►
From YouTube: IPFS Weekly Call 2019-01-28 🙌🏽📞
Description
IPFS Newsletter: https://tinyletter.com/ipfsnewsletter
B
All
right,
hello
and
welcome
to
the
ipfs
weekly
call
where
we
get
to
learn
about
the
exciting
things
are
being
built
in
the
ipfs
community,
so
hello,
everyone
I'd
like
to
start
off
with
thanking
Alan
for
taking
notes,
and
if
you
are
part
of
this
call,
if
you
can
just
put
your
name
in
as
an
attendee
that'd,
be
great
by
putting
in
the
Google
Doc
file
now.
B
So,
if
you
could
just
put
in
your
attendance
I'll,
be
great
I,
don't
think
we
have.
We
don't
have
any
announcements
listed,
so
we
are
going
to
jump
right
into
the
presentation
which
is
going
to
be
done
by
Matt
and
CTO
of
pinata
and
Fiona.
Pinata
is
an
API
where
you
can
store
and
use
content
on
top
of
the
ipfs
a
table
of
ipfs.
So
without
further
ado
Matt.
Would
you
like
to
take
it.
E
It's
expecting
a
little
bit
more
warmup
time
with
their
announcements.
All
righty
share
this
off
here,
see
my
screen:
yeah,
perfect,
all
righty.
So
a
little
bit
of
a
brief
history,
both
myself
and
my
co-founder
Kyle
we're
from
Omaha
Nebraska
got
lots
of
corn
lots
of
cows,
not
a
lot
of
web
3
development,
but
we're
trying
to
get
that
kick-started
here.
E
We've
been
in
the
web
3
space
for
almost
two
years
now
we
started
off
doing
blockchain,
consulting
and
prototyping
and
mainly
aetherium
based,
and
we
noticed
everybody
was
using
IP
FS
for
their
data
storage,
but
everybody
was
building
their
own
solution.
Kind
of
you
know
everybody
to
spin
up
their
own
nodes,
figuring
out
all
the
problems
that
to
work
through
themselves.
E
We
just
didn't
see
that
as
a
very
efficient
solution
going
forward,
so
we
wanted
to
come
up
with
something
that
allowed
specifically
etherium
developers
and
then
web
3
developers
to
post
content
on
ipfs
without
needing
to
host
their
own
node
manage
updates,
and
we
wanted
to
give
it
give
them
an
easy
API
to
work
with
in
the
process.
So
we
built
pinata,
we
launched
I,
think
G's
October.
Last
year,
around
the
time,
East
Berlin
was
not
sure
the
exact
date
we've
just
been
improving
it
since
then,
so
to
start
off
with
I'll.
E
Give
you
guys
a
little
bit
of
demo
of
what
our
service
is
and
kind
of
some
of
the
features
that
we
have
so
start
off
with
a
key
point
here
is,
we
can
add
files
to
ipfs
effortlessly.
So
let's
say
we
don't
have
an
account.
Let's
go
ahead
and
sign
up
for
it
here
and
let's
do
I'll.
Do
ipfs
call
pinata
cloud.
E
E
E
So,
like
I
said,
you
start
off
and
you
get
some
easy,
an
API
key
and
then
you
get
a
secret
key
and
we
like
to
include
examples
with
postman
is
usually
an
easy
one
that
people
can
be
familiar
with,
and
then
we
also
like
to
provide
JavaScript
examples
for
people
as
well.
There's
a
lot
of
web
3
developers
will
build
their
applications,
utilizing
JavaScript.
So
without
further
ado,
let's
give
it
a
little
bit
of
a
demo
and
see
kind
of
how
this
would
work
in
practice.
So
I'm
gonna
bring
this
postman
in
here.
E
I'm
gonna
take
a
soft
screen
for
a
second
grab,
my
API
keys,
and
we
can
put
them
in
here.
So
this
is
an
endpoint
that
we
have.
That
allows
you
to
count
tests
to
make
sure
that
your
authentication
works
so
and
go
ahead,
should
get
a
response
back
here
saying:
congratulations,
we're
communicating
with
the
pinata
API,
so
everything's
looking
good!
E
Now,
let's,
let's
do
a
little
bit
a
little
bit
more
fun
than
that.
So
let's
say
we
wanted
to
pin
an
actual
file
to
ipfs.
So
if
we
go
to
our
Doc's,
let's
check
that
out.
You'll
notice
here
we
like
to
include
little
photoshopped
images
of
our
of
our
logo
and
various
kind
of
stock
images.
My
co-founder
does
a
great
job
of
that.
So
we'll
come
in
here
and
we
kind
of
see
our
documentation
will
tell
people,
and
you
know
how
to
format
the
body.
The
request
file.
E
Give
you
a
little
bit
of
idea
of
what
you're
working
with
one
thing
that
we
do.
Allow
is
the
optional
adding
of
metadata,
so
we
noticed
that
people
were
uploading,
their
content
to
ipfs
and
it's
kind
of
hard
to
keep
track
of
what
they
were
all
opinion,
you're,
just
gonna
hash
back.
You
know
what
is
that
you
never
really
have
insight
into
what
those
hashes
are
actually
what
the
content
is
behind
them.
So
we
allowed
people
to
that
include
a
custom
name
for
their
pins
and
then
also
provides
some
custom
key
values.
E
So
you
can
really
name
them
whatever
you
want,
we'll
keep
track
of
those,
and
then
you
can
use
those
later
on
for
searching
for
your
content
through
our
system.
So
it
can
be
strings
numbers
dates,
it
really
kind
of
whatever
you
want.
So,
let's,
let's
check
out
the
postman
example.
Here,
let's
see
okay,
this
is
kind
of
what
we
want
to
look
like
in
the
interest
of
time.
I've
got
this
a
little
bit
a
pre
pre
demo
mode
ready
for
you
guys
to
go.
E
So,
let's
go
back
in
here
and,
let's
add
our
headers
back
to
the
pin
file
endpoint.
So
you
go
in
here
and
we
got
a
body
so
I'm
going
to
choose
my
favorite
file
here.
If
I
can
find
it
here,
we
got
or
a
big
fan
of
the
party
parrot
here
in
the
Midwest
I,
don't
know
how
that
got
started,
but
so
it's
a
fun
fun
little
gift,
so
we're
gonna
upload
the
party
parrot
and
you'll
see
here
you
get
this
full
screen
we're,
including
a
little
bit
of
metadata
here.
E
So
we
have
the
name:
what's
gonna
call
mr.
parrot
PNG
and
we
provided
some
key
values.
Let's
say
our
parrot
works
for
the
parrot
Cinco.
We
can
use
that
later
to
search
and
find
all
employees
of
parrots
and
company.
So
let's,
let's
give
us
a
little
bit
of
a
send
here
and
we
should
receive
a
request
back
awesome.
So
we've
got
a
hash
back
and
we've
got
the
size
of
that
object.
E
We
have
the
time
stamp
for
that
object,
so
we'll
store
those
in
our
systems
that
I'll
get
to
in
a
little
bit
what
you
can
do
with
that
data.
So
next
in
the
list,
let's
say
we
want.
We
already
have
content
that
we're
hosting,
let's
say
we're
decentralized
application
and
we're
hosting
our
own
IP
FS
tones.
But
we
wanted
to
duplicate
our
content,
give
it
a
little
bit
more
resiliency
on
the
ipfs
ecosystem
and
then
also
we
wanted
to
have
that
metadata
functionality
that
we
provide.
E
E
This
will
be
a
different
parrot.
This
is
gonna,
be
our
pirate
parrot.
So
I
have
this
pinned
on
a
different
note
in
this
ecosystem.
It's
the
party
parrot,
except
he's,
got
a
pirate
sword
and
he
also
works
for
the
parrots
and
company.
So
let's
click
send
here
and
this
will
go
out.
Our
nodes
will
find
the
content,
so
we
found
it
and
we've
gotten
the
hash
back.
It's
a
clarification
that,
yes,
this
is
what
we
wanted
to
pin
and
we
have
the
pin
size
and
again
our
timestamp.
E
E
Okay,
so
our
in
point
here
go
basically
all
of
our
constant
values:
we've
kept
those
as
specific
values
for
a
rest,
end
point,
and
then,
if
it's
a
wild
card,
you
just
include
a
star
there
and
then,
if
you
want
to
get
with
the
customized
stuff,
our
metadata
it'll
look
like
this.
The
documentation
goes
over
this
a
little
bit.
E
It's
pretty
pretty
extensive,
so
I'm
not
gonna,
sit
here
and
read
through
the
documentation
all
in
front
of
you,
but
the
gist
of
it
is
you
can
search
via
the
metadata
you've,
provided
you
can
also
search
for
things
like
all
of
our
end
points
here,
like
our
hash,
contains.
You
know
when
it
was
pinned
if
you've
unpinned
it
at
a
certain
date.
You
can
search
that
as
well.
E
You
can
search
for
pin
sizes,
you
can
search
for
things
that
are
pinned
to
unpin
doll,
and
then
we
also
allow
pagination
as
well
so
you're,
not
getting
a
thousand
entries
back
all
at
once
and
then,
like
I,
said
the
the
customized
metadata
stuff,
so
your
example
values
and
example
keys,
and
then
the
operations
that
you
can
utilize
to
search
for
things.
So
we
have
kind
of
classic
sequel.
We
use
Postgres
to
store
a
lot
of
our
metadata.
It
provides
pretty
fast
indexing
and
responses,
so
you
can
use
greater
than
you
know
greater
than
equal.
E
Basically,
all
these
classic
boolean
values
and
then
again
we
provide
response,
will
show
all
the
things
that
you've
got
that
matched
your
values
here.
Sorry,
if
this
is
a
zoom
in
too
fast
I
know,
sometimes
zoom
can
delay
a
little
bit.
I'll,
try
and
slow
down
the
scrolling
and
then
our
postman
example.
Again
we
got
that
and
also
we
provide
a
little
bit
of
a
JavaScript
example
for
what
that
might
look
like
in
practice.
So
let's
apply
that
in
postman.
E
Again,
let's
say
we
want
to
have
all
wildcards
and
then
the
only
thing
that
we
care
about
is
our
metadata
that
we
associated
with
it.
So
we
don't
care
about
the
name.
We
only
care
about
the
key
values.
Let's
say
we
want
to
search
for
the
avatars
of
everybody
that
works
at
the
parrots
and
company,
so
we'll
click
send
here
and
you'll
notice.
We
gotten
back
to
counts.
We've
had
two
uploads
that
match
those
that
criteria.
E
So
that's
that's
what
how
the
API
looks
like
when
you're
interacting
with
it
there.
We
also
provide
a
nice
kind
of
GUI
for
people
that
don't
want
to
utilize
the
the
API.
Maybe
some
of
your
executive
team
wants
to
look
at
things
on
a
little
bit
of
a
higher
level
or
it
upload
things
to
the
system
in
a
little
bit
of
a
higher
level.
So
for
uploading
we
have
our
pinata
upload.
Excuse
me
here:
unless
I
could
drink
water.
E
Okay,
okay,
so
we
can
upload
files,
just
normally,
let's
say
we're
going
to
upload
the
Christmas
parrot.
All
right,
you
can
show
your
your
file
size
before
you
upload
it
simple
upload
a
little
bit
of
a
progress
bar
there
and
you'll
notice.
It's
now
been
pinned
it's
on
PFS
our
servers
and
then
we
also
have
a
pin
by
hash.
So
on
the
on
the
GUI,
this
works
a
little
bit
different.
E
We
wanted
to
make
this
more
of
an
asynchronous
process,
because
it's
not
very
fun
to
you,
know,
click
pin
and
then
have
it
spin
and
then
wait.
You
know
five
minutes.
Maybe
it
takes
a
while
for
us
our
servers
to
find
your
content,
because
it's
hidden
off
someone
the
ecosystem.
So
we
have
the
concept
of
a
pinning
queue.
If
you
add
stuff
to
the
queue,
our
servers
will
go
and
find
your
content
asynchronously
and
then,
when
it's
done,
it'll
be
added
to
your
your
pins
content.
So
a
good
example
of
this.
E
This
is
a
book
that
I've
found
out
online.
There's
a
big
old,
ipfs,
hashes
subreddit.
It's
got
lots
of
good
hashes.
If
you
guys
ever
need
anything
to
test,
so
this
one
is
a
PDF
and
it's
called
the
ethics
of
Big
Data,
which
I
figure
is
a
very
appropriate
thing
for
people
to
be
reading
about
nowadays.
So
if
we
click
pin
here,
it'll
add
this
to
the
pin,
Q
and
you'll
notice.
We
have
the
name,
a
hash
and
it's
status
so
right
now
it's
searching
and
we're
going
to
repeatedly
check
back
in
and
wait.
E
E
Also,
it's
just
I
like
this
image.
A
lot
so
I
wanted
to
share
it
with
you
guys
my
co-founders
good
at
the
photo
show.
So
what
do
we
do
with
this
content?
Now
that
we've
pinned
it,
we
have
the
GUI
for
our
pin
explorer.
So
you
notice
we've
pinned
four
items
here.
We
have
shows
when
they've
been
pinned
the
current.
If
they're
unpinned
just
shows
earth
their
pin,
they
haven't
been
unpinned.
They've
currently
been
pinned.
E
We
have
all
of
our
PNM
G's
or
gifts
or
Big
Data,
and
we
also
have
a
gateway
that
we
host
as
well.
So
you
can
view
your
content
through
here
we
have
those
all
connected
with
our
nodes
directly,
so
it
should
load
content
a
little
bit
faster
than
if
you
were
to
go
through
a
third-party
gateway,
because
our
yeah,
it's
directly
connected
to
all
of
our
nodes,
again
got
a
parrot
herring
so
little
pirate,
so
yeah
there's
the
gateway.
It's
a
you'd
no
longer
want
to
pin
your
content.
Well,
that's!
Okay!
E
We
can
go
ahead
and
remove
the
pin
and
it
should
come
in
here
and
unpin
it.
Yes,
all
right!
So
that's
been
unpinned.
Our
servers
are
no
longer
pinning
it
and
you
will
no
longer
be
charged
for
that
content.
So,
in
your
account
section
you
can
see
a
little
bit
of
you
know
the
number
of
items
you
have
pinned
the
size
and
then
let's
talk
a
little
bit
about
our
our
pricing.
I
can
go
over
that.
Basically,
it's
it's
the
typical
kind
of
infrastructure
as
a
service
or
similar
to
like
Amazon
or
digitalocean.
E
You
don't
even
need
to
add
a
credit
card
and
then,
if
you
do
want
to
add
a
credit
card,
we
utilize
stripe
as
our
as
our
payment
provider
keeps
things
pretty
secure
and
easy
from
us
or
from
our
standpoint.
So
you
can
just
add
your
credit
card
here
and
go
from
there
yeah.
That's
that's
pretty
much
the
gist
of
our
system.
E
D
B
The
first
question:
p'malee:
does
your
name
field
need
to
be
unique?
No.
E
It
does
not
we
we
don't
require.
That
is
an
actual
as
a
as
a
requirement.
So
we
figured
that
if
people
wanted
to
name
things
the
same
thing,
you
know
we'd,
let
them
we
have
other
unique
identifiers
like
the
hash
to
keep
them
separate.
But
if
keep
things
unique,
you
know
obviously
don't
name
things
the
same.
E
Nowhere
is
so
W
asks
how
much
storage
does
pinata
have
available
nowadays.
Is
it
using
ipfs
cluster
to
connect
more
nodes,
/
disks,
okay,
so
storage
available
right
now
we
have
I
think
it
was
200.
Gigabytes
is
what
our
note
is
set
to
and
we
just
gradually
increase
that
as
more
users
add
content
to
our
system.
It's
pretty
flexible!
We
utilize
block
block
storage.
E
There
were
some
things
that
I
wanted
to
do
with
our
system
that
it
didn't
appear
that
ipfs
cluster
can
do
for
us
recently.
I've
learned
that
ipfs
cluster
can
do
that
now,
so
we
are
planning
on
implementing
ipfs
cluster
into
our
system.
In
the
upcoming
months,
I
was
just
a
decision.
I
made
early
on
that
suited
our
needs,
but
those
ability,
abilities
of
IPs
of
ipfs
cluster
seem
to
be
able
to
accommodate
that.
So
we're
definitely
looking
to
add
that
plans
on
adding
more
user
Keys
for
differentiated
access
tracking.
F
Yeah,
absolutely
into
what
I
mean
here
is
like.
If
you
are
like
just
building
gaps
right
like
you,
don't
want
to
sell
it
to
give
like
your
access
token,
for
your
P&L
service,
single
user.
That
opens
your
eared
app,
so
you
want
to
create
like
a
justice
bulky
that
is
just
for
that
user.
Would
that
require
today
I
creating
multiple
PDA
accounts
for
each
user
and
like
tracking
all
of
them,
or
is
there
a
subsystem
there
where
I
can
just
like
create
keys
on
the
fly
for
the
users
yeah.
E
Yeah,
that's
a
great
question,
so
no
current,
we
don't
have
that
you're.
Actually,
the
first
person
that's
asked
for
that.
So
that's
it's
a
good
thing
to
throw
up
on
our
roadmap
I
like
that
concept,
but
yeah.
So
right
now,
if
you
wanted
to
have
multiple
sub
accounts,
you
just
basically
create
multiple,
multiple
keys.
E
E
There
was
a
suggestion.
I
remember,
seeing
that
on
forums
somewhere
about
paying
other
gateways
that
seemed
kind
of
like
spamming,
the
network
and
kind
of
seemed
like
a
little
bit
of
a
almost
abuse
of
the
way
that
the
DHT
works.
So
we
we
haven't
been
doing
that
I
guess,
since
you
are
with
protocol
Labs,
is
that
a
sanctioned
activity
that
we're
allowed
to
participate,
I.
F
G
A
F
G
F
F
E
That
makes
sense
so
a
little
bit
from
our
strategy.
What
we're
going
for
is
we're
planning
on.
Basically,
one
of
our
one
of
our
goals
is
to
kind
of
increase
the
number
of
nodes
that
we
have
under
pinata
and
the
number
of
gateways
that
we
have
under
pinata.
So,
similarly
to
how
I
think
ipfs
io
has
a
bunch
of
different
gateways
all
around
the
world.
E
I
know
it's
masked
under
your
main
domain,
but
it's
my
understanding
that
they're
all
connected
under
the
hood
and
that's
kind
of
the
approach
that
we'd
like
to
take
so
have
a
bunch
of
different
gateways
and
kind
of
connect
them
all
via
the
the
the
swarm
connect
feature
that
ipfs
and
bootstrap
as
well.
Oh
lots
of
questions
coming
in
does
Bill
and
understand
deduplication
yeah.
It
sure
does
so
if
you,
if
you
upload
the
same
content
to
ipfs
more
than
once,
basically
we'll
just
count
that
as
the
same
entry.
E
E
E
Is
our
services
will
go
through
and
will
repeatedly
check
to
try
and
repin
that
content,
so
yeah
we'll
just
kind
of
keep
checking
until
we
find
it
and
then
hopefully
we'll
find
it
at
some
point?
Yes,
what's
in
the
works?
Let's
what
features
are
you
excited
about
Oh
briefly,
I'm
gonna
go
back,
it's
popped
in
my
head.
Molly
I'm
also,
if
you
don't
want
to
do
asynchronous
requests
for
painting
by
hash,
you
can
utilize
the
add
hash
to
pen
queue,
endpoint
and
that'll,
just
asynchronous
asynchronously
added
to
the
queue
Thank.
A
E
That's
it
was
kind
of
an
MVP
thing
right
now.
It
didn't
really
hurt
our
system
to
repeatedly
check.
We
have
it
set
up
in
a
way,
so
it's
not
really
creating
a
lot
of
overhead.
If
that
becomes
a
system
and
the
problem
in
the
future,
I
might
have
a
timeout
after
a
couple
days
of
checking,
but
so
far
it
hasn't
been
an
issue.
I
figured
out
kind
of
one
of
the
statistics
on
how
people
were
utilizing
the
system
before
implementing
any
changes
like
that,
let's
see
what's
in
the
works,
what
features
are
you
excited
about?
E
One
thing
we're
working
on
right
now
is
IP
NS
people
have
been
excited
about
hosting
their
own
websites
and
kind
of
dynamic
blogs
and
stuff.
Utilizing
ipfs,
well
I
PMS,
so
we're
looking
at
utilizing,
IP,
MS
and
then
also
DNS
link
as
well
to
give
people
a
little
bit
of
options
there
and
then
we're
also
looking
to
add
in
a
near-future
the
ability
for
you
to
kind
of
get
a
little
bit
more
control
over
where
your
contents
being
hosted
at
so
what
region
of
the
world
it's
being
hosted
at
where'd
we
go.
E
Are
you
planning
to
open
the
code
base
for
pinata
cloud
yeah
so
right
now,
it's
it
is
closed
source,
and
that
was
mainly
news.
Dirty
shoes
right
now,
I
am
a
one-man
dev
shop,
so
I
wanted
to
make
sure
that
we
had
some
external
kind
of
eyes
on
our
code
before
we
release
it
to
the
public
but
yeah.
It's
it's,
definitely
something
that
we
constantly
have
on
our
roadmap.
E
E
E
E
B
E
Yeah,
that's
that's
a
it's
an
interesting
one,
so
I
guess
the
main
one
I
think
this
was
talked
about.
I
think
this
was
talked
about
on
a
couple
of
github
issues
through
one
way
or
the
other.
One
of
the
things
I've
been
gunning
for
is
the
ability
to
track
bandwidth
on
a
per
hash
basis.
So
if
one
person,
if
one
hash,
is
utilizing
a
ton
of
bandwidth,
it
doesn't
really
make
sense
to
kind
of
artificially
inflate
everybody
else's,
get
gigabyte
cost
for
that
and
right
now
there
Kyle
bring
this
in
very
odd
and
then
yeah.
E
G
H
E
Absolutely
I
commented
this
on
on
the
I
think
it's
either
an
email
or
on
the
github.
We
do
use
a
lot
of
node,
so
if
there
would,
if
there
ever
is
an
IP
FS
cluster
JavaScript,
API
that'd
be
nice,
but
I
mean
that's
always
something
that
we
can
kind
of
start
to
create
ourselves.
Once
we
get
a
little
bit
bigger
in
terms
of
staff
and
I
hope
what
we
contribute
yeah
can
contribute
back
to
the
community
and
that
way,
if
you
guys
haven't
been
able
to
get
to
that
by
then
that'd
be
great.
E
Yeah
David
asks
the
Vedas.
Would
you
be
ready
to
handle
a
few
terabytes
so
yeah
our
we
utilize
digital
ocean
right
now
for
our
storage
and
those
block
storage
currently
handles
up
to
16
terabytes,
so
yeah
we
can
expand
up
to
16
terabytes
per
node
in
terms
of
the
performance
on
that
it
might
be
a
little
questionable
at
first
we'd
probably
have
to
upgrade
our
specifications
on
our
node
to
make
that
work
as
well.
E
C
B
Thank
you
so
much
for
joining
us
at
CFS,
weekly
coffee.
We
really
appreciate
it.
This
is
great
and
for
everyone
else,
thank
you
for.
Thank
you.
So
much
for
joining
us
we'll
be
here
same
time
next
week
and
if
you
are
actually
building
something
with
ipfs.
If
you
have
a
project,
then
please
let
us
know,
and
we
would
love
to
have
you
present
as
well.
Thank
you.
Everyone
and
I
will
see
you
next
week.
Bye,
bye,
Becker.