►
From YouTube: Saturn Launch - Ansgar Grunseid
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
Today
is
honestly
an
incredible
pleasure
for
myself
and
the
rest
of
the
team,
because
you
guys
will
all
be
able
to
tell
your
grandkids
and
however
many
years.
That
is
that
you
were
there
on
the
day,
October
27th
year
of
Our
Lord
2022
the
Saturn
launched
today
in
about
20
minutes.
We
are
going
to
launch
in
front
of
your
very
eyes,
but
until
then
I'm
going
to
take
you
on
a
little
journey
of
what
Saturn
is
why
we
built
it.
A
What
problems
it
solves,
so
you
can
get
just
as
excited
as
we
are,
if
not
more
excited
so
I'm
Ansgar,
grunzide
I
am
a
Saturn
lead
and
today
we're
going
to
walk
through
what
Saturn
is
a
new
CDN.
So
every
great
product,
every
great
project
starts
by
solving
a
problem,
starts
by
embracing
an
opportunity
and
we
have
a
monstrous
one
in
front
of
us
with
ipfs
and
filecoin.
Filecoin
is
an
incredible
storage.
Network
there's
already
been
assembled
17xbit
bytes
of
capacity
300
petabytes
used,
but
there's
a
problem.
A
A
We
can
upload
these
videos
through
these
videos,
but
how
can
we
build
an
app
like
YouTube,
where
a
video
can
go
viral
in
a
hundred
thousand
two
hundred
thousand
five
hundred
thousand
a
million
people
can
all
access
this
video
concurrently,
you
know
so
the
wave
the
way
filecoin
has
been
built
is
you
know
when
a
file
is
stored
across
Falcon's
Network,
it's
broken
into
pieces
and
sharded
across
storage
providers,
usually
somewhere
between
four
to
ten
six
to
ten,
but
that
doesn't
scale
to
YouTube
scale
to
web
2
scale.
So
it's
a
huge
opportunity.
A
How
do
we
unlock
web3,
YouTube
and
beyond
that
to
step
forward
forward
in
time,
more
close
to
like
where
we
find
ourselves
today,
there
are
already
tons
and
tons
of
people
who
want
to
build
apps
on
filecoin
and
can't
they're
already
storing
their
data
in
filecoin,
but
they
can't
access
it
readily.
So
this
is
the
opportunity
that
Saturn,
realizes
and
Embraces,
but
beyond
that,
like
every
other
great
product,
you
brace
some
sort
of
insight,
there's
some
sort
of
wave,
something
that
changed
from
the
past.
A
That
lets
you
do
something
that
wasn't
possible
before
the
Insight
that
Saturn's
had
and
the
wave
that
we
ride
is
cryptographic
incentives,
cryptographic
incentives
are
rewiring,
the
internet
they're
how
filecoin
was
able
to
engender
a
storage,
Network
and
Achieve
one
percent
of
the
world's
storage
in
less
than
two
years.
That's
an
incredible
feat
and
the
reason
they
were
able
to
do
that
is
not
because
they
sat
down
with
a
checkbook
and
raised
a
bunch
of
money.
A
Go
do
that
is
they
built
a
cryptographic
network,
the
network
we
all
work
on
and
what
other
people
came
with
their
storage
and
their
investment
and
they
added
their
capacity
Network
and
that's
what
we
want
to
lean
into,
and
so
what
Saturn
is
just
like
file
coin
and
just
like
other
cryptographic
networks
is
the
trustless
permissionless
network.
At
the
end
of
today,
anyone
around
the
world
will
be
able
to
download
Saturn
software,
run
it
on
their
server
and
join
the
network
without
asking,
for
my
permission,
Patrick's
permission.
A
Anyone
hears
permission,
and
the
other
key
piece
of
this,
of
course,
is
to
engender
growth.
All
of
this
as
well
and
in
addition
to
all
the
other
protocol
Labs
products,
it's
all
open
source,
it's
all
community-led
and
it's
all
open
source.
So
what
is
the
solution?
What
does
Saturn
look
like?
Well
we're
going
to
lean
into
cryptographic,
incentives
and
we're
going
to
lean
into
crypto
economics
to
engender
this
network,
and
we
have
grand
aspirations
for
this
network.
A
A
What
does
that
mean
right
now?
You
have
massive
networks
like
AWS,
Google
cloud
cloud.
Cloudflare
excuse
me
and
they're
the
ones
that
are
in
one
big
Silo,
accusing
these
massive
piles
of
gold,
and
what
we
want
to
do
is
take
that
giant,
pile
and
divvy
it
up
and
distribute
it
across
everybody
for
contributing
their
resources
to
the
network.
So
we
want
to
take
that
pile
of
gold
and
put
it
back
into
everyone's
pocket.
A
So
how
do
we
do
that
with
a
lot
of
different
pieces
of
the
network,
so
we're
going
to
run
through
these
various
pieces
here
quickly?
So
the
first
piece
is
the
retrieval
clients
and
one
of
the
key
insights
we've
had
about
how
to
build
a
retrieval
client
in
a
trustless
network
like
Saturn
is
verification.
It
solves
this
problem.
You
want
to
go
access,
a
picture
of
your
favorite
kitten
and
you
go
to
a
Saturn
node
and
you
say
please
give
me
the
my
favorite
kitten.
The
problem
is
the
owner
of
that
node.
A
Oh
God,
they
hate
cats,
but
they
really
love
dogs,
so
they
send
you
a
picture
of
their
favorite
puppy.
How
do
you
know
that
that
content
you
got
back
is
the
content
you
requested-
and
this
is
one
of
the
key
pieces
to
grow
a
trustless
network
just
like
filecoin.
You
need
to
solve
this
problem,
and
so
the
way
Saturn
solved
is
Problems
by
something
called
incremental
verification,
and
what
this
means
is
a
client
like
a
browser
like
a
game,
app
Etc
as
they
receive
content
from
the
network
from
an
untrusted
node
in
Saturn's
Network.
A
It
incrementally
verifies
that
file
to
say
this
is
the
file
I
requested
when
you
request
a
sid.
You
know
piece
by
piece
by
piece:
beautiful
kitty,
pixel
by
beautiful
kitty,
pixel
by
beautiful
kitty,
pixel
you're
getting
the
file.
You
request
it
back
now
the
next
piece,
these
no's.
What
do
these
nodes
look
like
Saturn's
network
is
bifurcated
in
the
beginning
into
two
pieces.
The
first
piece
is
Big
beefy
servers
and
data
centers.
A
We
call
these
L1
nodes,
and
so
these
are
servers
they're
going
to
be
running
in
Edge
locations
worldwide,
and
the
core
of
this
piece
is
more
or
less
a
little
Docker
container
that
we
wrote
and
you
spin
up
a
Docker
container.
It
runs
the
software,
you
join
Saturn's
Network
and
you
serve
content
to
the
network
and
you
remunerated
in
filecoin
in
return.
A
Now,
to
orchestrate
this
all,
how
do
we
organize
all
this
into
an
actual,
cohesive
Network?
Well,
that's
in
two
pieces.
The
first
is
what
we
call
the
orchestrator,
and
this
is
a
piece
of
software
that
says
how
do
we
match
L1
nodes
and
l2s
together
such
that
they
can
perform
and
serve?
You
know
users
around
the
world.
So,
for
example,
let's
say
you
are
in
Lisbon
as
Jermain
and
you
try
and
access
that
kitty
picture.
We
don't
want
you
to
go
all
the
way
around
the
world
to
download
that
kitty
picture
from
someone
in
California.
A
That's
slow,
so
the
orchestra
helps
ties
all
this
together
and
the
other
piece
is
the
crypto
incentive
piece.
You
know
you
need
to
remunerate
and
reward
nodes
in
the
network
for
fulfilling
requests
and
to
do
that.
The
very
first
version
we
have
is
all
of
these
nodes.
The
l1s,
l2s
and
retrieval
clients
submit
logs,
essentially
saying
I
talk
to
this
other
node
in
the
network
and
here's
what
happened.
We
exchanged
this
amount
of
data
with
these
kind
of
time,
stamps
Etc
and
these
logs
are
fed
into
the
payouts
and
the
earnings.
A
So
these
logs
determine
how
much
you
earn
in
filecoin
as
an
L1
or
as
an
L2
for
retrievals,
the
more
retrievals
you
serve
the
more
content
you
reserve,
the
faster
you
serve
it,
the
more
you
earn
and
the
other
piece
here
with
logs
is
also
anti-fraud.
So
this
is
another
big
piece
in
any
crypto
incentivized
Network
there's
going
to
be
fraud
now,
so
here's
a
a
diagram
of
what
this
kind
of
looks
like.
A
So
if
you
see
all
the
way
on
the
left
side
and
the
little
white
boxes,
which
you
have
a
client
that
could
be
a
browser,
it
could
be
a
native
app
like
steam.
You
know,
if
we're
downloading
games,
it
could
be
a
game
itself
and
what
the
retrieval
client
does
and
the
initial
architecture
is.
It
tries
to
find
the
nearest
L1
node
via
DNS,
so
it's
emits
a
request.
You
know
DNS
request.
It
says:
where
is
the
nearest
L1?
A
It
connects
to
the
nearest
L1
and
again,
an
L1
is
a
server
in
a
Data
Center
and
an
L1
has
a
little
local
cache
and
we'll
check
do
I.
Have
the
kitty
picture
my
cache?
If
so
return
it
immediately?
If
not,
they
cache
missed
to
a
swarm
of
l2s
around
it.
If
you
imagine,
imagine
the
L2
swarm
as
like
a
little
bit
torrent
cluster,
a
BitTorrent
swarm
around
l1s,
so
these
are
many
more
prolific
but
smaller
nodes
that
provide
backup
to
the
L1
nodes
and
when
an
L2
cache
misses.
A
So
if
a
request
comes
in
from
retrieval
client,
the
L1
doesn't
have
it
sends
the
request
to
l2s
l2s.
Don't
have
it,
then
it
falls
back
to
the
filecoin
storage
providers.
It
hosts
that
content
itself.
So
all
these
layer,
layers
of
the
network,
are
caching
layers,
just
like
your
CPU
have
a
level
one
cache
level,
two
cash
level,
three
Cash
Etc
Saturn
has
the
same
caching
infrastructure
and
you
can
see
the
orchestrated
and
the
logs
piece
off
to
the
side
here
now
the
cool
and
exciting
part
we're
going
to
run
through
a
demo.
A
So
all
this
yada
yada's
guys
talking
yeah
yeah.
Let's
see
some,
let's
see
it
so
real
quick
I
want
to
give
a
shout
out
that
before
launch,
we
have
a
bunch
of
initial
launch
Partners
who
have
been
running
l1s
and
we're
about
to
see
that
in
the
demo,
as
these
requests
will
land
on
some
of
their
servers
that
they're
running.
A
So
what
we're
going
to
show
here
in
this
demo
is
an
escalating
performance
improvement
from
ipfs,
so
we're
going
to
request
the
Sid
from
the
ipfs
network
and
then
we're
going
to
request
that
same
Sid
from
the
IPS
Gateway
and
then
we're
going
to
request
that
same
Sid
from
Saturn
and
if
everything
goes
according
to
plan,
my
laptop
won't
catch
fire
and
we'll
see
everything
getting
faster
and
faster
and
faster.
Let's
see
here
so
I
am
SSH
or
about
to
be.
The
SSH
connections
crashed
one
second.
A
So
we're
connecting
to
two
servers
here:
the
top
one
is
in
Sydney
Australia
and
the
bottom
one
is
in
Oregon,
and
so
these
are
VMS
that
we
just
spun
up
around
the
world
in
major
cities
that
we're
going
to
Benchmark
the
latency
of
the
two
networks.
So
we're
going
to
do
now
here
is
we're
going
to
broadcast
my
input
simultaneously
to
both
danger
mode,
and
here
is
the
asset
that
we're
going
to
request.
A
So
this
is
just
a
little
six
megabyte
video
file
of
a
little
well,
you
tell
me
what
that
is
to
be
honest,
but
this
is
the
asset
we're
going
to
request
so
here
with
this
first
command,
we're
going
to
come
here
and
there's
a
little
ipfs
Damon
running
and
these
VMS
is
it
possible
to
see
it
all
up
there?
Is
that
that's
just
gobbly
book?
Isn't
it?
Let
me
see
if
I
can
yeah.
Let
me
see
if
I
can
make
this
bigger.
A
You've
earned
yourself
free,
Portuguese
snacks,
upstairs
okay,
can
you
guys
read
that
more
clearly,
ish,
okay,
perfect?
So
here's
we're
gonna
do
so
what
this
command
does?
You
can
kind
of
read?
It
clears
the
ipfs
cache,
so
we're
not
going
to
pull
the
Sid
from
cache,
and
then
it
is
curling
the
little
local
ipfs
Damon
running
on
this
to
go
talk
to
the
IPS
Network
on
our
behalf
to
go
request
this
Sid.
A
So
here
it
goes
time
spent
the
node
in
Oregon
is
starting
to
download
the
file.
The
node
and
Sydney
started
to
download
the
file
and
then
at
the
bottom.
Here
it's
spitting
out
the
time
to
First
Bite.
So
it
took
1.6
seconds
for
the
ipfs
network
to
respond
with
the
first
byte
of
this
Sid,
and
the
top
file
in
Sydney
is
still
being
retrieved,
we're
going
to
move
on
okay,
so
that
was
the
IPOs
Network
in
times
of
time
to
First
Bite.
Now,
let's
test
the
ipms
Gateway.
A
So
here
we
go
finish
then,
with
a
half
a
second
or
530
millisecond
time
to
First
Bite
up
top
and
it
looks
like
the
bottom
one
hung
there,
we
go
okay,
so
you
can
see
the
ifs.
Gateway
still
has
some
room
for
improvement
and
again
this
is
why
we're
building
Saturn
to
improve
the
speeds
of
these
Networks.
A
So
now,
let's
go
to
the
Big
Kahuna
and
again,
what
this
is
doing
is
this
is
talking
to
Saturn's
Network.
So
this
is
Saturn's
network
has
more
points
of
presence
around
the
world
right
now
and
when
we
launch
in
a
few
minutes
here
as
more
and
more
L1
providers
jump
in
and
provide
capacity,
Network
we're
going
to
grow
tremendously,
3x
4X
the
number
of
points
of
presence
of
any
other
network.
Soon,
that's
our
Hope
by
end
of
year,
so
now,
let's
Benchmark
and
then
boom
done.
So
as
you
can
see.
So
what
just
happened?
A
What
did
we
just
watch?
Why
is
this
neat
and
new?
You
can
now
access
data
on
ipfs
files
on
ipfs
and
retrieve
SIDS,
as
you
just
see,
10
times
faster
in
the
ifs
Gateway
and
the
ipfs
network
itself,
and
this
will
go
live
here
shortly.
So
you
can
see
the
China
first
bite
in
the
bottom
in
Oregon
was
43
milliseconds
and
the
time
first
bite
up
top
was
a
22
milliseconds,
and
so
you
can
yeah.
Thank
you.
Thank
you.
Thank
all.
These
people
worked
hard
yeah.
Thank
you.
A
Thank
you.
So
again
the
reason
Saturn
is
outperforming.
These
other
services
is
because
Saturn's
footprint
is
larger
and
can
grow
larger
because
there's
crypto
incentivized
to
and
this
performance
will
only
improve
from
here
now
Diego.
Here
we
are
so
right
here
right
now
we're
going
to
launch
Saturn
before
this
moment.
A
You
could
go
and
run
Saturn
software
and
download
on
your
server
and
run
it,
but
it
would
join
test
net.
What
Diego
just
smashed
enter
on
was
you
no
longer
need
to
have
kind
of
an
allow,
listed
filecoin
address
to
join
the
production
Network.
So
you?
Yes,
you
everyone
here
in
this
room,
can
now
go
and
we'll
walk
through
as
like
to
set
up
a
node
here
in
a
second,
you
can
go
set
up
a
server,
install
Saturn,
share
bandwidth
and
earn
file
coin.
A
So
so
what
does
that
look
like?
In
the
beginning,
the
hardware
requirements
are
pretty
steep
and
we're
starting
them
steep
for
a
reason.
So
you
need
10
gigabit
per
second
upload,
one
terabyte
SSD
about
32
gigabytes
of
RAM
and
Ford
for
your
server
and
running
Saturn
on
it.
You
can
earn
anywhere
between
zero
to
three
thousand
dollars
a
month
or
about
zero
to
six
hundred
fill
a
month
and
again
how
much
you
earn
depends
on
how
much
you
share
with
the
network.
A
The
more
you
share,
the
more
you
earn
and
as
you'll
hear
from
Amin
and
Marie
in
a
bit.
There
are
other
metrics
here
that
Saturn
cares
about
as
well,
and
it
sense
as
well
so
tying
to
First.
Bite
is
a
huge
one.
The
faster
you
serve
that
content,
the
more
you
earn
the
longer
your
node
is
up.
You
know
and
available
to
the
network,
the
more
you
earn,
the
faster
you
upload
data,
the
more
you
earn.
A
So
all
these
incentives
are
baked
into
the
network
too,
and
this
is
I
guarantee
you,
your
favorite
website
of
2022.
saturn.network.
You
can
go
there
and
figure
out
how
to
register
and
set
up
with
an
L1.
So
now,
let's
take
a
look
at
what
running
an
L1
actually
entails.
So
the
first
thing
you
need
to
do
to
run
an
L1.
Is
you
need
a
falcon
12?
What
how
are
we
going
to
pay
you?
A
We
need
your
farcoin
wallet,
so
you're
going
to
set
up
a
filecoin
wallet
and
you
can
use
a
coinbase
wallet,
a
binance
wallet,
your
own
hosted
wallet.
You
can
do
whatever
you
want.
We
need
a
falcoin
wallet,
then
you
need
a
Linux
server
with
some
of
the
minimum
Hardware
requirements
that
we
specify
here
and
you
can
install
and
run
Saturn
by
Docker
or
by
ansible.
A
We'll
have
more
kind
of
like
chef
and
other
software
setups
before
to
help,
and
then
once
you
get
set
up
and
running
here,
I'll
show
you
some
of
the
things
you
can
take.
A
look
at
we'll
want
to
take
a
look
at
so
there
was
a
little
video,
and
so
once
you
know,
joins
the
network.
The
page
I'm
on
now
is
the
orchestrator.
There's
the
piece
of
software.
A
That
kind
of
takes
a
look
at
the
network
and
you
can
see
all
the
nodes
that
are
online
right
now
and
you
can
see
a
bunch
of
these
cool
gory
stats.
You
can
see
memory,
usage,
CPU,
usage
time
to
First
byte
and
you
can
see
like
the
weights.
You
know
in
DNS
for
how
people
are
routed
to
those
nodes
and
then
we'll
also
running
a
node.
You
have
access
to
a
dashboard
like
this,
where
you
can
see
your
earnings,
the
amount
of
bandwidth.
A
You
serve
a
number
of
retrievals
Etc
and
then
for
this
last
piece
here,
once
you
set
up
with
your
L1
and
running
an
L1
and
monitoring
your
stats
and
checking
out
your
node
as
participating
in
the
network,
you're
paid
out
monthly
to
your
filecoin
wallet
and
again
we'll
hear
more
about
the
payment,
Pipeline
and
anti-fraud,
from
the
ever
sagacious,
incapable,
amine
and
Maria
here
in
a
second
now,
where
do
we
go
from
here?
So
you
just
saw
the
fastest
ipfs
access
content
in
the
history
of
ipfs.
Where
are
we
going
from
here
so
we're
here?
A
The
beginning
and
big
things
always
have
Small
Beginnings,
so
we
just
launched
literally
30
seconds
ago.
Thank
you
Diego
and
up
next
we're
going
to
test
L2
node.
So
again,
these
are
the
desktop
nodes.
We're
going
to
test
those
in
testnet
come
quarter,
one
of
2023
or
March
2023.
We
want
to
scale
the
L1
Network
and
by
scaling
L1
Network,
to
put
things
in
perspective.
We
want
to
have
more
l1s
and
more
kind
of
big
heavy
points
of
presence
than
any
other
CDN
period.
So
right
now
we
got
60
something
points
of
presence
worldwide.
A
To
put
things
in
perspective.
Cloudflare
is
about
240
Amazon
has
about
200
Etc,
so
we
believe
we
can
blow
past
that
we
want
to,
but
that's
kind
of
one
of
the
first
order
level
goals,
Scalia
and
network,
and
then
we
want
to
achieve
sub
second
time
to
First
Bite
for
ipfs
Content.
Globally
and
again,
this
involves
getting
lots
of
nodes
all
over
the
world
and
they
have
the
crypto
incentives
to
grow
there
and
then
come
middle
of
next
year.
A
Now
the
last
little
piece
here
is:
we
are
hiring
our
oh,
our
hungry
band
of
Misfits.
So
we're.
If
you
know
anyone
would
be
interested
in
being
a
product
manager,
full
stack,
engineer
or
senior
distributed
systems.
Engineer
please
find
me
afterwards.
I'd
love
to
speak
with
you
and
then
I
want
to
give
a
big
shout
out
here
at
the
end,
it'll
be
Equitable,
we'll
do
a
little
random
sort
of
everybody,
but
I'm
going
to
enumerate
every
one
of
you.
You
guys
know
who
you
are
so
huge.
A
B
Great
talk
any
thoughts
on
any
thoughts
on
what
this
steady
state
might
look
like
for
the
service
providers.
I
mean,
are
you
envisaging
I,
don't
know
60
being
sort
of
professionals
with
their
own
Rigs
and
you
know
I,
don't
know
20
small
guys
in
their
basement
type
of
thing,
any
kind
of
modeling
simulation
along
these
lines,
and
what
does
that
mean
for
a
sort
of
frequent
Clarity
of
data
versus
the
long
tail
yeah.
A
That's
a
great
question,
so
the
first
answer
is:
we
await
with
beta
breath,
to
find
out
how
the
distribution
looks
like
you
know,
will
it
be
a
lot
of
people
who
own
their
own
data
centers
and
have
capacity,
and
they
added
the
Saturn
will
be
people
who
go
out
and
rent
the
servers
and
then
add
those
to
Saturn?
So
we
want
and
we
aim
for
people
to
make
a
profit
running
an
L1.
You
know
in
a
data
center,
obviously,
even
when
they're
renting
one
you
know,
but
some
data
centers
are
more
expensive
than
others.
A
C
Hi
just
curious,
like
10
gigabits,
seems
pretty
steep,
considering
what
terabyte
density
is
cheap,
32
gigabytes
Ram
is
cheap.
10
gigabits
is
quite
steep
compared
to
that.
So
that's
part.
One
is
like
why
why
not
like
one
or
two
point,
five
gigabytes
or
something
and
second,
why
do
a
differentiation
between
l1s
and
l2s
if
they
could
be
joined
as
just
l1s,
then
you
probably
wouldn't
need
really
steep
bandwidth.
A
Great
question
the
short
answer:
to
answer
your
first
one
about:
why
is
10
gigabits?
A
second
is
a
lot
of
Home
residential
internet
connections
now
get
up
to
a
gigabit
and
stretch
up
to
a
gigabit
and
imagine
you're
an
L1
with
a
Max
upload
capacity
of
a
gigabit,
a
second
and
then
you're,
sending
a
big
data
file
from
one
L1
to
a
user
who's
slurping
that
down
I
gave
it
a
second.
It
becomes
very
easy
for
one
or
a
few
uses
on
fast
download
connections
to
monopolize
1l1.
A
So
we
want
to
avoid
that
as
much
as
possible,
and
the
other
piece
of
this
is
is
we'd
rather
start
high,
with
higher
Hardware
constraints
and
lower
it
over
time
than
vice
versa.
If
we
start
low
and
find
out
that
just
the
performance
is
bad
because
l1s
don't
have
enough
upload
capacity,
it's
a
terrible
experience
to
kind
of
go
to
users
who
set
up
their
server.
Like
sorry,
you're,
not
really
tall
enough
to
ride
the
roller
coaster
anymore.
A
You
already
got
on
you
already
put
the
safety
harness
on,
but
halfway
through,
so
you
got
to
get
off,
you
know.
So
that's
not
a
good
experience,
so
we
were
going
to
start
high
and
then
we're
going
to
see,
but
we're
acutely
aware
that
that
requirement
varies
differently.
You
know
they're
across
the
variegated
regions
around
the
world
like
India,
and
you
know,
Asia
APAC
like
that
time
will
be
in
a
second
is
much
harder
requirement
to
satisfy
than
like
North,
America
and
Europe.
So
we're
acutely
aware
of
that.
A
C
A
Question
so
the
reason
even
those
two
aren't
combined
is
we
want
to
be
able
to
accept
any
capacity
to
the
network
that
we
can.
So
if
you
had
to
combine
the
two
you'd,
essentially,
that
would
happen
in
a
data
center,
but
we
want
to
let
people
at
their
home
networks
contribute
to
the
network
and,
you
know
be
part
of
the
file
coin
economy.
A
So
we
made
these
decisions
to
kind
of
break
apart
the
layers
of
Saturn
such
that
we
can
onboard
hardware
and
data
centers
and
take
full
advantage
of
that
and
onboard
Hardware
in
people's
houses
and
take
advantage
of
that,
and
we
want
to
cover
everything
in
between.
Does
that
make
sense?
Does
that
answer
your
question
great
question?
Yeah
Molly.
D
I
have
a
question
about
the
L2
roadmap,
so
you
said
L2
test
net
in
December,
L2
beta
in
March
and
then
going
to
to
GA
later
talk
to
me
about
L2
incentives.
When,
when
does
my
station
know,
that's
currently
served
a
hundred
thousand
Saturn
retrievals
I
believe
at
least
after
I
turned
it
on
last
night.
D
It
please
don't
start
your
machine
glad
to
help,
but
when
does
my
file
coin
count
start
to
go
upward,
do
I
get
to
earn
some
test
net
Phil
or
something
we.
A
Would
like
to
start
handing
out
tests
that
fill
as
soon
as
possible
before
that
paying
many
many
people
at
scale
with
filecoin
and
doing
that.
Well,
is
a
hard
problem,
and
so
we
not
sure
how
and
when
we're
going
to
start
paying
real
people
refill.
Yet
that's
probably
the
biggest
question
we
need
to
answer
answer
really
really
well,
because
money
is
changing
hands
before
launch.
So
it's
not
so
much
technically,
like
the
L2,
doesn't
work
and
can't
contribute
the
network
and
can't
with
no
requests.
That's
groovy
gravy.
A
What
we
need
to
figure
out
is
how
to
pay
you
for
that
groovy
grade,
and
so
that
is
one
of
the
last
big
kind
of
open
unsolved
questions.
Now
that
we're
actually
looking
into
so
I,
unfortunately,
don't
have
a
more
concrete
answer
when,
but
you
raise
a
great
Point
like
getting
tests,
fill
flowing
around
the
test
Network
for
this
stuff.
That's
a
next
natural
step.
D
E
A
A
You
know
so
the
ipfs
Gateway
is
already
serving
three
petabytes
of
bandwidth
every
month
and
as
we
just
saw
in
you
know,
the
demo
Saturn
is
much
much
much
faster
and
so
the
very
first
set
of
customers
are
going
to
be
meeting
those
people
where
we're
already
proven
demand
for
Content
off
the
iPhone's
Network
and
helping
them
and
figuring
out
like
what
is
worth
paying
for.
How
do
we
accelerate?
What
do
those
apps
look
like
what
kind
of
websites
things
like
that
and
then
beyond
that
part
of
it?
A
Is
we
want
to
build
a
faster
CDN,
it's
for
web
3
and
see
what
kind
of
things
do
people
want
to
build?
You
know.
Is
it
the
web
3
YouTubes
that
are
waiting
to
build
web3
YouTube,
because
it's
just
been
too
fast
and
just
plainly
not
possible
to
build
something
like
that,
because
Saturn
wasn't
a
big
piece
of
that.
So
we
have
a
place
to
start
and
that's
where
we're
going
to
start,
but
then
we're
going
to
try
and
figure
out
use
cases.
After
that,
you
know
we
have
other
ideas
as
well.
A
So,
for
example
like
when
nodes
join
a
blockchain,
you
know
there
are
lots
of
snapshots
and
these
are
pretty
big
and
heavy
to
retrieve
and
so,
like.
Maybe
that's
another
use
case,
but
so
we're
going
to
start
with.
What's
proven
where
the
proven
demand
is
and
then
really
figure
that
after
that
and
talk
with
people
say
what
do
they
want
to
build
the
most
on
web3,
where
they
just
can't
until
today,.
F
A
Yeah
so
the
way
incremental
verification
Works,
what
we've
built
so
far
and
we'll
build
client
libraries
for
you,
know,
native
apps,
and
we
built
one
for
the
browser
already.
The
way
it
works
in
the
browser
is
there's
a
new
API
called
the
service
worker
in
the
browser
and
for
those
of
you,
unfamiliar
with
the
service
worker.
If
you
see
a
browser
tab
page,
you
know,
there's
JavaScript
that
runs
in
that
page.
A
It
controls,
you
know
changing
the
colors
of
buttons
and
things
like
that,
but
there's
a
new
API
that
browsers
added
called
the
service
worker,
and
it's
this
like
Throne
that
sits
beneath
the
page.
It's
invisible,
it
can't
control
buttons.
It
can't
you
know
play
sounds,
but
it
has
a
different
superpower.
It
super
powers
and
it
sees
network
network
requests
that
enter
and
leave
the
page
and
it
can
do
processing
on
them.
So
we
leverage
this
service
worker
such
that
web
pages
that
want
to
be
accelerated
with
Saturn
want
to
have
like
one
click.
F
That's
cool
I'm,
wondering
if
then,
the
Saturn
network
has
like
it's
tracking
all
of
that
eventually,
so
that
l1's
l2s
they
will
know
all
these
like
statistics
in
terms
absolutely
yeah
notification.
A
Steps,
absolutely
you
know
you
can't
improve
what
you
can't
measure
and
we're
building
a
performance
product.
You
know
we're
building
the
Ferrari
here,
so
all
these
things
will
be
tracked
and
all
these
things
will
be
measured.
But
to
answer
your
question
about
the
verification,
perfect
great
great
question:
yeah.