►
Description
This week: Infura's IPFS tech stack and offerings
Read Infura's blog post: https://blog.infura.io/new-tools-for-running-ipfs-nodes-196de636f079/
and make sure to check this great tutorial: https://www.freecodecamp.org/news/hands-on-get-started-with-infura-and-ipfs-on-ethereum-b63635142af0/
The IPFS weekly call is open to all. It's a venue to highlight exciting work in the community and share what we know.
See: https://github.com/ipfs/community/issues/453
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
A
B
B
And
please
interrupt
at
any
time
with
questions
or
anything
that
you'd
like
to
know
cool
so
getting
into
this
is
we
mainly
work
off
the
HTTP
API?
So
obviously
this
is
the
ffs
Docs.
We
offer
a
subset
of
the
ipfs
commands,
and
so
here
they
are
on
our
Docs.
Here
you
can
see
the
different
it's
mainly
around
pinning.
B
Here's
the
example
I
feel
festive
here
at
I/o,
hello.
This
is
the
hello
world
example.
This
is
what
a
pinning
link
looks
like
with
our
service
and
some
quick.
How
we
contribute
to
open
source
here
is
what
we
use
internally
as
a
way
to
transfer
data
between
ffs
notes
and
storage.
It's
a
command
line
tool.
B
It's
actually,
it
works
really!
Well,
you
can
just
move
data
around
between
different
servers
and
we
use
it
internally
for
our
own
purposes
and
move
around
another
one,
sometimes
seeing
getting
data
from
type
you're
fast
and
how
it's
working
as
a
little
harder.
So
we
built
a
data
dog
plug
in
that
house
tracing
and
you
can
check
it
out
here.
B
We
participated
one
as
a
sponsor
just
recently,
and
I
wanted
to
see
what
people
could
build
with
our
ffs
projects,
and
I
can
talk
about
get
:
more
in
detail
to
if
anyone
has
questions,
but
basically
there
was
six
project.
Submissions,
two
really
cool
and
thought
I
want
to
note
here
is
one
where
content
creators
can
upload
anything
back,
two
FS
and
then
sign
it
with
a
theorem
project,
called
you
port,
where
it
ties
your
basically
control
word
in
theory,
my
address
to
the
registration,
and
so
you
actually
can
register
them
in
the
theorem
network.
B
Based
on
these
content,
uploads
thought
that
was
really
cool.
Another
one
was
again
on
the
theorem
there's:
a
open
protocol
for
wallets,
called
wallet
connect,
and
so
here
is
a
popular
project
where
the
use-ip
of
has
to
hold
data
and
basic
combines
it
with
your
wallet
connect
persona
is
the
controller
of
these
fgfs
contents,
thought
that
was
really
cool,
then
so
getting
into
how
in
fira
has
the
architecture
so
going
way
back
to
when
it
first
came
out
minimize
this,
so
it
started
out
very,
very
simple:
this
is
the
simplest
version.
B
B
B
Within
a
year
we
had
over
two
million
items
pinned
through
the
service,
and
we
had
a
very
I
guess
naive,
our
first
drafted
at
this
and
we're
running
its
problems
with
still
data.
We
weren't
I,
guess
garbage
collecting
or
we're
called
banishing
any
data,
and
so
the
next
original
architecture
has
a
bit
more
complexity.
B
B
This
probably
took
us
to
late
2018.
This
is
what
the
architecture
look
like
and
again
it
we
were
scaling
or
serving
more
traffic
than
this
architecture
allowed,
and
so
we
had
a
new
version,
which
is,
which
is
what
I'll
mostly
talk
about
here,
and
so
this
is
the
architecture
as
of
early
2019,
and
so
basically,
we
have
what's
called
this
ipfs
proxy,
basically
a
large
repo,
just
a
single
code
base,
which
I
can
do
all
these
things
and
so
basically
I'm
going
through
it.
B
Basically,
we
so
IP
n
s
is
the
pillars
of
protocol
name
service,
and
so
we
have
what's
called
an
echo
nest
tracker
and
then
a
pin
list
tracker.
So
for
for
each
user,
we
keep
a
track
of
what
storage,
what
blocks
are
being
stored,
everything
about
an
each
specific
user
and
then
what
we
like
to
so
our
idea,
a
guy
is
designed
where
it's
meant
to
act.
B
Like
a
local
note
like
if
you're
familiar
with
interacting
with
your
local
storage,
we
like
the
API,
it
should
be
very,
very
similar
or
identical,
so
we
don't
have
custom
API
calls.
We
have
pretty
much
just
everything,
a
subset
of
what's
in
the
ipfs
Docs
themselves,
if
there
are
taking
I
guess
input,
if,
if
you
guys
would
like
to
see
any
custom,
api's
and
custom
commands,
and
things
like
that,
we'd
be
interested
to
hear
about
that.
B
But
right
now
it's
it
looks
exactly
like
a
local
node
and
so
I
guess
going
on
yeah
we
so
the
go
FS!
That's
the
one
you're
used
to
the
public
go
up
your
best
client.
We
work
with
that
and
basically
this
mono
repo.
We
call
it
almost
like
sits
on
top
of
it
and
then
interacts
with
it
and
then,
of
course,
we
use
s3
Amazonas
s3
for
our
storage
and
so.
B
Yeah
I
think
that's
everything
about
this.
Here's
the
full
of
what
we're
working
on
eventually
allowing
in
the
next
version.
So
everything
you
see
here
is
obviously
a
lot
more
than
the
API
I
showed
in
the
beginning
currently
supports,
but
that's
what
we're
working
towards
pretty
much
all
these
commands
would
be
usable
in
a
version.
I
don't
have
a
release
date
to
announce,
but
we're
working
on
getting
all
this
into
our
API
I.
B
C
A
good
question
just
about
kind
of
the
the
user
flow
of
someone
who's
coming
through
here
and
wants
inferior
to
back
up
their
stuff.
Do
you
spin
up
a
new
co,
ipfs
node
for
each
one
of
those?
How
do
you
kind
of
segment
or
understand
the
I
mean?
Obviously
you
can
track
what
content
they're
pinning,
but
when
you
get
requests
for
that
data
back,
are
you
also
tracking
how
much
bandwidth
or
other
things
are
related
to
a
particular
user
as
part
of
your
your
service
or
it?
B
B
C
B
B
B
D
B
D
D
It
just
if
you
have
something
an
IP
LD
and
you
have
in
transaction
in
theorem.
You
can
actually
use
the
multi
codec
in
IPL
d2
natively
link
a
hash
bags
linked
addressing
in
IPL
d
to
include
that
transaction
in
or
your
node
and
all
right,
I
like
it
so
that
as
of
last
year,
there
was
support
for
it
as
a
plug
in
for
the
go
ipfs
and
I've
lost
track.
As
far
as
word,
where
it
is.
E
B
B
Yeah
that'd
be
awesome
and
a
yeah
we're
fully
remote,
and
so
we
have
that
I'll.
Stop
the
show
there
and
answer
any
questions
or
we're
actually
like
so
we're
in
the
process
of
really
making
a
new
ipfs
version.
The
diagram
I
just
showed
was
from
about
a
year
ago,
and
the
new
one
will
look
similar,
but
they'll
offer
a
ton,
more
I,
guess,
functions
and
the
features,
and
so,
if
there's
anything,
the
community
would
like
to
see
here
would
be
really
useful.
B
F
B
It
all
there
I
mean
so
there
is
a
so
what
we
just
kind
of
call
it
like
the
banasur,
which
so
are
so
users
can
take
down
content
that
they
pin
the
if
there
are
things.
So
if
one,
if
a
piece
of
information
isn't
being
accessed
I,
think
if
the
user
leaves
it
I
should
don't
want
to
say
anything
incorrect,
so
I
need
to.
We
can
write
a
some
in-depth
details
about
how
we
garbage
collect
and
cover
things,
but
yeah
so
I
might
I.
Don't
have
a
great
answer
for
you
right
now:
hey!
That's
it!
B
A
G
Interested
in
the
mono
repo
piece
of
the
diagram
that
looked
as
far
as
I
could
tell
was,
like
the
front
end
to
a
request
flow.
The
first
thing:
it
was
the
thing
labeled
idea:
fest
proxy
mono,
repo
cuz.
That
was
that
referring
to
a
bunch
of
ipfs
like
processes
that
share
a
repo
or
what
what
was
them,
what
which
repo
was
mono
in
that.
B
B
It's
I
think
the
name
itself
leads
to
you
believe
that
there's
a
certain
projects
position,
there's.
Basically
a
single
large
code
base
that
handles
everything
around
I
feel
faster.
It's
referring
to
a
git
repo,
not
a
ipfs
people,
for
example
correct
yeah,
it's
a
internally.
It's
a
private
code
base,
Suvir
a
fair
on
github
yeah,
understood.
A
B
I
can
so
basically,
there
was
some
initial
pain
points
around
so
we're
first
like
from
the
very
beginning,
art
the
first
architecture.
We
had
a
single
node
that
allowed
unbounded
rights
on
a
public
endpoint,
and
so
eventually
that
would
hit
a
eventually
make
that
unusable.
So
there's
over
two
million
items
in
the
permanent
storage
on
that
first
node
and
it
basically
bled
to
us
trying
to
do
stricter
right
right,
limiting
and
making
sure
that.
B
B
Requests
by
I
think
there's
a
there's
a
issue
but
I,
because
we
so
something
the
theory.
I'm
added
was
a
way
to
act
on
behalf
of
another
user,
if
basically,
a
relaying
sign
transactions,
and
so
we're
working
on
a
way
to
see
relay
signed
is
the.
If
a
user
wants
to
interact
with
their
best
content,
the
pin
content
boy
yeah
interacts
with
it
anyway.
B
Some
of
the
some
of
the
things
we
can
do
on
behalf
of
them
are
limited
and
so
we're
working,
possibly
in
a
PR
or
some
way
to
working
with
the
go
back.
You
guys,
client
and
allowing
this
and
things
like
that,
where
users
could
that's
like
you,
don't
have
signed
transactions
or
sign
off
on
a
certain
subset
of
interactions
that
we
could
take
on
their
behalf.
I
think
that's
one
of
the
because
right
now
do
we
there's
definitely
pain
points
around
things.
B
From
aren't
you
I,
don't
think
they
wanted
me
in
storing
users,
private
keys
it
be
basically
having
these
are
signed
messages
and
having
the
ipfs
Network
able
to
recognize
that
it
might
yeah.
It's
probably
some
sort
of
change
to
I
think
that
self,
but
that's
just
something
that
yeah
we'd
like
to
talk
to
a
protocol
labs
about
and
have
a
discussion.
Yeah.
C
B
C
C
The
IP
of
my
side,
but
something
something
I
know
that's
been
talked
about.
I
think
with
Matt
and
with
others,
is
just
a
way
of
being
able
to
have
a
pinning
service
that
backs
up
to
file
coin
for
more
kind
of
distributed
backend
and
then
offers
interfaces
on
night
if
I
side
for
pinning
content.
That
then
ends
up
stored
by
my
miners
in
the
file
coin
Network,
and
so
that
was
the
kind
of
interesting
space
that
I
wasn't
sure
whether
you
guys
have
started
digging
into
yet.
B
Gotcha
so
nice,
if
I
like
unused
Orage
people
can
earn
file
coin.
By
having
a
new
storage
I
mean
I,
guess
we
could
act
as
a
file,
a
coin
hoster
and
turn
it
that
way.
That's
like
a
business
or
yeah.
We
could
offer
like
a
client
that
people
could
use
to
storage,
but
I
guess
yeah
I
need
to
learn
more,
but
yeah
I
think
that
I'd
be
a
useful
service.