►
From YouTube: Proofs-of-Replication - Filecoin Research
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
B
A
A
This
thing
called
a
proof
of
retrieve
ability,
which
is
a
proof
that
you
actually
have
in
your
possession
a
certain
file
that
the
verifier
is
aware
of
and
approve
of
space
and
so
approve
replication
is
kind
of,
at
the
same
time,
a
proof
of
retrieve
ability
for
a
particular
file,
or
it
can
be,
and
also
a
proof
of
space.
So.
B
In
a
high-level
the
protocol
works
in
this
file
in
this.
In
this
particular
way,
the
prover
commits
to
saw
some
data
and
runs
an
encoding
on
this
data.
The
encoding
is
meant
to
be
slow
because
in
a
later
point
the
verifier
will
be
challenging
the
prover.
If
the
prover
is
able
to
respond
quickly,
it
means
that
they've
done
the
encoding.
If
the
prover
doesn't
reply
quickly,
they're
able
to
generate
some
data
on
the
spot
and
lie
about
the
storage
that
they
have
proved.
C
The
replication
are
a
critical
component
for
the
Falcon
Network.
Is
it's
the
the
main
algorithm
that
that
the
network
uses
to
make
sure
that
miners
are
behaving
correctly,
that
they
are
indeed
storing
the
data
that
they
promised
the
store
for
clients
in
a
way
proof
of
replication
is
the
the
heart
of
file
coin.
In
the
same
way,
that
proof
of
work
is:
is
the
heart
of
Bitcoin
file.
A
Coin
is
using
it
in
sort
of
an
extra
interesting
way,
because
proof
replication
is
not
just
a
proof
again
of
retriever
ability
is
a
proof
of
space.
It's
a
proof
of
useful
space
and
file
coin.
You
know
like
Bitcoin
or
aetherium,
is
a
decentralized
network
with
with
with
a
ledger,
and
it
uses
a
civil
resistance
mechanism
like
proof
of
work
in
Bitcoin
file,
coin
uses
proof
of
space,
but
while
bitcoins
proof
of
work,
just
kind
of
you
know
wastes
energy
file.
Coins
prove
the
space,
is
a
byproduct
of
the
system,
actually
storing
files.
Recently.
B
A
A
proof
of
replication
was,
was
a
very
you
know,
tricky
tricky
task,
remember
the
the
key,
difficult
thing
to
achieve
with
the
proof
of
replication.
Is
this
verifiable
proof
that
you're
using
space
to
store
a
particular
file?
Ok-
and
we
want
to
achieve
this
in
a
cryptographic-
sense
right
and
sort
of
the
the
best
possible
thing
that
you
can
do
kind
of
is
basically
achieve.
What's
called
it
a
tight
proof
of
space
and
it's
basically
a
proof
of
space
where
you
really
can't
save
anything?
A
Ok,
and
so,
if
you're,
if
you're,
if
you're
producing
a
proof
of
space,
where
you
really
can't
save
anything,
then
you
might
as
well
store
a
movie
file
right
and
then,
on
top
of
that,
you
can
give
a
proof
of
retrieve
ability
of
a
particular
movie
file
to
users.
But
from
the
from
the
perspective
of
the
the
proof
of
space
verifiers,
they
know
that
you
can't
save
anything.
A
So
you
might
as
well
use
a
space
to
store
what
you're
claiming
to
store,
and
so
that's
sort
of
the
the
property
you
wanted
to
achieve,
and
it
actually
took
us
a
while
to
iterate
and
and
make
these
proofs
of
space
tighter
and
tighter
and
tighter
until
we
got
to
something
which
really
had
the
security
property
that
we
wanted.
So
it
started
with
kind
of
just
a
basic
proof
of
space
based
on
this
and
graph
thing
called
a
depth
robust
graph
and
and
it's
something
it's
a
tool.
A
That's
been
used
a
lot
in
computer
science,
but
we
were
using
it
to
conserve
basic
proofs
of
space
and
then
they
weren't
quite
tight
enough
right.
So
the
adversary-
who
might
you
know,
save
you,
know
a
small
percentage
of
the
data
or
actually
even
half
of
the
data,
and
that
wasn't
good
enough
for
us
and
and
therefore
we
were
improving
on
it,
and
then
we
got
to
kind
of
combining
it
with
these
other
things
called
expander
graphs,
which
have
also
been
used
for
proofs
of
space.
A
So
we
took
these
two
different
techniques
that
were
being
used
to
construct
proofs
of
space
that
weren't
really
good
enough,
and
we
sort
of
combined
them
together
in
an
interesting
way
and
ended
up
with
something
which
was
the
tightest
proof
of
space
that
ever
existed.
And
so
it
was
not
only
necessary
for
file
claim,
but
it
was.
It
was
a
contribution
to
the
to
the
entire
literature
of
proof
of
space.
Since
is
the
tightest
perfect
space
that
there
is
and
it
was
actually
necessary
for
file
coins
application.
A
Any
type
of
space
can
be
used
to
build
a
proof
of
replication
with
inefficient,
unsealing.
Okay,
inefficient
data
extraction
fast
unsealing
is
harder
to
achieve,
and
that's
what
we
achieved
with
with
zigzag,
which
is
not
just
a
type
of
the
space.
But
a
type
of
space,
with
fast
data
extraction,
where
you're
using
the
proof
of
space
to
incentivize
file
storage,
and
you
want
to
make
sure
that
whoever
produces
this
proof
of
space
will
be
using
it
to
store
files
of
interest.
A
Then
you
need
a
guarantee
that
there's
no
reason
that
the
adversary
would
do
something
else
that
would
destroy
the
file
storage
format
and
therefore
lose
the
data
and
gain
an
advantage
in
passing
the
protocol.
And
so
that's
where
you
really
need
zigzag.
You
needed
to
be
as
tight
as
possible
so
that
you
know
there's
no
there's
no
cost
savings
for
not
following
the
honest
protocol
and
then
the
honest
protocol
is
the
only
way
to
ensure
retrieve
ability
or
lossless
encoding.