
out
onto
you
know
on
on
onto
a
public
aetherium
chain,
and
so
that's
those
are
kind
of
you
know,
broadly
speaking,
the
actually
p
actual
pieces
that
you
know
we're
we're
interested
in
here
and
it's
it's
how
we
can
kind
of
you
know
if
if
say
we
we,
you
know,
look
at
these
pieces
in
isolation,
so
to
speak.
We
can
kind
of
see
where
there
might
be.
A
Opportunities
to
you
know
provide
great
effect,
flexibility
there.
So
if
we
go
back
to
the
start
of
course,
you
know.
Nats
is
a
very
well
established
messaging
protocol,
but
you
might
find
down
the
line
that
you
know
other
enterprises
they
want
to
use
say
I
know
you
say
kafka
or
there's
some,
some
real
old-school
workforce,
like
you
know,
ibm
mq
or
you
know
some
something
like
that
there,
but
but
then,
in
terms
of
the
the
actual,
you
know
the
zero
knowledge
circuits
that
are
used
here.
This
is
using
socrates.
A
Currently,
but
again,
there's
there's
areas
there
that
we
might
want
to
support
other
ones,
but
also,
you
know
again
provide
other
language
supports
and
then
yeah
here
we
have
the
persistence
piece
here
as
well.
I
mean,
I
think
karl,
is
probably
worthwhile
you
as
well
going
into
a
bit
more
depth
just
for
context
around
around
all
of
this,
because
you
can
really
talk
about
how
it's
evolved
and
the
changes
that
have
happened
there
and
I
can
I'll
move
this
around.
As
you
know,
to
kind
of
support
you
for
sure,
yeah.
C
No
no
problem
yeah,
and
I
think
it
makes
a
lot
of
sense
sort
of
to
piggyback
on
what
you
said
about
how
we
need
to
look
at
the
entirety
of
this
of
this
effort,
and
you
know
each
of
the
each
of
the
parts
in
isolation
represents.
You
know
an
increment
or
an
iterative,
an
error,
an
iterative
step
that
could
be
taken
in
terms
of
client
development,
and
you
know
huge
shout
out
to
tomas
and
nethermind
and
his
team
for
the
work
that
they've
done.
C
You
know
as
well
to
you
know,
to
make
to
make
this
possible
right
to
make
this
architecture
possible.
You
know,
I
think,
the
the
real
driving
you
know
and
as
you
as
you
said,
like
the
the
radish
34
proof
of
concept,
you.
G
C
If
it
used
gas
that
used
whisper,
it
didn't
it,
you
know
there
was.
There
were
a
lot
of
shortcomings
and
hacks
to
it.
It
was.
C
It
was
a
great
proof
of
concept
you
know,
and
based
on
that,
you
know
one
of
the
big
things
that
that
that
I
noticed
was
that
you
know
that
it
was
using
a
timber
service
which
was
which
is
a
an
ui
blockchain
project
that
stores
merkle
tree
merkle
trees
in
a
database,
and
essentially
that
was
decoupled
from
the
from
the
guest
client,
and
I
thought
I
thought
it
makes
a
lot
of
sense
if
it
were,
if
it
weren't
decoupled
from
the
client.
C
If
actually
the
merkle
tree
database
was,
you
know,
lived
behind
like
a
baseline,
rpc
set
of
methods
or
like
a
baseline
rpc
module
and
that
first
you
know
that
first
module.
You
know
it
makes
sense
if
we,
if
we
allow
users
to
track
shield
contracts,
you
know
merkle
tree
per
shield.
If
you
will
all
you
know
on
the
client
itself,
because
the
client
is
already
getting
you
know
all
of
the
all
the
the
headers
and
and
processing
all
of
them.
So
why
do
that
twice?
C
C
I
C
A
blockchain
client,
why
not
consolidate
the
merkle
tree
onto
it
for
efficiency,
and
so
that's,
that's
sort
of
where
things
started
and
I
think
that's
kind
of
the
first.
That
would
be
the
first
step
you
know
in
in
thinking
about
a
new
client,
implementation
or,
or
you
know,
a
client
implementation
generally
and
tomash
can
speak
to
this
too.
You
know
if
you
think
about
like
nats,
that
that
was
the
replacement
for
for
whisper.
So
there
was
two.
There
was
two
issues.
C
There
was
the
fact
that
guess
was
used
in
the
fact
that
whisper
was
used
and
whisper
obviously
has
since
been
been
deprecated
we
after
we
after
the
decision,
was
made
to
sort
of
deprecated
in
our
in
our
in
the
baseline
protocol.
It
actually
was
deprecated.
Officially,
it's
no
longer
no
longer
being
worked
on.
So
nabs,
though,
is
fundamentally
different
than
whisper.
In
that
it's
it
allows
us
to
do
true
point-to-point
messaging,
where
whisper
is
a
gossip
protocol
and
it's
more
peer-to-peer
and
they're.
C
You
know
they're
subtle,
subtle
in
the
nomenclature
they're
around
point-to-point
and
peer-to-peer,
but
like
there's
quite
a
big
difference
and
the
privacy
aspects
of
peer-to-peer
messaging
and
the
gossip
protocols,
while
they're
useful
for
many
applications.
C
You
know
sending
sending
notifications
to
counterparties
in
a
in
a
business
process
is
not,
I
don't
believe
one
of
them.
You
really
need
to
do
that
off
chain.
You
need
to
be
able
to
connect
directly
to
you
know
what
your
counterparties
and
deliver
those
messages
directly
to
them.
You
use
the
org
registry
contract
to
look
up
their
endpoint
and
you
know:
there's
there
are
public
keys
associated
with
each
entry
in
the
org
registry,
and
you
know.
I
C
The
phone
book
so
that
like
nats,
wins
it
lended
itself
well
as
a
replacement
for
whisper.
I
think,
there's
you
know
opportunities
to
bake
to
make
naps
into
the
client
as
well,
and
then
it's
another
something
that
tomas
and
I
have
been
have
been
talking
about
so
much
I'll
I'll,
actually
I'll
pause
here
and
let
you
weigh
in
is
there
anything
that
I've
missed
here
that
you'd
like
to
correct
me
on
or
chime
in
on.
J
Yeah,
I
think
it's
straight
to
the
point,
so
it's
all
about
the
efficiency
and
streamlining
like
the
flow
of
tracking
the
data
of
exposing
the
data
in
a
very
familiar
way
for
all
the
ethereum
developers.
J
People
who
I
have
been
building
integrations
with
ethereum
clients
in
the
past,
so
providing
the
same
experience
for
talking
to
baseline
as
you
used
to
talk
to
ethereum
clients
in
general
when
asking
about
transactions
blocks
and
so
on
should
make
it
much
simpler
and
just
like
change
it
even
slightly
better
towards
a
very
quick
rapid
integration
with
the
systems
in
the
in
the
enterprises
and,
together
with
the
provide
platform,
everything
that's
in
there
with
very
quick
synchronization
times
to
keep
adding
the
nodes
and
making
sure
that,
like
you,
don't
have
outages
that
everything
actually
works.
C
Thanks
good
connor,
I
think
that
that
yeah
do
you
have
anything
to
add
connor.
I
guess
I'll
turn
it
back
over
to
you.
A
I
know
all
good,
oh
so
in
terms
of
the
the
actual
components
and
this
this
is
where
we'll
kind
of
you
know
try
to
just
you
know,
high
level
kind
of
think
about.
You
know
if,
if
if,
if,
if,
if
we're
trying
to
break
down
baseline,
like
you
know
previously,
we
we
have
the
kind
of
sequence
diagram
that
shows
the
different
flows,
but
then,
in
terms
of
the
actual
you
know,
considerations
of
you
know
what
what
the
specs
each
of
the
components
do
and
what
their
responsibility
is.
A
We've
grouped
them
together
here
and
so
hopefully
this
is
all
fairly
self-explanatory
to
people,
and
you
know
please
as
well.
If,
if
any
of
this
stuff
isn't
clear
to
anyone,
you
know
don't
hold
back
from
answering
questions,
because
you
know
we'd
much
rather
have
like
a
an
interactive
session
where
people
are
saying.
Well,
you
know
why.
Why
do
you
do
things
this
way
or
you
know?
Why
is
it
like
that?
A
Because,
although
there's
you
know
the
the
people
who
have
written
a
lot
of
the
stuff
are
here,
it's
it's
really
helpful.
If
we
can,
you
know,
educate
people
as
we
go
as
well,
because
you
know
it's
it's
it's
hard,
often
trying
to
do
this
stuff.
If
you
don't
have
someone
someone
to
ask
the
right
questions
to
so,
I
really
do
want
to
encourage
people
to
you
know,
stop
and
ask
any
questions.
A
However,
however,
you
know
mundane,
they
might
feel-
or
you
know
whatever
else
so
yeah,
just
just
just
chime
up
but
yeah.
So
as
we've
mentioned,
there
there's
the
actual
client
itself
and
then
we
have
the
the
the
the
smart
contracts.
Basically
that
are,
you
know
used
by
baseline
and
that's
yeah.
That's
there,
then
we
we
have
the
actual
privacy
circuits
as
well,
which
they
are
used
for
the
you
know
the
zero
knowledge.
This
is
the
zero
knowledge
proofs,
the
the
actual
api
as
well.
A
A
Does
that
kind
of
you
know
make
sense
to
everyone?
It's
like
you
know
what
what
this
actually
looks
like
you
know
to
help
contextualize
it
as
well.
We
can
you
know
you
can
show
you
on
the
actual
baseline
repository
repo.
A
If
I
can
figure
out
yeah
there,
we
go
you
you'll,
if
you,
if
you
have
a
look
there
in
the
code
too,
you'll
find
that
the
actual
modules
kind
of
they're
not
they're,
not
perfectly
aligned
with
that,
but
they
are
kind
of
similar
and
that
you
know
you
have
your
socrates
circuits
there
you
have.
You
know
some
an
example.
A
Reference
you've
got
the
core
piece,
and
this
is
where
you've
got
like
the
api
contracts,
messaging
persistence,
privacy
and
there's
also
additionally
types
there
as
well.
But
the
point
is:
is
that
you
know
we've
kind
of
broken
it
down
being
led
slightly
more
by
the
code
here
as
opposed
to
you
know
just
going
with
the
conceptual
part,
because
you
know
what
we're
trying
to
do
here
is
think
about
how
we
could
actually
move
this
forward.
B
A
Okay,
so
you
know-
and
and
this
is
like
a
you
know-
one
potential
approach
here,
but
I
think
it's
good
to
highlight
this,
and
then
we
can
start.
You
know
the
actual
session
part
going
where
we
can
start
going
into
a
few
more
details.
A
So
you
know
what
you
we
could
potentially
look
at
breaking
this
down
into
a
few
phases
where
you
know
you
have
your
actual
blockchain
client
piece
where
you
select
a
different
client
and
then
you
say:
okay!
Well,
let's,
let's
make
sure
that
we
can
run
the
core
contracts,
you
know
plugging
in
with
baysu
and
then
you
know
have
update
the
example
project.
So
it
supports
that
and
I'd
I'm.
A
I
propose
anyway,
that
that's
kind
of
like
you
know
I
suppose,
the
most
minimum
first
change
you
could
do
to
at
least
validate
it
there
now.
I
know
that
because
of
anything
things
like
the
merkle
tree,
that's
used
by
you
know
by
by
the
actual
protocol
itself.
It's
not
like
it
would
just
work
out
the
box
right
now
and
it's
just
probably
going
to
be
useful
to
jump
back
into
some
of
those
specifics
there.
A
K
Hi
everyone,
I'm
juan
blanco,
a
creative
ethereum,
with
ideas
to
donate,
to
certainly.net
integration
on
ethereum,
or
you
know
he
was
a
client,
but
I
left
that
alone.
K
So
on
the
questions
on
the
on
the
miracle
tree
and
then
because
the
idea
this
is,
this
is
going
to
be
the
integration
directly
with
the
with
the
public
main
net.
Is
that
correct?
K
So
if
targeting
just
bezu
is,
is
might
be
an
isu
and
we
probably
will
want
to
talk
to
to
the
cath
guys
as
well
and
just
to
kind
of
you
know,
put
a
prototype
or
as
well
and
also
I
was
thinking
in
areas
like
optimism
or
layer
2,
as
we
might
want
to
consider
that
a
layer
2
might
want
to
integrate
to
the
and
put
the
data
there,
which
later
on,
might
be
put
into
the
main
mate.
K
I
A
Oh,
that's
just
I
was
just
saying
yeah
it's
hard
to
keep
all
this
stuff
in
my
head,
so
maybe,
if
someone
can
just
add,
like
a
placeholder
into
the
google
doc
for
now,
because
I
have
a
feeling
if
I,
if
I
hit
something
in
here,
it
might
stop
the
presentation
and
you
know
I
don't
want
to
disrupt
things
just
yet
so,
but
yet
to
your
points
juan.
A
I
think
that
definitely
the
the
layer
two
questions
an
interesting
one
and
I'd
love
to
know
if
that's
actually
come
up
before
because
it's
you
know,
I've
not
heard
anyone
talk
about
that,
but
I
think
it's
it's
it's
very
relevant,
but
the
other
is.
I
think
it
would
be
really
good
for
carl
and
thomas
just
to
give
the
perspective
on.
You
know
why
an
ethereum's
being
used,
because
originally
death
was
actually
being
used
in
the
first
proof
of
concept
for
baseline.
J
So
I
mean
sure
yeah
as
often
we
we
have
this
confusion
of
names,
so
juan
is
actually
from
ethereum
another
mind.
Is
the
client
name
the
why
it's
used,
I
think,
to
add
choices
to
add
possibilities,
so
we
could
like.
We
always
were
thinking
that
dot
net
core
as
a
as
a
technology
for
the
client
is
very
enterprise
friendly.
So
many
many
enterprises
already
have
lots
of.net
developers
on
site.
J
They
there
are
integrations
libraries
to
ethereum
like
juan's
ethereum,
it's
a
fantastic
tool
to
just
connect
with
the
ethereum
clients,
and
if
you
have
also
the
client
that
you
can
expand
by
adding
plugins
by
feeling
familiar
with
the
code
and
many
enterprises
will
be
more
likely
to
feel
familiar
with
dotnet
than
anything
else,
then
it's
great
and
then,
when
you
actually
make
it
like
baseline
native
in
a
way
that's
adding
json
rpc
interfaces,
tooling,
that
is
for
baseline,
by
integrating
with
a
fantastic
stack
from
provide,
which
makes
it
easier
to
deploy
everything
together
to
launch
it.
I
J
K
So
and
then
there's
perspective,
it's
mainly
as
an
enterprise
I
might
want
to
to
to
go
and
check
the
what
has
happened
yeah
on
the
on
the
main
night
so
without
having
to
rely
on
a
third
party
vendor.
So
it's
interesting
later
on.
Maybe
you
could
defy
coming
along
and
then
you
just
want
to
link
a
proof
or
something
to
a
specific
token,
the
double,
a
etc
yeah.
K
Not
gonna
explain
the
the
use
case,
so
you
want
to
just
verify
unchained,
it's
like
okay,
this
this
is
in
here,
so
I
want
to
connect
and
then
yeah,
it's
mainly
connected
to
the
smart
contract
and
and
and
so
on.
That's
why
I
kind
of
driving
on
the
standard
of
the
client.
So
mainly
you
just
create
an
api.
I
create
my
my
applications.
You
can
start
doing
whatever
you
may
want
to
do
on
on
mainnet.
K
It's
another
use
case.
It
was
the
expense
of
the
of
the.
As
you
know,
as
more
and
more
enterprises
start
using
baseline,
you
might
want
to
use
the
player
to
just
to
reduce
the
cost
and-
and
maybe
it's
another
step
might
be.
Other
complexity
might
be
no
bit
required,
but
but
you
might
want
to
the
yeah
use
a
rollup
to
do
so
and
even
if
baseline
is
like
in
the
robot
being
soon
so
you
made
them
roll
over
that
roller.
K
Yeah
assassin,
you
know,
if
you
wouldn't,
I
understand
you
know
as
a
normal
enterprise,
you
say
hey,
I
can
hear
everything
I
just
draw
it
up
and
then
I
can
connect
etc.
Just
thinking
more
like
for
external
parties
who
you
know,
went
you're
doing
it
starts
saying
and
doing
voice
factoring
or
whatever
you
want
to
tokenize
something,
and
you
want
to
validate
other
proof
as
for
for
for
something
that
is
real
and
it's
been
validated.
K
So
yeah,
those
are
the
points
on
on.
You
know
why
why
everything
could
be
roll
up
and
why
the
idea
of
using
gas
or
or
optimism
or.
D
A
No,
I
think
I
think
yes,
I
mean
so
certainly
this
this
idea
of
using
roll-ups
is
some
just
yeah.
I.
A
A
very
valid
point,
because
so
certainly
I
know,
I
know
that
john
john
walput's
been
doing
some
kind
of
you
know
some
high
level
calculations
about
the
the
transaction
volumes
that
they'd
want
to
support,
but
yeah.
I
think
I
think,
given
that
you
know
in
the
next
one
you
know
pre-pre-two.
A
You
know
shards
anyway,
which
we're
looking
at
at
least
two
three
years.
You
know
it's.
It's
roll-ups
is
kind
of
the
scaling
solution
right,
so
it's
yeah.
It
seems,
like
sorry,
excuse
me
a
good
place
to
be
I'm
just
just
on
on
the
subject
of
the
actual
clients
as
well.
Now
I
did
see
that
we
had
is
a
nice
still
here
or
maybe
she's
dropped
out
up
until
a
few
minutes
ago.
Anyway,
I
thought
we
had
yeah.
We
did
have
an
ace
here,
but
she's
she's,
not
here.
A
Unfortunately,
the
the
the
reason
I
mention
her
is
because
she's
working
on
the
the
kind
of
baseline
standard
specification-
and
you
know
to
to
have
kind
of
that
more
generic.
You
know
interrupt
for
different
clients
as
well.
Of
course,
you
know
we
can
work
on
an
additional
client,
but
also
having
standards.
To
kind
of
you
know
make
very
clear
that
the
these
are
kind
of
this
is
what
they
need
to
support.
A
I,
what
nevermind
does
right
now
for
baseline,
so
yeah-
I
I
guess
was
this
something
to
follow
up
down
down
the
line
with
her
and
the
the
standards
works,
that's
happening
there,
but
that
will
certainly
be
one
way
to
make
it
clearer.
You
know
what
the
paths
would
be.
If
you
know
subsequent
clients
wanted
to
come
on.
I
think
that
you
know
such
as
whether
it's
gather
open,
ethereum
or
anything
else.
A
Yes,
so
then
really,
the
the
other
other
other
part.
Oh
an
ace
is
here
and
are
you
there.
A
Oh,
that's,
okay!
I
I
wasn't
sure
if
you
caught
any
of
that
there
are
you.
Did
you
hear
any
of
what
we
said
just
now
or
we.
A
This
year,
no,
no,
no!
No!
No!
I
saw
I
saw
you
were
here
and
I
thought
it
would
be
great
to
to
get
to
get
just
to.
You
know
your
view
on
this,
because
you're
driving
a
lot
of
the
standards
work
right
now
and
the
the
question
here
in
terms
of
what
we
were
discussing
was
you
know
right
right
now.
A
The
baseline
protocol
has
never
mind
as
its
kind
of
client
that
it
supports,
but
with
the
standards
work
that
you're
leading
within
the
group,
is
there
anything
at
this
point
in
time?
That's
looking
at
you
know
what
are
the
interfaces
that
a
client
actually
needs
to
support
for
baseline,
because
a
lot
of
across
all
of
the
work
that
you've.
L
Yeah
yeah
sure.
So
yes,
so
it's
basically
the
second
standard.
So
we
are
looking
at
interfaces
and
data
and
both
like
internal
and
external
as
well.
So
what
the
way
we
have
approached,
it
is
really
to
focus
on
what
are
the
category
of
interfaces
in
terms
of
what
are
the
capabilities
that
they
deliver.
So
if
we
focus
on
org
management,
for
example-
and
we
focus
on
I
registry,
what
are
the
functions
but
also
in
terms
of
the
data
model?
L
What
are
the
key
entities
and
the
relationship
between
those
entities,
so
different
clients
will
have
very
clear
specifications
about
how
all
of
those
should
be,
or
must
be.
I
should
say
must
so
should
be
implemented
and
we
have
started
the
work
and
a
lot
of
the
api
work
actually
is
inspired
or
is
built
upon
the.
What
was
released
in
v0.1
anyway
now
think
that
there
are
definitely
areas
of
improvement
and
things
to
be
discussed.
L
You
mentioned
socrates
for
privacy
secrets,
but
the
critics,
if
I'm
not
wrong,
only
support
zika
snark,
but
do
we
want,
from
a
standout
point
of
view,
to
really
limit
and
be
prescriptive
about
the
utilization
of
what
type
of
zero
knowledge
protocols?
The
answer
should
be
well.
No,
so
we
should
abstract
and
what
is
that,
focusing
on
the
zika
snark
related
interfaces
which
are
currently
in
the
different
reference
implementations.
L
One
discussion
that
we
could
have
is:
is
it
possible
to
abstract
even
further
those
interfaces
as
an
example,
so
we
ensure
support
for
other
type
of
zero
knowledge
protocols
on
the
persistent
side,
it's
not
currently
on
in
the
reference
implementations,
but
I
know
that
kyle
and
teams
are
actually
actively
working
on
that
so
once
it
will
be.
Yes,.
L
L
Service
so
once
it
will
be
pushed
to
the
repo,
we
will
basically
go
through
it
understand
what
are
the
functions
and
what
are
the
entities
that
power
this
entire
interface
and
start
the
conversation
on
the
stand
outside
from
that
key
point,
so
our
cycle
on
the
api
and
data
model
side,
rather
than
on
the
core
side
of
the
protocol,
is
always
looking
at
what
are
the
exit?
What
is
the
existing
implementation
as
a
visual
one
or
the
different
reference
implementation?
L
L
D
A
C
Persistence
package,
as
of
today
actually
has
it,
has
a
skeleton
in
there
and
I
used
if
you
want
to
look
at
that,
there's
some
there's
some
stuff
today.
L
I
do
have
a
question
while
you
are
changing
room
when
we
discuss
the
core
contract,
so
for
the
first
implementation,
the
shield
contract
was
really
one
of
the
cornerstone
of
the
implementation
right,
but
now
we
are
going
and
well
as
we
are
progressing,
we
realize
that
we
may
not
need
the
shield
contract
from
and
again
I'm
speaking
from
a
standout
point
of
view,
there
are
different
shielding
mechanisms
and
the
the
usage
of
a
shield
contract
is
one
implementation
of
those,
so
this
is
a
type
of
when
we
think
about
approaches.
L
For
you
know,
from
a
standout
point
of
view,
we
always
have
to
come
back
to
the
question
is
how
do
we
differentiate
between
the
implementations
that
yes,
allow
us
to
start?
The
conversation
in
terms
of
what
are
the
must
have
and
should
have
specification
versus
like
recommendation
like
guides.
This
is
what
nethermine
has
done.
This
is
how
they
have
done
it
to
date,
etc.
I
A
Sorry
right
all
set.
Thank
you.
Thank
you
for
those
additions
there
as
well.
It's
yeah
as
a
as
as
any
says
to
you
know
by
by
having
these
further
abstractions,
it's
kind
of
the
the
natural
extension
to
you
know
be
able
to
plug
anything
in
there.
So
moving
on
from
this
slide,
it's
kind
of
you
know
super
speculative
here,
because
we're
really
what
we're
saying
is:
okay,
we'll
say
if
we
plug
in
a
you
know
a
different
client
and
it
works
with
the
contracts.
A
We've
got
an
updated
project,
then
you
know:
do
we
want
to
start
looking
at
then
trying
a
different
sort
of
you
know
swapping
out
the
actual
privacy
circuits
that
we're
using,
and
you
know,
yeah,
here's
where
maybe
we
don't
want
to
go
with
you
know
is
it
you
know
it
does
zk
snarks.
Maybe
you
want
to
start
looking
at
starks
or
you
know
some
some
other
zero
knowledge
technology.
That's
here
having
that
flexibility,
there,
yeah
and
so
then,
and
and
then
finally,
you
know
we
can
get
into
different.
A
You
know
persistence,
modules
and
messaging
formats.
But
again,
if
there's
you
know
if
the
specification
kind
of
moves
faster
than
the
code
so
to
speak,
then
there'll
be
some
natural
abstractions
there,
which,
which
would
already
end
up
being
implemented
in
the
code.
So
it
would
make
that
kind
of
you
know
chopping
and
changing
a
lot
more
straightforward
phase.
Two
twice
so
I
mean
the
overall.
You
know
next
steps
here
anyway.
That
you
know
are
being
proposed.
A
Is
that
you
know
we
have
we
have
a
an
issue
created
and
there's
also
the
the
hackathon
starting
as
well
in
that
will
be
two
weeks
and
four
days
you
know
for
for
the
baseline
protocol.
So
you
know
what
would
be
a
great
output
from
the
session.
Now
is
if
we've
got
enough
information
there
so
that
you
know,
if
someone
decides
they
want
to
take
on
this
task
during
the
hackathon,
then
we
can.
Actually,
you
know,
they've
got
a
good
starting
point.
A
So
what
I'm
going
to
do
now
is
I'm
going
to
open
up
the
I
guess,
the
the
collaboration
space
so
to
speak,
the
you
know
the
the
kind
of
shared
doc
that
we
have
so
that
we
can.
You
know
I'll
start,
collaborating
basically
on
on
this
right.
So
can
everyone
see
what
I
have
open
there,
I'm
going
to
re-share
the
screen
just
so
that
it's
so
it's
it's
clearer
for
everyone.
Just
one.
A
A
Yeah
I
had
the
same
thing
happen
when
I
saw
on
a
call
the
other
day,
so
what
I
meant
to
show,
let
me
just
finish,
show
you
show
these
slides
again.
I
haven't
seen
that,
let's
see
how
many
people
haven't,
they
don't
get
around
to
changing
them.
Each
time
apple
released
a
new
release
of
versex
right.
So
I
I
was
just
kind
of
wrapping
up
the
last
couple
of
points
here,
which
was
you
know.
We've
got.
We
were
talking
about
a
road
map
with
the
potential
approaches
and
then
excuse
the
typo.
Here.
A
We've
got
phase
two
twice,
but
you
know
there's
the
potential
for
subsequent
phases
where
we
look
to
actually
have
different.
A
You
know
components
that
we
kind
of
swap
out,
because,
although
we're
talking
about
clients
and
languages
here,
there's
no
reason
that
we
can't
kind
of
then
move
beyond
it
to
start
provide
greater
flexibility,
for
I
know
the
messaging
layer,
the
persistence
layer,
like
the
you
know,
the
circuits
that
are
being
used
yeah,
and
so
that's
that's
really
what
that's
that's
that's
getting
to,
but
then
the
actual
the
next
steps
is
that
we
want
to
have
something
ready.
A
Really
for
the
for
the
hackathon,
so
that
there's
a
way
for
people
if
they
want
to
get
involved
in
this
and
use
the
hackathon
as
an
opportunity
and
there's
there's
somewhere
for
them
to
start
and
there's
enough
reference
material
that
it
becomes.
You
know
helpful
right,
so
I'm
going
to
now
share
the
correct
window
and
we
can.
I
guess,
the
the
more
collaborative
part
of
this
session
can
now
start.
A
I
I
presume
everyone
can
can
see
the
has
access
themselves
to
the
actual,
the
the
the
google
doc
here
for
notes.
A
So
what
what
we
have
in
here
again,
it's
you
know
it's
there's
not
much
in
there
right
now,
because
the
point
is:
is
that
during
this
session,
we're
going
to
start
putting
more
content
in
here,
but
what
we
have
included
a
reference
for
is
the
actual
the
the
github
epic
that's
been
created.
So
this
is
something
that
exists
in
the
the
public
baseline
protocol,
github
repo.
So
there's
you
know,
no
one
should
have
any
issues
accessing
it,
but
really
really
what
we're
doing
what
what
we
have
here
is.
A
Just
you
know,
we've
kind
of
fleshed
out
some
of
what
what's
been
discussed
here,
there's
not
really
more
info
any
any
more
information
here,
but
at
least
this
this
epic
can
kind
of
be
our.
You
know
our
reference
point
for
them,
the
subtasks
that
get
created,
which
could
be
the
items
to
look
at
for
the
actual
hackathon.
A
A
So
if
we
were
to
first
of
all,
I
mean
what,
in
terms
of
people
who
are
here
on
the
call,
you
know,
are
there
specific
areas
that
you
know
you'd
really
like
to
you
know
sort
of
be
prioritized
here.
You
know
what
what
we've
done
in
the
slide
deck
is.
We've
just
suggested
something
you
know
we've
just
suggested.
A
You
know
you
could
look
at
taking
the
the
blockchain
client
and
you
know
creating
a
new
example
project
that
uses
that
and
make
sure
that
the
core
contracts
work
with
it
you
know
is.
A
Is
this
you
know
a
sensible
initial
approach,
because
you
know
this
is
kind
of
just
off
the
back
of
some
discussions
that
kyle
and
myself
have
had
it's,
not
that
you
know,
there's
been
a
big
consensus
or
a
big
group
of
individuals
saying
you
know
this
is
the
area
that
you
know
we
think
would
be
best
to
start
so
you
know.
Does
anyone
have
any
opinions
or
views
that
they'd
like
to
share
here?
Okay
or.
L
Sorry,
I
have
more
of
a
question
so
from
a
standout
point
of
view
and
sorry,
I've,
I'm
looking
at
everything
through
that
lens
today,
but
if
we
think
about
portability
as
one
of
the
key
issues
well
not
issue
or
one
of
the
key
objectives
that
we
need
to
reach
is
there.
L
What
would
we
think
is
are
the
components
that
we
should
focus
on
to
ensure
that
a
client
is
able-
or
I
should
say
a
company
baselining
is
able
basically
to
port
their
existing
business
process
and
data
from
one
client
to
another.
L
I
L
A
Yeah,
I
I
think,
that's
that's
a
really
excellent
point
to
raise,
and
it's
just
in
terms
of
the
the
portability
kind
of
help
frame,
the
thinking
so
so
I've
I've
got
the
actual
the
session.
D
L
A
Yeah,
okay,
so
yeah
over
to
the
floor.
Does
anyone
have
something
they'd
like
to
start
with
in
terms
of
their
views?
I'm
going
to
stay
this
way,
so
it's
not
being
you
know
not
being
tainted
by
what's
been
suggested
so
far,.
C
Yeah,
I
think
the
I
baseline
rpc
interface
would
be
an
interesting
place.
C
Think
the
the
merkle
trees,
the
merkle
tree,
the
baseline
rpc
interface-
I
guess
that's
sort
of
part
of
the
the
api
implementation
component
or
maybe
maybe
you
would
consider
it
the
client.
It's
like
a
combination
of
the
client
component
and
the
api
implementation
component.
C
I
think
that
would
be
a
a
good,
a
good
place
to
dig
in
probably
to
start
yeah
and
tomas.
Can
you
know
tomas
can
give
some
good
insights
around
that.
K
It
can
be
stories
the
requirements,
so
they
can
be
matched
to
the
standards
track.
That
analyse
carries
so
mainly
why
this
requirement
of
the
miracle
tree
or
other
blockchain.
You
know,
even
you
know
what
we
do.
Everything
in
ethereum
could
will
need
to
do,
and
probably
that
that
translates
to
a
layer
to
it,
implementation
and
and
yeah
just
trying
to
it.
If
it's
a
full
requirement,
what's
the
requirement,
what
does
it
do?
K
J
Yeah
there's
already
a
section
of
the
standard
that
describes
the
baseline
rpc,
so
this
is
the
one
that
is
consistent
with
the
implementation
for
the
reference
implementation
one
and
it
will
be
evolving,
but
just
slightly
so
we
we're
introducing
a
few
wrappers
around
the
things
that
are
there
and
the
interface
itself
is
will
be
very
familiar
for
anyone
who
worked
with
the
ethereum
json
or
pc.
So
we
we
take
the
same
approach.
We
trick
the
block
parameters
exactly
the
same
way.
L
L
C
Know
the
one
thing
to
note
that
the
deploy
function
actually
is
already
deprecated
and
we've.
You
know:
we've
been
using
we've,
you
know,
we've
been
using
a
baseline
track,
that's
sort
of
the
the
key
I'd
say:
that's
you
know
the
key
right
now
because
you
like
it
in
terms
of
the
standard
itself,
yeah
you're
right,
it's
just
so.
C
It's
not
replacing
it
it's
a
baseline
track,
but
again
it
track
is
a
lot
more
abstract
right,
like
you,
can
think
of
of
that
the
track
method
to
apply
regardless
of
the
of
whether
it's
a
shield,
contract
or
trusted
execution
or
or
some
other
privacy
mechanism
right.
L
C
Basically,
we
were
using
baseline,
deploy
and
baseline
track,
but
now
we're
actually
just
using
track.
After
like
we
deploy
the
contracts,
you
know,
and
then
we
call
baseline
track,
which
is
really
what
what
initializes
the
merkle
tree
database
and
it.
E
C
Listening
for
updates
on
that
on
that
on
that,
on
that
shield,
contract.
J
Okay,
so
deploy
should
be
treated
as
a
as
a
tool
for
diagnostic,
so
any
enterprise
before
going
maintenance
that
can
use
deploy
for
very
quick
rapid
testing
of
the
private
chains
where
they
can
deploy
some
built-in
standardized
simplest
implementations
of
the
merkle
tree
and
so
on.
So
it's
not
part
of
the
requirement
for
the
clients.
It
may
be
treated
as
some
additional
tool
within
the
same
name,
space
that
is
available
in
netermind,
but
not
really
part
of
the
standard
requirement.
A
So
can
I
can
I
ask
that
if
someone
can
put
the
links
to
the
the
relevant
parts
of
the
specification
on
the
google
doc,
because
I
think
that
that
will
certainly
be
very
helpful
just
to
make
sure
everyone's
clear
on
this,
and
if
we
could
do
it
kind
of
right
at
this
point,
so
that
people
can,
you
know
open
it
up,
because
then
at
least
it's
easier
to
follow
the
discussion.
L
Yeah,
yes
sure,
so
one
thing
is
that,
as
I
said
in
the
opening
session,
we
have
the
building
blocks
and
the
high
level
requirements,
but
in
what
you
will
see
is
there's
a
draft,
so
I
will
include
the
link
now
requirements,
so
it
doesn't
it's
not
set
and
set
and
stone.
I
C
I
need
to
jump
connor.
I
just
wanted
to
say
thanks
for
facilitating
this,
and
I
will
circle
back
to
this
to
the
session
and
see
what
kind
of
notes
and
action
items
come
out
of
it.
But
yeah.
A
So
what
I
would
like
to
do,
though,
is
just
actually
where,
where
I'm
trying
to
get
to
here,
is
that
we
can
actually,
we
can
have
some
notes
that
you
know
can
then
make
it
into
the
actual.
Github.
Well
can
be
created
as
a
github
issue
that
so
someone
knows
exactly
where
they,
where
they
need
to
look
here
now
to
to
what
I'll
do
is.
If
we,
if
we
open
up
the
actual
the
link
that
is
has
given
us,
then
make
sure
that
everyone's
kind
of
clear
so.
A
So
is
it
basically,
this
the
baseline
rpc
calls
here.
L
Yes,
so
this
is
what
we
have
started
working
on,
but
well,
as
per
the
conversation
now
so
disregard
deploy.
So,
first
of
all,
the
structure
of
the
document
will
change
so
this
document,
the
api
data
model,
will
focus
on
all
the
interfaces,
their
description
and
we
will
need
to
provide
the
caveat
and
then
we'll
have
the
data
model.
So
for
each
what
we
call
requirement
categories,
for
example,
messaging
requirement,
categories,
persistence
or
org
management.
What
are
the
entities
so
work
groups,
user
organization,
those
type
of
things,
so
this
is
our.
A
Cool
and-
and
so
in
terms
of
thomas,
just
just
so
that's
kind
of
everyone's
clear
here.
What
you
did
basically
was
that
you
extended
ethereum
so
that
there
was
this
additional
json
rpc
api.
Basically,
that
could
be
it
could
support.
So
you
could
actually,
you
know,
have
a
merkle
tree
that
was
accessible
to
some.
You
know
something
well,
it's
accessible
to
someone
who
wasn't
a
process,
that's
not
actually
running
within
the
clients.
This
is
obviously
typically
the
case
with
ethereum
clients.
J
Like
we're
trying
to
to
to
keep
making
sure
that
cocktails
are
equally
great
and
going
forward
and
provide
for
the
net
car
community
so
as
for
the
extension-
yes,
we
even
we
even
made
it
like
a
plug-in.
So
it's
the
baseline
itself
now
can
be
just
dropped
into
the
nethermind
directory
of
plugins,
and
it
adds
a
a
few
components
to
nethermind.
It
adds
a
the
cli
extensions
for
some
more
like
diagnostic
developer,
tooling.
J
It
adds
the
rpc
methods
that
you
see
here
and,
as
mentioned
before,
the
deploy
one
can
be
treated
more
as
a
as
a
rapid
prototyping
locally
within
enterprises
without
connecting
to
mainnet
spawning
some
in-memory
database
and
testing
code
behaves.
J
J
So
it
actually
creates
something
that
starts
like
spawns
additional
processing
on
the
node,
so
it
tells
the
node
that
it
should
start
serving
information
about
the
baseline
trees
and
when
you,
when
you
start
tracking
the
baseline
tree,
you
will
have
a
much
faster
access
to
to
proves
to
verification
to
extracting
data
from
the
tree
and
and
pushing
the
data
to
the
tree.
J
Apart
from
that,
what
we
work
on
as
well
is
a
set
of
network
and
sync
extensions:
to
simplify
the
process
for
the
enterprises
to
sync
quickly
like
to
to
use
the
fact
that
we
have
nats
infrastructure
at
hand
that
we
actually
talk
to
enterprises
that
can
that
already
pay
for
data
storage
and
can
serve
us
that
data
much
quicker.
J
While
we
still
have
the
notion
of
trustlessness
and
verification
of
everything
we
can
actually,
instead
of
collecting
the
data
slowly
over
the
network,
how
it
is
on
the
public
maintenance,
I
mean
we
still
on
the
public,
mean
it,
but
we
can
actually
speed
up
some
processes
and
that's
it's
very
similar
to
to
multiple
approaches
from
various
clients
on
ethereum
network
now
to
either
usb
torrents
or
other
sources
of
data.
J
We
think
within
baseline
within
whether
another
minds
you
can
actually
sync
mainnet
a
bit
faster
soon,
just
by
using
those
solutions
and
and
using
also
provide
infrastructure
to
support
it
so
yeah.
I
think
this.
This
is
all
that's
coming
with
baseline
and
the
additional
configurations
and
the
extensions
that
make
the
nethermind
and
bay
and
provide
stack
much
closer.
So
we've
added
solutions
to
use
the
remind
client
and
use
the
like
transaction
payments.
Api
broadcasting
of
transactions
via
provide
payment
system.
So
you
can,
you
can
spawn
under
my
node.
J
You
can
send
the
transaction
there
without
the
signature,
without
the
without
holding
private
keys,
without
defining
gas
information
without
paying
for
the
gas,
and
that
transaction
can
be
sent
to
the
some
of
the
components
of
provide
stacks
here.
J
They
default
and
vault
that
handles
the
secrets
of
private
keys
and
so
on,
the
account
management
infrastructure
and
the
gas
payments,
so
the
transaction
is
actually
covered
for
you
and
you
can
work
with
providers
to
to
do
the
payments
processing
like
so
you
no
longer
have
to
worry
about
gas
about
private
keys,
and
these
are
two
things
that
usually
with
enterprises
are
for
the
most
problematic.
J
First
of
all,
why
do
I
have
to
hold
ease
that
is
highly
volatile
and
still
a
bit
unregulated
when
you
think
about
what
your
accountants
will
say,
what
your
compliance
will
say?
So,
there's
one
thing
and
the
other
thing
is
private
keys.
It's
like
it
feels
expensive,
problematic,
and
so
once
you
want
to
have
some
custodian
for
the
case,
all
of
that
comes
together
and
not
all
of
it
is
part
of
baseline.
But
it's
part
of
that
work
that
we've
done
as
bringing
various
companies
that
work
on
baseline
to
deliver
better
experience.
A
Yeah,
this
is
really
really
helpful
and
we,
with
with
the
actual
nether
minds
docs
as
well,
because
of
course
that's
you
know,
they're
probably
complementary
to
what's
in
the
specification
right
now
is.
Is
there
a
section
that
talks
about
some
of
these
extensions?
So
I
was
having
a
quick
look
there
just
to
see
if
we
can
stick
in
a
reference
to
a
specific
part
of
the
documentation.
Again,
it's
just
for
the
picture
of
completeness
in
the
actual
google
doc
that
we're.
J
Sure
I'd
be
happy
to
to
share
the
one
link
to
the-
maybe
maybe
not
only
but
like
somewhat
slightly
updated
version
of
the
docs,
but
I
think
you
could
still
use
it
for
for
spawning
the
entire
baseline
process
locally
by
launching
two
nether
mind:
nodes
that
create
in-memory
networks
and
making
them
deploy
the
trees
and
start
talking
to
each
other.
J
A
Yeah,
so
if
you
stick
it
in
the
we
have,
we
have
the
the
the
actual
notes
section
here
and
so.
A
J
Okay,
I
have
it
yeah.
A
So
so,
to
try
and
kind
of
sum
up,
what's
been
discussed
so
far
here
really
in
terms
of
the
you
know
in
in
order
to
support
a
different
client,
the
the
the
the
main
consideration
is
that
that
client
can
support
these
additional
json
rpc
api
calls.
A
I
think
that
would
be
a
fair
assessment
of
what's
been
said
here
with
respect
to
you
know,
juan's
question
earlier
on
and
points
about.
You
know
making
use
of
actual
layer,
2
chains
as
well,
and
this
this
is
more
a
question
just
for
you
know,
with
the
specific
to
the
actual
json
rpc
api
do
do
all
of
the
kind
of
you
know.
The
transaction
capabilities
go
through
this.
A
This
kind
of
same
json,
rpc
api
that
we've,
you
know
mentioned
here
or
are
there
some
other
routes
that
they
actually
go
in,
because
this
is
obviously
used
for
the
the
merkle
tree.
But
then
there's
also
you
know
transactions
happening
as
well.
J
Yeah,
that's
that's
much
nicer,
because
actually
the
transactions
are
using
exactly
the
same
apis
that
are
already
in
ethereum.
So
when
you,
when
you
call
ethereum
and
when
you
want
to
broadcast
transactions
by
ethereum
node,
you
will
call
ath
module,
which
is
exposing
two
methods.
One
is
called
send
transaction
and
the
other
one
is
called
send
draw
transaction.
J
So
one
is
a
transaction
without
signature
that
assumes
that
your
node
somehow
handles
the
signing
and
and
adding
yeah
adding
signature.
The
private
keys
this,
the
other
one,
is
just
pushing
to
the
node
transactions
already
signed
ready
for
broadcast.
So
so
the
latter
doesn't
change
at
all
the
the
former,
the
one
that
requires
the
note
to
handle
your
private
keys.
Aloe
demands
to
to
forward
that
responsibility
to
provide.
J
So
this
is
this
enterprise
flow,
where
you
push
the
transaction
that
express
what
you
really
want
to
do
with
the
transaction
and
that
the
transaction
is
wrapped
into
the
gas
payment
and
distributed.
But
you
use
exactly
the
same
api.
There
are
no
new
apis
added
at
all.
It's
just
a
matter
of
configuration
of
the
node
itself.
A
With
the
does
this
baseline
in
its
current
state
anyway,
you
know
use
one
over
the
other
intent
to
in
terms
of
send
transaction
or
send
raw
transaction.
J
No,
it's
like
you,
it's
it's
total
freedom
of
choice.
It's
if
you
want
to
pay
for
the
transaction
yourself.
If
you
can
manage
private
keys
yourself,
and
I
believe
some
companies
will
want
to
do
that
and
they're
totally
fine
to
use
the
standard
flow
with
the
transaction
pool
and
broadcasting.
J
A
Yeah,
okay,
okay!
So
to
to
to
answer
the
question
that
bond
raised
there,
then
you
know
as
long
as
these
level
layer,
two
scaling.
If
the
layer
two
scaling
solution
presents
a
standard,
json
rpc
api,
then
it
wouldn't
be
problematic
at
all
to
to
work
with
it.
K
So,
in
a
way
what
they
will
need
to
implement,
obviously
you
need
to
have
the
miracle
tree,
the
storage,
the
we
that
will
need
to
be
sync
across
different
nodes,
or
just
just
a
specific
for
performance.
It's
just
it's
just
a
single
node,
that's
distorted
data
for
performance,
okay,
and
then
you,
you
expect
you
what
you
have
said,
a
cent
transaction
which
will
be
normally
used
when
you
have
to
account
unlock
or
something
like
that.
K
You
mean
the
in
the
same
way
that
quorum
works,
that
you
have
your
authenticate
with
your
account
or
something
that
will
that
will
delegate
the
the
payments
or
whatever.
Obviously
that
will
be
done
in
a
private
network.
So
it's
not
like
you
opening
your
note
to
everyone
to
use
right,
yeah
like
and
obviously
the
bold
element
that
provides
provide.
K
M
J
K
K
K
J
Mean
like,
like
probably
for
any
ethereum
node
on
the
public
maintenance
that
you
would
host.
Obviously
it
would
be
publicly
visible,
but
the
json
rpc
itself
would
be
just
private
for
you,
so
you
would
ensure
that
the
firewall
protects
it.
So
the
same
case
is
here:
it
provides
some
additional
capabilities
to
support
you
with
the
standard
queries
around
baseline
and
potentially
the
transaction
orchestration
extensions,
but
it's
it
doesn't
require
a
setup
of
private
network
or
any
special
tokens
or
anything
like
that.
K
Yeah
yeah,
okay,
yeah
see
that's
this
yeah.
That's
a
very
nice
explanation.
It
just
simplifies
the
baseline
or
orchestra.
It's
gonna
work
using
a
public
node
that
connects
to
with
your
enterprise.
J
I
think
the
best
intuition
would
be
when
you
think
about
ethereum
nowadays,
obviously
very
often
you'll
be
asking
about
the
logs
right,
so
you'll
call
ifget
logs
and
in
a
way
you
don't
this.
This
interface
is
not
necessary
for
the
node
to
exist
right,
so
you
can
always
listen
to
the
blocks
incoming
extract
the
logs
yourself
index
them
on
the
side.
So
what
baseline
rpc
does
it's
it's
a
similar
work
it
it
handles
for
you
all
the
events
that
are
very
specific
to
baseline
and
simplifies
for
you
declaring
methods.
J
It
provides
you
already
with
the
with
the
storage,
with
extraction
with
verification
of
those
things
you
can
write
them
all,
all
of
them
on
the
side
and
they're
like
not
necessary
for
the
node
to
operate,
but
they
do
just
a
lot
of
work
for
you.
You
don't
have
to
do
that
yourself.
Yeah.
K
I
K
So
that
could
be
kind
of
part
of
the
the
standard
as
well.
You
know
the
and
just
trying
to
build,
as
you
may
need
you
know
to
that.
Was
kind
of
that
was
the
point
originally,
as
you
may
need
to
verify
that
the
third
party
provider,
the
is
sending
your
transactions
and
such
a
victim
do
they.
You
know
this
is
an
audit
capability
making
sense.
K
J
K
J
J
K
Yeah,
so
on
the
layer,
2
scenario
in
in
a
layer,
2
scenario:
oh
yeah,
that
kind
of
will
be
slightly
trickier,
so
just
put
the
nether
mine
as
the
the
main
node,
and
then
you
have
a
optimism
as
the
layer
two.
K
So
I
see
supposed
the
this
mark,
the
seal,
smart
contract
and
so
as
a
user
can
send
me
the
transactions
to
optimism
that
will
be
relocked.
Eventually,
events
won't
be
triggered
yeah
because
you're
doing
it
roll
up,
and
then
you
just
mainly
store
in
state.
Well,
you
didn't
start
this.
You
just
just
store
the
call
data
and
then
that
later
on
can
be
verified
in
the
role
of
etc.
K
So,
in
that
scenario,
optimism
on
its
own
would
need
to.
If
you
want
to
provide
this
mechanism,
we
need
to
validate
it
itself
and
for
a
third
party
to
audit
and
maintain.
Obviously
they
don't
have
access
to
the
to
the
seal.
This
smart
contract.
We
have
to
go
through
the
optimism
lawyer
later
later
or
rebuild
even
more
fun,
all
the
all
the
data
of
the
using
the
input
in
the
in
in
in
mainnet
and
and
then
right.
Okay,
I
won't
go,
but
just
it's
just
thinking
all
the
problems,
but.
K
J
Yeah,
I
think
that
seriously
optimists
will
not
like
will
not
provide
you
the
state
of
dating
transactions,
but
it
can
actually
provide
you
with
a
list
of
call
data
and
by
using
exactly
the
same
rules
you
can
you
can
in
a
way
on
layer
one
and
by
having
those
baseline
extensions
start,
acting
as
a
verifier
of
what
the
optimistic
roll-up
pushes
by,
but
just
running
those
and
checking
yes,
as
baseline
participant
checking.
If
everything
makes
sense
so
yeah,
you
can
do
that.
I
A
Okay,
so
I
think
you
know
we
certainly
you
know
over
the
course
of
the
session
so
far
we
we
have
clarity
on
you
know
what
would
be
the
right
next
steps,
and
I
think
you
know
what
what's
what's
great
here
as
well.
Is
that
you
know
it's
very
clear
that
you
know
helping
using
the
specification
to
help
guide.
This
is
going
to
be,
you
know
very,
very
helpful
and
make
make
life
a
lot
easier.
A
Did
anyone
else
really
have
anything
else
that
they
wanted
to
bring
up
or
discuss
or
any
other
thoughts,
because
I
I
think,
in
terms
of
you
know
what
we
set
out
to
achieve
with
respect
to
this
session,
was
you
know,
figuring
out
what
the
path
forward
would
be,
and
I
I
would
say
that
you
know
we
are
there
basically
and
then,
of
course,
you
know
you
can
follow
the
specification
to
you
know
to
try
switching
to
a
different
client.
A
Then
probably
the
sensible
step
after
that
is
to
look
at
using
a
different
circuit,
and
then,
after
that
you
can
start
chopping
and
changing
the
more
kind
of
mundane
pieces,
such
as,
like
your
persistence
layer,
your
messaging
layers,
because
these
are
sort
of
you
know
very
well
established
they're,
not
kind
of
you-
know
areas
which
people
won't
be.
A
A
You
know
these
are
these
spaces
are
evolving
a
lot,
and
so
it's
not
something
that
we
can
take
for
granted
that
you
could
just
pull
out
one
thing
and
put
another
one
in
the
same
way
that
you
know
you
could
take
that
sadness,
you
know
switch
it
for
you
know
with
another
points,
point
messaging
protocol,
or
you
know,
switch
from
one
persistence
there
to
the
other
yeah.
So
we
was
was
there
anything
else
that
anyone
wanted
to
cover
off
or
use
the
session
to
discuss?
A
Now,
because
we
do,
we
do
still
have
35
minutes
again.
If
you
know,
if
there
isn't
anything,
then
you
know
we
don't
have
to
keep
the
session
open
either
either,
but
yeah
I'd
love
to
put
it
out
to
the
floor.
I
A
Alternatively,
if
anyone
has
any
questions
along
the
lines
of,
why
is
this
like
this
in
baseline?
I'm
sure
you
know
the
the
this
could
be
a
chance
to
talk
about
that
as
well.
K
Another
question,
as
we
were
talking
about
different
components:
that's
that
question.
It's
a
general
comment
as
we
progress
through
those
and
then
explore
different
alternatives.
We
still
have
these
conversations
as
well
like
in
this
similar
kind
of
submit.
I
think,
as
more
people
get
interested,
they
will
bring
more
ideas
to
the
to
this.
We're
like
talking
about
messaging.
K
K
You
know
so
that
that's
kind
of
key
on
the
on
the
standard
element
and
but
I
think,
we'll
discover
as
we
move
along
and
and
we
have
different
proprietary
needs
as
well,
like
I
said,
with
ibm
and
and
then
all
at
the
prices
you
know
as
well,
which
will
definitely
won't
have
open
move
the
enterprise
messaging
mechanism
because
they
will
make
sense
to
them.
So
so
that's
going
to
be
connor.
You've
been
doing
a
lot
of
messaging
in
the
past.
You
probably
can
add
a
lot
of
that
in
here.
A
Yeah
yeah.
Well,
I
I
think
yeah
john
walpur,
you
know
constantly
talks
about
make
it
all
as
boring
as
possible.
So
you
know
it
looks
like
a
normal
technology.
You
know
not
something
shiny
and
new,
which
you
know
it's
it's
harder
to.
You
know
it's
always
takes
lot
longer
times
for
adoption
for
adoption
and
enterprises.
A
K
Yeah,
but
you
still
need
to
have
that
current
mistake,
like
hey,
I
want
to
put
investment
in
here
because,
obviously
it's
going
to
return
me
all
this
and
improve
my
messaging,
but
later
on,
it's
like
whoa.
Look
at
all
this
new
new
disruptive
technology
that
can
join
into
into
it.
So
I
keep
saying
that
that's
the
non-boring
part,
that's
the
that's
where
the
huge
ira
will
come
later
and
yeah.
A
K
But
yeah
yeah,
I
think
just
covering
this
for
now
is
fine
messaging,
will
come
later,
like
any
thing
that
you're
gonna
be
joining.
You
know
just
keeping
the
chat
going
and
deeds
and
the
company
phone
book
is
going
to
be
very
interesting.
A
Definitely
and
there's
I
mean
in
terms
of
the
actual
sessions
there's
trying
to
think
when
it's
actually
taking
place
the
right
there's
a
session
on
the
global
phone
book
problem.
Yes,
sorry
perfect,
so
there's
a
project
rather
than
a
problem
yeah
which
which,
which
is
coming
up
but
yeah,
look.
You
know
if
I
I
think,
unless
anyone
else
has
anything
else,
now's
a
good
time
for
us
to
wrap
this
session
up
I'll
type
up.
A
Just
some
of
the
you
know,
just
we've
kind
of
got
the
core
information
here,
but
there's
a
few
further.
You
know
we
want
to
ensure
that
the
out
there
is
just
kind
of
actual
output
from
this,
and
so
we'll
just
summarize
what
we
have
in
the
notes
here
and
then
we'll.
Let
everyone
jump
off
and
go
and
join
in
with
one
of
the
other
sessions
if
they
want
or
get
on
with
them.
You
know
what
whatever
else
they
wanted
to
do
with
their
day.
So,
firstly,
just
quickly.
A
Thank
you
to
everyone
and,
for
you
know,
who's
been
here
special
thanks
as
well
to
thomas,
because
you
know
you
kind
of
you
know.
I
think
kyle
pulled
you
in
and
you
know
it's
been.
You
know
really
really
valuable
to
get
your
input
here
as
well
and
is
in
here
but
she's
been
doing
a
fantastic
job
with
the
specification
kyle.
Of
course
you
know
the
work
he's
done
on
baseline
and
also
the
work
that
provider
doing
is.
A
You
know,
thomas
pointed
out
too:
there's
listen
to
some
very
cool
things
happening
there
and,
of
course,
one
for
your
contributions
to
the
the
discussion
as
well.
It's
just
it's
been
really
good
and
yeah.
It's
you
know
the
ethereum
got
mentioned,
and
it
is.
It
is
a
great
library
for
net
developers
and
likewise.
A
On
the
client
side
too,
so
a
shout
out
to
people's
stacks.
A
A
H
A
A
Hey
john,
so
the
sessions
we
we
wrapped
it
up
a
few
minutes
ago.
H
Action
item
yeah.
H
I'm
trying
to
I'm
kind
of
going
through
all
of
them
right
now
and
making
sure
folks
are
creating
activities
so
that
it
doesn't
become
just
all
education.
A
Yeah
yeah
sure,
so
so
I'm
in
the
process
right
now
of
typing
up
the
notes
from
the
session,
but
but
I
mean
what
was
really
great,
though,
was
that
anis
was
there
as
well
and
thomas
from
nethermind,
and
so
they
they
were
able
to
provide
some
really
useful
background.
A
And,
firstly,
you
know
anise
was
you
know
pointing
out
how
the
actual
specification
can
help
drive
this
this
work
and
you
know
we're
able
to
clarify
which
part
of
the
spec
and
people
can
use
in
order
to
actually
kick
off
this
piece
of
work.
So
it's
really
cool
in
that
it's
actually
pulling.
You
know
specification
work
in
as
well.
As
you
know,
the
code
that
exists
right
now,
so
that's
kind
of
an
added
bonus
of
it.
A
The
other
part
as
well
was
you
know,
juan
blanco,
brought
up
this
whole
notion
of
well.
You
know
what
about
baselining
on
like
a
layer,
two
solution,
if
someone
you
know
wanted
to
you
know
whether
they
didn't
want
to
use
main
efforts.
Oh.
H
Absolutely
that's!
That's
a
big
chunk
of
the
of
breakout
number
five
right
there
or,
depending
on
how
you're,
counting
but
yeah
the
baseline
scaling
works,
scaling
the
work
step.
Throughputs!
That's
all
going
to
be
about
layer,
twos,
the.
A
Yeah
I
mean
the.
The
conclusion
we
reached
anyway
was
that
it
should
actually
be
trivial
if,
if
they're
presenting
just
the
standard,
you
know
ethereum
json
rpc
api
for
transacting
with
them,
then
you
know
you
can
you
can
just
use
it
there?
But
of
course,
if
you're
trying
to
spam
multiple
networks,
then
yeah
it's
going
to
get
a
bit
more
complex.
H
The
risk
I'm
taking
here
that
we're
all
taking
is
well
it's
not
a
big
risk,
I
mean
so
we
have
a
lot
of
dead
air.
It's
not
a
big
deal
but
trying
to
make
sure
that,
like
even
after
this
live
session,
we
you
know
people
pop
back
in
have
conversations
that
keeps
going
that
we're
activating
people
to
to
do
to
have
discussions
throughout.
You
know
the
period,
and
hopefully
I
mean
ideally,
if
somebody
like
on
the
other
side
of
the
planet.
That's
waking
up.
H
D
H
So
that'd
be
good,
did
you
guys
figure
out,
I
mean
did
do
you
feel
like
you've,
come
up
with
any
issues
that
so
far,
that
will
that
we
can
turn
into
github
challenges.
A
Yeah
yeah,
I
mean
what
you
know.
What
well
I
mean,
I
think,
there's
a
very
clear
one
and
so
off
off
the
back
of
this.
There
was
kind
of
I
think
I
think
two
so
there's
various
ways
we
can
kind
of
you
know,
break
break
this
down
in
terms
of
you
know,
chopping
up
the
baseline,
so
that
you
can.
You
know
maybe
like
this.
The
simple
things
which
would
be
you
know,
use
a
different
persistence
layer
or
use
a
different
messaging
layer.
You
know
where
these
are.
A
You
know,
there's
lots
of
different
solutions
to
choose
from
they're
tried
and
tested
well
understood,
but
I
think
that
the
the
two
areas
that
there's
probably
that
are
least
clear
right
now,
but
that
are
important
for
you
know
growing
the
overall
baseline
ecosystem
are
one
is
the
actual
client.
So
if
someone
doesn't
want
to
use
nevermind,
they
want
to
say
basu
what
needs
to
be
done
and
off
the
back
of
this
session
it.
H
Makes
sense
by
the
way
you
guys
ought
to
get
back
together
tomorrow
when
sam's
back
he's
got
a
ton
on
this.
H
Yeah,
sam
stokes
he's
he's
he's
on
on
pto
today.
He
says:
census
regrets
but
he's
coming
back
and
he's
got
a
ton
on
on
ways
to
get
multiple
clients
working
baselined.
A
Okay,
awesome:
well,
what
I'll
do
is
I'll
I'll
actually
point
him
to
these.
These
notes
as
well
that
have
written
that
I'm
writing
up,
because
my
view
was
that
you
know
we
can
take
these
and
basically
create,
like
you
know,
a
an
issue
off
the
back
of
the
epic
for
this.
So
you
know,
at
least
if
we
forget
his
input
as
well,
then
you
know
that's
that's
fantastic
and.
H
That's
great,
maybe
maybe
put
him
you
know,
put
a
at
sign
for
him
in
the
slack
for
this
channel
yeah
and
tell
him
tell
him
what
give
him
a
sense
of
time
stamp
like
when
how
far
back
in
time
he
should
go
to
to
get
your
on
the
live
stream
and
he
can
just
review
what
what
what
you
guys
discussed
on
the
live
stream.
A
H
Just
tell
him
show
him
how
you
know
tell
him
what
time
frame
to
look
for
actually
he'll
be
able
to
scrub.
I
mean
it's
pretty
obvious,
because
you
know
nobody
like
like
right
now.
People
will
hear
us
talking,
but
it'll
be
pretty
obvious,
where
the
cuts
will
be
actually
be
pretty.
E
A
Yeah
exactly
exactly,
but
the
other
thing
as
well
was
that
you
know
a
and
a
niece.
You
know
highlighted
this
was
well.
You
know
what
what
if
someone
doesn't
actually
want
to
use
like
a
zk
snark,
you
know
it
might
be
that
they
they
want
to
use
something
else,
and
so
being
able
to.
You
know,
have
some
flexibility
there
on
the
privacy
side.
That's
that's
going
to
be
another
key
consideration,
so
you
know
I
would
you
know,
think
that
off
the
back
of
this
session,
it's
like
those
are
kind
of
the
two.
A
You
know
bigger
impact
areas
to
focus
on
because
they're
they're,
almost
like
the,
where
there's
more
uncertainty,
because
they're
not
so
well
established.
Like
you
know
the
other
kind
of
you
know
the
inter-process
communication
by
and
how
these
other
bits
sort
of
fit
together.
You
can
read
that
in
a
different
language,
if
you
need
to
whatever
else
you
know
it's
just
kind
of
fairly
standard
work,
but
you
certainly
in
this
with
the
world
right
now.
You
don't
have
lots
of
choice
with
clients
and
you
don't
have
lots
of
choice
with
different
zero
knowledge.
H
Period
so
that's
yeah,
thomas
walton
polka.
Can
those
guys.
A
I
know
tom
yeah
so
yeah.
I
think
that
you
know
those
are
probably
the
two
best
kind
of
actions
off
the
back
of
this
session,
which
we
can.
H
You
know
be
great
to
see
those
issues
in
underneath
that,
under
the
under
underneath
this
particular
epic,
on
the
on
the
roadmap
and.
H
Boy
the
brass
ring
is
for
for
all
of
these
sessions
and
you
know,
presumably
this
isn't
the
last
we're
gonna
have
activity
in
this
channel
right,
there's
more
more
to
come.
A
H
H
So
yeah
you
know,
hopefully
other
people
will
get
inspired
by
this
I
mean
that's
really.
The
goal
of
all
of
these
is
to
inspire
people
to
actually
generate
produce,
work
right.
A
In
the
repo
yeah
exactly-
and
I
think
I
mean
what
what
iu
was
hoping
that
off
the
back
of
this,
we
can,
then
you
know
if
we
can
have
these
issues
and
then,
if
maybe
we
could
get
some
funding
from
the
bitcoin.
You
know
for
the
hackathon
to
actually.
H
A
H
D
H
Bounty
within
that
hackathon
that
we
we've
set
up,
so
anybody
can
come
in
there
and
put
more
bounties
on
issues,
but
I
we're
funding
at
least
eight
or
up
to
eight.
H
Looking
at
the
end
of
friday,
at
we'll
we'll
get
together,
everybody
and
say
all
right:
what
are
the
issues
that
that
are
worthy
of
bounty
and.
H
Is
this
being
good?
How
is
this
working
for
you
connor?
Is
this
the
the
how's
this
going
for
your
company
and
you
know
where's
the
what's
the
brass
ring
for
you
guys.
A
So
I
I
I
think
at
this
point
in
time
like
we
don't
have
anyone
using
baseline,
but
I
I
think
that
the
the
opportunity
the
baseline
you
know
signifies
is
significant.
H
H
For
like
food
brands
and
stuff,
and
I'm
like
yeah
yeah,
you
can,
and
it
was
just
a
light
bulb
for
me,
I'm
like
oh
yeah
this.
This
is
kicking
right
because
it's
not
like
people
are
wondering.
Oh,
let
me
get
an
education
about
blockchain.
This
is
like.
Oh,
I
have
a
consistency
problem
here
and
here
I
need.
I
need
to
fix
it.
H
D
I
D
A
Abs,
absolutely
yeah,
so
I
mean
yeah,
I
mean
there's,
certainly
some
use
cases.
I've
got
in
mind,
but
it's
it's
more
just
yeah.
It's
the
whole
thing
of
connecting
with
the
you
know,
tracking
down
the
people.
You
know
what
I
do
want
to
start
doing
is
putting
some
content
out
there
around
baseline
as
well,
because
that's
that's
kind
of
you
know
an
area.
That's
well.
D
H
H
Ultimately,
that's
that's
you
know.
That's
what
we've
got
to
generate
is
more
more
for
the
specs
and
more
for
for
the
repo.
You
know
we
want
to
get
a
lot
more
people
contributing
enabled.
That's
that's
really.
What
I'm
hoping
is
that
this
whole
session
is
going
to
be
about
is
enabling
more
people
to
contribute
more.
D
H
I
H
We
were
just
talking
about
like
what
what
kind
of
act
you
know
like
what
kind
of
issues
you
know
like
codable
issues
we
could
generate
out
of
all
of
this.
I
think
it
might
be
worthy
of
some
time
tomorrow
when
sam
stokes
is
back
to
to
jump
into
this
room
again
with
some
folks
that
are
really
you
know,
keen
and
because
he's
got
some
really
good
ideas
about
this
stuff
too.
K
From
an
eye's
perspective,
having
a
baseline
api
from
the
net
about
java.net
side
when
things
are
settling
there
or
will
are
settled,
it
will
be,
will
be
great.
I
think.
K
Well,
well,
for
the
baseline
api,
it
will
be
the
the
from
the
smart
contract
integration,
exactly
constant
standard
to
the
to
the
cl
rpc,
and
so
that
will
be
the
yeah
just
the
most
simplistic
and
an
api
layer
from
an
example
point
of
view.
Ideally,
I
think
we
should
have
the
reference
examples
from
java
and.net
in
a
similar
way
that
we
call
the
javascript
one.
A
I'm
here
still,
I
mean
I'd,
definitely
like
to
have
that.
It's
it's
it's
more
about
just
you
know
getting
the
work
done,
because
I
I
I
don't
have
anyone
on.
I
don't
have
the
capacity
on
my
team
at
the
moment
to
put
someone
on
it.
I
I
would
love
to
get
someone
there
to
just
you
know,
get
it
running
on
top
of
basically.
I
H
H
I
A
And
this
this
is
where,
like
you
know,
with
the
session
just
now,
we
kind
of
figured
out
what
the
phases
would
look
like,
because
in
the
first
part,
is
literally
just
get
it
working
with
hyperledger
basu,
and
then
you
know,
there's
there's
another
piece
which
is
the
you
know.
You
know
providing
flexibility
on
the
underlying
circuit
that
it
uses,
but
that
that
doesn't
have
to
be
done,
obviously
as
a
prerequisite
and
then
the
other
part
is
yeah.
A
You
know
basically
rewrite
what
is
the
javascript
stack
on
it
in
in
for
the
jvm
and
you
know.net,
so
those
are
kind
of
those
additional
tasks
there.
H
So
that's
going
to
improve
deployability.
What's
it
going
to
do
for
an
ethereum,
I
mean
like
what
where's
that
where's,
the
where's
the
brass
ring
for
that
for
you.
H
Well,
you
know
I'm
always
trying
to
find
ways
for
participants
in
the
community
to
achieve
their
goals
through
the
community.
Doing
you
know,
working
with.
G
H
K
Well,
you
know,
like
I'm
part
of
the
standards
and
I'm
looking
at
everything.
So
to
me
it's
as
an
output,
it's
just
it's
having
services.
If
we
know
if
we
know
that
something's
going
to
be
stable,
then
when
people
are
going
to
come
to
me
to
sue
for
support
from
the
open
source,
they
can
say
I
we
can
agree
that
this
is
something
solid
and
to
me
that
will
be
the
you
know
that
can
mainly
guide
them.
K
You
know
I
can
that's
why
I'm
asking
so
many
questions
right,
you're
working
a
lot
with
microsoft,
as
I
remember
right,
but
I
used
to,
but
obviously.net
is
heavily
used
by
all
the
mic:
microsoft,
users,
okay,
so
so
mainly
in
the
same
way
as
people
will
integrate
with
web3j.
H
So
so
so
what
all
right?
So
I'm
trying
to
think
so
if
I
have
a
net
built,
erp
type
application
or
some
other
kind
of
system
of
record
and
it's
in
net,
then
the
then
the
ethereum
comes
into
play.
K
Yeah,
yes,
the
same
as
we
did
with
sap
integration
like
we
did
when
when
I
was
in
consensex,
which
is
mainly,
I
say,
p,
to
sap,
with
a
smart
contract
in
the
middle.
So
if
you
remove
the
smart
contract
in
the
middle
and
then
you
might
say
like
okay,
I
want
to
submit
some,
you
know,
part
of
the
purchasing
process
and
and
so
on
and
use
space
like
in
the
middle
and
obviously
have
some
verification
of
the
proof
customer
submitted.
K
Maybe
you
might
want
to
create
those
bobby
interfaces
in
sap
that
provides
the
information
of
the
proofs
or
whatever.
K
So
those
are
the
areas,
the
simple
areas
and
later
on,
like
a
week
discuss
eventually
things
will
move
into
the
the
main
main
tokens
will
move
to
the
main
mate,
and
then
you
will
want
to
have
a
full
circle
here,
as
you.
H
May
you
know
parochially
the
the
main
or
the
the
baseline
baseline
structure
doesn't
have
a
lot
of
smart
contracts.
Right
I
mean
it's
it
or
tokens,
it's
just
three
of
them
right
and
they're
always
the
same
they're.
You
know
well
as
we
as
we
mature,
then
they'll
change,
but
they
we're
talking
about
the
org
registry,
the
shield
contract
and
the
verifier
yeah.
H
Everything
else
is
is
in
either
containers
of
code
that
are
have
no
bearing
I
mean
there
aren't
blockchain
at
all
right,
they're,
just
code,
dotnet
code
or
java
code
right
and
all
you're
doing
is
signing
those
and
saying
yep.
The
same
code
here
is
was
executed
over
there
and
I
think
that's
an
interesting
place
to
to
think
about
right
where,
if
I've
got
a
container,
that
has
some
business
logic
off
chain
right,
but
I
want
to
baseline
that
container
and
the
data
right.
So
I
don't
want
to
just
baseline
that
purchase
order.
H
I
want
to
baseline
the
business
logic
that
goes
that
I'm
sending
you,
along
with
the
purchase,
order
right
and
that
can
say
I'm
going
to
docker
that
up
send
it
to
you
if
I'm
on,
if
I'm
not
on
a.net
system
and
you're
on
a.net
system,
you
know
to
baseline
it
they're
going
to
have
to
be
signed
the
same
and
they're
in
different
execution
environments,
which
is
why
containers
are
awesome
now.
H
K
K
Mm-Hmm
but
that's
wouldn't
you
start
moving
to
to
serverless
and
start
trying
to
do
something
with
the
microsoft
phone
guys
with
functions
so
well,
cryptolysis
to
maybe
grip.
H
Let's
say
yeah
or
or
yeah
or
you
get
lambda
functions
right.
H
That's
kind
of
an
interesting
idea
right,
so
you
just
say:
okay,
we're.
If
I
have
some
business
logic
that
goes
along
with
some
work
step
like
here's,
my
master
service
agreement
here,
the
the
conditionals
and
the
loops
and
stuff
that
you
know
that
involve
that
particular
piece
of
logic,
and
I
want
all
subsequent
work
steps
to
follow
the
rules
of
that
container,
and
I
don't
want
those
rules
on
a
blockchain
and
I
don't
want
to
write
them
into
a
into
a
bunch
of
zero
knowledge
circuit
math.
H
Then
you
need,
we
were
calling
this
way
back.
We
were
calling
this.
You
know.
I
need
to
send
you
a
code
book
right.
That
is
the
same.
Almost
I
remember
I
was
watching
a
an
old
apollo
on
the
apollo
missions
and
and
how
you
know,
computers
in
those
capsules
or
in
the
in
the
in
those
spacecraft
in
the
60s
and
70s
how
they
operated
right.
They
were
just
yeah,
effectively
code
books,
you
know
good,
go
call
line,
62.
H
All
right
now
go
live
call,
you
know
and
then
line
62,
we'll
call
line
45
and
you
know
that's
how
they
work,
and
so
you
kind
of
want
that.
You
want
we're
all
calling
line
62
and
we
all
know
that
line.
62
is
the
same
line
and
you
sort
of
want
that
to
be
either
containerized
or,
as
you
say,
maybe
a
lamp
like
a
lambda
type
function.
H
K
Yeah
yeah
that's
kind
of
what
I
was
doing
with
the
netapps
a
while
ago,
which
is
the
the
the
registry
of
the
components
that
were
purposed
to
ipfs
and.
K
So
so
mainly,
this
is
the
prototype,
the
idea.
Well,
you
can
you
can
look
at
it.
I
can
show
you
it's
meant
to
be.
I
have
an
ens
race
3,
where
I
have.
K
K
So
and
then
you
can
download
the
components
and
and
then
obviously
those
could
be
well
ideas
that
those
will
be
verified
by
multiple
parties,
the
same
as
as,
if
you
have
a
jars
or
you
have
what
they
are.
You
know,
etc
and
that
that
can
define
your
execution
layer.
So
I'll
give
you
a
demo.
If
I
cancel
my
screen.
H
H
E
H
K
But
yeah
I'll
give
you
a
demo,
because
it
was
kind
of
the
same
as
with
this
discuss
a
while
ago
about
about
rules
and
genes
and
then
and
the
work
that
aaron
was
doing
about
because
those
who
are
also
part
of
enas.
K
H
Oh
there
it
is,
I
found
it
that
was,
I
could
hardly
concentrate
in
one
vlog.
I
was
in
there
listening
to.
Can
you
still
hear
me
yeah
yeah
yeah?
I
was
oops.
H
K
A
All
right
we've
got
a
one
of
these
issues
has
been
created.
H
Oh
cool
thanks,
connor,
looking
forward
yeah
we'll
we'll
do
it
we'll
do
a
triage
at
the
end
of
this
and
tomorrow
and
make
sure
that
we,
you
know,
bottle
up
all
the
juicy
issues
and
we'll
make
some
decisions
about.
I
got
to
figure
out
some
kind
of
regime
for
selection,
so
we'll
figure
that
out
too.
H
You
feeling
good
about
this
so
far.
Is
this
useful
yeah.
A
Yeah
well,
no!
No!
I
guess
I
was
surprised
that
we
managed
to
gonna
get
through
it
all
quickly
and
just
get
to
a
great
result.
You
know,
and
I
and
I
think
the
reason
you
know
we're
just
lucky
that
we
had
to
ask
from
nethermind
there
and
an
ace
as
well,
because
you
know
me
and
kyle.
You
know
obviously
we're
running
the
session,
but
I'm.
A
H
Say:
hey,
you
know
because
there's
I
know
we
have
several
people
from
india
and
that's
a
growing
group
coming.
You
know
there's
several
universities
that
are
popping
in
and
out
carnegie
mellon's
here.
So
you
know
just
make
sure
that
they
know
to
go
and
take
a
look.
D
D
A
How
how
have
you
finally
found
the
sessions
so
far
here.
N
Good
good
yesterday
was
quite
technical.
I
I
am
plucked
like
at
some
point
during
the
the
zero
knowledge
proof
circuits,
but
yeah
it's
been.
It's
been
very
interesting:
yeah
yeah,
yeah.
N
D
A
Yeah,
okay,
okay
cool-
because
maybe
I
can
I
can
take
you
through
some
of
the
stuff
we
were
talking
about
yesterday.
N
D
A
How
are
you
doing
good
cool
cool
yeah
if
you've
got
time
now,
that'd
be
awesome
to
have
a
chat
because
really
today
like
well,
we
had
a
really
productive
session
yesterday,
but
there
was
like
some
follow-on
questions
and
I
was
having
a
chat
with
john
later
on
and
he
he
mentioned.
Basically
that
would
be
good
to
kind
of
sync
up.
D
A
So
I'll
for
your
benefit
and
also
zavia's
as
well
like
I'll
I'll,
share
my
screen,
and
then
we
can
kind
of
go
from
the
perspective
of
what
we
actually.
You
know
what
was
being
discussed
yesterday
right.
So
let's
just
make
sure
there
we
go.
Okay,
so,
like
you
know,
the
the
the
original
sort
of
goal
of
intent
behind
this
session
was
to
talk
about.
A
You
know
the
actual
areas
for
where
we
could
really
broaden
the
support
for
baseline
beyond,
like,
as
is
currently
the
case
with
nethermine
plus
javascript,
and
actually
open
it
up
to
work
with,
like
hyperledger
basu
and
potentially
on
the
the
jvm
stack
as
well
for
some
of
the
core
pieces,
because
I
think,
if
you
kind
of
you
know
if
you,
if
you
cover
off
those
areas,
you've
covered
off
a
lot
of
enterprise
and
an
ace
was
also
you
know
she
pitched
in
just
about
the
specification
and
how
you
know
the
specifications
obviously
going
to
be
a
really
great
reference
point
for
people
to
work
from
when
it
comes
to
doing
you
know
other
implementations
and
support
there,
and
so
we
kind
of
we
got
down
sort
of
through
the.
A
Let
me
go
to
the
notes
here
from
the
session
yesterday,
but
I
mean
you
know.
I
think
the
one
of
the
takeaways
anyway
was
that
you
know
you've
got
this
baseline,
rpc
specification,
that's
emerging
and
you
know
any
ethereum
compatible
client
that
exposes
this
rpc.
Well,
I
guess
baseline
rpc.
A
It
shouldn't
be
too
much
work
then
to
actually
integrate
it
with
baseline,
but
what
what
menis
was
saying
too,
is
that
you
know
there
is
that
question
mark
around
the
actual
circuits
that
are
supported,
and
you
know
you
where
the
weather,
potentially
we
actually,
you
know,
want
to
look
outside
the
realm
of
say
just
said
case
snarks
or
you
know
other
other
sort
of
you
know
privacy
protocols,
and
I
guess
this
was
an
area
where
we're
thinking.
A
Okay
is:
is
there
kind
of
scope
at
this
point
for
creating
a
actual
issue
to
you
know
to
discuss
really
the
steps
that
would
be
required
to
actually
plug
into
you
know
some
other
privacy
technology?
I'd.
Imagine
I
guess
you
know
say
you
know
a
zero
knowledge
technology,
but
you
know
I
got
the
impression
you
potentially
had
some
thoughts
around
this
or
you
know
it's
an
area
that
you've
kind
of
scratched
the
surface
on.
F
Yeah
so
the
past
couple
weeks,
I've
been
working
been
talking
to
some
members
of
the
hyperledger
basu
team
that
work
at
consensus
and
trying
to
get
baseline
working
with
a
base
suit,
client,
yeah
and
the
the
approach
that
we're
taking
is
I've
been
working
on
a
what
I've
called
a
baseline
relay
server
right.
D
F
It
would
basically,
it
would
act
as
the
json
rpc
server
and
it
would
sit
in
front
of
the
client
so
that
it
would.
It
would
act
as
your
web3
provider
right
right
and
it
contains
the
logic
for
all
the
baseline,
json
rpc
methods
and
then
any
other
methods
that
are
hit,
such
like
any
of
the
traditional
ethereum
json
rpc
methods
that
it
receives.
It's
just
going
to
blindly
forward
to
the
client
that's
sitting
behind
it.
A
Okay,
okay,
I
said:
there's
already
did
anyone
from
the
the
basu
team
actually
joined
the
summit?
You
know.
A
We
did,
we
didn't,
have
anyone
yesterday
talking
about
it,
so
it's
yeah.
It's
it's
good.
I
mean
certainly
good
to
know
because
I
I
did
have
some
discussions
previously
with
oh
yeah,
david
mitchell.
A
We
we
see
involved
on
those
discussions
on
that
side.
We
said
other
members
of
the
t,
the
bc
team.
F
Yeah,
it
was
david
and
tim
baco
are
the
the
two
that
I've
talked
to
from
the
basically
okay.
A
Cool,
obviously,
I
think
it'd
be
good
to
get
their
kind
of
input
on
that,
because
of
course,
if
stuff's
already
happening
here,
then
it
makes
less
sense
to
they
kind
of
do
some
additional
pieces
on
it,
and
so
so
sorry,
what
were
they
calling
that
component,
like
just
baseline
relay.
F
Yeah
yeah
baseline
relays,
yeah
it's
it's
just
a
micro
service
or
docker
container
that
sits
in
front
of
the
the
client.
So
it
could
it's
designed
to
work
with.
You
know
it
could
sit
in
front
of
a
bay
suit,
client
or
you
could
have
infuria
behind
it
or
the
idea
would
be
any
any
web.
3
provider
could
sit
behind
it
and
it
would
add
the
baseline
functionality
to
it.
A
Okay,
with
that
with
that
relay,
actually
contain
the
merkle
tree
within
it
as
well.
F
D
F
So
so
the
idea
is,
it
has
a
separate
the
way.
I've
I've
done
it.
I
guess
it
could
work
other
ways,
but
I've
designed
it
to
have
a
mongodb
container
that
runs
alongside
of
it
and
that's
where
the
merkle
trees
actually
get
stored
and
yeah.
It
contains
all
that
data
and
all
the
updates
happen
within
that
database.
F
A
Okay,
cool,
that's
that's
just
that's
useful
to
know
we'll
get
them
then,
to
you
know,
provide
provide
some
context
for
the
the
actual
issue
that's
been
created,
then
the
the
other,
the
the
other
thing
to
touch
on
would
be
the.
Where
are
we
yeah?
It
was.
Was
this
notion
of
yeah
this
the
support
of
other
circuits?
I
know
that
there
was
like
a
separate
session
on
that,
specifically
in
terms
of
the
circuit
improvements.
A
I
I
wasn't
able
to
make
that
that
session
and,
of
course,
there's
still
discussions
happening
there
at
the
moment.
But
it's
is.
Is
there
currently
like
a
path
sort
of
planned
out
in
terms
of
you
know,
providing
support
for
you
know,
potentially
you
know
other
types
of
circuits,
or
is
it
more
just
about
right
now,
more
focused
on
just
incrementing?
What's
already
there,
I
know
john
had
mentioned
aztec
as
well.
As
you
know,
one
that
was
being
looked
at
potentially
by
the
aztec
team.
F
Yeah,
I
don't
know
so
much
about
other
zero
knowledge
technologies
outside
of
I
guess,
socrates,
which
is
the
one
that's
in
use
right
now,
but
I
mean
the
the
baseline
rpc
interface
is:
is
pretty
agnostic
in
terms
of
what
I
don't
think
it
really
cares.
What
zk
technology
is
is
sitting
behind
it
as
long
as
as
long
as
the
verifier
contract
on
chain
matches
the
the
interface
that
it's
expecting
to
interact
with
it
should
work
like
the
the
actual
verification
process
and
generation
of
proofs.
F
As
long
as
you,
you
have
the
same
inputs
and
output
types,
the
I
guess,
the
baseline
relay
service
and
the
the
ethereum
client
that
you're
using
really
shouldn't
care
about
what
zero
knowledge
technology
you're
using.
A
Yeah
at
this
point
in
time,
has
anyone
made
any
noises
or
if
anyone
else
on
the
call
has
got
any.
You
know
views
here
if
there
are
specific
other
ones
that
there's
interest.
A
L
I
I
don't
know
if
it's
if
it
has
been
discussed
before
but
you're
mentioning
about
two
interfaces
that
are
currently
in
core.
So
there
are
the
isaac,
snark
circuit
provider,
and
it
is
not
compilation.
Artifacts
and
my
question
to
sam
is
oh
to
anybody
else.
Sorry,
so
I'm
not,
but
I
think
you
may
know
maybe
a
bit
more
you're
closer
to
this.
Are
there
any
plan
to
kind
of
abstract
those
to
the
point
that
conor
was
making
earlier
to
to
be
a
bit
more
zero
knowledge
protocol
agnostic
in
core
itself?
L
F
Think
they're
they're
abstracted
a
bit
already,
but
it's
definitely
possible
that
it.
It
needs
a
further
layer.
I
can
show
you
the
let's
see.
Can
I
share
my
screen
real,
quick.
F
F
So
the
verifier
is
the
one
like
right
now.
Zocrate
is
the
way
it
works.
Is
it
produces
this
verifier
contract?
For
you,
I'm
not
sure
how
other
zero
knowledge
technologies
work,
but
either
they
they
would
need
to
create
this
themselves
or
I
guess
you
could
manually,
go
through
and
create
a
verifier
contract.
F
F
So
I
don't
know
about
other
other
zero
knowledge
technologies.
If
they
have.
You
know
the
same
same
data
format
here
in
terms
of
proof
and
public
inputs.
This
is
the
way
zocrates
works.
You
you
feed
the
verify,
function
or
method
these
inputs,
and
it's
just
going
to
return
a
true
or
false
and
tell
you
if,
if
that,
what
you
input,
if
the
proof
was
verified,
you
get
a
true
back.
If
not
you
get
a
false
and
that's
that's
pretty
much
all
the
the
feedback
you
get
in
terms
of
on-chain.
L
Okay,
thank
you,
I
think
well,
I'm
far
from
being
the
most
appropriate.
G
L
To
discuss
zika
circuit,
but
based
on
the
conversations
that
I've
heard
there
may
be
a
need
for
other
parameters
to
be
included
for
specific,
zero
knowledge
protocols,
but
the
base
it
remains
the
same
because
they
are
all
based
on
like
private
and
people
you
can
put,
etc,
etc.
So
yeah,
I
can
see
that
it
is
abstracted
enough
to
cater
for
zika,
snack
and
maybe
other
protocols,
but
I
don't
know
that
for
few
other
protocols
that
were
mentioned,
that
would
be
sufficient.
L
So
I
guess
from
a
specification
point
of
view
and
not
well,
it's
not
even
specification
from
a
standout
point
of
view,
so
I'm
really
referring
to
the
api
and
data
model
baseline
standard
or
is
it
standard?
Do
you
think
that
the
current
state
would
be
and
would
state
that
we
have
to
use
the
zika
snark?
We
have
to
use
zika
snack
basically
and
not
other
technologies
or
zika
protocols.
F
F
Yeah,
I
think
it
can
certainly
be
modified
to
to
work
with
other
ones.
I
mean.
Currently,
we've
only
worked.
I've
only
worked
with
socrates,
thus
far
so
I
I
don't
think
I
know
enough
to
tell
you
what
would
need
to
be
changed
to
accommodate
for
other
other
types
of
zero
knowledge.
F
Thank
you,
but
potentially
one
change.
If
you
know,
if
that
interface
doesn't
doesn't
work
for
other
zero
knowledge
technologies,
maybe
we
could,
as
part
of
baseline.
You
know
we
couldn't
just
use
the
the
verifier
contract
that
zachary
spits
out,
but
maybe
we
have
an
intermediary
step
that
I
guess
changes
the
verifier
contract
modifies
it
a
bit
to
to
match
a
different
interface
that
you
know
works
for
several
different
technologies.
L
Yeah,
yes,
that
makes
sense.
One
approach
could
also
be
from
a
standard
point
of
view.
Is
that
the
requirement
is
that
a
privacy
interface
must
be
implemented,
but
we
don't
impose
any
zero
knowledge
protocol,
but
we
do
say
if
you
are
using
zk
snark,
you
have
to
use
isaac,
snack
circuit
provider
and
isaac
is
now
compile
compilation,
artifacts,
etc,
etc.
L
So
if
we
say
there
are
a
list,
if
we
say
that
are
a
list
of
zero
nudge
protocols
supported
and
those
interfaces
are,
must
have
for
any
client
that
want
to
go
and
use
those
supported
protocol.
But
for
any
other
protocol
they
will
have
to
basically
have
their
like
do
their
anti-offices
from
scratch
and
do
everything,
as
you
said,
like
write
their
own
very
fire
contract
and
everything.
F
Yeah,
it.
I
F
You
know
if,
if
there
was
a
common,
a
common
on-chain
function,
it
was
calling-
and
you
know
it
would
work
with
all
these
different
technologies
instead
of
having
to
you
know,
I
guess
have
have
different
code
in
there
and
it's
having
to
to
call
different
methods
depending
on
what
zero
knowledge,
verifier
type
you're
using
yep.
A
Awesome
so
I
mean
from
the
the
takeaway
there
is,
I
guess
it's
it.
It
might
be
that
it
would
work
with
a
different
protocol,
but
we
need
to
kind
of
well.
I
guess,
choose
one
and
see
right
and
if
we
can
get
there,
then
yeah.
You
know
we'll
we'll
understand
more
if
the
interfaces
need
to
be
adapted
or
not
out
of
interestingness
in
terms
of
further,
like
the
specification
like
does
where
you
have
like
these
interfaces
defined
in
solidity
code.
Do
you
know
at
this
point
in
time?
L
In
more
definitely,
we
will
talk
in
more
abstract
terms,
so
any
example
of
code
from
core,
so
visual
ones
that
has
been
released
in
august,
will
be
for
non-normative.
Those
will
be
non-normative
examples,
so
we
are
really
focusing
right.
Now
on
the
irish
history
imessaging
service,
I
volt
ib
baseline
rpc,
I
blockchain
service,
the
I
persistent
service
and
the
sickest
inoculated
interfaces
that
we
mentioned.
So
those
are
the
core
interfaces
that
apparently
will
like
must
be
implemented,
and
so
the
goal
is
to
ensure
that
we
have
in
the
specification
all
the
right
functions.
L
If
you
remember
yesterday,
I
mentioned
the
deploy
function
and
some
sorry
thomas
on
the
course
said.
Oh
actually,
we
have
already
evolved
faster,
so
we
need
to
ensure
alignment
and
we
need
to
work
very
closely
on
the
team
actually
doing
the
development
and
understanding
why
some
functions
have
been
deprecated
replaced
or
those
type
of
things,
and
what
are
the
caveat
for
using
those
as
well
if
they
exist,
because
one
gap
that
we
may
have
in
the
documentation
right
now
is
that
understanding
the?
L
L
L
F
L
The
core
specification:
you
really
try
to
abstract,
even
the
app
logic
chain
service.
If
we
think
about
it,
this
will
not
be
a
must
have
the
way
we
may
approach,
it
is
to
say
if
you
use
a
dlt,
that
is
a
blockchain,
then
you
must
implement
this,
but
it
should
not
be
like
a
must-have
requirement,
because
the
way
we
have
approached
the
baseline
protocol,
it
should
be
a
dlc
agnostic.
L
So
this
is
where
we
are
going
with
the
core
specification.
As
a
result
of
this,
we
need
to
align
the
api
specification
in
the
same
mindset
so
because
we
need
to
be
consistent
in
what
we
are
saying.
So,
as
I
will
say
that
baseline
can
is
only
exclusively
on
blockchain.
Therefore,
we
will
say
that
a
client
must
implement
the
blockchain
service
regardless
or
we
say
should
if
we
use
blockchain,
you
know
so
those
are
the
type
of
you
know,
literal
details
that
we
think
about
on
the
specification
side
and
that.
L
L
Well,
there
are
others,
so
I
remember
andreas
mentioned
a
couple
of
other
dlt
distributed
leisure
technologies
which
handled
blockchain,
and
he
wasn't
knowledgeable
enough
to
be
able
to
challenge
this
point.
So
I
I
won't
go
there,
so
I
would
update
to
this.
I
won't
be
able
to
give
you
an
example,
but
this
is
why
we
had
the
conversation
and
yeah.
L
O
L
But
then
again,
this
is
when
we'll
get
into
the
detailing
actually
establishing
the
level
of
requirements.
So
everybody
on
the
current
show
you
know
what
we
are
looking
at
must
should
may
may
not
should
not,
must
not
a
key
word.
L
L
O
A
It's
just
it's
really,
it's
a
really
interesting
thought
point
in
terms
of
what
direction
it
goes.
L
O
L
L
O
I
was
looking
at,
I
guess,
speaking
of
changing
names
for
interfaces.
Excuse
me
I'm
having
a
second
breakfast
here.
Looking
at
the
I,
the
zkp
interface
and
like
I
was
working
through
some
scenarios
where
I
was
like.
Do
I
even
need
socrates
or
snark
music,
smart
type
service,
and
does
that
mean
I
can't
baseline?
O
L
L
To
your
open
brand,
if
we
say
that
one
core
component
of
the
protocol
is
a
privacy
is
a
privacy
component,
we
definitely
need
to
have
a
privacy
interface.
It
doesn't
matter
what
pro
zero
knowledge
protocol
and
actually
yes,
there
are
no
portfolios
required.
I
was
you
know,
going
sideways
and
say:
oh,
but
do
we
even
need
zero
knowledge
protocol?
Can
it
be
another
type
of
cryptographic
photographer,
but
let's
just
figure
out
so
I
I
do
think
that
we
need
a
privacy
interface
and
we
need
a
zero
knowledge
protocol
now.
L
I
strongly
think
that
we
should
not
say
that
we
need
zika
snack,
because
why?
I
don't
see
any
justification
for
this.
To
be
honest,
yeah.
O
Yeah,
I
I
hear
you
I
often
find
I
guess
my
habit,
which
is
I
don't
know,
good
or
bad,
is
to
reduce
just
to
distill
ideas
and
things
down
to
their
essence,
but
it's
like
at
some
point
we're
going
to
end
up
just
describing
in
our
standards
like
the
facets
of
modern
computing
yeah.
You
know
that's
right,
it's
like,
oh,
what
does
it
have
to
be
blockchain?
Well,
no,
just
give
me
anything
and
then
basically
are
just
this.
O
We
just
have
the
most
generic
interfaces
that
could
be
applied
to
any
technology
anywhere
and
then
we've,
basically,
you
know
boiled
away
everything,
that's
interesting,
so
we
don't
necessarily
want
to
go
too
far.
Oh.
O
F
Oh
yeah.
L
F
Added
something
to
the
notes
section
for
this,
this
breakout
room-
I
can
show
it
on
my
screen.
These
are
the
the
jason
rpc
methods
that
I
included
in
the
baseline
relay
service.
I've
been
working
on
so
most
of
these
are
the
same,
but
a
few
variations
from.
I
guess,
what's
in
the
repo
right
now,
and
what
what
nethermine
had
done
in
their
first
iteration,
namely
we
we
got
rid
of
the
baseline
deploy
method
because
it
was
essentially
just
re-implementing,
some
of
the
eth
underscore
syn
transaction.
F
You
know
the
the
traditional
way
of
deploying
contracts
on
chain,
and
then
you
know
a
few
other
variations.
I
haven't.
I
haven't
added
the
full
list
of
parameters
and
everything,
but
maybe
this
is
a
a
good
starting
point
for
beta
to
see
what
what's
been
going
on,
and
I
I
think
the
I
baseline
rpc
interface,
pretty
much
just
has
a
method
for
each
one
of
these
methods,
kind
of
a
one-to-one
mapping
that
would
that
would
call
these.
L
Yes,
that's
helpful,
but
you
add,
before
the
eye
blockchain
service.
I
believe.
F
Oh,
I
block
chain
service.
F
K
K
I
don't
know
if
that
will
make
sense,
is
later
on
or
they're
more
real,
specific
functionality
to
baseline,
because
this
is
just
the
helpful
for
the
the
smart
contract
kind
of
creative
in
the
tree,
yeah
yeah,
so
something
like
baseline
merkle
or
something
like
that-
the
kind
of
categorize
this
into
what
it
is
because
we
have
as
well
the
deploy.
K
I
know
I've
seen
it
somewhere
just
before
and
yeah,
and
also
you
you.
I
don't
know
if
this
covered
this,
but
you
may
have
different
seals
that
are
being
deployed
as
well.
I
don't
know
I
really
am
just
looking
at
the
methods
at
the
moment,
so
I
don't
know
if
this
covert
specific
scenario.
K
K
L
K
Well,
at
the
moment,
no,
but
looking
in
the
future,
maybe
you
know
the
seal
contract
might
be.
You
know,
you
know
be
the
one
of
the
kind
of
drivers.
L
That's
true,
yeah
yeah,
that's
a
good
point
that
you
are
making
and
that's
a
point
that
was
raised
before
is
there
are
other
shielding
mechanism?
The
shield
contract
must
not
be
must
have
requirement
of
the
baseline
specification
or
baseline
standard.
So
from
that
point
of
view,
we
are
fully
aligned
with
what
you
say
is
that
if
the
high
baseline
rpc
sole
function
is
to
interact
with
the
merkle
tree
within
the
shield
contract,
but
the
shield
contract
itself
is
not
required.
L
How
can
we
say
that
the
eye
baseline
rpc
is
a
required
interface
so
that
that
yeah?
That's
a
very,
I
think
when
the
time
will
come
to
have
this
conversation
during
the
standard
and
specification
work
group.
K
O
Yep
also,
I
don't
know
if
this
is
bringing
up
it's
related,
but
the
shield
contract
excel
itself.
We
might
consider
a
different
name.
I
mean
I
like
it,
but
it's
not
it's.
It
was
well,
its
origins
were
nightfall
and
in
nightfall
they
have
shield
contract
and
that
implementation
it.
It
literally
does
shield
it
shields
the
movement
of
tokens
behind
the
facade
of
this
commitment,
that's
used
as
a
proxy
and
in
in
baseline
at
least
the
current
implementation
like
there's.
No,
it's
it's
technically
using
that
same
mechanics,
but
it
doesn't.
O
Quite
it's
not
quite
a
shield.
I
mean
I
guess
on
some
level
it
is,
but
not
in
the
same
way
that
that
their
shielding
token
transfer
in
in
nightfall
or
night
light.
L
Yeah
it
it
is
relevant
because
if
we
are
saying
that
we
inherited
the
shield
contract
as
a
mechanism
to
hide
the
transaction
yeah
like
movement
of
transfer
and
values,
etc,
and
that
we
progressed
on
this,
so
we
inherited
from
nightfall
and
then
we
improved
on
it
to
accommodate
baseline
needs.
And
now
we
are
saying,
but
oh
there
are
other
way
to
do
that.
And
if
we
look
at
some
existing
eip
or
other
standards,
we
may
realize
that
that
may
not
be
in
the
future.
L
The
recommended
approach
right
so
are
there
currently
any
small
question
for
everybody?
Are
there
any
currently
good
candidates
either
in
terms
of
who
can
stand
out
or
proposals
that
you
may
have
seen
that
are
a
good
candidate
just
to
replace
the
shield
contract
or
to
complement?
I
don't
know,
I'm
just
raising
the
point.
A
L
Okay,
so
because
I'm
aware
I'm
not
too
close
of
the
conversation,
but
andre
has
mentioned
that,
oh
actually,
andrea
and
kyle,
I
think
they
were
having
a
discussion
about
they
have.
L
They
are
either
investigating
or
working
on
something
that
make
the
shield
contract.
How
can
I
say
not
deprecated,
that's
definitely
not
the
right
word,
but
not
as
cornerstone
to
the
old
baseline.
You
know
framework,
as
it
was
even
few
months
ago
or
few
weeks
ago.
To
be
honest,
I
don't
you
have
any.
O
I
was
just
going
to
say
that
the
shield
contract
it
has
like
a
structural
function
in
terms
of
like
what
it
does
in
baseline,
but
it
doesn't
doesn't
provide
the
shield
right
as
like
in
the
context
of
my
fall,
but
I'd
be
curious
to
see
if,
like
you
could
like,
if
you
basically
I
don't
know
what
I'm
gonna
say,
we
could
think
of
again
maybe
going
back
to
what
we're
talking
about
with
the
blockchain
like
what
is
the
job
it's
performing,
and
is
it
actually
shielding
or
is
it
doing
something
else?
O
Is
it
I'm
more
of
a
workflow
manager
than
a
shield
anyway,
I'm
curious
to
see
what
you
said:
kyle
and.
L
O
G
Archer,
hey
I'm
mace.
This
is
mark
hi
mark
this.
This
is
the
only
session
that
is
live
right
now
that
I
can
tell
at
the
opening
plenary,
there's
only
three
people
in
it,
even
though
it
does
show
it
does
look
as
though
there's
some
movement
now,
but
this
is
the
only
session,
that's
having
any
live
content
right
now,.
L
That
the
shield
contract
is
not
a
must-have
requirement
as
part
of
the
standard,
and
I
was
like
what
would
be
the
shielding
mechanism,
because
I
remember
you
mentioned
something
but
yeah
I
couldn't
remember
or
I'm
not
knowledgeable.
You
know
enough
sorry
to
share
with
the
rest
of
the
team.
So
if
we
don't
use
the
shield
contract,
what
could
be
another
way
to
achieve
the
same
function?
Basically.
M
Okay,
so,
first
of
all
it's
it's.
P
All
the
discussions
are
obviously
rooted
in
in
in
the
current
reference
implementation,
which
is
a
particular
implementation
of
a
of
a
of
a
possible
possible
implementation
of
different
implementation
patterns
right.
So
the
shield
contracts
basic
function
is
to
preserve
privacy
right.
P
No,
I
mean
that,
that's
yes,
but
it's
yes,
the
shoe
contract
is
is,
is,
is,
is
there
too
for
the
for
the
for
the
for
the
group
for
the
group
participants
and
the
and
the
and
and
the
anchoring
of
the
of
the
of
the
proof,
and
it
connects
to
the
to
the.
O
The
merkle
tree
contract,
it's
an
anchoring
point
like
the
verifier
contract
right,
that's
generated,
at
least
in
this
current
implementation
right,
that's
what's
enforcing
the
correctness
on
chain
of
the
off-chain
computation
and
the
shield
is
one
of
the
glue
that
you
know
anchors.
That
implementation
there
to
specific
other
parties
that
are
participating.
You
know
like
is
tying
it
to
the
rest
of
data
elements
that
are
on
on
chain
or
other
data
points
and
other
data
documents
that
are
in
a
right
workflow.
So
I
was
like
proposing
it's
more
of
like
a
workflow
manager.
D
P
P
Because
it's
not
because
it's
just
simply
not
scalable
right
so
from
it
from
scalability
point
of
view,
you,
wouldn't
you
wouldn't
want
that
you
would
want
to
have
a
you
would
want
to
have
a
a
you
want
to
have
a
contract
governance
contract
that
manages
the
the
the
anchoring
and
and
and
verification
of
transactions
on
chain.
But
that's
that's
about
that's
about
it
right
I
mean
if
you,
if
you
think
about.
O
P
The
the
that
can
depend
right,
if
that
depends
on
your
operational
model
right.
Do
you
have
a
single
operator
or
or
or
do
you
have
a
network
of
of
operators
that
that
that
run,
the
that
run
the
same
stack,
yeah,
right
and
and
so
can
multiple
ones
write
into
into
that
smart
contract
architecture
or
just
one
or
a
set
of
right?
So
so
that
that's
the
I
I
think
there
is
there's
there
is.
There
is
no
need
to
have
any
unchained
representations
of
users.
P
O
You
route
and
that's
kind
of
like
I
feel
like
something's,
come
up
in
other
discussions,
especially
yesterday
with
the
car
taking
the
ckp.
It's
like
if
you're
heading
at
tokenization,
with
this
like
invoice
factoring
and
all
that,
then
there's
a
lot
of
things.
You
kind
of
have
to
do
to
get
your
data
and
everything
into
the
I
use
the
term
on-ramp
of
tokenization
and
but,
if
you're
not
and
ultimately,
tokenizing,
and
I
feel
like
yeah,
there's
a
lot
up
for
grams
in
terms
of
what
you
can
do.
A
lot
more
flexibility.
P
So
if
you
so
that
we,
we
discussed
that
yesterday
in
scaling
and
our
our
word
not
solution,
but
our
our
our
approach
would
be.
You
know
it's
like
privacy,
preserving
yeah,
use
zkp's,
but
use
use,
zkp
rollups
and
keep
everything
on
chain
and
the
blocks
that
you
would
normally
write
on
chain.
P
P
And
then
you
have
the
data
availability
problem,
but
if
you're
not
tokenizing,
then
who
cares
right.
O
Yeah
well
so,
in
that
roll-up
strategy,
like
are
you,
I
guess
I
know
conceptually
the
roll-up
but
like.
Is
it
almost?
What
is
managing
the
roll-up
and
what
gets
rolled
an
operator
like
yeah?
But
I
mean
I
guess
yeah.
So
is
there
a
single
operator,
that's
sort
of
moving
towards
a
more
centralized
or
quasi-centralized
like
yeah.
P
O
P
P
Yeah,
so
so
so
you
can
have
different
roll-up
strategies
right,
but
if
you
want
to
preserve
privacy,
which
I
guess
people
want
to
your
your,
your
primary
purpose
is
is
going
to
be,
is
going
to
to
to
to
use
to
use
the
zk
zk
ck
roll
ups,
because
that
preserves
privacy,
nothing
is
is,
is
because
you
will
have,
I
mean,
think
of
it.
This
way.
You'll
have.
P
It
is
highly
unlikely
so
from
an
operational
point
of
view
that
that
you'll
have
a
baseline
stack
for
each
for
each
pair
for
each
bilateral
pair
of
of
entities
right
because
it's
just
not
it's
just
not
capital
efficient,
I
mean
you
could.
P
But
that's
you
know
it's
not
it's
from
from
a
scalability
from
a
cost
point
of
view.
It's
just
not
doesn't
make
sense.
So
you'll
have
an
operator
that
has
like
a
thousand
clients,
customers
right
and
they
use
baseline
and
they
they
pump
their
transactions
through
that
and
by
by
rolling
them
up.
You
get
to
you
know,
transaction
costs
that
are,
you
know
less
than
a
penny,
significantly
less
than
a
penny
like
a
fraction
of
that
right,
because
you're
anchoring
10
000.
P
O
Are
the
second?
Yes,
so
in
that
sense,
though,
you're
trading
off
transaction
speed,
for
for
what,
though,
because
like
in
in
in
how
baseline
is
at
least
structured
in
current
implementations,
or
maybe
even
going
back
as
radishes
like,
you
have
multiple
parties
who
are
watching
a
particular
commitment
and
listening
to
events
from
a
particular
smart
contract
so
that
they
can
stay
in
sync.
So
is
the
trade-off
efficiency,
for
you
know
the.
What
was
the
term?
Is
it
consistent
consistency
or
time
to
consistency.
I
O
No,
not
necessarily
when
you
have.
How
do
you
get
the
it's
kind
of
like
a
side
effect
in
terms
of
how
it's
designed
right,
because
it's
not
a
direct
data
exchange
between
parties,
it's
sort
of
like?
Oh,
I
listen
for
events
on
chain
for
data
that
I
am
already
paying
attention
to
in
my
my
local
system.
P
Okay,
so
I
sent
you
a
message,
hi
right,
so
what's
our
agreement,
our
agreement
is
that
the
messages
we're
sending
back
and
forth
to
either
are
the
truth
right.
So
if
I
say
hi,
you
can
be
assured
that
I
actually
said
hi
right.
How
do
you
know
linkedin
could
have
changed
what
I
said
yeah
right
so,
but
if
I
were
to
baseline
that
message,
then
you
know
exactly
that.
I
said
hi
right
so
and
then
you
you
write.
Okay.
I
acknowledge
that
by
writing.
P
Back
I
have
acknowledged
that
state
and
I
re
I
update
the
state
with
my
with
my
subsequent
message
right.
So
that's
the
back
and
forth
that's
happening
and
and
and
linkedin
can't
do
anything
about
it,
except
shutting
down
the
the
messaging.
So
the
only
thing
that
I'm
doing
is
I'm
creating
anchor
points
that
I
put
on
chain
at
one
point.
So
we
can.
We
have
a.
We
have
a
you
know,
there's
there's
something
that
we
can
point
to
and
say:
hey.
You
know
it's
like.
P
O
Yeah,
well,
it's
just
sort
of
the
the
mechanics
of
like
how
you
actually,
let's
say
in
that
scenario
like
you
sent
hi
to
me,
and
I
need
to
verify
that
you
actually
sent
hi
to
me.
So
I
have
to
go
and
take
the
content
that
you've
given
me
hi
and
any
other
metadata,
along
with
it
and
run
that
through
are
agreed
upon
hashing
or
zero
knowledge
scheme
right.
P
O
O
I
P
Always
out
of
band
right,
so
if
you're,
if
you're,
if
you're
trading
with
someone
in
a
decks
right,
it's
like
it's
like
you,
you
you
you
you
know,
especially
if
you're,
if
you're,
if
you're,
if
you're
an
fa,
you
check
check
who
you're
trading
with
often
so
it's
like
yeah
this,
this
identity,
fair,
that's
what
the
global
phone
book
is
for
right.
You
wanna!
You
wanna!
Do
that
you
to
you!
You!
You
verify
identities
out
of
band.
O
P
There's
a
there's,
a
there's,
a
there's,
a
phone
book
right,
your
your
contact
list-
or
you
know
it's
like
the
linkedin
universe
of
of
of
verified
identities.
Quote-Unquote
but
let's
say
linkedin-
were
to
use
a
a
phone
book
and
had
a
kyc
process
that
that
a
short
correctness
of
identity,
which
the
global
phone
book
here
is
is
is
track,
is
trying
to
achieve
so.
O
P
And
then
yeah
I
mean
I
mean
yes,
you
can
you,
can
you
can,
do
you?
Can
trade
zkp's
off
off
chain?
You
know
as
as
much
as
you
want.
You
don't
need
that
right.
It's
just
you.
You
just
have
a
messaging
system
and
then
you're
trading
back
zero
knowledge,
proofs
back
and
back
and
back.
O
P
You
know
everyone
needs
a
hobby
right.
Everybody
needs
a
hobby
right
exactly
so
so
it's,
but
it's
like
it's
like
you.
You
don't
need
anything
on
chain
in
principle,
if
you
trust
each
other
and
and
and
you
you
have,
you
know
your
your.
Your
you've
agreed
upon
what
you're,
what
your,
what
your
vsm
does
on
each
side
by
you
know
having
an
agreed-upon.
P
P
Zing
z-z-I-n-k,
that's
from
better
matter
labs
than
cairo
that
is
from
starkware
and
then
noir,
which
is
from
stack,
so
they
are
they're.
All
so
noir
is
sort
of
like
fully
anonymous
trans
roll-ups
and
then
cairo
is
roll-ups,
but
with
starks
and
and
zinc
is,
is
ck
snarks
with
plonk.
O
P
Okay,
interesting
so
those
are,
those
are
all
new.
Those
are
the
fact
that
they
can
do
that,
because
now
you
have
new
commitment
schemes
and
with
new
commitment
schemes
you
can
have
a
larger
number
of
constraints.
With
larger
number
of
constraints.
You
can
you
can
map
more
complex
project
projects,
programs
right
into
into
into
into
circuits
and
then
generate
commitments
from
that
which
are
then
just
translated
into
bytecode,
which
is
then
so
it's
it's
literally
like
the
evm.
P
It's
just
the
architecture
of
the
of
the
of
the
virtual
state
machine
is
not
technical.
Technically
speaking,
it's
not
a
neumann
architecture.
It's
an
it's
an
it's
a
it's
a
it's
a
polynomial
commitment,
architecture,
okay,
right!
So
because
you're
just
verifying
constraints,
right,
you're,
you're,
you're,
you're,
verifying
cryptographic,
commitments,
probabilistically!
P
Yeah,
that's
it's!
It's
a
bit!
It's
a
bit
of
a
bit
of
a
bit
of
that
sort
of
like
it's
like
like
math
2.0.
It's
it's!
It's
not
it's!
It's
highly
non-trivial
right.
O
P
So
not
that
that
is,
that
is
an
implementation
choice.
You
can
do
it
this
way,
but
you
don't
have
to
there
are
more
efficient
ways
of
doing
that.
Let's
put
it
this
way
right,
so
it's
like,
if
you're,
if
you're,
trying
to
minimize
your
unchained
footprint,
then
if
you
don't
want
to
tokenize
the
absolute
minimal
contract
that
you
need,
I
can
I
can
send
you
the
code
for
that.
I
have
that.
P
M
Actually,
actually,
on
on
on
github,
let
me
let
me
it's
public,
it's
it's
within
the
element
of
implementation
of
of
site
tree.
Second,
just
need
to
find
it.
N
I
O
Still
struggling
with
currently
trying
to
just
wrap
my
head
around
like
what
what
and
how
and
how
you
represent
in
your
business
workflows
in
a
stark
or
snark
based
circuit,
or
I
guess,
whatever
zinc
and
cairo
or
noir,
have
as
their
abstraction
layer.
That
allows
you
to
define
logic
right
program
if
they
use
a
native
language
or
some
dsl
or
something.
M
Yeah
I
can,
I
can
hold
on.
Let
me
let
me
let
me
ping,
you
the
let
me
ping
you
something
else.
O
Like
you
know,
often
you're
struggling
with
a
mix
of
data
and
state
and
rules
around
when
that
can
change
in
context,
and
it's
like,
I
guess,
that's
really:
what
comes
how
much
context
of
the
business
process
you
push
into
these
control
structures
and
because
of
how
they
have
to
run.
O
You
know
in
that
they
need
a
lock
they,
you
know
almost
like
a
black
box.
I
think
well,
anyways
cryptographic.
Proofs
are
largely
black
boxes
to
most
everybody
other
than
the
people
who
wrote
them
just
because
the
level
of
understanding
is
is
so
high.
I
think
so.
P
Right
I
mean
literally
it's
it's,
it's
like
it's
like
a
year
over
a
year,
there
were
some
real
big
cryptographic,
breakthroughs
that,
like
the
aztec
guys
did
and
and
and
also
the
the
the
alien
and
and
and
his
bunch
over
at
starkware,
so
that
that
is
that's
like
just
in
the
last
year,
and
that
is
not
just
coming
into
in
into
production
right,
but
but
that's
clearly
the
way
forward
right,
because
you
you
can,
with
with
these
general
prover
services
right
you
you,
you
can
now
just
write
code
as
you,
I
put
the
link
in
there
to
to
like
the
zinc
framework.
P
You
can
just
write
code
as
you
normally
would
right.
It's
like
a
javascript
type,
rust
type
language
right
and
then
it's
it
works
very
similarly,
and
then
you
know
you
have
a
compiler,
you
know
and
the
compiler
does.
This
does
all
the
all
the
all
the
magic
you
know,
converting
it
into
circuits,
doing
commitments,
and
then
you
know
converting
that
those
commitments
into.
P
It's
the
code
is
in
is
in
contracts.
Okay
and
there's
you
can
you.
M
Can
run
this
so
there's
there's
that's
yeah!
That's
a.
M
P
Is
this
is
simply,
this
is
a
this
is
simply
a
a
contract
that
allows
you
to
to
to
to
anchor
a
registry
entry.
M
Right
so
committer
the
anchor
hash
and
what
block
height?
P
You
can,
if
you,
if
you
run
truffle,
you
know
you
can
just
take,
take
the
code
as
is
and
and
and
use
you
know
you
can
deploy
it,
you
can
you
can,
you
can
run
run
it
locally.
P
There's
there's
an
explainer:
you
can
just
follow
that
and
and.
M
P
It's
just
it's
just
a
simple:
it's
just
a
simple
contract
force
for
then
staking
and
if
you're,
if
you're,
trying
to
spam
the
contract.
So
it's
you
have
to
be
a.
You
have
to
be
an
operator
for
the
contract
to
be
able
to
write
to
it,
and
if
you
write
to
it
more
than
once
a
block,
then
you're
getting
getting
your
ass
kicked
you're
you're
you're,
getting
your
your
stake!
Slashed.
O
P
Yes,
so
this
is
then
just
a
just
an
anchor
hash
the
anchor
hash
could
be.
It
could
be,
then,
just
a
uri
that
points
back
to
your
to
your
data,
storage,
yeah.
P
Yeah
this
is
this
is
really
generic.
If
you
want
to
anchor
a
hash,
this
is
like
and
and
there's
there's
there's
it's
it's
a
very
straightforward,
simple.
You
need
to
register
and
if
you
want
to
register
you
need
to
stake
and
then
you
can
only
submit
there's
only
one
transaction,
a
block
allowed
for
each
for
each
for
each
for
each
operator.
P
So
and
yeah
and
again
that
could
be,
you
know,
do
governance
contract
and
then
you
know
you
can
update
that
value
right.
It's
it's!
It's
set
in
the
governance
contract
right.
You
know
it's
like
big
deal
right,
so
you
you
can
you
can
you
can
you
can
you
can?
You
know,
have
a
have
a
have,
a
function
that
sets
that
variable
sure.
O
P
Yeah
and
you
make
it
make,
it
can
only
be
called
from
the
from
from
the
addresses
and
the
governance.
So
you
can.
You
can
expand
that
functionality,
but
this
is
sort
of
like
the
most
minimalistic
anchoring
contract
that
provides
enough
security
to
to
so.
I
ran
this
through
like
through
myth
x
and
it's
it's
all
secure.
P
I
mean
I
haven't,
found
anything
in
there
right.
There's
no.
M
D
O
P
P
Staked
you
know
an
anchoring
contract
that
allows
you
to
to
to
provide
minimal,
or
you
know,
strong
security
guarantees.
P
O
You,
okay!
I
wonder,
though,
if
you
wanted
to
so,
if
you,
if
I
wanted
to
say,
use
this,
this
becomes
I'm
just
thinking
kind
of
loud
here,
and
so
it
becomes
my.
What
replaces
the
shield
contract
and
I
have
everybody
in
my
work
group
stake,
but
I
still
want
to
use
a
verifier
contract
in
conjunction
with
this,
based
on
sort
of
business
logic,
and
I
could
probably
put
that
in
when
you,
when
you're,
registering
your
anchor
hash
just
to
call
out
to
that
contract,
to
verify
that,
because
the
hash
is
the
commitment.
P
Maybe
if
you
yeah,
if
you,
if
you
want
to
verify
yeah
yeah,
if
you
want
to
verify
that
but
again
think
of
think
of
this
as
a
as
a
you
know,
if
you
want
to
write
an
anchor
hash
right,
if
you
you,
you
would
have
to
say
what
type
of
hash
is
it
right
is
it?
Is
it
just
a
just
a
uri
hash,
or
is
it
a
zk
hash?
Okay,.
P
And
then
you
need
to
know
what
what
type
of
type
of
type
of
commitment
scheme
you
are,
what
you
know
what
framework
you're
using
yeah.
So
if
you,
if
you
want
to
generalize
this
for
like
one,
you
know
one
to
rule
them
all,
then
you
would
have
to
to
to
add
more
fields
here.
Right
right
wanted
to
do
that
to
make
it
like.
But
then
again
it's
it's.
It's
the
the
the
you
know
you
wanna
you
want
to.
P
You
know
if
I,
if
I
would
probably
use
a
different,
I
would
use
this
contract
for,
for
I
would
actually
write
the
smart
contract
architecture
as
a
as
a
so
since
this
is
not
upgradable
right
and
you
want
to
upgrade
that.
I
would
actually
write
this
as
a
diamond
pattern.
P
M
I
I
D
P
So
you
you
can
you
can,
then
you
can
have
a
you,
can
then
decide.
You
know
you
can
have
different
different
facets.
One
is
sort
of
like
for
non-tokenizing
and
then
others
is
for
tokenizing
right.
So
you
have
a
you,
have
a
document
that
that
that
is
taught
so
you're
sincere.
You
know
it's
like
the
the
the
nice
thing
about
that
is.
P
And
you
include
that
right,
so
you
can
just
check
a
you're
telling
me
that
that
hash
is
in
that
anchor
hash,
and
you
can
then
verify
that
you
can
could
theoretically
verify
that
and
also
in
zero
knowledge.
O
P
Right
so
so
the
the
so
that
there's
there's
there's
there's
the
sd
sd's
general
approval,
prover
frameworks,
you
know,
mature
you'll
be
able
it
will
just
be.
You
know
it
will
just
be
like
solidity
right
and
the
evm,
so
you
have
multiple
contracts
you
can
do
you
know
you
can
do
everything
the
same
way.
You
will
you'll
have
upgradable
right
commitments
and
things
like
that
because
you
can
you
can
you
can
com?
P
I
mean
because
you're,
because
the
the
the
the
the
nice
thing
about
like
with
zk's
darks,
you
can
have
either
you
can
say
I
can.
I
can
do
and
it's
just
a
performance
question.
I
I
don't
have
to
do
a
trusted
setup
and
then
I
it's
just
it's
just
the
size
of
the
of
the
of
the
of
the
polynomials
that
I'm
using
the
number
of
constraints.
P
I
I
set
a
max
once
I
do
my
first
once
I
do
the
first
commitment
and
that
sort
of
like
sets
it
and
I
can,
I
can
add,
additional
things
to
that
to
those
general.
So
I
don't
need
a
trusted
setup.
P
If
I
use,
if
I
use
a
if
I
use
a
hidden
number
groups,
as
my
as
my
as
my
as
my
groups
for
the
for
the
fully
elliptic
curve,
the
for,
if
with
plonk,
I
still
do
it
do
a
trusted
setup,
but
I
just
need
to
do
one
and
it's
extensible.
So
I
can
use
that.
Add
one
more
commitment
to
that
and
then
I
have
a
new
trusted
setup.
So
that's
very
flexible
and
you
can
make
that
very
large.
P
You
can
do
like
10
to
the
22
commitments
where
your
typical
commitment
schemes
are
like
10
to
the
10..
So
you
have
a
lot
of
space
left
for
for
a
lot
of
programs
and
then
and
or
for
starks.
P
You
know
you
can
do
the
same
thing
similar
to
then
you
don't
need
to
trust
the
setup,
but
your
proofs
are
much
larger.
So
it's
like
it's
it's
a
it's
it's
and
then
then
it
becomes
optimizing.
The
size
of
the
proofs.
It's
really
where
this
is
all
what
what
it
boils
down
to.
P
P
Like
you
know,
if
I,
if
you
know
I,
I
get
like
one
megabyte
for
like
4
500
transactions
and
that's
kind
of
like
the
size
of
a
of
a
block
right
now,
and
I
I
think
the
mainnet
would
be
not
happy
if,
if
you
just
like,
oh
I'm,
just
paying
you
like
five
grand
to
just
anchor
that
transaction-
and
you
know
like
oh.
M
P
Wait
I
have
like
you
just
like
blew
up,
blew
up
mainnet
because
you're
paying
more
than
you
know,
it's
like
you're
censoring
everything,
because
you
can
only
do
like
what
so
you
know
where
that's
going
right.
So
it's
it's
it's
it's!
It's!
The
the
the
the
the
blocker
for
zk
rollups
is
the
size
of
the
of
the
of
the
of
the
data,
even
though
it's
called
data
right.
P
But
it's
still,
I
don't
know
what
like
16
gas,
a
byte
or
something
like
that,
and
you
know
it's
like
you're
you're,
if
you,
if
you
just
if
you
just
if
you
just
open
up
the
the
the
floodgates
you
know
to
write
like
I
don't
know
like
two
megabyte
call
data
blocks
also
onto
chain
which
is
like
4
500..
It's
like
I
mean
you're,
going
to
it's
like
you're,
creating
like
a
you,
know,
you're
creating
the
largest
distributed
database
in
the
world
right,
you're,
you're
right,
it's
like
especially
for
stuff.
P
You
don't
need
yeah
right.
It's
like
you
really
only
need
this
stuff.
If
you,
if
you,
if
you
want
to
create
an
on-chain
asset,
yeah
right,
otherwise
it's
like
it's
like.
Don't
do
it
don't
put
it
on
chain,
it
doesn't
make,
makes
no
sense.
O
P
O
P
I
P
That's
perfectly
fine,
that's
that's
what
the
current
current
architecture
can
support
and
that's
what
it's
good
for,
and
you
know
you
do
that
every
you
know
once
in
a
blue,
blue,
blue
moon.
That's
perfectly
fine
right
and
you're
willing
to
pay
the
operator,
like
you
know,
10
grand
for
that
one
right,
because
it's
so
much
so
much
so
much
money
right,
not.
P
O
D
P
Right
that
can
one
can
do
that
yeah
right,
you
do
100
transactions
a
day.
That's
like
you
can
do
that
right.
Now,
it's
not
a
problem
right,
so
you
can.
You
can
do
all
like
10
million
and
up
transactions
done
right.
It's
like
not
really!
You
know
it's
like
it's
like
it's
it's
it's!
You
know
it's
not
that
it's
not
that
many
as
soon
as
you
go
lower.
You
know
obviously
there's
significantly
more,
but
in
the
extremely
high
value
range
there
are
few
yeah.
P
So
it's
it's
not
sustainable
for
for
a
healthy
provider
ecosystem
right.
So
if
you
want
a
healthy
provider
ecosystem
where
people
can
actually,
you
know
make
money
like
with
dexes
and
things
like
that,
then
yeah,
you
need
a
a
you
need,
a
a
you
need
scalability
that
allows
you
to
to
to
do
you
know
a
couple
billion
transactions
a
day
across.
You
know
across
the
entire
ecosystem,.
O
Yeah,
that's
an
interesting
point.
You
make
about
just
thinking
about
the
ecosystem
of
providers
of
this
service.
I
think,
to
date
mostly
just
focused
on
people
who
need
the
service,
but
not
the
providers
of
the
service,
the
quote-unquote
providers.
In
this
this
scale,
this
maturity
level
are
consultants
who
can
build
bespoke
systems
that
use
this
technique.
P
Right
exactly
yeah,
but
that's
that's
just
I
mean
this
is
just
at
the
very,
very,
very,
very,
very,
very
beginning
right.
So
this
is
this
is
going
to
get
significantly
more
sophisticated
because
there
is
there's
there
are.
You
know
it's
like
I'm
discussing
this
within
the
within
a
telco
context,
and
there
are
like
billions
of
dollars
out
that
are
that
are
just
from
settlement
alone
that
are
that
are
that
are
pain
in
the
ass
to
to
reconcile
right
and
their
people
are.
O
You
talk
about
that
just
briefly.
We
can.
I
guess,
because
this
is
being
recorded,
live
streamed,
but,
like
understanding
that
business
use
case
is
something
I'm
I'm
not.
I've
heard
about.
P
I'm
like
so
so
so,
if
you're
making,
let's
say
if
you're,
if
you're,
if
you're,
if
you're
in
vietnam
and
you
you
go
to-
I
don't
know
the
consensus
website
right.
So
that's
with
your
with
your
smartphone
right.
So
what
happens?
I
I.
I
have
a
contract
with
my
local
ictsp,
so
isp,
you
know
it's
like
whatever
that
is
and
whoever
that
is
in
vietnam
right,
but
they
don't
have
circuits
in
in
the
u.s
right
where
it's
like,
based
on
it's
hosted
by.
P
P
Let's
say
I
don't
know
at
t
right
right,
so
then
it
hops
from
the
local
provider
to
t
to
consensus.
So
then
your
local
provider
and
at
t
need
to
settle,
because
at
t
is
not
doing
that
for
free
right.
So
so
they
have
a.
They
have
an
msa
between
them
where
they
have.
You
know
where
they
have
orders
right,
they've
specified
so
orders
or
specific
products
and
products
are
based
on
throughput
latency.
P
You
know
what
have
you
not
they're
they're
different
products,
product
parameters
right
is,
it
is
dedicated?
Is
it
dedicated
circuits?
Is
it
virtual
circuits?
It's
like
you,
know,
mpls
versus
you
know,
so
all
the
all
the
good
stuff
right
and
then
they're
packed
into
an
order
and
and
then
once
the
circuits
are,
are,
are
done
being
used
because
you
know
traffic
drops
and
what
have
you
not?
And
so
then,
then
then,
there's
billing
right.
So
that
happens.
P
So
it
says:
hey
you,
use
that
much
and
then
so
18t
tells
tells
the
vietnam
the
local
provider
in
vietnam.
Hey.
You
used
that
much
data
from
this
from
this
much
product.
You
owe
me,
I
don't
know
50
grand.
P
The
the
vietnamese
provider
said:
oh
no,
no,
no!
No!
It's.
Like
I
closed
that
circuit.
You
know
it's
like
they're,
like
I
don't
know
like
five
circuits
and
and
they
have
a
dispute
over
over
when
one
circuit
closed
right,
so
the
entire
invoice
is
held
up
right
because
the
vietnamese
guy
said
no.
No,
it's
like.
I
owe
you
40
grand,
not
50,
and
then
18t
says
no.
You
owe
me
50
and
then
they
need
to
reconcile
the
whole
thing
right.
P
So
now,
if
you,
if
you
were
to
baseline,
that
right,
you're
you're,
you're
you're
doing
the
order
and
then
you
update
the
order
based
on
when
a
circuit
is
closed,
either
way
right
and
then,
if
if
they
get
them
vietnamese
guys
said,
oh
by
the
way,
I'm
updating
my
order
now,
because
let
the
circuit
close
it
would
go
to
a
t
and
say
it's
like.
Oh,
yes,
that's
true
or
no,
no,
that's
not
true
right.
So
the.
E
O
So
you,
basically
the
dispute,
happens
closer
to
real
time
rather
than
in
a
delayed
billing.
P
P
So
if
there's,
oh
there's,
we
closed
these
five
circuits
yesterday
and
then,
even
if
it's
delayed
and
they're
like
at
these
times
and
then
you're
like
no,
you
didn't
it's
like.
Yes,
we
did
right
so
before
you
actually
build
something
right
and
once
you
build
it,
you
need
to
accrue
for
it.
Do
you
know
what
I
mean?
So
it
has
a
direct
impact
on
your
on
your
on
your
on
your
balance
sheet.
M
I
O
Does
that
mean
that
the
hardware
like
in
terms
of
specifics,
like
the
actual
telco
hardware,
needs
to
support
the
ability
to
talk
to
mainnet,
for
example
like
no?
No,
like
what
level
does
this?
Do
you
baseline?
I
guess
that's
the
question.
O
P
O
So
the
circuit
is
physical,
but
then
also
conceptual
in
the
business
use
case.
It's
like
right.
P
P
P
P
So
so
you're
so
you're
you're
at
the
entry
and
at
the
end
exit
points
to
the
actual
physical,
the
actual
pipe
right
right,
you're
you're,
you're,
you're,
you're
associating
you
have.
You
have
virtualized
routers
right,
so
you
don't
have
a
physical.
You
have
a
physical
router,
but
you
actually
have
virtualized
routers
a
layer.
P
P
P
Yes,
so
at
that
layer,
you're
you're,
like
is,
is
this:
is
this
virtual
network
active
or
not,
and
then
you're
like
it's?
It
went
poof
on
our
side
right
because
either
on
purpose
or
accidentally
right
and
if
you
know
hey,
it,
went
down,
it
went
down
for
like
10
minutes,
and
then
we
got
it.
We
found
it,
we
got
got
it
back
up,
but
we
only
used
that
much
data,
not
that
much
right
or
it
was
not
active
at
that
point
in
time.
So
you
can
so
these
are.
P
These
are
the
the
type
of
things
that
happen
all
the
time,
they're
they're
all
paying
the
ass
wow
right
and
you
can
you,
you
know,
take
that.
Take
that
to
take
that
to
cars.
Take
that,
to
you
know,
iot
devices
right,
it's
it's
it
all!
Any
digital
server
service
has
that
has
that
attached
to
it
the
settlement
of
data
services
right
this
just.
P
O
P
Yeah,
you
can
operate
yeah,
you
can
do
that
and
you
can
you
can.
Then.
If
you
want
to,
you
can
do
the
you
can
do
the
metering
and
the
the
device
actually
controls
the
the
billings
data.
I
mean
you.
Can
you
can
push
everything
you
can
push
everything
to
the
edge
if
you
want
to,
but
that
is
a
everybody
talks
about
it.
But
it's
like
it's
for
most
people,
that's
a
that's
a
conceptual
bridge
too
far.
P
O
Or,
like
maybe
hop
on
to
elon's
new
satellite
delivery
of
data.
O
Yeah
like
for
free
and
free,
netflix
and
hulu,
if
you
you
just
know,
was
it
starlink
or
something
like
that
or
yeah.
O
Yeah
yeah,
that's
exciting,
I
mean
yeah.
I
always
look
interested
to
hear
use
cases
for
baseline
and
you
know
a
lot
of
it's
been
from
when
we
started
working
on
it
a
year
ago.
As
for
supply
chain,
which
yeah
I'm
just
an
engineer,
suddenly
I
mean
I
just
know
what
people
tell
me
about
it,
but
always
curious
to
find
out
other
ones
different,
different
angles.
P
Yeah
so
so,
there's
a
there's,
a
there's,
a
white
paper
coming
out
from
the
metro,
ethernet
forum
mef,
which
is
one
of
the
three
big
three
big
telco
industry
groups,
so
t4
tm
forum,
gsma
and
and
meth,
and
now
there's
a
fourth
one
that
just
started:
cbn.
P
Okay.
So
but
that's
those
are
that's
that's.
What's
that's
that's
the
main
thing
in
in
that,
then
there
are
smaller
ones
like
orion
and
so
forth.
But
that's
that's
it's
that's!
There's
the
those
are
like
the
the
big
global
ones
where
all
the
big
big
big
players
are
are
running
around.
O
P
Yeah,
so
anybody
wants
to
talk
about
scaling,
continue
the
the
scaling
discussion.
It
starts
in
five
minutes
over
in
breakout,
breakout
room
room
room
five,
so
I
I
gave
you
the
gist
already
with
the
what
the
what
the
result
of
the
of
the
discussion
was.
You
teased
it
a
bit
earlier
a
teaser,
but
you
can
find
everything
is
documented.
P
P
The
last
thing
we're
talking
about
throughput
versus
privacy
versus
security,
and
then
we
are
going
to
put
that
into
an
evaluation
matrix,
so
the
basic
one
it
already
exists,
so
the
matter
guys
have
have
already
done
that
and
then
we
need
to
also
put
requirements
together.
So
some
someone
should
should
then
work
on
that
based
on
what's
in
the
mind
map,
and
then
we
need
to
come
up
with
a
github
with
a
github
story
for
to
for
for
for
the
hackathon
to
build
that.
M
Cool
all
right,
I
am
hopping
off
great
chatting
talk
to
you
guys.