►
From YouTube: Opportunities for 2020-2021 -Juan Benet & Pooja Shah
Description
Juan Benet kicks off Day 2 of the IPFS Pinning Summit, describing some of the many opportunities that will exist for companies building Web3 infrastructure in 2020-2021. Pooja Shah also provides context on how Filecoin fits in the picture.
Read more about the inaugural IPFS Pinning Summit on the website: https://ipfspinningsummit.com/
For more information on IPFS
- visit the project website: https://ipfs.io
- or follow IPFS on Twitter: https://twitter.com/IPFS
Sign up to get IPFS news, including releases, ecosystem updates, and community announcements in your inbox, each Tuesday: http://eepurl.com/gL2Pi5
For more information on Filecoin
- visit the project website: https://filecoin.io/
- or follow Filecoin on Twitter: https://twitter.com/Filecoin
Get Filecoin community news and announcements in your inbox, monthly: http://eepurl.com/gbfn1n
A
Thanks
very
much
for
coming
to
the
summit.
Yesterday
we
had
a
fantastic
day
and
we
were
about
to
have
a
another
great
day
today,
so
reflecting
on
on
yesterday,
we
had
a
number
of
fantastic
talks,
really
talking
about
kind
of
the
the
president
of
pinning
services
and
what
groups
are
doing
and
what
they're
thinking
of
doing
and
in
which
directions
to
head
and
what
today's
about.
We
talked
a
bit
about
making
this
about
the
future
and
thinking
about
the
opportunities
that
are
gonna
come
for
services
during
2020
and
potentially
2021.
A
B
A
We
will
hear
more
about
kind
of
how
those
might
be
integrated
into
applications
and
then,
and
then
we'll
also
see
hear
about
potential.
Falcone
integrations
and
power
gain
great,
so
quick
motivation
on
on
why?
Why
there's
interesting
opportunities
at
the
moment?
So
first
off
2020
and
2021,
is
kind
of
a
critical
period.
There
are
lots
of
key
applications
and
projects
being
built
and
interest,
improving
the
ecosystem
as
a
whole.
So
this
this
means
all
kinds
of
developer.
Tools
are
arriving
that
are
they're,
transforming
how
applications
and
users
of
ipfs
end
up
using
the
system.
A
So
so
this
means
that
in
this
time
period
as
new
developer
tools
are
being
made
or
new
systems
that
are
built
either
under
ipfs
or
over
activists,
or
things
like
Liberty,
improving
or
things
like
Falcone
and
so
on.
All
of
these
developer
tools
and
systems
coming
are
changing
the
nature
of
how
people
store
data
and
move
data
without
DFS,
and
so
it's
a
it's
pretty
interesting
time
for
infrastructure
providers
in
in
the
ffs
ecosystem,
because
it
means
that
there's
there's
more
opportunities.
A
The
product
landscape
is
changing
and
there's
there's
suddenly
things
that
weren't
possible
before
are
now
possible
or
suddenly
new
new
abilities
are
unlocked
for
for
infrastructure
providers,
and
so
this
is,
you
can
think
of
it,
as
as,
for
example,
noting
the
the
release
of
things
like
buckets
from
textile,
where
now
we
can,
we
can
see
that
those
kinds
of
things
getting
integrated
into
into
other
services.
So
we
heard
the
announcement
yesterday
about
about
integrations
into
into
some
applications,
but
this
is
something
that
many
planning
services
could
do
right.
A
So
many
tech
services
in
script
could
now
offer
replication
replication
over
buckets.
This
was
just
one
example.
There
are
a
lot
of
different
kind
of
superlatives
emerging
in
the
ecosystem,
that'll
improve
how
how
people
use
use
of
EFS.
The
second
reason
is
that
the
ecosystem
is
growing
very
fast
right
now.
A
One
interesting
note
here
is
because
this
is
an
ecosystem
undergoing
a
lot
of
growth
and
because
I
people's
is
still
very,
very
nearly
like
relative
to
the
mainstream
right,
so
we
still
have
not
yet
across
the
chasm
to
the
to
the
broader
and
mainstream
mainstream
world.
There
said
just
a
huge
market
available
for
adoption.
A
The
last
thing
that
I
wanted
to
to
wrap
into
into
today
was
that
I
frequently
get
asked
how
Falcon
will
fit
into
the
picture
and
how
painting
services
can
participate
in
in
that
in
that
ecosystem
and
how
fat,
when
I
profess
will
interrupt,
and
so
on.
So
I
wanted
to
create
an
opportunity
today
to
to
discuss
some
of
that
and
and
wrap
some
of
the
opportunities
that
can
come
out
of
that
along
these
other
ones.
That
are
kind
of
less
specific
great.
A
So
so,
with
that,
let's,
let's
dive
in
I
kind
of
Brooke
Brooke
it
apart
into
three
three
different
sections
number
one
is
use
case.
Specialization
number
two
is
expanding
product
offerings
and
three
three
sparkling
and
I'm
gonna
be
sprinkling
recommendations
along
the
way,
so
I'm
not
gonna,
have
kind
of
like
a
recommendation
section
at
the
end,
I'll
rather
kind
of
be
mentioning
mentioning
recommendations
here
and
there.
A
So,
for
example,
somebody
making
web
3
application,
their
knees
are
going
to
be
very
different
than
somebody
trying
to
to
publish
or
distribute
video,
and
so
thinking
about
those
use
cases
and
thinking
about
how
your
IP
is
or
what
your
performance
has
to
change
to
match.
Those
needs
will
help
you
target
like
these
different
different
kinds
of
markets.
A
So,
let's
dive
into
into
some
examples.
So
you
know
we
talked
about
well
what
three
AppData
before
so
this
includes
all
of
the
front
ends
of
applications
or
or
the
actual
user
generated
data
within
them,
so
think
of
in
an
audience
here
as
an
example,
think
of
the
playlists,
the
the
records
about
the
songs,
the
play,
the
views
and
all
that
kind
of
stuff.
A
So
all
of
that
data
is
structured
as
in
a
way
that
you
can
now
replicate
directly
in
a
fake
service,
and
this
is
for
the
most
part,
the
majority
of
the
of
the
pending
service
focuses
around
us
another.
Another
example
of
this
is
pretty
pretty
similar,
but
you
can
think
of
web
3
apps
that
are
primarily
just
a
very
light
front-end
around
some,
some
aetherium
smart
contractors
or
some
other
contract
in
another.
A
The
come
on
path
right
now
is
people
using
a
pinning
service,
which
eventually
kind
of
backs
up
to
to
some
regular
cloud
like
AWS
and
so
on,
and
a
lot
of
these
folks
these
that
creators
would
prefer
to
get
their
their
data
stored
in
a
blockchain,
because
it's
kind
of
critical
to
their
contracts
to
have
the
whole
thing
fully
decentralized
and
so
a
lot
of
the
ecosystem
growth
that
we're
seeing
is
in
the
web
through
space.
So
this
is
kind
of
like
a
well
well-run
use
case.
I
think
we're
covering
this
one.
A
This
one
pretty
well,
there's
probably
still
things
to
improve
there,
but
but
overall
this,
this
is
well
done.
Well,
some
of
the
stuff
that
I
think
we
haven't
focused
on
that
much,
but
I
think
our
present
really
good
opportunities
for
for
finding
services
are
things
like
developer
assets
so
think
of?
How
might
we?
A
How
might
we
store
or
and
replicate
all
kinds
of
developer
assets
for
use
in
in
the
production
of
software
I
and
in
tune
the
pinning
servers
to
be
able
to
back
up
all
of
that
kind
of
stuff?
Today,
people
tend
to
use
github
for
a
number
of
things
and
tend
to
use
some
package
manager,
registries
and
so
on,
but
there's
also
a
lot
of
stuff
that
doesn't
can't
easily
go
on
there
or
we're
using
a
press
in
conjunction
with
those
things
can
can
improve
things.
A
So
we
we've
heard
about
the
example
from
from
kind
of
using
using
a
professor
move
around
containers
and
and
speeding
up
the
distribution
at
data
center.
So
all
of
those
use
cases
are
going
to
need
some
tooling
improvement
and
some
performance
improvements
to
really
flesh
them
out
and-
and
this
is
where
having
it
for
providers
that
are
focused
in
these
use
cases
will
will
help,
understand
them
very
well
and
be
able
to
help
the
people
building.
Applications
on
this
actually
have
a
really
good
good
experience
and
yeah
we
I'm
gonna,
suggest
also.
A
Things
like
BOTS
are
a
good,
a
good
example
for
if
infrastructure
providers,
where
you
can
run
run
a
bot
that
I
can
will
do
some
some
processing
or
pinning
or
something
and
those
bots
can
be
tightly
integrated
with
with
some
platform
like
github
or
get
lab
or
slack,
or
things
like
that,
and
when
we
think
about
developer
asses
I.
Think
of,
like
all
of
the
various
different
static
assets
that
you
might
not
push
into
into
get
think
of.
A
Think
of
all
the
binaries
and
and
the
bills,
the
intermediate
kind
of
artifacts
that
are
created,
think
of
packages.
Think
of
containers.
Think
of
VMs,
all
of
that
kind
of
stuff
and
being
able
to
distribute
it
with
with
ipfs,
would
be
really
really
sweet,
but
it,
but
again
it
doesn't
need
some
some
kind
of
infrastructure
tooling
to
improve.
Then
there's
things
like
thinking
about
cncd
pipelines,
where
all
of
the
logs
of
as
he
sed
service
could
be,
could
be
move,
move
this
way
or
archived
in
ipfs.
A
And
then
there
are
things
like
integrations
with
with
larger
data
tools.
So
a
lot
of
software
and
sub
involving
large
data
pipelines
and,
in
those
cases,
a
lot
of
the
traditional
tools
or
things
like
it,
just
don't
work,
and
so
you
would
like
to
be
able
to
version
version
control
a
lot
of
a
lot
of
these
larger
pieces
of
data
and
back
them
up
somewhere.
A
But
but
that
kind
of
tooling
is
not
not
very
well
fleshed
out.
Things
like
get
all
of
fats
and
others
exists,
but
but
they
they
don't
fit.
A
bunch
of
those
use.
Cases
that
well
alright.
Moving
on
to
video,
so
a
pinning
service
oriented
towards
towards
video
would
would
probably
switch
would
really
require
a
different
set
of
performance
guarantees,
so
you'll
probably
have
much
higher
bandwidth.
A
Usage
and
we'd
have
to
focus
on
very
low
latency
delivery
of
the
first
few
first
few
bites
and
then
kind
of
after
that,
just
making
sure
that
there's
kind
of
like
a
reliable
treatment
and
and
so
on,
and
that
might
might
just
look
a
bit
different
than
than
some
of
the
other
use
cases.
Also.
This
might
just
be
a
lot
of
data,
so
most
of
the
pinning
services
that
that
that
I
know
about
are
not
yet
storing
large
quantities
of
video
and
the
moment
that
an
application
is
considering
storing
large
quantities
of
video.
A
Then
price
becomes
a
really
big
deal,
and
so
it's
a
possibility
there.
That
focusing
on
the
this
use
case
would
require
kind
of
thinking
about
the
cost
model
and
different
way.
Streaming
is
a
particularly
interesting
use
case.
We're
thinking
of
infrastructure
for
Strings,
it's
a
whole
different,
different
ball
as
well,
when
think
about
live
video
and
transcoding,
and
and
storing
all
these
informated
out
of
artifacts
and
serving
them.
That
might
be
interesting,
but
also
it
could
be
relaying.
A
Then
there
there
are
things
like
large
datasets
things
like
public
datasets.
You
know
large
enterprise
data
and
so
on.
A
lot
of
these
use
cases
look
very
similar
and
they're
they're,
not
they're,
not
that
hard
to
do
with
the
tooling
that
is
already
has.
But
there
there
is
kind
of
a
sophisticated
set
of
constraints
for
when
for
the
different
kinds
of
domains,
so,
for
example,
machine
learning,
Orient's
around
certain
kinds
of
datasets
with
certain
kinds
of
sizes
and
so
on,
which
might
be
very
different
from
biology.
A
A
One
of
the
things
that
I've
been
thinking
about
alongside
this
is
that
perhaps
a
really
really
good
way
to
to
connect
the
work
that
were
that
we're
doing
in
FS,
with
all
the
archive
stuff
and
in
I
thought
going
with
a
with
Falcone
discover
which,
which
launched
yesterday
you'll
get
to
there's
a
lighting
talk
video
today
about
it
would
be
to
to
just
create
bounties
for
finding
really
good
valuable
data
sets
that
people
want
on
on
ipfs
and
then
just
funding.
The
as
a
bounty
funding
the
the
backing
up
of
it
to
a
pinning
service.
A
So
imagine
we
figure
out
some
some
dates,
as
we
figure
out
what
it
would
cost
to
pin
them
on
on
on
one
of
the
pinning
services,
and
then
we
just
put
a
bounty
out
for
for
actually
processing
that
data
set
and
putting
it
into
the
pinning
service
for
for
say,
six
months
or
twelve
Masterson
that
and
yeah,
so
I'm
very
likely
gonna
do
this.
So
so,
if
you're
interested
in
this
kind
of
problem
or
you've
been
wanting
to
kind
of
hunt
hunt
for
some
days,
that's
an
ad
them
typing
past.
A
We
talked
yesterday
a
bit
about
games
in
game
distribution
platforms.
This
is
another
use
case
that
I
think
is,
is
kind
of
ripe,
there's
still
a
lot
of
tooling
that
the
money
to
get
built
built
out
for
it
for
different
components,
but
we're
now
at
the
point
where
we're
distributing
game
assets
could
be
done
much
faster,
much
faster
with
that,
because,
especially
thinking
about
all
the
different
kind
of
bandwidth
utilization
that
you
can
get
in
kind
of
the
the
same
kind
of
peer-to-peer,
Victorian
sort
of
way.
A
You
know
in
the
old
days
Blizzard
used
to
use
by
trying
to
distribute,
patches
and
and
large
large
software
updates,
I'm,
not
sure
if
they
still
do,
but
it
certainly
is
extremely
difficult
to
use
a
lot
of
that
tooling
if
you're
not
familiar
with
it.
Something
like
I
professed
gonna
make
this
dramatically
easier
and
better,
but
this
is
going
to
require
some
some
thinking
about
the
whole
distribution
of
flow.
A
Thank
you
think
of
a
bit
of
a
bit
of
it
like
a
package
manager,
but
the
if
people
are
interested
in
games
and
I
think
fess,
and
this
might
be
like
a
very
interesting
direction
to
head,
because
these
kinds
of
use
case
has
generated
a
huge
amount
of
data
and
so
forth
for
a
pinning
service
that
kind
of
specializes
in
this
direction.
They
might
find
like
a
very
large
large
market
there.
It
might
require
a
large
list
in
terms
of
developer
tooling,
but
this
would
be
potentially
really
interesting.
A
There
are
a
number
of
companies
around
the
block
chain,
space
already,
building
very
high
quality
games
and
and
I
think
that
they,
they
are
still
kind
of
building
a
normal
web
2
oriented
pathways
because
they
don't
know
that
they
can
rely
on
one
kind
of
infrastructure
providers
to
sort
out
their
data
needs.
So
this
this
is
I,
think
it's
a
it's
a
pretty
interesting
opportunity,
I
think
potentially
to
very
good
use
case.
A
Use
cases
to
to
start
with
would
be
decentraland
and
crypto
voxels
if,
if
a
pending
service
wants
to
go
and
backup
all
of
their
data
and
kind
of
do
a
bit
of
the
work
with
them
to
to
figure
out
how
to
structure
it,
how
to
back
it
up
how
to
pin
it
and
so
on,
then
then
it
would
start
proving
out
the
use
case
and
then
maybe
some
blog
posts.
After
that,
a
lot
more
games
and
I
start
doing
this.
A
There
are
a
number
of
browsers
that
are
that
are
starting
to
add
IPO
support,
but
in
some
cases,
especially
in
mobile
they're,
opting
to
start
with
using
a
gateway
for
for
resolution.
Instead
of
putting
a
note
in
the
browser
itself,
this
probably
for
resource
requirements
and
so
on,
which
makes
it
a
lot
of
sense.
In
this
case,
we've
we've
heard
that
it
would
be
extremely
useful
to
be
able
to
have
infrastructure
providers
that
can
just
run
these
gateways
for
them
and
our
experience
in
doing
so
and
can
run
them
with
extremely
high
availability.
A
So
this
is
I
think
an
extremely
useful
opportunity
for
folks
that
are
interested
in
that
kind
of
service.
All
right.
So
that's
the
set
for,
for
the
use
case,
specialization
segment,
there's
of
course,
more
use
cases,
but
this
is
just
kind
of
a
flavor
of
how
a
pending
service
might
might
kind
of
orient
in
one
direction
and
go
after
a
specific
kind
of
sub
market
and
and
and
help
out
there.
A
So
this
might
include
things
like
manage
notes
like
being
able
to
easily
deploy
networks
of
nodes,
doing
all
of
the
different
kind
of
like
multi-region
support.
If
somebody
wants
to
have
nodes
in
a
bunch
of
different
regions
in
the
world,
figuring
out
all
the
metrics
and
then
Analytics
to
be
able
to
present
all
that
stuff
to
two
users
really
well
similar
to
this
being
able
to
deploy
private
networks
where
people
want
to
maintain
a
completely
separate
network
and
run
an
application
in
a
separate
way.
A
A
A
We
don't
really
traffic
by
default
inside
of
ifs,
because
that
would
be
like
a
pretty
bad
bandwidth
usage
for
for
users,
but
this
is
something
that
could
become
a
service
right,
and
so
we
could
totally
see
services
springing
up
where
I
used
or
I
profess
hires
the
traffic
really
as
a
service
and
just
moves
all
of
their
traffic
through
and
kind
of
exposes
a
public.
A
public
address.
A
So
much
that
there's
some
kind
of
a
need
for
better
statin
analytics
around
how
the
network
is
operating.
So
if
somebody's
has
a
lot
of
content
and
they
want
it
to
be
to
be
rideable,
actually
understanding
how
well
distributed
over
the
network,
it
is
and
and
so
on,
isn't
useful
set
of
features.
So
parties
that
have
a
lot
of
data
and
want
to
make
a
lot
of
it
available
might
opt
to
you
to
use
services
that
provide
really
high
quality
experience.
A
Here
then,
there's
also
think
thinking
about
developer
tooling
for
ipfs
developers,
specifically,
which
means
think
of
CI
a
and
C
D
pipelines
in
a
peer-to-peer
the
deal
with
all
the
peer-to-peer
stuff
that
might
be
different.
So
things
like
configuring,
the
network
in
a
specific
way,
either
connecting
to
private
networks
that
are
isolated
or
connecting
to
the
public
network,
run,
run
tests
being
able
to
have
like
a
we
just
released
test
ground
as
a
project,
and
that
project
is
this
large-scale
peer-to-peer,
lab
kind
of
kind
of
software.
A
A
lot
of
groups
have
mentioned
that
they
would
be
interested
in
running
it
and
using
it.
This
is
an
opportunity
for
an
infrastructure
provider
that
would
just
like
to
wrap
this
up
in
kind
of
like
a
one-click
deploy
for
for
a
whole
set
of
test.
Suites,
that's
some
stuff
that
we,
we
don't
have
battles
to
to
work
on,
but
it's
really
available
and
that's
run
its
wholly
open
sources
or
anybody
can
can
can
do
this.
A
I
think
that's
that's
kind
of
a
small
business
like
Travis
setting
there
for
somebody
to
create
I
think
everywhere
across
the
stack,
a
stronger
focus
and
an
analytics
and
metrics,
and
visualizing
of
the
of
this
would
be
extremely
useful
as
a
differentiator.
A
lot
of
people
want
to
be
able
to
to
know
what's
going
on
in
their
nodes
and
especially
in
some
cases
being
able
to
get
alerts
when
certain
kinds
of
things
happen.
A
So
certain
kind
of
Swift
conditions
on
the
network
causing
some
kind
of
behavior
people
would
like
to
be
able
to
to
be
alerted
on
those
those
cases
and
then
then,
of
course,
there's
there's
all
kinds
of
data
processing
that
people
want
to
do
before
pinning.
So
this
they
start
looking
into
looking
like
how
important
kinds
of
files
or
or
data
sets
integrations
to
to
other
data
processing
tools
and
systems.
A
A
Being
able
to
do
do
to
do
the
kind
of
computation
that
so
there's
a
way
of
using
I
profess.
Were
you
can
start
thinking
of
the
code
of
the
data,
ask
code
and
think
of
running
transformations
over
over
the
ability,
graphs,
so
being
able
to
have
a
a
mo
model
where
you
could
do
things
like
MapReduce
over
at
confess
or
for
things
of
that
nature?
A
That
might
be
a
really
interesting
platform,
and
so
that's
how
it
starts
moving
away
from
from
finding
services
and
starts
going
into
kind
of
kind
of
lambda
style,
computing
and
so
on.
But
it
might
be
like
a
pretty
interesting,
interesting
service
and
related
to
that.
There's
there's
a
ton
of
need
lately
in
the
machine
learning
community
for
for
data
processing
pipelines,
especially
once
I
take
into
account.
A
You
know
artifact
memorization,
where,
where
you're
doing
these
super
expensive
computations
and
you
would
like
to
not
rerun
them
again,
if
you
can
and
continue
addressing
gives
you
the
ability
ability
to
do
this.
So
so
there's
some
there's
a
few
services
out
there
called
that
are
trying
to
do
this.
I
BFS
could
be
an
extremely
extremely
useful
addition
to
this,
and-
and
so
that's
something
I
hope
any
service.
Might
my
optical.
A
Think
textile
and
others
have
have
made
these
being
able
to
kind
of
have
that
that,
as
a
service
is
something
that
that
some
groups
might
might
want-
and
we
talked
about
collaborative
pins
a
bit
yesterday.
But
it
strikes
me
that
this
is
one.
This
is
one
thing
that
that
a
lot
of
groups
in
mainstream
data
science
would
really
care
about.
There's
there's
a
lot
of
very
large
datasets
where
people
would
love
to
be
able
to
share
the
costs
of
storing
and
distributing
these
because
they
don't
in
a
lot
of
cases.
A
Just
kind
of
having
community
funded
registries
of
data
would
be
would
be
an
extremely
useful,
usual
thing
to
do,
and
so
this
starts
veering
away
from
from
from
just
broad
pinning
and
starts
turning
into
kind
of
like
a
github
for
for
data.
This
is
something
that
the
number
of
groups
have
have
talked
about
over
the
years
as
wanting,
but
what's
really
missing
is
a
group
that
can
shoulder
like
the
very
large
quantities
of
data
involved.
B
Awesome
so
hi
everyone
I'm,
really
excited
to
be
here.
Talking
about
final
coin
at
the
ipfs
think
something
I
think
there's
some
really
awesome.
Integration
opportunities
and
so
I
wanted
to
start
a
little
bit
by
setting
the
stage
for
what
sorts
of
use
cases
that
file
coin
is
aiming
to
solve,
and
then
one
I
think
is
going
to
tie
together
and
see
how
violent
fits
into
the
larger
ecosystem
and
then
we'll
dive
into
some
more
like
concrete
opportunities
for
how
to.
B
I'm
not
sure
how
many
of
you
have
been
following
the
Falcon
roadmap
very
closely,
but
we're
about
to
launch
me
met
in
just
a
couple
months
and
already
the
file
Queen
ecosystem
is
vibrant
and
thriving.
There
are
tons
of
opportunities
across
the
stack
at
the
research
layer.
There
are
opportunities
for
application
developers.
Service
providers
is
just
probably
what's
most
relevant
to
our
conversation
today
and
really
for
anyone
who
cares
about
efficiently
priced
geographically
distributed
storage.
B
So
I'd
like
to
start
by
talking
about
why
file
coin
exists.
What
problems
are
we
trying
to
solve
and
which
use
cases
do
we
want
to
enable
I'd
like
to
focus
in
especially
on
some
of
the
core
features
and
file
point
that
make
these
use
cases
even
stronger
and
then
I'll
touch
on
where
we
are
as
a
project
community
and
ecosystem
and
then
hand
it
back
to
one
file.
B
While
coin
aims
to
be
this
resilient
and
trustworthy
foundation
for
all
of
humanity's
knowledge
and
we're
getting
there
by
solving
a
few
key
problems
that
exist
with
centralized
storage
infrastructure
today.
So
first
file
coin
is
a
decentralized
cloud
file.
Qin
creates
a
marketplace
where
storage,
clients
and
providers
can
interact
directly
and
thus
file.
Point
eliminates
the
reliance
on
reliance
on
a
few
centralized
cloud
storage
providers
and
instead
creates
a
decentralized
cloud
file.
B
Coin
users
retain
full
control
over
their
data,
so
you
get
to
make
your
own
decisions
on
what
happens
to
the
data
that
you
store
on
the
file
coin
network
file
coins
in
efficient
storage
market
users
on
the
network,
get
to
set
the
prices
that
they're
willing
to
pay
and
accept
for
storage.
So,
instead
of
having
to
pay
the
prices
that
are
set
by
corporate
pricing
depart
users
have
the
freedom
to
enter
into
economic
contracts
that
are
most
affordable
for
them.
One
thing:
that's
really
unique
about
file
coin.
B
Is
this
feature
of
verifiable
storage,
so
file
Queen
requires
that
storage
providers
repeatedly
generate
cryptographic
proofs
of
storage
to
verify
that
when
they
say
that
they're
storing
your
data,
they
actually
are
these
verifiable
tamper.
Proof
cryptographic
proofs,
enable
file
point
to
actually
work
as
a
decentralized
protocol.
That
also
has
data
storage
guarantees.
So
you
no
longer
need
to
trust
an
intermediary
organization
to
know
that
your
data
is
being
stored
safely.
B
That's
a
core
feature
that
is
enabled
by
the
protocol
and,
lastly,
file
Queen
puts
the
Zeta
bytes
of
idle
storage
capacity
that
exists
in
the
world
to
work
there.
There
are
massive
amounts
of
storage
capacity
that
are
sitting
around
unused
in
data
centers,
basements
and
hard
drives
all
over
the
world,
and
this
amount
of
data
is
growing
every
single
day
file,
Queens
a
marketplace
that
allows
this
storage
capacity
to
be
hired
by
customers,
creating
efficient
use
for
an
otherwise
underutilized
asset.
B
For
the
last
year,
we've
been
working
with
usc's
Shoah
foundation
to
store
the
world's
largest
genocide,
testimonial
archive
on
the
file
coin
network.
There
are
people
in
the
world
today
who
deny
that
certain
atrocities
and
genocides
like
the
Holocaust
were
ever
committed.
The
show
of
foundation's
mission
is
to
develop
empathy,
understanding
and
respect
by
sharing
these
stories
of
testimony.
B
More
broadly,
the
visual
History
Archive,
as
it's
called,
includes
testimony
from
more
than
55,000
survivors
across
nine
genocides
in
62
countries,
and
it's
more
than
nine
petabytes
worth
of
recordings
that
will
ultimately
be
stored
on
the
file
coin
mean
that
some
of
this
data
is
all
being
stored
on
a
file
quintet
set.
The
show
Foundation
chose
to
work
with
file
point
because
of
our
use
of
these
tamper-proof
cryptographic.
Proofs
of
storage
file.
Queen's
proofs
allow
users
to
trust
that
the
videos
that
they're
viewing
haven't
been
tampered
with
that
they're
the
original
truth.
B
Another
example
comes
from
our
work,
collaborating
with
a
number
of
scientific
and
cultural
institutions
to
store
valuable
massive
datasets
on
file
coin.
It's
kind
of
staggering,
but
I
recently
learned
that
we
store
we
generate
more
than
two-and-a-half
quintillion
bytes
of
data
every
single
day
for
context.
This
is
something
like
more
than
2000
shipping
containers
full
of
870
megabytes
cd-roms.
B
It
includes
294
billion
emails,
64
billion
whatsapp
messages,
500
million
tweets,
but
it
also
includes
useful
scientific
data
such
as
climate
science,
reports,
disease
tracking
maps,
connected
card
coordinates
and
more
while
much
of
this
data
is
stored
and
preserved.
It's
also
the
case
that
millions
of
gigabytes
worth
of
useful
scientific
data
and
other
types
of
data
are
routinely
discarded
every
day
and
if
I
look
when
we
believe
that
this
is
a
lost
opportunity
for
our
society
if
stored
and
analyzed,
these
data
could
yield
really
precious.
You
know
and
priceless
research
and
discoveries.
B
We
won't
file
point
to
be
the
home
for
this
sort
of
important
information.
So
over
the
last
few
months-
and
one
alluded
to
this
earlier-
we've
been
working
with
several
institutions
to
preserve
open
access
data,
we've
already
begun,
onboarding
petabytes
worth
of
data
onto
the
file
point
test
set,
and
we
have
a
pipeline
for
hundreds
of
petabytes
of
data
to
follow
on
the
file,
claiming
that
we
just
announced
this
project
called
file.
Queen
discovery
yesterday
on
the
file
claim
blog
and
we'll
have
someone
doing
a
lightning
talk
on
this
project
as
well
later
today.
B
The
last
example
that
I
want
to
highlight
comes
from
storing
web
3,
app
and
user
data
on
file
coin
and
again
one
touched
on
this
a
little
bit
earlier,
but
I
wanted
to
talk
specifically
about
the
file
point
angle
in
this
use
case.
So
file
coin
we
think,
is
an
especially
good
a
solution
for
DAP
projects
in
the
ipfs
ecosystem
and,
more
broadly,
most
about
three
applications.
Today,
split
their
storage
across
a
number
of
solutions.
For
example,
web
3
photo
sharing
applications
and
I've
grabbed
an
example
from
textile
photos.
B
Here
they
stored
their
business
logic
on
networks
like
aetherium,
their
front-end
applications
on
IP
FS
and
their
user
data,
either
an
IP
FS
or
a
web
to
service
provider
like
Amazon
Web
Services.
The
idea
is
that,
with
file
coin,
these
types
of
applications
can
store
all
of
the
relevant
data
directly
on
the
file
coin
network:
a
crypto
native
web
3
solution,
direct
bridges
from
smart
contract
platforms
to
file
pointer
also
already
in
the
works,
and
this
should
make
that
data
storage
on
file
point
even
more
seamless
in
the
near
term.
B
Seeing
this
type
of
real-world
impact
of
the
file
coin
project
is
especially
mode
8
motivating
for
me
and
I
personally,
just
so
excited
to
find
more
of
these
use
cases
and
see
ways
that
file
point
can
uniquely
serve
them,
but
even
with
some
of
these
early
use
cases,
I
can
already
start
to
see
this
future
or
a
file
point
is
really
gonna
change.
The
way
that
we
interact
with
data
just
day-to-day
the
file
point
community
is
kind
of
in
a
Cambrian
explosion
period.
B
Right
now,
we've
finally
got
all
of
the
core
protocol
and
research
development
work
done
and
we're
starting
to
see
real
businesses
and
applications
and
powerful
user
experiences
being
built
on
the
network.
So
this
is
the
next
part
of
the
discussion
that
I'd
like
to
get
to
where
where's
five-point
today
as
a
project
community
and
ecosystem.
B
As
I
mentioned
earlier,
the
core
tools
and
building
blocks
are
now
in
place
and
so
we're
starting
to
see
lots
of
activity
with
developers,
miners
and
storage
clients
or
other
types
of
users.
The
file
cream
project
is
active
development
across
200
repos,
including
node
implementations,
file,
kind,
proofs,
developer,
tooling
and
AP
is,
and
also
many
community
products
and
applications
since
launching
our
first
development
networks.
In
early
2019,
we've
seen
our
open
source
developer
community
grow
from
around
20
active
contributors
to
more
than
1,200
total
cumulative
contributors.
B
B
At
the
web
3
summit
last
year
we
launched
the
file
coin,
dev
Grants
Program,
to
provide
some
light
ecosystem
support
to
communities,
two
teams
that
were
building
applications
and
tooling
on
file
coin,
and
today
we've
run
three
waves
of
the
Deborah's
program,
awarding
around
forty
five
grants
and
total
one
and
a
half
million
dollars
to
teams
within
the
ecosystem.
We're
currently
accepting
applications
actually
for
dev
grants
wave
four
and
you
can
learn
more
about
the
program
on
the
web
site.
B
We
launched
the
first
feature:
complete
file
coin
test
net
in
December
2019,
and
this
has
been
really
exciting
since,
since
you
know
just
a
few
months
ago,
we've
seen
this
network
grow
from
zero
petabytes
to
almost
five
petabytes
of
storage
on
the
network
and
there's
almost
35
but
petabytes
a
storage.
That's
waiting
to
join
the
network.
B
There
are
also
hundreds
of
miners
that
are
actively
participating
every
day
on
the
file
coin
test
net,
and
this
is
really
interesting
because
you
know
these
are
serious
organizations
with
significant
funding
and
resources
they're
whole.
You
know,
businesses
that
are
already
starting
to
be
built
on
the
mining
side
on
file
coin,
and
these
teams
are
really
committing
hard,
so
we've
seen
actually
in
total,
something
like
I,
think
thousands
of
active
miners
that
have
participated
on
the
test
and
over
the
last
few
months.
This
is
not
an
incentivized
Network.
B
So
all
of
this
energy
leads
us
to
have
strong
expectations
that
the
network
is
going
to
grow
by
several
orders
of
magnitude
in
size.
When
we
actually
launched
the
main
net
in
general,
we've
seen
a
tremendous
amount
of
enthusiasm
within
the
velcroing
mining
community.
Some
miners
have
started
to
announce
on
you
know,
places
like
Twitter
pretty
publicly
that
they
have
exabytes
of
storage
capacity
that
are
waiting
to
join
form.
B
A
net
launch
file
coin
meetups
continue
to
take
place
around
the
worlds
weekly,
and
there
are
also
miners
that
have
posted
videos
of
racks
of
storage
space
that
are
ready
to
go
for
a
file
coin.
Lunch
there's
a
video
here,
I'll
skip
playing
it
today,
but
we
can
share
out
the
slides
and
it's
like
a
really
fun
video
to
watch
as
someone
who
works
more
in
the
product
side.
B
I'm,
currently
most
excited
by
all
of
the
developer,
tooling,
and
applications
that
have
started
to
build
on
the
file
clinical
system
and
building
on
final
coin
is
now
easier
than
ever.
We
have
comprehensive
documentation,
there
are
numerous
API
library,
development-oriented
test
networks,
hosted
nodes
and
all
sorts
of
other
developer
infrastructure
that
are
necessary
to
easily
build
on
file
coin.
Several
teams
have
already
started
using
this
developer
platform
to
build
community
friendly,
tooling.
So
some
examples
of
this
are
block
explorers
by
things
like
IP
FS
force,
ipfs,
Union
and
D
Tech.
B
This
is
a
chain
visualizer
built
by
AE
studio,
and
this
is
kind
of
a
preview
of
a
dashboard.
That's
going
to
be
built
for
mining
nodes
built
by
a
team
called
node
Factory
this.
This
is
a
product,
I'm
actually
really
excited
about,
and
you're
going
to
hear
a
lot
more
about
it.
Today,
Andrew
from
textile
presented
on
threads
yesterday
and
he's
going
to
be
presenting
with
Aaron
from
the
textile
team
on
power
gate
specifically,
but
textile
is
a
team
that
we've
all
heard
about
a
lot.
B
So
this
is
our
attempt
to
lower
that
barrier
to
entry
and
inspire
delight
and
some
of
our
early
file
coin
users
I.
Think
we're
also
going
to
give
lightning
talk
on
this
one
later
today,
but
this
product
is
also
interesting
and
I.
Think
there's
some
some
cool
opportunities
for
paying
services
to
leverage
these.
These
types
of
gooeys,
as
dashboards
too
exposed
to
their
users,
said,
choose
to
use
file.
Coin
storage.
B
All
of
these
products
and
many
more
are
going
to
be
live
within
the
next
few
months
and
just
a
reminder
that
you
know,
while
the
file
open
ecosystem
is
vibrant
and
growing.
This
is
just
the
beginning,
and
especially
within
the
world
of
pinning
services
and
other
infrastructure
that
ties
into
file
coin.if.
So
early
there
are
so
many
opportunities.
This
space
is
super
wide
open
and
we're
really
excited
to
start
working
with
mission
aligned
developers
who
are
looking
to
join
the
ecosystem
and
start
building
now,
and
that's
all
thank.
A
A
You
still
pay
the
same
amount
for
for
storing
in
the
long
term,
but
when
you
pay
a
different
amount
for
retrieval
and
you
end
up
getting
it
in
different,
very
different
time
scales.
So
you
know
as
low
as
a
few
minutes
to
as
high
as
many
hours.
If
three
is
a
lot
more
more
narrow
and
kind
of
most
most
cleaning
services,
today
kind
of
match
the
the
cost
spans
of
the
latency
bands
of
that,
because
that
is
a
being
kind
of
like
the
sweet
spot
of
what
most
cloud
storage
is
doing
right.
A
So
there's
some
kind
of
magical
ideal
that
we
can't
quite
reach,
but
where
everything
is
kind
of
trying
to
go
and
and
everything
is
sort
of
a
trade-off,
you
can,
you
can
either
spend
more
money
and
potentially
go
faster
or
you
can.
You
can
try
to
have
a
more
cost-effective
service
and
go
slower
and,
of
course,
as
advances
in
technology
happen,
we
can
kind
of
chip
away
at
this.
A
At
this
kind
of
frontier
and
get
closer
and
closer
to
the
ideal,
and
so
the
what
I
want
to
represent
here
is:
is
this
interesting
overlap
that
that
are
the
that
I
realized
a
few
a
few
weeks
ago,
which
is
that
currently
right?
Pinning
services
are
perfectly
positioned
to
be
about
the
best
retrieval
miners
in
the
Falco
Network,
and
so
what
we're
two
old
miners
are.
A
Are
this
this
kind
of
class
of
service,
where
the
the
main
role
is
to
keep
content
really
kind
of
hot
meaning
able
to
be
retrieved
very
quickly,
where
kind
of
latency
really
really
matters,
but
not
that
much
content.
So
think
of
you
know
in
the
orders
of
of
a
few
bytes,
which
is
roughly
around
the
size
of
what
most
building
services
are
doing,
and
so
that's
and
so
for
a
lot
of
the
FAQ.
When
use
cases
we're
we're
gonna
be
storing
petabytes
impediments
and
stuff.
A
We
can't
keep
all
of
that
that
that
hot
and
then
there
and
there's
a
kind
of
a
weird
reason
why
storage
here
in
in
this
graph
in
the
in
in
blue,
actually
can't
retrieve
faster
like
what
wide
resource
take.
So
long
to
retrieve-
and
that's
a
that's
a
feature
of
the
of
the
proofs
that
we're
using
we're
just
it
falls
out
of
the
cryptography
where
there's
a
very
expensive
operation
to
do
on
top
of
the
data
to
make
sure
that
the
parties
do
have
it.
A
And
if
you
want
to
be
able
to
get
the
data,
you
have
to
effectively
decrypt
it
out,
and
that
takes
some
time
as
advances
in
the
proofs
happen.
Storage
miners
are
going
to
be
able
to
inch.
You
know
further
and
further
left
it's
still
with,
with
around
the
same
kind
of
cost
costs
other
they
have,
but
ideally
being
able
to
to
provide
a
much
faster
service
once
those
kinds
of
proofs
are
in
place,
but
at
least
for
the
beginning
of
the
network,
and
probably
for
the
next
six
to
twelve
months.
Really
fast.
A
Maybe
maybe
they
won't
get
to
like
the
massive
cost
reductions
that
we
hope
for,
but
it,
but
it
is
possible
and
so
we're
gonna,
see
kind
of
storage.
Miners
inching
inching
towards
the
left
over
the
over
the
coming
years.
At
the
same
time,
for
two
miners,
we
wonder
if
you'll
manage
to
be
moving
towards
closer
and
closer
to
the
user,
to
reach
certain
certain
capabilities
that
modern
CDNs,
don't
have
so
modern
CDNs
kind
of
stop
at
the
at
the
ISP
level.
But
it's
likely
there
were
two
old
miners.
A
The
following
can
be
you
know,
other
people,
you
know
your
neighbors
in
in
storing
data
and
serving
it
to
you
very
quickly
from
either
that
next
house,
over
or
potentially
a
different
apartment,
building
or
a
different
apartment,
the
same
building.
It's
that
kind
of
edge
level
of
of
delivery.
Isn't
really
it
doesn't
really
exist
yet,
and
so
we
hope
that
over
the
next
few
years
we
can.
A
A
So
it's
kind
of
like
an
interesting,
interesting
feature
that
fell
out
of
of
we're
kind
of
pitting
services
are
now
and
where
we
expect
these
these
these
these
groups
to
go.
Of
course,
this
is
extremely
difficult
to
do.
The
lines
are
Amazon
and
Amazon
is
one
of
the
strongest
organizations
to
being
able
to
optimize
things.
So
a
lot
of
this
is
still
like.
A
We'll
see
what
happens
but
kind
of
the
big,
the
bigger
thesis
of
Falcone
is
that
a
market
can
operate
much
better
than
than
a
centrally
planned
centered
plan
economy
and
so
we'll
see
if
the
path
or
miners
are
able
to
get
the
same
kind
of
cost
reductions.
Let's
say
Bitcoin
miners
God
with
with
all
of
the
hardware
installations,
and
they
go
great.
A
So
the
this
is
kind
of
another
way
of
visualizing
is
on
the
same
data
which
is
on
the
left,
is
sort
of
the
diagram
that
Google
cloud
uses
to
describe
its
kind
of
core
data
centers
and
then
like
the
edge
caching
and
so
on,
and
you
can
think
of
think
of
that,
as
like
the
storage
miners
in
the
center
and
then
kind
of
the
ritual
at
the
edges.
I'm
kind
of
very
well
geographically
distributed
and
with
low
latency
delivery
to
the
user.