►
From YouTube: Cosmos by Billy Renenkamp
Description
Berlin Ethereum Meetup 2019/06/18
Interoperability Edition
A
A
You
had
very
few
options.
If
you
want
to
do
something
that
was
slightly
different
from
Bitcoin,
usually
you
would
just
fork
the
Bitcoin
codebase
and
tweak
some
of
the
parameters.
It
was
the
first
wave
of
coins
they
basically
just
changed
the
inflation
rate
or
whatever
the
the
pre
mines
that
they
could
be
the
one
who
rolled
them
all.
But
if
you
wanted
to
make
any
sort
of
fundamental
changes,
you
had
to
really
deeply
dive
into
a
very
hard
to
use
code
base.
A
A
A
You're
all
aware
of
how
aetherium
works
you
have
this.
This
amazing
ability
to
write
custom
logic
inside
these
smart
contracts,
a
generalizable
state
machine,
turning
complete,
which
let
the
sort
of
app
designers
design
the
apps
and
not
worry
about
networking
consensus
benefit
from
the
the
shared
state
of
aetherium
to
take
care
of
those
problems
and
focus
on
what
your
app
actually
wants
to
do.
Of
course,
this
comes
with
its
own
limitations.
You
are
all
using
the
same
resources.
A
A
A
That
sort
of
gave
birth
to
a
slide
that
doesn't
exist
in
here,
which
is
sort
of
the
idea
of
Internet
of
blockchains
I'm
gonna
hop
back,
but
basic
idea
is
that
it
makes
a
lot
more
sense
for
each
application
to
have
its
own
blockchain
have
its
own
security
concerns.
Have
its
own
resource
allocations,
have
its
own
state
things
like
that,
but
to
not
limit
a
blockchain
to
itself
by
creating
a
form
of
communication
between
block
chains,
so
that
you're
not
restricted
to
your
own
space
I'm
just
gonna.
A
This
is
gonna,
be
a
gift
for
the
whole
night,
and
we
think
this
is
really
great
because
you
get
a
lot
of
the
sort
of
traditional
computer
environment,
app
development
system
that
you're
used
to
we've
developed
an
SDK
to
help
that
process
take
place
faster
and
easier.
The
SDK
comes
with
a
lot
of
sort
of.
We
call
them
modules,
which
would
be
features.
A
You'd
expect
out
of
a
basic
block
chain,
we've
sort
of
solved
that
original
problem
or
that's
our
goal,
which
is
that
it's
very
difficult
to
build
block
chains
when,
if
you
ever
tried
to
again
fork
or
use
this
original
Bitcoin
code
base,
you
ran
into
that
problem.
So
if
we
can
provide
a
solution
for
networking
and
provide
a
solution
for
consensus
which
is
called
tenorman,
then
we
hope
that
you
can
do
similar
to
the
etherion
point
of
view
just
focus
on
the
application
state.
A
A
A
You
wouldn't
expect
for
everyone's
website
to
be
all
on
just
one
server,
although
you
might
be
kind
of
seeing
that
with
Amazon
Web
Services
in
a
way,
but
you
expect
a
bunch
of
independent
players
in
the
world
and
the
ability
for
those
different
computers
to
talk
to
each
other
to
do
different
things
to
send
packets
in
between
them.
So
we
think
this
architecture
has
been
working
fairly
well
for
the
Internet.
We
want
to
try
to
bring
a
similar
model
to
the
way
blockchains
communicate
to
each
other.
A
The
Kosmos
SDK
I
mentioned
comes
with
a
bunch
of
pre-made
modules
for
you,
the
core
of
it
being
tender
mint,
which
is
our
fast
finality
consensus,
algorithm
IBC
has
it
relies
on
tenorman,
but
it's
not
restricted
by
tenorman.
It
is
simulated
tcpip.
It
doesn't
really
matter
which
machine
it
comes
from
as
long
as
it
adheres
to
this
standard.
A
So
if
you
want
to
use
some
other
fast,
venality
consensus
mechanism,
we
fully
encourage
that
if
you
want
to
use
some
other
blockchain
architecture
in
system
like
substrate,
we
fully
encourage
that
we're
sort
of
betting
on
the
idea
that
there
will
be
an
interchange,
similar
to
an
internet,
a
bunch
of
different
block
chains,
doing
a
bunch
of
different
things
for
a
bunch
of
different
use
cases.
So
we
just
want
to
help
make
that
vision
come
to
life
by
creating
tools
to
make
blockchain.
A
So
whether
or
not
you
want
to
use
the
SDK
or
you
have
your
own
home
r
own
home
rolled
solution,
that's
great,
whether
it's
again
substrate
or
something
else
like.
As
long
as
you
have
this
agreed-upon
standard
of
communication
between
them,
we
can
really
open
up
what's
possible
between
these
different
environments.
A
So
dinner
min
is,
is
our
sort
of
it's
been
called
the
golden
standard
for
consensus
algorithms,
but
we
think
it's
just
sort
of
well
tested,
robust,
out-of-the-box,
ready
to
go
everything
and
the
company's
always
been
sort
of
practicality.
First,
taking
safety
over
liveness
sort
of
starting
with
the
idea
of
saw
independent
chains
as
like
the
building
blocks
that
you
would
want
for
a
growing,
interconnected
ecosystem.
A
Instead
of
trying
to
invent
the
whole
thing
at
once,
you
know:
could
you
imagine
if
the
internet
wasn't
possible
unless
you
had
invented
all
of
the
services
we
have
today
from
day
one?
So
we
want
to
sort
of
focus
on
building
blocks
and
let
the
the
architecture
rise
from
there.
Some
of
the
other
features
that
come
with
the
SDK
are
staking
modules.
So
this
is,
if
you
wanted
to
have
say
a
public
blockchain.
You
want
to
have
validators
who
could
join
come
and
leave
at
will.
Maybe
there's
a
fixed
set
of
them.
A
100
is
what
we're
currently
using
on
cosmos
hub,
there's
a
proposal
to
increase
it
250,
but
sort
of,
depending
on
your
security
requirements.
What
you
want
to
do
with
your
blockchain,
you
might
make
decisions
about
exactly
how
the
staking
is
configured.
Maybe
you
want
to
have
proof
of
authority.
Maybe
you
have
a
very
restricted
set
and
you
only
want
to
be
a
single
validator,
because,
whatever
reason
we
don't
make
judgements
on
that,
we
just
offer
tools
to
let
you
configure
it,
how
you
would
wish
we
also
have
a
governance
module.
So
this
is
liquid
democracy.
A
When
you
delegate
your
tokens
to
validators
that
delegation
can
be
used
as
the
voting
power.
You
allow
that
validator
to
vote
on
your
behalf
or
you
can
use
your
tokens
to
vote
differently
than
the
validator
you're
delegating
to
so.
We
think
this
really
opens
up
a
lot
of
possibilities
with
governance.
This
includes
the
ability
for
unchain
parameter
changes.
A
So
it's
something
like
in
between
soft
governance
and
hard
governance,
where
it's
not
strictly
code
upgrades,
but
it
is,
parameters
on
the
running
system
can
be
completely
under
control
from
the
the
governance
system
and
that's
sort
of
out
of
the
box.
Ready,
rewards
and
fees
often
used
in
accordance
with
governance
or
staking
IBC
I
mentioned
earlier.
A
I've
been
working
the
last
couple
weeks
on
a
non
fungible
token
module.
If
you
have
any
opinions
or
thoughts
about
that.
I'd
also
love
to
talk
to
you
about
sort
of
what
that
standard
looks
like
when
you're
moving
to
an
ecosystem
of
interconnected
block
chains,
instead
of
a
single
one,
I
think
there's
a
lot
of
really
interesting
opportunities.
There
PEGI
is
a
module.
A
Typically,
you
might
have
some
idea
of
like
eight
confirmations
or
eight
blocks
need
to
pass
until
it's
really
secure,
but
they're
actually
never
really
secure,
because
they're
probabilistic,
you
know
you
just
get
into
infant
testimony
small
chances
of
a
chain
reorganisation,
but
theoretically
it
could
always
get
reorganized.
So,
depending
on
your
your
security
threshold,
you
might
have
a
peggy
module,
who's
running
or
or
a
zone
whose
entire
job
is
just
to
make
synthetic
assets
like
I,
think
by
Nance
announced
they
would
be
doing
this
week.
A
So
once
they
see
that
say,
ether
has
been
locked
up
inside
of
a
smart
contract
for
12
confirmations,
they'll
mint,
a
rapt
ether
or
synthetic
ether
inside
their
chain,
but
give
it
over
to
that
original
person's
address.
So
they
can
move
along
inside
the
wider
ecosystem
of
the
inter
chain.
With
that
synthetic
version
of
ether,
if
they
ever
wanted
to
go
back
to
specifically
ether
version
of
it,
they
could
go
back
the
same
way.
This
also
assumes
an
East
one
point:
X
architecture,
eath
2.0,
would
have
fast
finality.
A
So
there's
no
reason
not
to
expect
that
they
would
support
IBC
and
to
be
able
to
interact
with
them
as
if
they
were
a
cosmos
chain
from
the
get-go
EVM
we've
been
calling
a
thurman,
it's
now
I
think
being
worked
on
by
chain
safe.
This
is
a
module
that
would
run
inside
of
the
SDK
so
that,
if
you
wanted
to
say,
move
over
your
solidity
contracts
directly
onto
your
own
chain,
you're,
just
the
only
contracts
running
there
to
have
your
sort
of
increased
resources
might
save
you
some
money
on
auditing.
A
We
think
this
is
sort
of
a
cool
feature
and
opens
up
a
lot
of
possibilities,
but
I
would
also
argue
that
the
idea
of
application-specific
blockchain
opens
up
a
lot
more
possibilities
than
just
reproducing
a
really
restricted
environment
like
the
EVM.
If
anybody
is
familiar
with
developing
on
Ruby
on
Rails
or
something
like
that,
it
feels
a
lot
closer
to
that.
A
So
these
are
sort
of
some
other
projects
that
are
building
with
us.
Iris
net
is
another
hub
in
Asia.
Lino
is
a
using
bandwidth,
fees
and
micro
services.
They
have
something
like
twitch
PewDiePie,
the
world's
most
popular
YouTube
star
is
now
exclusively
using
lino
for
all
of
his
games
streaming.
So
that's,
probably
the
highest
user
base
of
any
blockchain
app
out.
There
I
think
it's
number
9
in
the
App
Store.
It's
just
been
used
for
tips
right
now,
but
they're
moving
over
to
have
bandwidth
in
the
future.
A
That
process
is
that
you
probably
want
to
start
up
with
something
like
a
permission
set
and
eventually
distribute
it
once
you
have
a
real
value
they
need
to
care
about.
This
is
a
little
bit
like
the
security
mechanism
for
private
keys.
You
know,
and
people
get
really
angry
if
you
tell
them
to
write
their
mnemonic
phrase
six
times
if
they
don't
have
any
money
on
the
wallet,
so
wait
until
somebody
actually
maybe
has
a
little
bit
of
money
on
there
and
then
go
through
the
whole
security
protocol.
A
I
think
it's
a
much
better
user
experience
and
has
like
much
better
like
0
to
60,
just
gonna
jump
through
to
the
last
slide.
So
now,
the
next
step,
besides
IBC,
is
really
getting
into
what
are
all
the
different
modules
that
are
gonna,
be
out
there
and
creating
a
good
upstream
system
so
that,
if
you
start
using
the
Kosmos
SDK,
you
immediately
have
access
to
a
lot
of
high
quality
modules.
I
mentioned
I've
been
working
on
the
NFT
module
this
week,
but
anything
that
you
could
imagine
could
possibly
be
a
module.
A
There
was
a
Wazza
module
this
week
and
at
the
hackathon
was
really
exciting.
There's
a
couple
stable,
client
projects,
a
couple
decks
projects.
It's
really
exciting!
That
cosmos
is
kind
of
getting
to
the
point
that
aetherium
has
been
in
for
the
last
couple
years,
which
is
that
they
get
to
start
having
the
really
interesting
conversations
about
applications.
And
how
do
you
use
it?
What
a
users
really
want?
How
are
they
users
going
to
use
it
instead
of
the
really
boring
conversations,
which
is?
A
A
So
maybe
I
should
touch
really
quickly.
The
cosmos
hub
as
a
blockchain
is
just
meant
to
be
a
router
of
IBC
packets
so
similar
to
the
internet.
Having
a
and
spoke
model,
it's
a
lot
more
efficient.
If
everybody
decides
on
specific
hubs
to
route
their
traffic
through
then
to
have
a
connection
with
every
single
computer
in
the
world
and
so
cosmos
as
a
blockchain
just
wants
to
be
a
packet
router
for
all
the
different
block
chains
that
might
exist.
I
mentioned
iris
net
earlier
they're,
a
hub,
that's
in
Asia.
A
There
are
partner
of
ours
and
they
also
want
to
be
a
hub
for
a
different
group
of
block
chains,
different
category
whatever
it
is
you
you
have
these
options
and
that's
kind
of
one
of
the
fundamental
aspects
of
the
cosmos
idea.
The
interchanges
is
autonomy
and
control
about
how
and
when
you
route
your
information
who
you
and
I
operate
with
you
don't
have
shared
security.
You
have
autonomy
to
decide
who
you
interact
with
and
what
your
security
threshold
is
for
interacting
with
them.
A
So
we
hope
to
offer
that
as
a
service,
the
Kosmos
hub
keep
a
very,
very
high
security
threshold
to
allow
high
quality
projects
to
interoperate
I
think
that's
about
it.
I
don't
know.
If
that
was
fast
or
slow.
Is
there
any
questions
now?
Are
we
doing
that
later?
I
know
you're
not
supposed
to
give
timelines,
but
you
gave
one
earlier
I
just
kind
of
didn't
hear
it.
B
Basic
question
to
understand
the
interoperability
between
block
clients.
If
I
want
to
transfer,
let's
say
one
either
from
aetherium
to
the
Bitcoin
block
claims
and
the
number
of
either
and
the
ethereum
blockchain
has
to
be
decreased
and
the
equivalent
on
the
Bitcoin
block
plane
has
to
be
increased,
and
so
how
does
it
work
as
a
buffer
account
between
or
where
the
money
comes?
All
of
the
other
money
goes
on
the
one
side
and
the
other
money
comes
to
the
other
side.
The
same
question:
if
you
transfer
an
asset
so.
A
The
Bitcoin
aetherium
example
might
be
a
little
more
clear
if
you
think
about
going
from
bitcoins
who
aetherium
just
because
the
scripting
language
and
bitcoin
is
a
lot
harder.
So
there's
two
ways
you
can
think
about
it.
One
would
be
one
of
these
pegs
owns
I
mentioned
that
peg
zone
has
a
validator
set
who's
watching
the
Bitcoin
blockchain
all
the
time.
A
They
have
a
designated
address
there,
where,
if
you
were
to
move
Bitcoin
into
that
address,
they
would
wait
some
safe
number
of
confirmations
until
they
think
there's
no
more
chain
reorganization
and
then
they
wouldn't
meant
a
synthetic
Bitcoin
on
their
chain.
So
their
chain
is
connected
to
probably
cosmos
and
maybe
another
hub
or
something
like
that
and
from
there
you
are
able
to
move
that
synthetic
bitcoins
somewhere
else
say
you
wanted
it
to
move
it
to
aetherium
directly.
A
You
could
move
it
via
a
hub
or
directly
to
another
peg
zone,
who
is
pegging
with
aetherium
they're
doing
the
same
thing
that
I
described
a
Bitcoin
but
on
a
theory,
and
they
have
some
contract.
Where,
once
you
send
money
into
it,
they
wait
for
eight
confirmations.
They
make
a
synthetic
version
of
it
on
their
own
chain.
In
this
situation
they
would
have
a
very
special
one,
which
is
that
they
would
mint
synthetic
Bitcoin
on
the
etherion
blockchain,
with
the
understanding
that,
after
that,
bitcoin
has
moved
around
on
a
theorem.
A
It
might
make
its
way
backwards
via
that
peg
zone
via
the
hub
peg
zone
Bitcoin.
If
you're
talking
about
actually
transforming
Bitcoin
to
ether,
you're
gonna
need
an
exchange.
You're
gonna
need
a
market
maker.
You're
gonna
need
somebody
who's
willing
to
to
take
that
that
order,
and
that
could
take
place
in
a
number
of
places
once
it's
already
on
the
interchain.
Maybe
you
take
it
over
to
the
finance
chain.
You
sell
it
over
there.