►
From YouTube: IPFS All Hands 🙌🏽📞February 25, 2019
Description
IPFS Newsletter: https://tinyletter.com/ipfsnewsletter
A
A
A
It
looks
like
we
don't
have
any
announcements
for
today,
but
if
you
are
part
of
this
call,
if
you
could
please
put
down
your
name
in
our
notes,
that
would
be
great,
so
I
just
put
in
a
Google
Doc,
just
put
in
your
name,
they'll
be
great,
and
so
today
we
are
going
to
hear
from
Cael
Teeter
from
Microsoft
he's
a
senior
software
engineer
and
he
also
works
on
blockchain.
A
A
This
view
this
she
was
fine.
Okay,
now.
B
A
B
B
Why
we
built
it
and
the
pieces
we
built
to
be
honest
or
really
just
wrapping,
what
what
you
all
have
kind
of
put
together
as
part
of
the
ipfs
protocol
and
Lib
p2p,
and
so
just
kind
of
want
to
walk
through
that,
and
then
tell
you
a
little
bit
about
where
we
see
this
going.
I'll
get
some
feedback
from
you
guys
and
just
give
you
some
insight
to
that.
B
That's
really
easy
for
people
to
kind
of
spin
up
invite
people
into
the
network
have
them
join
in
and
it's
a
work
in
progress.
I
mean
we
continue,
because
this
technology
is
is
so
new
that
we're
continuing
to
kind
of
evolve,
network
governance
and
how
people
join
into
those
networks
even
to
some
extent
now
I
mentioned
public.
B
How
these
things
are
starting
to
branch
into
the
public
space
like
public
aetherium
and
things
like
that
from
a
decentralized
ID
perspective,
there's
also
a
lot
of
public
work
going
on
there,
but
one
of
the
bigger
ones
and
I'll
talk
I'll
just
pick
on
core
I'm
here
today,
which
is
a
basically
a
fork
of
etherium
that
allows
us
to
do
things
like
private
transactions.
So
essentially
we
have
our
network
running
with
guests.
You
know
just
go
with
your
in
client
running
then.
B
On
top
of
that
we
have
a
private
transaction
manager
who's
going
to
manage
these
private
transactions
so
essentially
participant
one
wants
to
do
a
private
transaction
with
participant
to
use
disease
keys
in
this
Enclave
of
the
private
transaction
manager
to
encrypt
that
I.
Send
that
to
him
and
then,
when
the
block
shows
up
in
the
blockchain,
only
participant
two
is
allowed
to
basically
decrypt
and
view
that
data.
B
So
that's
that's
typical
companies
like
that,
but
if
you
look
at
what
IP
FS
could
bring
to
the
ring
there
and
it's
something
we
run
into
actually
quite
frequently
as
I
mentioned
before,
in
a
normal
small
use
case.
Technically,
everything
works
right.
We
have
a
smart
contract
written
in
solidity
that
essentially
has
some
state
in
it.
But
then
we
started
branching
out
and
I
think
one
of
the
first
ones
we
we
kind
of
picked
on
was
supply
chain
and
in
supply
chain
management.
B
There's
bill
of
ladings,
there's
all
kinds
of
other
ancillary
documents
and
data
that's
coming
along
with.
You
know
stuff,
that's
going
on
in
the
supply
chain,
so
we
can
certainly
track.
You
know
kind
of
the
work
flow
what's
happening
in
supply
chain,
but
then
we
had
to
basically
store
documents
off
chain,
and
so
you
know
the
first.
You
know
naive
approach
that
was
well.
We
can
just
you
know:
encrypt
files
put
them
in
some
storage
share
them
to
other
people,
and
when
you
have
like
two
parties
yeah,
it
may
work.
B
But
once
you
start
getting
beyond
that
into
you
know,
maybe
1520
even
beyond
that
participants
that
doesn't
scale
and
so
quickly.
The
members
are
saying
like
how
do
we
we,
like
the
blockchain
technology
for
the
state
management,
workflow
kind
of
thing,
but
we
don't
like
is
the
fact
that
we
can't
use
our
files
like
PDF
files
and
things
like
that
to
come
along
with
it.
So
how
do
we
do
that
in
a
scalable
way?
That
makes
it
it
kind
of
fits
with?
B
So,
basically,
let
me
just
jump
out
there
and
show
you
what
we
have
and
feel
free
to
jump
in
here
guys
if
you
have
questions
or
if
I'm
going
too
fast
or
anything
so
essentially,
I
mean
azure.
Here
right
now
we
basically
have
a
catalog
here
and
we
have
ipfs
published
up
there.
We
published
it
on
behalf
of
you
guys
right
now,
so
we
basically
and
I'll
show
you
the
source
code
behind
the
scenes.
B
What
we
did
there
we're
basically
just
wrapping
the
IP
FS
client
in
a
docker
container
and
then
doing
some
orchestration
to
spin
it
up
to
end-users.
They
come
into
the
marketplace
like
the
commander
here
and
they
would
just
say:
I
want
to
create
something
new
as
part
of
that
they
could
type
IP,
FS
and
I.
Have
it
in
beta
here
right
now,
just
because
some
of
the
features
we
don't
have
fully
lit
up,
I'll
show
you
what
those
are
in
a
second,
but
essentially
you
can
hit
create
here
and
then
it'll
basically
walk
you
through.
B
How
big
do
you
want
the
network
to
be
how
many
nodes,
how
fast
you
want
it
to
be,
and
it'll
essentially
go
build
a
bunch
of
IP
FS
nodes,
make
them
private
by
flipping
the
flag
in
lupita,
P,
and
also
you
know,
building
a
swarm
key
and
setting
all
that
orchestration
up.
So
basically
automating
all
those
pieces,
so
you
don't
have
to
deal
with
it
now.
B
I
have
one
stood
up
here
already,
which
essentially
is
just
a
couple
nodes
I
think
for
this
demo,
I
just
had
to
so
I
have
one
in
the
East
Coast
right
here
and
then
I
have
one
in
the
UK.
You
can
see
a
bunch
of
other
stuff
gets
built
around
it.
Most
of
this
is
just
for
the
VM.
So
there's
some
IPs
here
there's
a
virtual
network
that
gets
formed
these
virtual
networks.
We
have
a
feature
in
Azure
that
allows
us
to
join
them
easily
peering.
B
So
if
it
comes
up
here,
second
I'll
show
you
that,
but
essentially
what
it
does
is
allows
us
to
build
a
private
tunnel
using
our
dark
fiber
links
between
the
data
centers,
so
we're
taking
a
second.
We
can
run
through
there
and
basically
you
can
build
these
peering
between
them
so
that
they
can
communicate
just
as
they're
on
the
same
network.
You
can
see
these
are
connected
right
now
and
then,
essentially,
we
have
our
neck
cards.
We
have
some
load
balancers
here
the
load
balancers.
B
They
can
communicate
through
that
one
it'll
just
spread
it
across
our
nodes
in
case
one
of
them
goes
down,
however,
any
of
that
kind
of
stuff
we
have
availability
sets
in
here
to
protect
the
data,
so
basically
they're
durable,
VMs,
that's
pretty
much
it
there
I
mean
when
they're
stood
up,
it's
pretty
basic.
Here's
the
actually!
If
you
look
at
the
there's
a
docker
container
running
on
this
one,
this
is
two
node,
so
this
guys
in
the
East
Coast
this
guy's
in
the
UK
do.
B
B
Hopefully,
that's
a
little
better
yeah,
so
basically
we
just
have
a
docker
container.
You
can
see
it's
running
out
of
one
of
our
private
docker
repos,
so
this
is
part
of
our
build
process.
I'll
show
you
in
a
second
how
that
works
behind
the
scenes,
but
essentially
we
have
a
docker
repository,
that's
running
in
Azure.
We
call
it
the
azure
container
registry
and
then
essentially,
it's
spinning
up
these
nodes.
If
you
look
at
the
logs
for
it
just
to
show
you
what
it
looks
like
when
it
boots
up.
B
Basically,
it's
gonna
essentially
spin
up
download
all
the
bits
for
ipfs.
It's
basically
then
going
to
do
this
bootstrap,
it's
gonna,
remove
all
the
public
nodes
and
then
it's
going
to
generate
a
swarm
key
and
then
it's
going
to
go
ahead
and
boot
up
using
that
key
and
basically
restrict
itself
only
to
private
on
this
interface,
and
this
is
running
right
now.
So
we
have
if
we
did
like
and
primarily
we
just
have
people
come
through
HTTP,
so
I'll
just
show
you
it
locally
here,
but.
B
You
can
see,
I
have
another
peer
here,
which
is
the
guy.
This
is
running
in
the
UK
right
now
and
here
it's
so
what
happened
out
there
whenever
we
spin
these
up
is
and
probably
I
should
walk
through.
Just
show
you
what
it
looks
like
as
you
create
a
new
network
or,
if
you're,
joining
an
existing
one.
What
ends
up
happening
is,
after
the
deployment
happens,
so
you'll
spin
up
a
bunch
of
nodes
here,
you'll
get
some
output.
You
can
use
to
share
with
others.
B
So
basically,
we
have
our
boot
node,
so
we're
spinning
up
a
new
network.
This
is
what
other
people
would
talk
to.
We
have
our
swarm
key
and
then,
if
we
had
a
VPN
gateway,
we
would
do
that.
So
let
me
show
you
what
that
looks
like
a
little
bit
so
for
creating
a
new
one.
Basically
just
tell
it
if
we
want
to
use
you
know,
premium,
SSD
or
regular,
SSD
or
hard
drive
or
cost
savings,
or
any
of
that
kind
of
stuff
that
you
want
to
do.
B
B
The
networks,
where
there's
some
interesting
settings
here
once
it
loads
up
here,
so
essentially
here's
our
compute
nodes.
We
limited
to
ten
right
now
and
that's
primarily
just
a
function
of
how
we
have
some
of
the
template
being
done
right
now,
but
we
could.
We
could
certainly
lift
at.
We
just
haven't
had
a
demand
for
more
than
you
know.
Five
or
six
usually
is
that
is
the
norm.
You
can
basically
pick
the
size
of
the
VM.
How
many?
B
How
much
CPU
and
memory
is
there
and
then
you
could
say
I
want
to
spin
up
a
new
network.
I
want
to
join
an
existing
one.
So
if
you
join
an
existing
one,
you'll
pick
a
member
ID
which
essentially
is
just
used
for
network
governments.
So
if
I
spin
up
a
new
Network
and
then
I
say
hey
and
bite
you
to
it,
I'll
say
you
are
member
number
two
or
you
remember,
number
three.
B
They
would
pick
that
when
they
come
in
and
they
would
also
add
the
swarm
key
into
boutonnat
IP
address
on
that
I
gave
them
out-of-band
we're
looking
to
automate
that
and
I'll
talk
about
that,
and
some
of
the
next
steps
that
we're
looking
to
take
with
this,
but
for
now
it's
kind
of
an
out-of-band
process
so
that
people
have
their
networks
protected.
So
basically
they
just
manually
invite
somebody
to
that,
and
then
you
can
see
here.
You
can
actually
spin
up
a
full
VPN
tunnel
if
you
want
to
do
that,
which
is
basically
IPSec
tunnel.
B
B
Cool,
so
let's
jump
back
over
here
quickly
sake,
we're
15,
so
I
wanted
to
talk
a
little
bit
about
you
know:
that's
that's
some
of
the
basics,
so
just
basically
spinning
up
a
private,
IP,
FS,
knodel
and
a
sure
in
an
automated
form,
joining
a
network
creating
a
new
one
and
that's
kind
of
what
I
talked
about
is
the
first
box
here
of
decentralized,
shared
storage.
So
now,
when
people
spin
up
their
quorum,
cluster
or
hyper
ledger
fabric,
cluster
or
court,
a
cluster-
and
they
say
I-
want
to
have
some
Alabama
storage
for
that.
B
They
can
also
come
in
here
and
use
this
template
and
basically
spin
up
that
decentralized
network.
Invite
the
other
members
that
they
want
to
join
into
that
network
as
part
of
their
whole
blockchain
solution.
Now
some
of
the
things
that's,
you
know
quickly
bubble
up
when
you
start
doing
this
stuff
is
privacy
and
encryption,
there's
a
big
one
that
people
as
well
as
key
management
kind
of
goes
along
with
that,
and
so
what
I
mean
by
that
is
these
files.
B
B
It's
automatically
saying:
oh
I
got
another
file
that
goes
with
that
I'm
going
to
use
this
key
for
encryption
when
the
receiving
party
gets
it
he'll
use
his
key
from
the
Enclave
to
decrypt
it
something
we're
working
on
with
with
quorum
to
see
to
make
that
real,
because
the
reality
is
they
need
to
be
able
to
keep
these
at
file
level.
Basically,
security.
B
The
network
invitation
is
still
a
little
much
out-of-band
right
now,
so
essentially,
as
people
are
onboarding
into
that
network,
it's
very
much
like
I
have
to
contact
you
through
email
or
something
like
that
and
say
hey.
You
know
you
want
to
join
my
network.
I've
stood
up
the
decentralized
network
as
part
of
it.
That's
something
we
have
automated
and
other
templates
with
email.
B
We're
just
gonna,
yeah
I
think
we
should
add
an
optional
section
at
the
end
of
that
it
just
says:
hey,
while
you're
spinning
up
your
your
blockchain
Network
go
ahead
and
spin
up
some
ipfs
nodes,
I'm
just
making
an
optional
component,
so
they
don't
have
to
do
two
different
deployments.
They
can
just
do
it
as
part
of
one
so
I
mean
I
have
some
links
here
and
things
to
this
stuff.
I
showed
you
here
and
we
continue
to
put
sample
code
up
here.
B
We're
gonna
be
doing
some
hackathons
here,
leading
up
to
a
theory,
oh
and
we'd,
like
to
kind
of
build
some
more
of
this
stuff
around
that
as
well
to
help
people.
You
know
with
specific
industry
vertical
things
like
supply
chain
and
showing
the
complete
demo
with
you
know:
here's
how
I
set
up
my
network
with
quorum
and
here's
how
I
PFS
linked
into
that
here's
the
whole
flow.
So
people
don't
get
lost
because
there's
a
couple
templates
right
now
to
spin
that
up
and
then
I
have
a
user
voice
out
there
for
for
anybody.
B
That
has
any
feedback
and
that's
pretty
much
what
we've
been
working
on.
So
we
put
this
out
there
we
get
we'd
be
getting
some
hits
on
it,
so
we
have
some
telemetry
coming
back
around
people
using
it,
and
we've
got
a
lot
of
interest
from
ferment.
Aside
saying
this
is
this
is
what
we
need,
except
for.
We
also
want
that
privacy
aspect.
So
if
you
can
add
that
on
to
be
great,
so
that's
that's
what
we're
at
I
love
to
hear
any
questions
comments.
A
All
right
it's
time
for
questions.
If
you
have
any
questions
or
comments,
if
you
could
put
them
in
a
chat
that
would
be
great
and
kill.
Do
you
mind
if
you
could
stop
sharing
sure
ask
them?
Thank
you.
Yeah
I
can
see
everyone's
faces
now.
Okay,
so
thank
you
very
much
for
their
presentation
and
let
us
start
with
questions
and
we
will
start
with
Johnny
crunch
Johnny
crunch.
Can
you
please
tell
us
what
is
your
question?
Yes,.
C
B
A
B
B
If
you
see
my
screen,
this
is
kind
of
like
the
internals
like
how
our
templates
are
kind
of
built
out.
This
is
our
docker
build
over
here
and
basically
the
steps
that
go
along
with
that
and
then
essentially,
whenever
it
gets
deployed
essentially
on
the
VM.
We
run
this
bash
script,
so
this
installs
docker,
we
essentially
lock
the
version
in
here,
so
we're
locked,
a
4:17
right
now
and
then
essentially,
when
we
pull
those
bits
down.
Actually
there
is
part
they're
packaged
as
part
of
the
docker.
B
That's
right,
so
they
get
pulled
in
into
this
and
we're
just
doing
a
double
you
get
here
right
now,
centaur
beta,
we
don't
compare
the
md5
hash
or
any
kind
of
thing
on
that.
So
maybe
that's
something
we
should
do
in
a
future
to
guarantee
the
bits
that
we're
pulling
from
ipfs
are
legitimate
and
then
essentially,
the
swarm
key
that
gets
generated.
I
have
a
little
application
inside
of
here,
it's
a
little
node
based
application
that
generates
the
swarm
key
and
then
a
swarm
key
is
essentially
only
only
exposed
on
my
network.
B
If
you
go
and
look
at
the
template
output,
it's
treated
as
like
a
quote-unquote
secret,
which
basically
means
it
doesn't
show
up
in
any
of
our
log
files
or
any
of
that
kind
of
stuff.
So
it's
there.
It
exists
as
the
network
spinning
up
and
I
can
choose
to
share
it
with
people,
but
as
far
as
like
from
a
privacy
standpoint,
we
don't
leave
that
hanging
around
or
out
there
publicly
exposed
kind
of
thing.
B
So
there's
been
a
lot
of
work
in
it
specifically
I've
been
working
with
both
the
consensus
group,
so
christian
Rovin
and
those
guys
from
you
port,
as
well
as
Daniel
Buckner
on
our
side,
who's
part
of
the
D
ID
and
a
diff
as
a
whole.
An
ahkam
is
the
other
trio.
I
mean
there's
others
involved
in
the
space.
Obviously,
but
those
are
the
ones
I've
been
mostly
focused
on
and
the
idea
is.
B
B
This
way
as
people
are
onboarding
to
like
these
ipfs
node
clusters
or
even
the
quorum
or
you
know,
hyper
ledger
networks
or
whatever
we
can
leverage
that
to
say:
hey
he's
in
accounting
or
he's
in
the
warehouse,
those
kind
of
things.
So
whenever
that
person
gets
fired
or
moves
on
from
the
company,
we
don't
have
problems.
You
know
coming
back
there
and
essentially
the
decoupling
of
the
key
mechanism
as
part
of
the
di
D
spec.
C
So
I'm
working
on
this
right
now
and
did
you
have
any
documentation
or
a
pointer
to
insure
interfacing
with
the
Enclave
with
the
Corp
specifically
or
Microsoft
yeah.
C
B
There
of
you
sure,
if
you're
part
of
the
diff
or
any
of
the
D
ID
you've
seen
we
just
released
the
so.
Basically
the
attenti
cater
app
that
we
had,
which
was
like
the--.
We
basically
are
extending
that,
and
so
they
put
a
beta
out
of
that.
So
I
can
I'll
send
a
link
to
that
as
well.
If
you're
interested
to
check
it
out,
it's
mostly
for
mobile,
you
know,
use
the
Enclave
in
the
phone
to
also
do
the
key
storage
and
signing.
C
Cool
and
just
a
shameless
self
plug
is
I
just
put
in
the
notes.
The
D
ID,
the
IP
ID,
and
you
planetary,
identifier,
is
the
ID
methods
back
is
up
and
we
certainly
welcome
feedback.
It's
a
draft
and
I'm
going
to
Barcelona
into
the
week
to
actually
to
discuss
it
with
the
working
group
with
w3c
awesome.
A
And
all
these
links
will
be
available
in
the
show,
notes
and
I'll
also
put
them
in
the
YouTube
video
once
this
video
goes
up
so
we'll
have
all
that
stuff
available,
and
next
question
from
dan
dan
would
like
to
know.
Is
the
source
code
available
for
what
you
just
presented
to
us
today?
So
can
we
find
a
source
code
yeah.
B
I'm
happy
to
share
the
source
from
the
template.
To
be
honest,
it's
not
really
hidden!
So
when
you
deploy
stuff
in
Azure
the
last
step
right
before
you
created,
it
says,
download,
template
and
parameters
most
of
the
sources
there,
but
I'm
happy
to
share
it.
We
don't
have
it
in
a
public
repo
as
of
yet
but
there's
nothing
secret
there
so
I'm
happy
to
share
it
with
the
group.
Ultimately,
we
want
to
make
that
open
source.
B
You
know
just
get
that
out
there,
but
it
does
take
a
little
bit
of
work
inside
Microsoft
for
me
to
get
all
that
stuff
done.
So
we're
we're
just
getting
started
here.
No
one
happy
to
share
it
privately.
So
if
anyone
needs
it,
you
know
right
now,
I'm
happy
to
share
it
with
this
group.
I
can
just
archive
it
and
zip
it
up
or
something
until
we
get
it
in
a
public
repo.
A
B
Most
of
its
encrypted
right
because
most
enterprises
or
companies
that
are
working
in
space,
it's
like
a
need-to-know
basis
for
transaction
details,
and
so
they
would
much
rather
opt
people
into
that.
Then,
by
default,
have
it
shown
to
everyone
that
can
vary
some,
but
that's
the
general
rule
is
that
people
want
to
see
that
stuff
encrypted
and
basically
even
as
needed
basis.
They
allow
people
to
see
it
kind
of
thing.
A
Okay,
we
have
time
for
one
more
question:
do
you
have
another
question?
I
have
a
question
throughout
the
presentation
you
did
mention
encryption,
but
I'm
wondering
Kail.
Do
you
have
a
wish
list
of
what
kind
of
features
you
would
like
to
see
coming
from
ipfs?
A
B
A
good
question,
yeah
I
mean
to
be
honest,
it's
been
really
easy,
working
with
the
binaries
and
things
that
the
builds
that
I've
gotten
out
of
you
guys
today,
I
guess
the
bigger
question
I
would
have
is
the
feature
I'm
using
there
in
Lib
p2p,
which
is
basically
that
flag
to
tell
it
to
do
private
as
well
as
so
noticing.
What
public
is?
That's
labeled
experimental
I'm
wondering
what
the
roadmap
is
there
for
that
is
this
something
the
you
know.
B
A
Depth,
do
we
have
anyone
who's
working,
a
limp?
You
peanut
could
answer
that
question
on
a
call.
If
not
I,
we
can
definitely
follow
up
help.
Okay,.