►
From YouTube: NFT.storage + Web3.storage
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
Hello-
everyone,
I'm
david,
I
think
it's
my
third
time
potentially
introducing
myself.
I
am
the
product
lead
for
two
products
that
protocol
labs
develops
to
make
things
easier
to
use
ipfs
and
powercoin
in
your
applications.
So
I'm
here
to
tell
you
about
them
today
and
for
all
y'all
who
are
hacking
out
there
and
are
interested
in
using
very
simple
to
use
services
to
actually
have
your
the
data
you're
storing
out
on
the
web
be
decentralized,
then
please
do
check
us
out.
A
They
are
sister
storage
products
and
services
on
one
hand
that
provide
the
ease
of
use
and
performance
that
we're
all
kind
of
used
to
these
days
with
traditional
cloud
storage,
the
whole
two
lines
of
code,
high
performance,
high
availability,
that
sort
of
thing,
alongside
being
able
to
use
as
your
primitives
and
how
you
address
data
ipfscids,
as
well
as
decentralized
storage
on
the
filecoin
network,
where
you
don't
actually
have
to
trust
us
that
we're
storing
your
data
for
the
promised
amount
of
time.
A
You
can
always
just
check
the
proofs
out
on
the
powerpoint
blockchain
and
actually,
last
time
I
gave
this
talk.
There
was
an
entire
like
50
minute
session
before
explaining
what
ipfs
and
filecoin
do
at
the
core.
So
it's
a
little
bit
light
on
the
details
here
on
that.
So
I'll
try
to
inject
some
of
that
in.
A
A
So
framed
a
little
bit
differently,
maybe
from
the
user's
point
of
view,
nft,
storage
and
web3.storage
on
the
surface,
give
you
simple
familiar
and
performant
interfaces.
So,
on
the
storage
side,
you
install
your
client
library,
you
can
store
things
real,
quick
things
get
uploaded
to
cloudflare
edge
workers.
So
it's
highly
available
fast.
A
What
ends
up
happening
is
that
folks,
who
are
just
at
a
hackathon
and
interested
in
using
ipfs,
they
end
up
having
to
get
really
really
in
the
weeds
on
stuff
they
have
to,
and
you
know
like
pure
to
pure
things,
don't
always
go
right.
You
know
you
can
have
a
firewall
and
then
more
generally,
you
know
a
lot
of
the
technology
is
in
motion
for
your
specific
use
case.
You
have
to
get
really
deep
into
like
the
data
structures
of
things
on
the
file
coin
side.
A
So
we
try
to
save
users
from
all
that
and
the
interface
in
between
the
ease
of
use
and
what's
going
on
underneath
the
hood
are
cids.
A
It
so
I'll
talk
a
little
bit
more
about
this
in
in
a
second.
But
that's
really
what
differentiates
nft,
storage
and
web
3d
that
storage
as
products
versus
more
traditional
cloud
storage
providers
and
then
once
you're
kind
of
on
the
services
today
they
are
very
simple
and
get
your
data
on
there
and
you
can
build
a
lot
already
on
top
of
that,
but
there's
all
sorts
of
web3
innovations
that
are
coming
along
both
from
the
ipfs
and
filecoin
communities.
A
So
if
you
look
at
the
bottom
here,
there's
all
sorts
of
exciting
stuff
coming
from
filecoin
that
will
increasingly
decentralize
services
like
ourselves,
even
out
of
the
picture
where
you
can
interact
more
directly
with
decentralized
storage
networks,
as
well
as
a
secret
sauce
that
we're,
including
in
our
services,
to
make
it
so
that
you
can
pretty
much
build
anything
that
you
can
today,
but
using
trustless
web3
technologies.
A
I've
kind
of
glossed
over
this
thus
far,
but
nft.storage
and
web3.storage
are
two
different
products
and
they
share
similarities
at
the
service
and
infrastructure
layer,
but
they're
really
two
different
products
with
two
different
goals
and
I'll
I'll
double
click
into
that
a
little
bit
more
and
yeah.
You
can
maybe
read
through
the
bullets
here,
but
I'm
gonna
voice
it
all
over
in
the
following
slides,
so
I'll
start
with
what
they
do
share.
So
with
that
intersection
of
the
venn
diagram.
A
The
goal
for
for
both
these
products
is
to
provide
production-ready,
decentralized
storage,
so
web3
web
3
is
really
in
the
public
eye.
Today,
applications
have
real
users.
These
applications
sometimes
have
tens
and
hundreds
of
billions
of
dollars
invested
in
them
or
valued
in
them,
and
you
can
still
go
to
a
top
nft
marketplace
and
wait
five
minutes
for
all
the
images
to
load
on
there.
A
You
know
we're
in
a
trial
period
and
there
is
more
and
more
attention
and
there's
a
lot
of
really
passionate
community
members
who
are
willing
to
be
patient,
but
the
faster
we
can
get
to
where
we
are
today
with
the
web.
More
generally,
the
faster
that
others
will
be
able
to
take
advantage
of
what
web
3
is
building,
and
you
know
that
in
the
end,
the
the
vast
majority
of
folks
aren't
making
decisions
on
how
decentralized
something
is
or
is
not
there.
A
They
just
want
things
to
work
so
end
users
and
developers
both
demand
from
their
storage
solutions,
specifically
fast
and
reliable
storage.
Cdn
speed,
retrieval
infrastructure
that
can
scale
to
whatever
they
need
and
just
simply
the
ability
to
store
any
sort
of
data.
And
you
know
it
could
be
data
size
or
file
types
or
like
video
and
have
it
formatted
in
a
way
where
you
can
stream
it
and
all
that
stuff.
And
you
know
it
just
got
to
work
in
the
end.
A
So
what
we
do
is
structure
our
infrastructure
to
lean
into
a
lot
of
existing
cloud
infrastructure
done
in
a
way
where
we're
not
locked
into
them,
and
we
don't
have
to
trust
them.
But
taking
advantage
of
the
fact
that
hey
you
know,
a
lot
of
cool
technologies
have
been
built
to
be
able
to
solve
a
lot
of
the
problems
of
performance
and
reliability,
and
that
sort
of
thing.
I
still
think
it's
really
cool
that
you
can
snap
your
fingers,
and
you
know
the
amount
of
storage
that
you're
paying
someone
for
is.
A
Can
elastically
10x
right
like
gone,
are
the
days
of
a
restaurant
having
to
run
their
website
in
the
server
in
the
back
of
back
room
and
they
get
like
half
an
hour
of
some.
You
know
developers
time
a
week
to
like
debug
any
issues
and
they
can
be
down
for
multiple
weeks
as
a
result,
and
things
like
that,
so
we
want
to
lean
in
into
the
innovations
that
exist
to
provide
that
sort
of
service
level
to
our
users.
A
At
the
same
time,
we
can't
just
recreate
aws
or
whatever
we
need
to
maintain
trustlessness
in
the
system,
so
gave
a
quick
overview
of
ipfscids
before,
and
I
would
encourage
you
to
read
more
in
depth
about
it,
but
you
know
where
what
it
comes
down
to
is
as
long
as
you
can
trust.
The
cryptography,
you
don't
have
to
trust
the
services
you're
using
so
with
webfreestorage
and
nft.storage
content.
A
Addressing
and
filecoin
proofs
are
the
star,
so
the
first
two
kind
of
sub
bullets
in
the
second
bullet
up
here
they
reference
what
content
addressing
does
for
you.
So
before
uploading,
any
data
to
nft
storage
and
web2.storage,
you,
you
compute
the
cid
of
that
data
locally.
So
you
take
the
content,
your
compute,
it's
hash
and
you
know
that's
the
hash
of
the
data.
That
is
a
property
of
the
data
itself,
no
matter
who
gives
that
data
to
you
in
the
future
or
whatever.
A
You
can
always
verify
that
and
that's
what
you're
going
to
end
up
referencing
the
data
with
when
you
pull
it
down
from
ipfs,
and
then
I
guess
that
is
the
trustlessness
created
from
like
that
data
layer
of
of
using
our
services,
there's
still
the
infrastructure
layer
where
you
know,
maybe
you
don't
have
to
rely
on
us
or
be
locked
into
us
to
be
able
to
reference
your
data
in
a
universal
way,
but
you're
still
relying
on
us
to
be
up,
and
how
do
you
know
that
your
data
is
safe?
A
Well,
if
we
go
down,
we
do
store
all
the
data
in
filecoin
deals
on
the
filecoin
network,
six
plus
replicas
across
independent
storage
providers-
and
you
can
always
at
any
time
check
the
filecoin
blockchain
to
verify
that
this
data
is
safe,
so
really
quickly.
A
Filecoin
is
a
type
of
ipfs,
but
it
comes
with
some
extra
secret
sauce
where
filecoin
storage
providers
who
say
they're
storing
your
data
have
to
periodically
submit
cryptographic
proofs
validated
by
the
network
and
end
up
on
the
blockchain
that
your
data
is
being
stored
for
a
specific
duration
of
time.
The
exact
data
with
that
exact
storage
provider
so
don't
have
to
trust
us
check
the
filecoin
blockchain.
It's
all
in
the
cryptography
that
your
data
is
safe
and
then
yeah.
A
Once
the
data
is
on
the
network,
you
can
pull
it
down
in
a
multitude
of
ways.
You
just
locally
can
verify
it's
correct
if
you
need
to
make
things
fully
trustless
and
one
cool
thing
about
ipfs.
Another
cool
thing
about
ipfs
is
that
you
can
store
more
copies
of
the
data
wherever
you
like,
you're,
not
locked
into
us
in
any
way
store
the
data
in
as
many
places
as
you
need
to
to
feel
comfortable.
We're
just
able
to
provide
a
simple
onramp
for
you
to
get
the
data
onto
the
network
in
the
first
place.
A
Zooming
in
a
little
bit
more
on
nft.storage
and
how
it
differs
from
web3.store,
so
we
talked
more
about
the
similarities
so
far.
Nft.Storage
the
mission
initially
was
to
really
drive
home
ipfscids
as
the
standard
in
nfts
and
the
way
we
sought
to
do.
This
is
kind
of
like
two-pronged,
so,
first
making
it
super
easy
and
simple
to
be
able
to
generate
ipfscids
for
your
data.
So
as
a
reminder,
nfts
don't
store
all
their
data
on
chain
in
the
nft
itself.
A
They
make
references
to
off-chain
data,
and
it's
really
important
for
that
off-chain
reference
to
be
a
unique
property
of
the
data
itself.
So
no
one
can
ever
argue
with
you
that
that
data
is
actually
the
data
that
your
nft
is
referring
to.
So
making
this
as
easy
as
possible
will
help
make
ipfscids
kind
of
the
primitive
here
and,
at
the
same
time,
the
other
side
of
the
problem
with
nfts
and
and
the
off
chain
data
side
is
that
the
data
actually
has
to
be
available.
A
So
if
you
ask
the
ipfs
network,
hey
here's,
my
cid
give
it
back
to
me.
You
should
be
able
to
get
it
back
and
we
want
to
be
one
of
the
folks
who
are
able
to
provide
a
trustless,
decentralized,
multi-generational,
decentralized
storage
option
for
nfts,
so
by
storing
it
into
filecoin
and
five
coin.
Deals
are
finite,
so
today
we're
renewing
those
deals
manually
and
we
monitor
the
file
coin
chain
to
kind
of
make
sure
that
at
least
x
number
of
copies
are
on
the
data
at
any
point.
A
But
the
plan
in
the
future
is
to
increasingly
decentralize
ourselves
instrument
things
like
smart
contracts,
we'll
do
this
listening
for
you
always
spin
up
new
file.
Coin
storage
deals
to
keep
the
the
number
of
copies
in
the
redundancy
up
to
a
certain
point,
even
as
deals
expire
or
go
offline
and
fund
this
all
through
an
endowment
in
a
dow.
A
So
that's
kind
of
like
the
high
level
vision
of
nft.storage,
and
it's
really
transformed
things
into
the
mission
being
enabling
nft
persistence
as
a
public
good
and
really
the
potential
of
nfts
is
to
be
an
immutable
data
graph
of
the
web.
When
you
think
about
what
the
internet
initially
was
vision-wise,
it
was
trying
to
create
a
graph
of
data
out
there
that
was
publicly
accessible
and
there's
institutions
like
the
internet
archive
and
others.
A
Looking
to
preserve
that
sort
of
thing,
we
think
nfts
are
going
to
be
a
big
part
of
the
story
moving
forward
so
with
nfts.
All
the
parts
that
are
on
chain
are
obviously
immutable.
We
just
want
to
make
sure
that
that
off
chain
component
can
still
maintain
that
data
graph,
because
the
nfts,
without
that
option
component,
really
lose
a
lot
of
their
value.
A
A
We
are
able
to
do
that
today,
because
filecoin
storage
is
extremely
cheap
today,
it's,
I
think,
something
like
one,
one
thousand
some
number
of
zeros
in
scientific
notation-
I
don't
know,
but
cheaper
than
s3
storage
per
per
unit
of
data
stored-
and
the
other
thing
I
wanted
to
mention
was
we
are,
in
the
background,
doing
something
called
nifty
save
where
we're
scraping
the
chains
out
there
that
contain
nfts
and
persisting
that
off
chain
data.
A
A
So
now
you
have
an
a
minting
transaction
for
an
nft
that
is
referencing
an
http
url
and
we
have
separately
stored
a
copy
that
is
content
addressed.
The
plan
is
to
to
publish
a
data
structure,
mapping
all
nfts
to
the
cids
of
them.
So
now,
folks,
who
use
an
http
url
originally,
but
now
that
server
has
disappeared
or
someone
forgot
to
pay
their
storage
bill
or
whatever
they
can
reference.
This
data
structure
to
get
back
their
nft
data,
so
that's
kind
of
the
the
end-to-end
mission
here.
A
It's
we're
iterating
more
over
time
to
kind
of
fulfill
this
vision.
But
if
you
are
storing
any
nft
data
out
there-
and
you
just
want
an
easy
to
use
experience
that
you
know
maybe
you're
excited
about
our
mission-
maybe
you're
a
little
skeptical,
whatever
no
cost
to
try
it
out
and
store
your
data
everywhere
else
that
you
need
to
to
make
yourself
feel
comfortable.
A
And
moving
on
to
the
web3.storage
side,
it's
again
very
similar
product,
but
it's
it's
the
difference
here
is
it's
much
more
generalized.
It's
really
there
for
any
storage
use
case
out
there.
So
it's
no
longer
trying
to
store
data
as
a
public
good,
but
any
other
storage
use
case
out
there
check
out
web3.storage
and
you
can
really
build
whatever
on
top
of
us.
It
can
be
an
end
user-facing
application.
A
You
can
build
developer,
tooling
or
even
instrument,
another
storage
service
on
top
of
us
and
use
us
as
that,
dumb
storage
layer,
we're
looking
to
unlock
the
potential
of
web3
and
ipfscids,
bringing
web3
more
to
the
web
and
want
to
enable
all
the
creativity
out
there.
One
terabyte
is
always
free
for
new
users.
A
Today,
it's
also
free
beyond
that.
There's
some
weird
thing
where
you
like
request:
more
storage
from
us,
and
we
check
out
that
you
have
like
a
real
project
and
again
filecoin
is
cheap.
Filecoin
storage
is
cheap
today,
so
we
could
do
this,
but
we're
moving
to
a
more
transparent
model
in
the
future
likely
where
users
take
filecoin
to
unlock
more
storage.
So
in
case,
that's
a
little
bit
confusing
to
you,
as
you
check
us
out.
A
You
know
make
sure
that
so-and-so
has
permissions
to
do
what
or
granting
permissions
we
just
launched.
You
can
authorization
and
nft.storage
we'll
bring
that
over
to
web3.storage
soon,
and
then
you
can
see
new
data
data
access
patterns,
potentially
where
individual
end
users
all
have
web3.storage
accounts
and
now
they're,
using
ucans
to
authorize
applications
to
use
that
storage
on
their
behalf
and
now
that
model
of
data
data
ownership
for
end
users
is
kind
of
unlocked
versus
you.
Trusting
that
this
application
is
going
to
do
all
the
storage
for
you
yeah.
A
I
won't
go
too
deep
into
the
other
stuff,
but
a
lot
on
the
horizon
to
be
really
excited
about
here
and
I'll
close
on
a
quick
teaser
for
those
who
are
looking
to
do
some
hacking
this
week
on
how
to
use
us
today
go
visit.
The
website
of
the
product
that
feels
like
it
fits
your
use
case,
create
an
account
and
make
a
key
install
the
client
library.
A
You
know
you
get
a
cid
by
one
line
of
code
and
you
store
it
with
another
line
of
code,
and
then
you
retrieve
that
data
at
any
point
using
that
cid
easiest
thing
to
do
would
be
to
use
a
an
http
gateway,
but
depending
on
your
application,
you
might
want
to
spin
up
your
own
ipfs
note
to
do
that
or
use
other
retrieval
methods.
A
Yeah
check
us
out,
we
got
websites,
we
got
docs,
we
got
discords,
follow
at
nft,
underscore
storage
on
twitter.
I
run
the
account,
so
you
know
it'll
be
fun
yeah
and
I
think
that's
all
I
had
if
folks
in
the
audience
had
any
questions
happy
to
answer
them.
Otherwise-
or
I
don't
know
if
you
have
a
chat
for
people
streaming,
if
there's
like
a
chat
for
questions,
but
otherwise
I
can
move
on
to
the
next
stuff.
B
A
Do
you
mind
defining
what
you
mean
by
invalidated
oh
validation
or
validated
locally?
How
so
yeah?
How
are
third-party
applications?
Gonna
use
it
yeah.
So
that
would
just
happen
at
the
like
the
application
layer
so
depending
on
your
structure.
So
like
one
simple
example
of
folks
using
nft.storage
is
openc
uses
nft.storage
when
you
use
their
freeze
metadata
feature
when
minting
an
nft.
You
click
on
that
openc.
A
Has
you
compute
that
cid
locally,
but
then
openc
will
keep
a
copy
of
what
that
cid
is
in
their
like
databases
and
stuff
like
that.
That
cid
also
ends
up
on
in
the
minting
transaction
for
the
nft.
So
now
it's
like
open
and
anyone
can
see
what
the
cid
is
associated
with
your
given
nft.
So
now
everyone
kind
of
knows
what
that
is,
and
it
becomes
public
at
that
point.
A
But
one
kind
of
crazy
thought
exercise
is
thinking
about
all
like
the
weird
data
access
patterns
that
you
can
enable
using
cids
at
the
core.
So
this
is
something
that
looks
very
traditional
with
the
additional
component
of
trustlessness
of
the
transactions
on
the
blockchain
telling
you
like
what
this
nft
was
associated
with
the
cid
authorized
by
so-and-so
like
public
key
or
whatever,
but
yeah
I
mean
it.
You
can
get
kind
of
crazy
with
all
the
things
that
you
can
can
unlock.
B
Just
to
follow
up
is
there
anything
keeping
track
of
the
various?
I
guess
no's,
I'm
not
really
sure
how
I
feel
to
structure
the
various
nodes
that
have
validated
this
cid
or,
like
basically,
anyone
who's
who's
used
it
at
some
point.
A
So
not
really
I
mean
you
can
look
in
the
ipfs
network
like
ips,
has
a
few
different
ways
to
provide
content
to
each
other.
So
that's
all
public!
You
can
see
what
nodes
are
broadcasting.
What
content!
A
I
guess
I'm
not
a
deep
expert
on
this,
but
I
guess,
like
a
node
could
lie
about
what
content
they
have
and
then
try
to
maliciously
send
you
the
wrong
content,
but
again,
that's
kind
of
the
point
of
being
able
to
verify
things
locally
got
it.
Thank
you.
Yeah.
C
Hi,
I
was
just
wondering
if
ipfs
is
working
on
anything
for
decentralized,
structured
data,
like
databases.
A
Yeah,
so
I
guess
in
terms
of
what
ipfs
quota
code
is
working
on
like
it's,
a
big
open
source
community
with
a
lot
of
folks
building
on
it,
and
there
are
definitely
people
you
know
who
are
they're
trying
to
build
this
use
case.
I
can
talk
a
little
bit
about
what
our
team
at
nft,
storage
and
web
through
storage
are
doing
where
yeah
we're
totally
looking
to
enable
database
databases
to
be
built
on
top
of
us.
So
one
kind
of
tangible
thing
we're
doing
right
now.
A
Right
now,
most
ipfs
infrastructure
speaks
in
ways
of
like
pins
in
ipfs
and
a
pin
needs
to
be
a
complete
graph
of
data,
but
we're
moving
to
infrastructure
where
you
can
actually
have
like
incomplete
graphs
of
data
and
be
able
to
serve
up
that
data.
So
you
can
start.
You
know,
unlocking
database
use
cases
on
top
of
us
by
like
focusing
on
like
the
diffs
and
that
sort
of
thing
so
yeah
ipfs
is
super
broad.
All
it
means
is,
I
can
access
data.