►
From YouTube: nft storage for NFTHack
Description
Hugo demonstrates a new service https://nft.storage - free decentralized storage for NFT data on IPFS and Filecoin.
Built for NFTHack participants (https://nfthack.ethglobal.co/).
B
Get
this
out
of
the
way:
can
you
see
the
nft
storage
awesome
home,
page
cool,
so,
okay?
So
last
week
I
presented
this.
This
was
still
a
little
bit
bare
bones
right
now,
yeah
we
have
almost
all
the
functionality
working
plus
some
awesome,
docs
and
I'm
gonna
show
it.
So
this
is
pretty
much
the
same.
We
did
some
tweaks
in
the
copy.
B
We
tweaked
some
of
the
examples
that
we
have
here
and
the
first
new
thing
that
I'm
going
to
show
you
is
the
documentation
for
the
the
client
library.
So
everything
all
the
api
is
documented.
B
We
are
using
the
the
typescript
types
that
we
that
we
have
in
the
library
this
works
in
node.
It
works
in
the
browser
seamlessly
and
the
users
the
devs
from
the
nfg
hack,
can
just
go
here.
Check
out
the
examples
check
out
the
parameters
check
out
all
the
api,
how
they
need
to
send
the
the
files
in
its
method.
It's
all
explained
here
with
examples,
it's
all
very
simple
and
they
have
all
the
communication
they
need.
So
here
it's
all
generated
from
the
cold
easy
to
update.
B
If
we
need
to
do
some
some
changes,
while
the
hack
the
hackathon
is
going
on,
we
can
do
it
and
it
will
all
be
reflected
instantaneously
in
the
in
the
website.
So
another
new
thing
that
we
have
is
the
an
open
api
schema
for
the
http
api
itself,
so
this
is
also
available
for
the
users.
B
B
B
And
you
execute,
hopefully
this
will
work
yeah,
that's
it!
We
have
our
file
here,
our
links
here
everything
went
well,
you
got
the
response
setters
and
they
don't
need,
don't
even
need
to
write
any
code.
Then
they
can
try
the
api
out
in
the
docs.
They
can
do
the
same
thing
for
the
other
endpoints
you
can
just
list
everything.
B
These
are.
Are
all
my
files
and
yeah
everything
for
the
the
foreign
points.
Each
type
is
described
completely.
The
nft
one
is
the
most
important
one.
So
everything
explained
like
the
pinning
data.
This
is
related
to
the
pinning
services
api,
all
the
files,
if
the,
if
they
upload
upload
the
directory,
all
the
links
this.
Actually
we
remove
this.
So
this
is
this.
B
I
have
a
pr
to
remove
these
links
because
we
think
we
we
shouldn't
actually
give
them
like
this,
so
we're
going
to
improve
the
documentation
to
explain
to
the
users
how
to
create
these.
These
urls
either
the
ipfs
one
or
the
http
one,
because
we
don't
want
them
to
just
take
these
from
the
response
and
put
it
in
in
the
blockchain
directly.
We
just
want
them
to
put
the
cid
there
and
then
they
can
create
whatever
url
they
need
either
the
ipfs
one
they
can
use
whatever
gateway
they
like,
and
it's.
B
B
This
is
an
old
version,
so
it's
missing
the
deals
field
which,
after
this
call
I'm
gonna,
push
it
to
production,
but
we
already
have
the
the
five
one
bits
schema
already.
It
just
needs
to
be
pushed
pushed
to
production
yeah.
So
one
last
thing
it
would
be
to
just
show
you
I
don't
know.
If
last
week
we
already
had
the
upload
bits,
but
yeah
you
can
also
upload
directly
through
the
interface
and
yeah
it
shows
up
here.
B
You
have
your
keys,
you
have
your
files,
you
have
your
docs,
everything
is
nice
and
well
explained
some
easy
to
follow
examples
and
the
comprehensive
documentation
already
there,
and
we
are
also
updating
right
now.
The
terms
of
service
that
we
just
got
from
legal
and
yeah,
that's
it.
If
you
have
any
questions,
please.
A
I
guess
I'm
curious
between
hugo
and
yusuf
like
in
terms
when
it
comes
to
best
practices
for
what
people
should
link
in
to
the
nft
metadata
body
and
how
they
should
be.
It's
like
great
put
it
on
ipfs
absolutely
best
practice,
then
what
you
know
ideally
you're,
not
choosing
a
specific
gateway.
Url
that
you
know,
locks
you
into
a
particular
provider
and
then
that
link
can
break
in
the
future.
Ipfs
colon
slash,
slash
is
more
agnostic
than
that,
but
doesn't
work
in
all
browsers.
So
it
sounds
like
hugo.
A
You
guys
are
recommending
pure
cids
to
be
the
thing
that
you
stick
in
the
nft,
whereas
I
think
yusef,
your
tutorial
recommends
ips
colon
slash
so.
A
C
What
we
recommend-
I
I
don't
know
I
just
went
with
ips
column-
slash
uris,
without
really
giving
it
a
whole
lot
of
thought,
but
because
I
wanted
to
avoid
http.
But
it
does.
It
is
problematic
because,
like
you
have
to
have
special
support
for
that,
I
guess
but
which
you
would
also
need
for
bear
cids.
But
so
I
guess,
if
you're
having
to
special
case
something,
maybe
it's
better
to
just
use
the
cid
like
you're,
saying
and
then
instruct
people
to
build
the
uri
out
of
that.
B
B
Lots
of
discussions
about
like
the
nft
assets
itself,
plus
the
metadata
there's
also
like
inside
the
metadata,
it's
supposed
to
be
a
an
uri
they're,
pointing
to
either
a
preview
of
the
actual
nft
assets
or
the
actual
assets.
But
it
would
be
useful
or
like
one
of
the
discussion
that
we
had
it's
like.
How
do
we
upload
everything
like
data
assets
plus
metadata
in
one
go
into
ipfs?
B
We
would
need
to
update
the
metadata
with
the
right
cid
of
the
root
folder.
That
gets
us
to
a
weird
point
that
we
need
to
figure
out
and
yeah,
and
we
need
to
kind
of
see
if
it's.
If
this
is
acceptable
for
the
devs.
That's
I
think
it's.
The
biggest
thing
that
we
need
to
learn
from
this
hackathon
is
how
to
engage
with
people
in
these
regards
how
to
integrate
them.
B
What's
the
best
way
to
do
it
in
ibfs
and
at
the
end
of
it,
maybe
we
have
a
story
to
tell
that
actually
explains
step
by
step.
Okay,
if
you
want
to
use
ibfs,
this
is
the
best
way.
This
is
the
way
we
we
recommend
and
it's
the
most
agnostic,
most
scalable,
most
decentralized
way.
We
can
provide
to
you.
This
is
it
we
have
docs
here
and
you
can
follow.
A
Yeah,
my
slight
nudges
to
I
agree
with
iraqi's
comment
in
the
chat,
but
using
the
ips
colon
seems
like
a
good
future
forward
approach
whereby,
when,
when
every
browser
is
like
brave
and
resolves
these
things
directly
or
opera
or
other,
that
people
will
be
able
to
look
directly
within
like
the
nft
metadata,
and
these
links
will
be
linkable
and
and
choose
their
gateway
or
local
node
of
choice
kind
of
directly
without
them
having
to
do
like
an
extra
rewrite
step.
So
I
feel
like
that
should
be
the
optimal
one.
A
But
we
should
make
sure
that
we
have
that
then
pushes
us
to
have
good
support
whenever
you're.
You
know
for
ipvs
colon,
slash
in
lots
of
form
in
lots
of
places.
E
There's
also
a
feature
on
the
table
that
we're
considering
after
the
hackathon
to
just
mint
the
not
not
meant
to
not
mince
the
nft
but
actually
create
the
metadata
for
them
essentially
and
then
so,
when
we
do
that,
we
could
like
always
do
this
right
and
not
have
this
problem.
A
Yeah
a
great
point
like
I'm
curious:
what's
what's
next,
you
guys
have
sprinted
to
this
deadline
to
in
order
to
get
this
up
and
ready
for
people
for
for
today,
but
I
imagine,
there's
probably
a
lot
of
stuff
that
didn't
make
it
on
the
sprint
cadence.
So
what's
what's
next
for
for
this
project,.
E
I'm
trying
to
keep
it
a
little
bit
loose
just
so
that
we
can
take
in
all
the
feedback
and
then
look
through
it
and
prioritize
it,
but
right
now
we're
like
actively
sort
of
in
the
hackathon
channels
like
taking
in
feedback
and
putting
it
into
a
docking
notion
that
anybody
can
see
but
yeah.
I
think
that
feature
is
probably
the
the
number
one
one
and
then
there
are
broader
questions
around
like
what
are
the
limitations
of
the
service?
How,
broadly,
can
we
give
it
out?
E
A
Nice,
and
are
we
actively
going
to
be
making
file
coin,
deals
with
the
nft
data
throughout
this
weekend?
Is
that
like
an
active
piece
of
this
as
well
or,
is
that
a
secondary
bit.
D
Correct
this
is
active,
I
mean
we
don't
have
enough
nfts
just
yet
to
even
try
to
batch
anything,
because
there
are
just
too
few
still.
We
expect
this
to
ramp
up
during
alan's
talk
tomorrow,
the
presentation
but
yes
effectively
our
sla
for
this,
like
24
48
hours
after
that,
your
stuff
will
show
up,
and
we
will
be
mentioning
that,
through
towards
the
end
of
the
hackathon
and
after
the
hackathon,
they
should
go
back
to
the
same
api
endpoints
and
check
out
where
they
are
still
planted
on
falcon
and
stuff.