►
Description
Every other week, the Retrieval Market Builders get together to share progress in their projects in a demo format. We want to sincerely thank all of our collaborators for demoing their developments and helping to improve the FIL Retrieval Markets one demo at a time.
June 22, 2022, Demo Day: In this video, you can find quick demos from -
• CryptoNetLab on Retrieval Pinning
• Ken Labs on Pando ID
Follow Filecoin!
Website: https://bit.ly/3ndAg44
Twitter: https://bit.ly/3ObND0x
Slack: https://bit.ly/3HKfFy7
Blog: https://bit.ly/3HFZFNv
Reddit: https://bit.ly/39N4Jmv
Telegram: https://bit.ly/3bkP8Ly
Subscribe to our newsletter! https://bit.ly/3Oy8J9j
#filecoin #ipfs #libp2p #web3 #nft
A
Hello
and
welcome
to
this
retrieval
market
demo
day.
Today's
date
is
wednesday,
the
17th
of
august
2022
and
we've
got
a
couple
of
very
exciting
talks
today.
Firstly,
we're
going
to
hear
from
crypto
netlab
about
the
retrievability,
retrievability
oracle
work
and
then
we're
going
to
hear
from
ken
labs
about
their
latest
work
on
pando
and
potentially
a
few
other
things.
If
anyone
has
any
questions,
please
just
put
them
in
the
chat
and
we
can
speak
about
them
afterwards.
B
Yes,
perfect?
Thank
you.
So
what
I'm
going
to
talk
to
you
about
is
this:
the
new
protocol
that
we
have
from
cryptonap
the
retrieval
pinning
protocol-
and
this
is
one
of
the
many
services
that
we
think
are
necessary
to
have
for
the
user
to
be
able
to
really
use
the
centralized
storage
network
in
their
everyday
life
as
useful
tools
and
the
question.
B
B
So
let
me
start
with
a
high-level
recap
of
about
decentralized
storage
network.
Usually,
what
happens
there
is
that
we
are
one
user
that
store
the
the
data
with
one
or
more
providers
the
nodes
of
the
the
centralized
storage
networks,
and
there
is
a
proof
of
storage,
usually
mechanism,
meaning
that
no
one
can
lie
about
storage.
The
providers
can
prove
that
they
are
storing
only
if
they
are
really
storing
and
clients
cannot
say.
We
have
things
about
honest
providers
if
they
provide
their
historians.
B
Clients
cannot
save
the
country
okay,
so
we
are
all
happy
with
this,
but
about
retrieval
there
is
not
really
any
proof
about
retrieval.
If
it's
happening
not
happening
is
usually
just
said
that
you
know
there
are
these
copies
of
the
data
around
and
there
is
at
least
one
honest
provider
that
has
the
copy
of
the
date
and
will
serve
the
data
when
the
client
requests
that
that's
what
we
want
to
solve.
We
want
to
go
from
just
assuming
there
is
one
honest
guy
to
have
more.
You
know
a
strong
argument
supporting
that.
B
B
B
B
B
If
he
sees
some
reason,
there
is
a
problem
between
the
the
client
of
the
provider
and
there
is
a
dispute
going
on,
because
one
of
the
two
says
none
that
the
file
is
not
the
the
the
the
retrieval
is
not
going
as
as
prom
said,
then
it
can
come
in
and
slash
one
of
the
two
one
they
can
can
discover.
Who
is
lying
and
slash
in
the
sense
of
make
some
action?
That
is
the
punishment
for
the
lying
dishonest
part.
B
We
implemented
this
third
solution
in
an
efficient
way.
This
is
what
is
the
retrieval
pinning
protocol.
If
you
want
to
know
more
about
the
the
design
we
have
a
light
paper,
you
can
access
it
by
the
qr
code,
and
here
I
will
give
you
just
the
high
level
intuition
how
we
do
this
and
the
key
ingredient
is
that
we
have
a
fixed
referee.
Nectar
of
referees
referees
are
special
nodes
that
are
online,
watching,
basically
not
really
just
listening.
B
If
there
is
a
dispute,
if
there
is
a
dispute,
they
are
calling
and
we
have
to
solve
the
disputes
right
now
we
have
five
and
we
need
only
to
trust
three
of
them
they
all
together.
They
are
our
oracle
right
and
five
is
just
the
number
that
we
are
using.
Now
it's
a
good
compromise
between
efficiency
and
security.
You
can
think
about
any
other
number
there,
and
once
that
you
have
that
the
protocol
is
simple,
client
and
provider.
B
So
this
is
they
both
sign
a
provider
and
and
client
at
the
bot
sign.
That
is
this
deal
that
is
public
on
chain
later
on
they
go
and
do
they
they
they
their
own
stuff,
in
a
sense
that
each
time
that
the
client
wants
to
access
the
file
task
to
the
provider,
the
provider,
if
gives
back
the
file.
If
this
is
everything,
is
good.
We
don't
see
anything
else
in
our
protocol,
but
if,
at
some
point
there
is
a
problem,
the
client
can
appeal.
B
So
the
client
can
say
to
the
to
the
referees
and
can
say
to
the
referees:
hey
help
me,
because
this
provider
is
not
giving
back
me
the
the
date
he
promised
to
retrieve,
and,
of
course,
this
client
can
lie.
This
could
be
just
a
malicious
client
trying
to
get
an
honest
provider
slasher
for
some
any
reasons
right,
maybe
competition.
You
know
you
don't
know
so
the
the
clients,
the
sorry
the
referees
can't
trust
the
client,
what
they
do.
B
I
don't
want
to
go
to
details
here,
but
we
have
all
the
mechanisms
to
do
this
in
a
reliable
way,
but
for
the
client
and
the
providers.
If
you
want
to
know
more,
go
to
the
paper,
don't
want
to
give
details
now
to
to
boring.
Maybe
it's
so
interesting,
I
think,
to
see
the
hub,
so
you
can
play
with
this.
These
protocols
implement
and
then
I
ask
you
and
suggest
you
to
play
and
give
us
feedback
as
much
as
you
can
and
what
I'm
going
to
show.
B
You
now
is
the
video
of
what
happens
on
the
client
side,
so
the
client
interacts
with
our
product
through
an
app
and
let's
see
what
happens
so.
First
things
you
have
to
connect
with
your
wallet
and
once
that
you
get
inside
the
the
app
you
see
all
the
deals
that
you
have
done
already.
If
it's
not
the
first
time,
of
course,
that
you're
using
this
for
each
deal,
you
can
click
and
you
have
information
about
the
deal.
So
the
things
that
we
have
seen.
B
This
is
a
preview
of
the
file
and
you
also
see
all
the
information
about
price.
The
the
this
is
the
file.
Let's
see,
I
did
fsc
id,
which
provides
the
deal
with
the
blah
blah
blah
deliberation,
and
so
on
note
that
there
is
also
an
nft
logo
here,
because
they
are
visualized
as
nfps.
We
think
this
is
very
useful
for
for
for
for
using
this
bootstrap,
the
use
of
this
application,
and
you
have
the
information
about
this
protocol,
the
referees,
how
many
nodes?
How
many
are
the
referees?
B
How
many
appeal
you
can
do
because
of
course
we
are
all
these
parameters,
the
audience
say
you
too
much.
But
yes,
the
deal
in
the
chapter
question.
Yes,
the
deal
is
an
nft,
it's
visualized,
then
it
is
the
deal
is
public
information
chain
and
we
also
create
an
nfc
that
visualize
that-
and
I
was
saying
you
have
all
the
information
here
about
the
protocol
parameters.
B
How
many
appeals
you
can
do
we
have
a
and
we
have
a
maximum,
of
course
to
avoid
you,
know
classic
attacks
and
how
many
rounds
so
many
times
the
providers
that
try
with
the
client
where
the
referee,
so
I
try
to
to
recover
the
file
for
you,
blah
blah
blah.
These
are
protocol
parameters
that
can
be
changed,
but
are
public
here
and
what
else
we
have.
B
Of
course,
you
can
create
new
deal.
The
most
important
part
right.
Creating
a
new
deal
is
quite
simple:
you
just
upload
the
file
this
create
the
ipfs
id.
Let's
see
the
cid
select,
the
provider
select
the
duration,
payment
and
collateral,
and
in
this
base
mode
all
these
values
are
basically
suggested
by
the
app
itself.
So
the
client
just
choose.
B
B
You
can
play
a
bit
more
with
our
with
our
app
meaning
that
all
the
parameters
that
are
needed
for
the
deal
now
are
free
to
choose
by
the
client,
so
the
app
doesn't
suggest
any
default
value,
but
you
are
free
to
play
the
market
as
much
as
you
want.
Of
course,
there
is.
There
are
more
lists
in
this
case
and,
as
you
see,
all
these
cars
are
showing
this
in
the
warning.
The
box
warning,
but
you
can
do
this.
B
B
I
think
we
have
just
seen.
Also,
as
you
see
you
can
okay
here,
it's
happening
that
once
the
deal
is
active,
that
is,
the
provider
on
the
other
side
is
also
signed.
The
what
happens
is
that
you
can
ask
for
an
appeal,
so
the
client
can
go
back
here
at
any
point.
If
something
is
happening,
something
is
not
working
with
the
retrieval
and
create
the
appeal.
This
is
just
the
the
button
here.
B
If
you
click
on
the
on
the
deal
you
see
how
much
of
like
you
see
the
process
of
the
appeal
from
the
fading
director
and
then
you
will
get
the
file
back
or
if
you
don't
get
the
file
back,
the
provider
is
going
to
lose
the
entire
collateral
that
they
put
on
stake
at
the
beginning
of
the
when
he
signed
the
deal.
Basically,
so
this
is
the.
This
is
the
side,
the
side
of
the
client.
I
want
to
show
you.
B
What's
the
provider
for
the
provider,
we
designed
a
command
line
interface
and
the
code
is
public
on
github
here,
the
qr
code.
That
brings
you
directly
to
the
repo
where
you
can
download
it
and
just
play
as
a
provider
with
this
app,
and
I
want
to
show
you
a
couple
of
videos.
Tell
me
patrick
if
I'm
over
time
or
if.
B
Thank
you
practice.
I
also
also
will
show
you
a
couple
of
video
about
how
the
provider
come.
Online
looks
like
now
and
the
first
step
that
the
provider
has
to
do
if
he
wants
to
play
with
his
protocol.
He
has
to
sign
up,
of
course,
sign
up
mainly
meaning
he
has
to
send
an
address
where
we
can
connect
with
him
referees
and
all
the
others
party
can
connect
with
him
to
exchange
the
files
and
all
other
messages.
This
is
not.
Oh
sorry.
I
want
one
slide.
This
is
the
sign
up.
B
Sorry,
yes,
and
you
see,
this
is
basically
sending
the
public
address
and
end
point,
save
it
and
by
saved
correctly
to
the
disk
and
provide
that
added
corrector.
This
means
that
this
provider
now
is
in
the
list
of
providers
with
this
address,
and
you
know
what
in
the
app
we
will
show
up
with
this
name,
if
you
choose
one
or
just
with
the
address
and
next
step
for
the
provider
after
the
successful
sign
up,
is
the
decided
policy.
So
what
is
the
policy?
B
So
these
are
all
the
parameters
that
the
the
provider
can
choose
and
set
out.
Of
course,
the
smart
contract
has
some
default
value.
We
suggest
default
value,
so
if
the
provider
does
do
nothing,
there
are
default
values
do
that
for
for
for
security,
we
say
here,
but
we
suggest
any
provider
to
set
up
their
own
values,
and
here
you
see
the
recap
of
basic
of
the
policy,
so
the
minimum
price
is
how
many
tokens
per
byte
per
seconds
is
the
minimum
for
providing
the
retrieval
steps
that
I
want
me
to
paid
as
a
provider.
B
This
is
the
minimum
the
max
size.
So
I
don't.
I
don't
accept
for
a
deal
a
size
that
is
larger
than
this.
I
don't
accept
bills
that
are
longer
than
one
year,
and
this
is
the
collateral
multiplier.
So
I
don't
accept
deals
that
ask
me
to
put
as
collateral
than
1
000
times
more
than
how
much
I
get
paid.
This
is
the
the
policy
and
now
that
we
have
basically
the
setup
of
the
provider
provide,
is
ready
to
work.
B
We
will
see
now
what
happens
on
this
command
line
when
we
asked
to
do
when
we
asked
you
to
accept
a
deal-
and
this
is
a
video
of
this
stress
test-
that
we
are,
we
we
did
and
we
are
doing
and
on
the
left.
Sorry,
on
the
right.
You
see
basically
the
the
code
for
the
stress
test
that
is
going
to
create
deals
for
this
provider
on
the
left
and
left-hand
side.
B
You
will
see
the
this
is
basically
the
the
terminal
of
the
of
the
of
the
provider
and
right
now
the
cash
is
empty,
meaning
there
is
no
proposal
for
for
for
this
provider,
and
you
will
see
now
that,
once
that
this
goes
on
chain
that
the
transaction
for
these
deals
is
is
accepted,
the
the
provider
will
the
command
line
of
right
will
start
to
process
the
request
of
doing
of
accepting
this.
That's
exactly
what
you
see
now
dealer
expires.
So
there
are
some
the
new
deal
proposal
creator
processing.
B
There
are
some
recap
for
the
providers
file,
statistics
about
size,
blocks,
hush,
etc,
etc,
and
deposit
needle
recap,
and
can
accept,
meaning
that
everything
is
in
the
policy.
The
provider
policy
and,
of
course,
for
accepting,
as
you
see
the
provider,
has
to
sign
that
is
accepting
sending
a
transaction
online.
B
B
This
idea,
that
was
this
is
one
year
ago,
was
just
an
idea
to
go
from
just
a
prototype
to
a
real
product
that
can
be
tested
by
users
and
hopefully
can
launch
later
on
this
this
year.
What
we
are
discussing
now
and
really,
if
your
feedback-
please
please
tell
us-
is
we
think
that
for
a
product
it
will
be
much
of
more
than
just
only
one
referee
network,
so
we
would
like
to
have
a
different
referees
network
with
maybe
different
parameters
and
and
then,
let's
say,
service
agreement
parameters.
B
Of
course
we
want
to
contact
the
strategy
provider.
There
was
a
question
before
I
think
yeah.
So
let
me
this
is
a
good
point
to
ask
peter
question:
it
doesn't
interface
on
fire,
I
mean
the
interface
of
fresh
icon
will
be
go
to
file
kindness
storage
provider
and
ask
and
to
sign
up
the
simple
way
I'll
show
you
before
and
start
to
test
this.
We
haven't
done
that
yet
because
we
it
was
too
early.
It
wasn't
that
all
the
code
wasn't
crazy.
We
are
going
to.
B
We
are
planning
to
do
that
next
month
for
starting
from
next
month
minimum,
and
we
can
set
up
parameters
to
help.
You
know
the
providers
to
to
to
use
this
in
the
in
the
best
way,
and
also
we
have
to
find
the
referees
refer.
The
five
the
referees
not
right
now
are
just
all
controlled
by
us,
but
that's
not
the
way
you
want
to
do
this
right.
You
want
to
ever
decentralize
it
as
much
as
you
can
decentralize
solutions.
B
So
we
really
would
like
other
institutions
teams
project
to
to
run
one
of
the
of
the
of
the
nodes
that
implement
a
rfa.
So
we
are
thinking
about
making
a
refereeing
program
to
onboard
people
that
may
be
interested
to
to
run
this
again
contact
us.
If
you
have
interest
here,
then
of
course
this
is,
as
patrick
saying
is
in
the
chat.
B
This
is
for
now
is
written
for
working
on
ethereum
and
it's
live
on
a
rinkabyte
one
of
the
ethereum
test
nets,
but
we
want
to
think
about
how
we
can
bring
this
to
work
with
the
falcon
virtual
machine.
Now
that
we
are,
we
have
one,
and
this
requires
some
maybe
request
something
and
and
of
course,
the
design
work
and
last
but
not
least,
is-
and
this
is
x
a
bit
like
this
is
all
for
very
focused.
The
three
first
three
points
very
focused
on
this
project.
B
So
we
would
like
to
generalize
the
interface
that
you
have
seen
for
adding
the
retrieval
pinning
but
also
other
services.
Other
cell
services,
for
example,
could
be
automatic
renewal.
What
we
call
the
perpetual
storage
contractor,
meaning
that
I
do
a
storage
contract
for
six
months,
but
I
want
to
have
a
method
that
actually
automatically
renew
the
storage
deal.
Everything
as
long
as
there
are
some
tokens
found
phones
in
a
in
the
in
a
wallet
in
a
contract.
C
Yeah
I
had
a
question
hi,
I'm
george.
By
the
way
I
work
with
the
microwave
team
and
we're
looking
at
payments
in
the
retrieval
market
by
state
channels,
payment,
charles
and
if
I
understood
your
presentation,
which
is
very
nice
by
the
way
correctly
you're,
adding
like
an
additional
economic
incentive
for
providers
to
like
actually
provide,
so
they
can
get
slashed
punished
if
they,
if
they
don't
provide
the
file.
C
How
do
you
see
that
incentive
mechanism
working
alongside
the
existing
thing,
where,
if
you
don't
provide
the
file
you
don't
get
paid
by
the
client
like?
Is
it
going
to
be
like
a
much
heavier
penalty
than
not
being
paid
for
like
very
key
files,
or
something
like
that
like?
How
do
you
see
those
two
things
living
alongside
each
other.
B
Yeah,
as
you
said,
is
we
think
that
missing
the
reward,
the
payment,
the
normal
payment?
It's
not
it's
not
enough
as
punishment,
often
because
maybe
the
payment
for
retrieval
it
is
not
clear
how
much
the
driver
is
going
to
pay
to
be
paid.
If
you
think
about
always
do
the,
I
think
it's
useful
to
understand.
This
is
always
doing
a
comparison
with
the
storage
and,
for
example,
file.
Client
right.
So
in
filecoin
storage
is
not
really
paid
by
the
client.
B
There
is
the
block
level
or
the
subsidized,
the
the
the
storage
right,
and
if
you
fail
to
provide
storage
like
specifically
in
five,
can
you
feel
the
the
daily
check
that
is
called
windowpost?
Of
course
you
will
miss
for
24
hours
sunblock
reward,
but
we
also
need
to
penalize,
because
that
is
not
enough.
B
You
are
actually
breaking
this
love
with
the
provider
and
that's
not
enough
because
sorry
with
this,
like
there's
love
with
the
with
the
client
and
that's
not
enough
in
the
crypto
economic
mechanism,
so
there
are
some
penalizations
that
is
going
on
and
we
think
we
need
the
same
for
retrieval
to
really
incentivize
the
provider
to
care
about
retrieval
as
much
as
they
hear
today
about
storage.
Does
it
make
sense.
C
B
Definitely
definitely
that's
why
they
should
not
be
slashed
if
there
is
like
a
small
dropping
service
and
that's
why
the
the
this
referee
connector
has
to
ask
to
the
pro
we
do.
We
design
the
the
appeal
process
of
the
referee
in
a
way
that
we
give
more
than
one
attempt
to
the
provider,
to
prove
that
he's
honest,
of
course,
if
you
have
to
that's,
why
it's
difficult
you
can
just
lash
every
time,
just
the
client.
Yes,
yes,
definitely.
A
A
B
So
we
trust
there
is
in
a
way
we
trust
the
referee
to
check
each
other
kind
of.
So
if
enough
of
them
are
honest
and
usually
it's
half
of
them
what?
But
this
is
a
parameter
that
we
can
play
with
if
a
good
fraction
of
them
are
honest,
then
we
are
fine.
One
one,
like
a
small
fraction
of
dishonest
referee,
can't
make
anyone
lose
collateral
money.
A
D
D
D
D
However,
the
vast
majority
of
relationships
between
people
organizations
and
things
have
simpler
requirements
when
alice
and
bob
want
to
interact.
There
are
exactly,
and
only
two
parties
in
the
world
who
should
care
alice
and
bob
instead
of
arbitrary
parties
needing
to
resolve
their
diades.
Only
alice
and
bob
do
peer.
Deities
are
perfect
in
these
cases.
D
D
D
Nice
use
case,
any
unknown
party
can
request
the
id
document
resolved
to
a
public
service.
For
example,
blockchain
service
pandora
is
not
that
public,
so
in
this
case
pendle
and
the
peers
that
persist
their
data
into
pando
or
consume
from
pando
should
interact
with
a
known
party
only
and
the
id
pair
exactly
meet
the
requirement.
D
D
D
D
D
Step
in
pando's
plan
is
that
we
will
implement
the
id
peers
layer.
2.
pando
will
store
the
id
docs
on
the
dock
chain
provider
and
consumer
use.
The
secret
key
generated
by
the
ecdh
exchange
to
encrypt
the
data
during
data
synchronization
we'll
develop
a
kms
as
a
part
of
pando
for
key
management,
and
we
need
to
discuss
more
details
about
the
implementation.