►
From YouTube: The Baseline - Livestream
Description
The Baseline Protocol Community's weekly show, featuring developer interviews and time for questions and answers. Get involved at https://baseline-protocol.org
C
D
A
C
A
A
So
I've
been
doing
a
bunch
of
math
over
the
last
week,
just
trying
to
kind
of
get
a
bead
on
that
story?
There's
a
publication
that
wants
an
article
about.
You
know
what
does
2030
internet
look
like
and
the
the
you
know
the
part
of
that
question
has
to
do
with
with
you
know
what
is
what
is
the
blockchain
do
for
the
internet,
and
how
can
we
predict
things.
A
A
A
D
A
Right
and
I'm
like
wait
a
minute
wait
a
minute
I
can't
do
read,
writes
on
us
on
a
local
ssd
that
fast
right.
So
there's
got
to
be
detail
underneath
that
right,
so
that
that
was
that's
really.
The
question
is
what
the
what
are
the
three
factors
that
make
you
know:
that'll
that'll,
make
any
enterprise
be
able
to
not
worry
about
the
end
of
the
the
end
of
the
of
the
the
sidewalk
or
the
end
of
the
blockchain
in
terms
of
getting
stuck
in.
A
A
Well,
the
capacity
of
the
mainnet,
the
final
layer
of
the
internet
state
machine
will
inc,
will
double
every
x
years,
right
kind
of
like
moore's
law
right
number
of
transistors
on
a
on
a
chip
can
double
every
two
years
and
or
18
months,
and
until
until
you
hit
quantum
problems
right
which
we
have
hit,
but
it
got
us
a
long
way.
So
what
is
that
sort
of
continuous
improvement,
or
is
that
ever?
Is
that
really
even
the
right
analogy
to
make?
A
Silicon
lithography,
you
could
say
yeah
we're
going
to
have
this
many.
You
know
transistors
per
yeah
and
and
we
can
improve
the
process,
a
lot
of
science
still
in
there.
We,
you
know,
I
think,
right
now
we're
gallium
arsenide
in
terms
of
you
know,
substrates
and
all
these
things
massive
innovation,
massive
improvement,
but
but
all
within
a
single
line
of
thinking
right,
it's
a
hard
but
straightforward
problem.
A
So
by
seven
you
know.
If
moore's
law
was
articulated
in
64,
you
could
articulate
you
could
you
could
be
confident
in
the
trajectory
of
that
prediction
by
mid
70s
and
then
by
what
78
bose
was
already
working
on
the
headset
and
by
a
decade
later,
there's
one
for
aviators
that
you
have
to
plug
into
power
a
decade
after
that
you
can
run
them
on
batteries,
and
I
suspect
that
they
knew
more
or
less
within
a
few
years
what,
when
they
could
pull
off
a
commercial
product
that
people
would
pay
for.
A
A
A
We
know
that
we
can
do
about
1.8
million
baseline
events
a
month
right
now
with
no
optimizations,
no
batching,
no
mutex,
no,
no
sharding,
no
eth2,
no
nothing.
If
eth2
delivers,
100x
processing
throughput,
then
we're
getting
closer.
If
we
have
another
100x
with
batching
and
other
techniques
that
we're
working
on
with
you
know,
karthik
soloperam
is
is
really
all
over.
What's
next
after
that,
and
can
we
say
well
yeah
shards?
Will
you
know
shards
will
have
problem
with
shards.
A
Is
that
you
get
marginal
diminishing
returns
as
you
add
shards,
but
can
you
say
well
we
can
we
can.
You
know
we're
going
to
go
64
or
64
shards
and
we'll
get
to
10
28
shards
within
this
many
years
ish,
and
what
will
that
do
or
is
that
even
the
way
to
think
about
it
is
sharding?
So
sharding
is
one
way
right
because
reads
and
writes
are
one
problem:
bandwidth
is
another,
so
we
know
bandwidth
is
increasing.
A
D
A
D
D
So
so
effectively
what
you're,
what
you're
doing?
It's?
Actually,
not
that
that
that
very
different
from
from
from
from
from
the
evm
in
a
way,
except
that
the
underlying
architecture
is
not
of
a
neumann
architecture.
It's
an
it's
a
it's!
A
circuit
circuit
based
architecture
right
so,
but
in
the
end
the
tool
chain
is
the
same.
You
have
you
have
a
higher
level
language
that
allows
you
to
that.
D
That
is
a
that
you
code
into
a
program
and
that's
the
general
problem
statement
for
zero
knowledge,
proofs
that
you
have
a
a
that
you
have
a
program
but
which
you
want
to.
You
have
inputs
and
you
have
an
output,
and
you
want
to
prove
something
about
the
output
that
without
revealing
or
what
that
is
right.
What
the?
D
What
the
knowledge
that
you
put
into
into
that
program
is
so
so
what
they,
what
they
did,
they
created
and
they're,
not
the
only
one
but
start
where
they
created
their
own
language,
similar
to
solidity
matter.
Labs
is
doing
that
also,
similarly
they're,
just
using
a
different
set
of
cryptographic
primitives
and
they
turned
this
into.
A
D
A
Yes
and
the
upside
transactions
are
the
upside
of
that.
Is
it
even
further
obfuscates,
what's
going
on,
which
is
part
of
the
baseline
principle
right,
which
is
yeah.
Anything
that
goes
on
change
should
look
like
noise
should
sound
like
a
metronome.
You
should
be
leaking
no
meta
or
actual
information
about
your
activities
or
your
business
relationships
right.
So
that's
good.
A
So,
even
if
it
further
distances
the
maintenance
at
some
point
so
say
say,
you've
got
say
you
and
I
do
a
million
purchase,
orders
and
other
kinds
of
events
a
day
between
each
other
and
other
counterparties,
and
we
go
through
the
baseline
pro
process.
We
message
it,
those
you
know,
go
through
messaging,
you
sync
them
up.
You
prove
you,
you
generate
the
signatures.
Signatures
come
back
in
zero
knowledge
circuit
of
one
type
or
another,
will
validate
that
and
then
right
now
that
goes
straight
to
the
chain,
the
main
ethereum
in
the
case
of
bri
one.
A
C
D
D
There
are
some
disadvantages,
but
let's
start
with
the
advantages
right,
you
start
with
a
positive,
and
then
you
you,
you
do
the
to
the
cons
right.
Let's
start
with
the
pros,
so
the
pros
is
obviously
you
can
you
can
you
can
do
some
some
some
neat
parallel
processing,
which
you,
by
definition
you
can't
do
with
on
on
on
chain
right?
It's
it's
it's
one
after
the
other
has
to
be
the
another
pro
is
that
you
don't
have
to
pay
50
50
per
transaction
in
in
gas
cost
right,
because
you
now
you
well.
D
Right
exactly,
rather
than
doing
it
doing
paying
paying
50
million
dollars
for
a
million
transactions.
You
end
up
paying
you
know,
you're
putting
putting
you
know,
let's
say
even
just
a
thousand
transactions
per
block
or
fifteen
thousand,
so
there's
there's
there,
there's
significant
cost
savings
right,
so
you're,
then
you're
talking
and
you're.
Talking
about
you
know,
maybe
you
know
five
thousand
dollars
versus
50
million,
so
your
your
your
transaction
cost.
Suddenly
it
goes
down
to
a
couple
of
cents
transactions
versus
50
50.
D
Another
nice
nice
thing
about
that
is.
Is
you
you
you
because
now
you
can
pay
more
your
your
transaction
is
going
to
get
confirmed
faster
and
is
more
likely
to
get
to
get
mined
so
that
so
you're
you're
actually
you're
actually
playing
the
game
that
the
miners
want
you
to
play
and
you're
you're
you're
you're
taking
advantage
of
scale
yet
again,
the
the
advantage
of
of
of
these
zero
knowledge.
Proofs
is
what
the
name
says.
There
is
zero
knowledge
right.
So
it's
like
it's
like.
D
A
D
B
B
I
work
at
kinko's
in
college
in
high
school
right
at
the
end
of
the
day,
because
I
was
had
to
run
the
register,
you
had
to
run
the
tape
and
reconcile
everything
and
you
know,
tie
it
up
with
a
bow
and
fill
out
your
report
and
at
the
time
this
is
how
long
it
was.
We
had
to
fax
that
report
into
the
home
office,
so
they
could
get
their
data
right.
So
it's
you
know
the
equivalent
kind
of
thing
right.
D
D
And
you
know
they're
they're,
all
so
the
the
consensus
that
the
company
then
presents
that
that
everything
is
correct
is
is,
is
now
provided
by
the
main
net,
because
it's
decentralized,
it's
a
lot
more
secure
than
the
than
the
whatever
oracle
or
sap
system
that
that
that
they
had
with
a
with
a
with
an
admin
and
test
one.
Two,
three
password
right.
E
E
E
Between
enterprises,
so
just
the
the
the
potential
already
now
with
the
with
the
given
technology,
the
given
setup
to
to
use
baseline
the
baseline
stack
to
exchange
messages
and
then
to
orchestrate
them
to
more
complex
workflows
where
you
aggregate
some
parts
of
these
messages
as
intermediate
steps
that
then
get
baseline.
Nobody,
nobody
is
telling
us
that
we
have
to
baseline
every
atomic
step
of
a
business
process.
We
have
to
baseline
the
atomic
steps
of
a
workflow,
but
the
definition.
What
what
bundle?
E
E
Most
of
them
want
to
have
a
monthly
printed
out
paper
with
the
us
dollar
price
per
transaction,
and
it
has
to
be
predictable
because
integration
processes
are
long-lasting
processes
and
it
it
doesn't
help
if
your
integration
works
in
november
and
will
will
not
work
in
december
because
of
some
crazy
cryptocurrency
price
movements.
So
these.
A
A
E
We
want
to
take
the
advantages
out
of
it,
but
somehow
you
have
to
smooth
out
the
disadvantages,
especially
in
the
in
the
early
phases
of
bringing
in
new
technology
to
enterprises,
because
they're,
just
not
they're,
just
not
interested
in
in
the
risk
of
taking
the
full
control
on
custody
on
cryptocurrencies.
You
have
to
bring
solutions
for
all
that.
E
So
if,
finally,
we
are
at
the
stage
where
we
can
really
convince
enterprises
or
potential
customers
about
the
baseline
technology,
and
they
all
see
the
the
benefit,
they
love
the
easy
integration
and
onboarding
stuff,
but
we
also
have
to
give
them
some
smooth
out
solutions,
at
least
for
the
for
the
next
say.
I
don't
know
one
or
two
years
to
deal
with
the
stuff
that
they
don't
know
yet,
and
this
is
mainly
about
keeping
private
keys
for
crypto
assets.
D
D
Why
wouldn't
you
just
give
them
a
set
of
apis,
an
api
key
and
then
be
done
with
it
and
and
and
hand
them
over
over
a
monthly
monthly
monthly
bill?
They
don't
really
care
right.
I
mean
they,
don't
even
care
what
it
runs
on
right.
You
could.
You
could
have
like,
like
like
a
hundred
thousand
people,
doing
doing
the
stuff
in
in
in
vietnam
right
and
they
they
don't
care
right.
D
It's
it's
they
honestly,
don't
so
so
as
long
as
you
as
you
is
that
you
know
that's
baseline
as
a
service
right,
I
mean
that's,
that's
what
you
do
right.
It's
it's!
It's
a
it's
a
class!
It's
a
cloud
service!
It's
a
lambda
function!
It's
a
you
know!
It's
it's!
You
dump
something
in!
You
get
something
out
what
you
want
and
that's
that's
wonderful.
They
don't
care
how
it
works.
E
Yeah
that
that's
another
learning
of
these
first
very
exciting
months
of
baseline-
that
at
least
in
in
my
perhaps
a
little
bit
biased
perspective
of
sap
integration-
that
it
is
a
huge
amount
of
old-school
integration
work
to
do
with
with
these
pocs
until
it
actually
comes
to
the
point
of
baselining
something.
But
this
is
also
good
because
if
there
have
been
integration,
challenges
that
have
not
been
tackled
in
the
in
the
recent
years
and
now
baseline
is
the
reason
to
finally
tackle
them.
It's
wonderful!
E
A
Yeah,
it's
business
process,
automation.
I
was
just
reading
an
article
the
other
day
that
you
know
suggested
that
really
where
it's
at
right
now,
everybody's
into
automating,
their
processes
and
now
automating
them
across
company
lines,
and
that
is
the
big
change
right.
You
know
the
headlines
still
are.
You
know
blockchain
blockchain
blockchain.
Sometimes,
although
I
did
hear
a
publisher
just
the
other
day
say
we
don't
want
any
more
blockchain
book
projects
right.
A
You
know
a
friend
of
mine
told
me-
and
I
was
like
yeah,
of
course,
because
we're
growing
up
and
there's
already
too
many.
You
know
prediction
books
about
blockchain
and
you
know
oh
here's
a
use
case
and
there's
a
use
case.
I'm
like
man,
I
think
it's
after
five
years,
we've
gotten.
We
know
what
the
use
cases
are
and
tom-
and
I
were
talking
earlier
this
morning,
tom,
like
tom
lyons
from
from
zurich
right
are.
You
are.
B
B
Some
of
them
have,
I
think,
you've
been
touched
on.
There
was
one
question
from
daryl
with
respect
to
you
know:
how's
the
kona
thing
going
without
using
names
or
anything
like
that.
I
think
stefan
kind
of
alluded
to
some
of
the
the
research
and
the
findings
that
they're
getting
there.
I
don't
know
if
kyle
and
stefan
want
to
say
anything
else
about
that.
C
Yeah
sure
I
would
say
the
kona
project
is
going
great.
The
the
teams
have
really
come
together.
You
know,
we've
got
a
good
cadence
going.
We
there's
a
lot
of
integration.
Work,
that's
been
just
being
done
at
the
same
time
that
a
lot
of
the
the
baseline
stack,
sort
of
hardening
and
production
readiness
work
is
coming
together.
C
You
know
in
the
kona
environment
say
that
we've
got
a
pretty
robust
baseline,
enabled
workflow
that
we've
modeled
and
is
being
worked
on
sort
of
on
the
fast
track
to
being
in
a
a
state
where
we
could
take
take
it
into
the
actual,
your
actual
production
environment,
the
ideas
you
know
to
launch
that
in
q1
early
q1
silvan.
You
want
to
hop
in
and
say
anything
else
there.
I
think,
there's
some
really
awesome.
You
know
stuff.
C
C
Need
yeah?
No,
no,
we
would
be.
I
wouldn't
go
too
far
there
but
yeah.
I
think,
there's
you
know
it's
really
great
to
see
the
understanding
that
kona
brings
to
the
table
of
the
space.
You
know
they're,
really
they
they
know.
They
know
what
questions
to
ask.
So
it's
not
like
a
lot
of
blockchain
projects.
A
B
Does
does
the
team
have
any
perspective
on
how
enterprises
are
viewing
tokenization
today
right
with
microstrategy
announcing
their
their
big?
You
know
they're
going
to
start
investing
in
bitcoin.
Are
we
seeing
more
adoption
of
that
or
interest
in
tokenization,
and
does
this
I'm
going
to
take
all
of
daryl's
questions
and
try
to
wrap
them
up,
and
does
this
imply
that
they're
going
to
have
to
work
more
with
exchanges
and
if
so,
what
kind
of
exchanges
and
that
kind
of.
A
D
A
A
Does
it
well,
you're,
just
confirming
consistency
between
state
machines
right
and
that's
all
you're
doing
on
a
record
that
you
know
on
a
record
by
record
basis,
which
could
probably
andreas
take
us
back
into
round
two
of
that
conversation,
because
I
think
that
we
left
a
couple
things
on
the
table
not
for
now.
Let's
get
some
questions
in,
but
I'd
love
to
talk
more
about
what
does
it
mean
when
you
start?
You
know
if
effectively
you're
talking
about
rolls
roll
ups
and
batching
and
that
you've
got
a
day
cadence
on
these
things.
B
Yeah,
but
also,
if
you
think,
about
some
of
the
demos
that
are
being
worked
on
right,
where
we're
looking
to
take
a
procurement
flow
with
an
invoice
or
a
set
of
invoices
and
tokenize
those
to
make
them
a
an
investable
our
collateral,
for
you,
know,
financing,
and
that
kind
of
thing
to
your
point.
It's
two
separate
things:
right:
you're,
baselining,
the
information
and
then
you're
tokenizing.
It
right.
A
Right
and
the
boring
thing
there
is
you're
just
reducing
the
amount
of
effort
that
the
factor
the
person
that's
going
to
give
the
cash
to
the
supply
company
that
needs
the
money
they
you
know
right
now.
They've
got
to
call
the
buyer
and
make
sure
that
they're
actually
going
to
pay
and
they've
got
some
risk
there,
and
maybe
the
buyer
fat
fingered
one
fewer
zero
in
the
in
the
amount
that
there's
that
they
thought
they
were
having
to
pay.
So
when
the
check
comes
in
checks,
we
stole
those
right.
A
When
the
check
comes
in
it's
it's
10.
You
know
it's
factor
of
10
less
than
what
they
were
expecting
and
there's
all
these
things
that
happen
and
if
you
just
say,
yeah,
here's
the
record,
it's
baseline,
you
can
still
call
them.
If
you
want,
you
have
a
lot
lower
risk.
I
mean
it's
that
boring
right
and
it's
just
a
one-step
issue.
E
D
E
E
I
think
the
same
happen
will
happen
with
tokenization
the
the
ones
that
are
really
interested
in
tokenizing
the
invoice.
Today
they
are
very
ahead
of
all
their
competitors
because
they
actually
think
about
what
tokenization
means,
but
until
it
becomes
a
serious
thing,
a
widespread
thing
with
all
enterprises.
A
C
A
Yeah,
so
you've
got
this.
This
record,
this
invoice
in
a
sap
system
and
you
by
tokenizing
you're
reading
the
record
and
then
putting
it
on
a
blockchain
for
for
whatever
yeah
and
and
mainly
for
that,
so
that
you
do,
you
have
the
benefits
of
anti-double,
spend
and
double
execute,
and-
and
you
know
that
everybody
has
the
same
thing
but
there's
another
place
for
baselining,
is
to
say
in
the
process
of
converting
that
record
into
a
blockchain
item.
A
D
B
Thank
you.
So
we've
got
another
kind
of
a
newbie
question.
How
would
an
enterprise
go
about
adopting
the
baseline
protocol?
Is
this
a
technique
that
they
can
start
themselves
or
you
know,
should
they
reach
out
to
unibright
and
provide
envision
and
everybody
in
the
community?
What
what
does
the
team
recommend.
A
Can
I
say
yes
and
yes,
they
can
reach
out
to
the
companies
in
the
community,
such
as
you
know,
brighton
and
and
provide
and
others,
and
they
can
also.
I
think
you
know
somebody
say
somebody
said
well.
Is
this
ready
for
productization,
I
said
well
or
production
readiness?
Well,
a
competent
team
can
take,
can
take
the
baseline
pattern
and
build
it
into
into
their
own
systems
in
production
on
their
own.
If
they're
competent
today
right
so
there's,
no,
you
don't
have
to
wait
around.
It's
not
like
it's
there's
no
magic
sauce
here.
A
B
Right
exactly
they
have
to
have
a
competent
team.
They
have
to
be
competent,
not
only
in
you
know,
I.t
and
integration
type
topics,
as
stefan
pointed
out,
but
also
blockchain
and
baseline,
and
you
know,
there's
no
better
way
to
accelerate
that
learning
than
to
work
with
someone
like
a
provider
or
unibrighter
or
one
of
the
others
right
to
really
kind
of
jumpstart.
That
there
is
a
lot
of
are
a
lot
of
questions
with
round
with
respect
to
data
security.
You
know
we
obviously
we're
close
to
baseline.
B
We,
we
know
how
a
lot
of
this
works,
but
maybe
for
some
of
the
newer
viewers.
How
are
companies
data
protected?
And
you
know
it
seems
to
be
a
big
blocker
as
well
for
for
sitsky
mensker?
It's
like
you
know.
Nobody
wants
to
put
their
data
on
the
blockchain.
Let's
just
remind
folks
how
that
works
with
baseline.
A
Yeah,
so
that's
a
it's
a
great
question
and
the
good
simple
answer
is
no
data
is
going
on
the
blockchain.
You
know
a
proof
of
consistency
is
going
on
the
blockchain.
You
can
use
that
as
a
key
and
a
key
value
lookup,
but
you
know
you're.
Basically
what
we're
I
think,
one
of
the
best
things
about
baselining
is
that
we're
not
in
some
kind
of
weird
cold
war
with
traditional
database
companies,
the
more
baselining
there
is,
the
more
those
companies
are
relevant,
which
is
good.
So
you
know
you're.
A
If
you
have
good
data
security
in
your
company,
then
that's
what
yeah
and
and
you're
compliant
with
things
like
gdpr
and
that
sort
of
thing
then
that's
all
you
have
to
do
you
don't
have
to
you.
Don't
have
any
new
threat
surface
with
baselining
you're
using
the
main
net
for
very
one,
very,
very
simple
thing,
and
it's
you
know
it's
more
than
just
state
hashing
your
database.
So
let's
be
clear,
it's
not
just
doing
that.
You
could
do
that
with
tyrion
or
chainpoint,
or
anything
like
that
right
now.
B
I'll
go
into
the
next
question.
You
know,
andreas!
You
touched
on
this
a
little
bit
earlier
with
respect
to
oracle's,
but
you
know
how
would
oracle's
be
used
with
baseline
and
you
know
obviously
chain
link
just
published
an
article,
a
medium
post.
I
believe
you
know,
is
one
oracle
enough
for
or
how
do
you
guys
envision
oracle's
working
with
the
baseline
protocol.
E
Yeah-
and
perhaps
I
can
tell
something
about
it,
because
we're
often
asked
when
we're
talking
about
system
of
records
and
integration
like
sap,
and
then
we
often
get
the
question.
Okay,
that's
where
you
need
our
oracles
right
and
in
the
meantime
of
like
having
answered
these
questions
several
times,
I
tend
to
say:
no,
because
there
are,
I
think
you
have
to
still
make
the
you
have
to
distinguish
between
process
and
data.
I
mean
this.
E
This
is
the
storytelling
towards
baseline
as
well
misunderstanding
blockchain
as
a
database
and
now
more
understanding
it
as
a
middleware
component,
and
I
think
the
same
holds
true
for
understanding
external
data
to
processes,
an
sap
system
in
itself.
It's
it's
more
than
just
a
database.
It's
already
there
are
workflows
inside
there's
goes
different
departments.
You
have
different
document
flows,
all
all
inside
the
box,
so
oraclizing
this
complete
state
machine,
it
would
just
be
it
would
wouldn't
be
impossible.
E
For
example,
if
you
have
an
insurance
case
that
relies
on
some
specific
weather
conditions
in
a
certain
area,
because
then
a
certain
insurance
kicks
in
then,
of
course,
you
need
a
trusted
source
of
truth
and
not
one
single
source
of
truth
telling
you
how
the
current
temperature
is,
but
for,
for
example,
for
a
purchase
order
that
one
responsible
person
creates
in
sap
and
clicks
on
a
button.
It's
not
so
much
use
of
oracalizing
this,
because
it's
just
the
single
source
of
truth
is
already
within
that
that
state
machine
so
short
answer
to
just
baseline
processes.
B
B
A
D
B
C
Think
so,
yes,
I
even
I
even
think
we
could
use
the
existing
baseline
stack
with
just
a
little
additional
extension
to
that
to
baseline
the
actual
messages
like
or
how
do
you
have
a
running
consensus
of
hashing
on
them,
each
nat's
message
and
use
a
very,
very
lightweight
layer.
Two
solution
extended
just
right
there
in
line.
B
A
A
You
know
if
you're
going
to
batch
things
up,
that's
great,
but
you
got
to
also
worry
about
sequencing
because,
as
you
batch,
if
things
get
into
that,
that
need
to
be
verifiably
sequenced,
one
after
the
other
and
they're
all
in
the
same
batch
and
they
pass
in
the
night.
You
got
an
issue.
There's
ways
of
dealing
with
that.
A
But
now
you've
got
that
particular
issue
to
deal
with,
whereas
if
you
were
just
I
mean
if
we
could
just
brute
force,
if,
if
the
it
was
the
quantum
ethereum
main
net
right,
you
know
like
the
magical
super
thing
in
the
sky
that
had
no
limits
in
terms
of
throughput
and
and
latency
and
everything.
Then
you
could
just
you
know
brute
force,
one
event
after
another
and
you
would
have
no,
you
wouldn't
have
to
be
clever
at
all
about
how
you
laid
those
down
as
it
is.
That's
not
the
case
right.
A
You
know
there
are
laws
of
physics
here
in
terms
of
reads
and
rights.
I
think
it's
not
even
about
consensus.
I
remember
talking
to
a
friend
who
I
I
won't
mention
the
name,
but
he
was
claiming
you
know
these
huge
numbers
of
transactions
per.
Second.
With
this,
his
particular
consensus
algorithm-
and
I
said
all
right:
what
happens
when
you
add
any
kind
of
data
layer
to
that,
so
you
can
say:
oh
no,
it
goes
down
to
less
than
100
transactions.
A
A
second
I'm
like
okay,
okay
right,
so
you
got
to
know
what
you
know
the
devils
are
in
the
details.
So
in
this
case
it's
you
know
it's
really.
How
many?
How
many
things
can
you
write?
How
many
things
can
you
read
at
any
given
time
on
a
singleton
and
yeah,
so
you
can
batch,
you
can
be
clever,
but
the
question
is:
how
far
can
you
go
before
you
run
on
a
rope
before.
A
Well,
you
know,
let
me
be
a
little
bit
yeah.
I
still
don't
you
know
if
you
go
too
far
with
the
whole
layer,
two
thing,
then
you
have
to
ask
why
not
just
stick
with
what
we've
already
got.
You
know
set
up
an
mq
between
different
companies.
Ariba
tried
to
do
effectively
that
be
the
global
integration
bus
in
the
90s,
and
we
want
to
give
them
that
much
power.
A
So
you
know
if,
if
you're,
just
if
we're,
if
you
what
you
don't
want
to
say-
and
I
think
this
is
important-
you
don't
want
to
say
to
people-
hey,
don't
set
up
all
these
different
integration
buses
for
every
every
time
you
set
up
a
new
partner
right,
the
mainnet's
going
to
do
that.
For
you,
that's
the
point
of
blockchain
or
baseline
right.
You
have
this
always-on
state
machine
that
resists
tampering,
and
you
don't
have
to
set
up
right.
You
can
pay
as
you
go.
A
D
Don't
think
it
has
to
be
the
answer,
so
I
would
I
would.
I
would
disagree
with
that
and
for
the
following
reason:
the
the
the
so
communication
integration
that
has
been
standardized-
and
that
is
that
is
is
is-
is
already
well
known,
for
example,
in
telecommunications
right
you
can
there
there's
there's,
there's
fully
automated.
You
know
some
companies
have
gone
out
to
fully
automated
quote
ordering
and
now
even
even
a
settlement.
The
big
challenge
is
so
so
the
communication
piece
is
standardized.
D
Automated
wonderful
everything
goes
back
and
forth
where
they
are
struggling
with.
Is
the
fact
that
you
know
it's
like
it's
like
it's
like
harry,
set
up,
set
up
the
msa
and
and
everything
in
in
sap.
One
way-
and
you
know
it's
like
it's
like
mature-
set
up
set
up
the
same
thing
in
oracle
slightly
differently.
So
that's.
A
D
Yes,
and
you
can
right
and-
and
you
can-
you
can
have
the
entire
document-
the
entire
commercial
document
upon
which
everything
is
based
synchronized,
so
you
can
validate
the
constraints
right.
That
translates
then
into
zero
knowledge,
proofs,
the
constraints
against
what
you're,
what
you're,
what
you're
attempting
to
do
right,
you're,
you're,
generating
that
and
you're
like
hey?
No,
actually
that
doesn't
work
against
the
constraints
you
just
put
in
inputs
that
are
not
that
are
not
they
don't
work
right.
Your
zero
knowledge
proof
gets,
gets
back
and
right,
so
not
good.
D
A
Let
me
just
copy
right
there,
because
I
think
that
is
you
just
you
just
touch
the
moment
of
of
of
clarity
on
what
baselining
is
all
about
and
what
the
state,
why
it's
a
standard
and
not
something
else,
because
it's
really
agreeing
that
we're
going
to
run
that
check
before
we
do
other
processes
right.
So
it's
it's
that's
why
it's
a
technique
and
not
some
fancy
technology
right.
A
A
B
A
I
was
gonna
agree
with
that.
You
know
baseline
hyphen
protocol.org,
the
slack
inviter
is
right
there.
Then
we
just
hit
700
and
we're
over
200
actives
at
any
given
moment.
It
looks
like
by
looking
at
the
analytics
there.
It's
pretty
good
and
it's
going
up
so
please
come
in
join
the
party,
it's
a
very
friendly
crowd
and
people
with
really
great
technical
chops,
who
are
very
happy
to
to
to
talk
with
you.
You
just
have
to
reach
out
one-on-one
all
right.