►
Description
Topics covered:
1:25 - Introduction
3:55 - What is Flash loan, Flash mint?
5:57 - Use cases
12: 30 - Prerequisites protocol to be able to have a flash loan & flash mint?
19:50 - The main rationale for creating a standard flash loan?
22:05 - If any of the current providers want to use it, do they have to migrate the contract?
23:50 - How useful this proposal is?
26:18 - What makes flash loans easier/harder for a platform to adopt?
30:05 - Built-in safety. What would it mean for there to be an unsafe flash loan, if the loan has to be paid inside the same transaction?
34:00 - How a flash loan could be used to make refinancing easier?
36:45 - The purpose, place, or meaning of audits are in the DeFi space for a user? What does audit cover/not cover?
40:30 - When do you think this proposal would be ready to move to the last call?
EIP - https://eips.ethereum.org/EIPS/eip-3156
Discussion to - https://ethereum-magicians.org/t/erc-3156-flash-loans-review-discussion/5077
Slides - https://docs.google.com/presentation/d/1Z8klLzd1P5sky1pmRzg7nFMCjgKMrAnvjI-eSWitD9w/edit?usp=sharing
Follow at Twitter:
Alberto Cuesta Cañada @acuestacanada | Pooja Ranjan @poojaranjan19 | William Schwab @William94029369 |
Check out other exciting EIPs in Peep an EIP series at https://github.com/ethereum-cat-herders/PM/projects/2
Contact Ethereum Cat Herders
Discord: https://discord.gg/tzYmDmF
Twitter: https://twitter.com/EthCatHerders
Medium: https://medium.com/ethereum-cat-herders
GitHub: https://github.com/ethereum-cat-herders/PM
Email: support@ethereumcatherders.com
Website: https://www.ethereumcatherders.com/
A
Hello
and
welcome
to
the
eap,
dedicated
series
p
penney,
where
we
discuss
an
ethereum
improvement
proposal,
understand
its
importance
and
talk
about
how
it
is
going
to
make
the
ethereum
ecosystem
even
better.
I'm
your
host
pooja
ranjan
and
we
are
joined
by
william
shaw
from
ethereum
cat
herders
maker,
the
eip
editor
and
some
community
members.
Along
with
the
author,
of
course,
today
we
are
going
to
talk
about
an
interesting
eip
3156
flash
loan.
A
This
is
a
fairly
new
proposal
for
a
rapid,
growing
use
case
and
ethereum
ecosystem,
that
is
decentralized
finance,
also
known
as
d5,
to
provide
an
overview
of
a
maybe
future
standard
for
flash
loan.
We
are
joined
by
one
of
the
authors,
alberto
kuseyta,
who
is
also
the
technical
lead
at
yield,
a
d5
project
and
usm
a
stablecoin.
A
B
Hi,
so
I'm
alberto
cuesta,
canada,
everyone
gets
to
run
the
name.
That's
fine,
so
I'll
be
a
little
bit
brief
about
me.
I'm
a
little
bit
on
the
on
the
senior
side,
so
I've
been
in
it
for
quite
a
while.
I
don't
want
to
talk
too
much,
I'm
currently
the
lead
engineer
at
yield
a
paradigm
backed
fixed
rate
lending
platform.
B
That's
what
I
do
for
for
during
the
day
and
during
the
night
I
take
side
projects
and
and
hack
away
with
them.
Those
side
projects
currently
are
usm
farm,
which
is
a
stablecoin,
the
one
that
you
mentioned
with
10,
which
is
the
the
ether
wrapper.
It's
a
new
version,
which
is
also
the
project
that
brought
me
to
the
erc3156,
the
flash
landing
I
did
flash
lending
first
for
for
yield.
That's
where
I
learned
from
austin
williams
and
georgios
what
the
pattern
was.
B
Then,
when
I
got
involved
with
the
new
ether
rapper,
I
said
hey,
I
don't
know
how
to
do
flashlight
and
let's
put
this
into
as
well,
and
then
discussing
that
with
fukulogu
from
here
we
said
there
should
be
a
standard
for
this.
Let's
do
a
standard.
Let's
do
an
erc,
it
will
be
fun,
and
so
here
we
are
and
before
doing
all
this
fun
stuff,
I
was
a
lead
blockchain
instructor
at
pigeon
skills
and
now
disappeared
and
partnered
at
the
blockchain
consultancy.
B
A
B
B
A
lot
of
people
will
have
heard
about
flash
loans,
but
still
the
pattern
is
a
little
bit
complicated
and
people
often
get
it
confused.
So
flash
loans
at
its
simplest
are
just
cryptocurrency
loans
that
are
repaid
before
the
end
of
the
transaction,
meaning
that
in
real
world
terms,
are
loans
of
zero
seconds
duration.
I
give
you
the
money
but
you're,
giving
it
back
to
me
right
now.
B
I'll
explain
what
that
is
useful
for,
but
the
I'll
explain
first.
Why
is
that
so
amazing,
the
15
second
loan
that
is
repaid
within
an
ethereum
transaction.
It's
actually
amazing,
because
it
means
that
there
is
zero
risk
for
the
lender
to
lend
that
money.
When
lenders
lend
us
money,
they
need
to
have
some
kind
of
guarantee
that
we
are
going
to
repay
the
loan
and
that
introduces
a
lot
of
friction
in
lending
in
flash
loans.
B
Since
it's
all
within
an
ethereum
transaction,
there
is
no
risk
if
the
if
the
borrower
fails
to
repay
the
money
within
those
within
that
transaction.
The
loan
never
happened.
The
the
competition
doesn't
get
included
in
the
blog,
so
it's
actually
zero
risk
for
the
lender.
I
give
you
the
money,
but
if
you
don't
give
it
back
to
me,
it's
fine.
I
never
gave
you
the
money
in
the
first
place
magic.
B
B
It
turns
out
that
you
can
do
a
bunch
of
stuff
first,
you
can
use
it
to
exploit
other
contracts
great,
so
that's
the
most
known
use
case
and
when
we
were
starting
with
the
with
west
10
and
this
erc,
we
had
to
fight
a
little
bit
against
the
perception
that
all
that
you
can
do
with
flash
loans
is
actually
to
exploit
other
smart
contracts,
which
is
I
use.
A
flash
loan
in
effect,
is
kind
of
free
money
for
the
borrower
as
well,
because
often
there
are
no
fees.
B
B
Developers
might
not
have
thought
that
someone
might
have
a
few
million
die
or
more
than
that
to
play
with
so
initially
it
was
used
a
lot
for
exploits
and
it
still
is
to
see
flashlights
in
most
of
the
exploits
that
you
see
currently,
but
that's
just
a
kind
of
use
case
that
we
are
trying
to
protect
ourselves
with
with
this
erc
as
well.
So
we
are
trying
to
make
sure
that
flash
loans
are
not
used
for
exploits
as
much
as
we
are
trying
that
flash
loans
will
be
used
for
productive
uses.
B
Okay,
exploiting
is
fine,
but
we
don't
do
it
during
our
day.
Job
second
use
case
very
famous
is
arbitraging.
You
will
know
some
ban
a
bunch
of
people
from
twitter
that
they
have
their
arbitraging
boats
and
they
run
liquidations
or
they
move
money
between
move
crypto
between
exchanges
or
things
like
that.
B
B
That's
not
bad!
That's
a
positive
use
case
because
basically,
it
makes
the
exchanges
have
a
more
reliable
and
consistent
price.
So,
in
a
sense,
that's
quite
useful
is
still
not
the
use
case.
That
makes
me
more
excited
about,
which
is
a
use
case.
That
will
be
a
lot
closer
to
people
when
they
actually
start
seeing
scenic
use,
which
is
refinance
such
a
boring
sounding
name
for
finance.
No,
it's
actually
quite
cool,
so
refinance
is,
for
example,
what
def
e
saver
does.
B
When
you
go
to
the
platform
and
says,
could
you
please
move
my
assets
from
my
positions
from
avi
to
compound,
or
vice
versa?
Basically,
if
one
platform
is
giving
you
a
better
interest
rate
for
your
savings
than
another,
you
can
move
assets
across.
That
gets
difficult.
If
you
actually
did
something
like
putting
your
collateral
in
in
compound
take
a
loan
of
some
other
asset
like
die
and
then
use
the
dice
somewhere.
You
don't
have
to
die
to
get
your
assets
back
and
it's
difficult
to
move
around
with
a
flash
loan.
B
You
can
see
that
in
depth,
saver,
I've
personally
coded
and
privately
they're,
not
public,
yet
a
couple
of
similar
solutions
for
for
yield,
and
I
can
tell
you
that
it's
a
very
powerful
thing
when
refinance
use
cases
are
more
widely
used.
You
will
see
that
assets
in
the
ethereum
blockchain
start
being
a
lot
more
liquid.
B
So
when
it's
a
lot
cheaper
to
actually
move
your
assets
and
your
depth
and
your
positions
between
platforms
around
you
will
start
seeing
a
lot
of
use
and
all
of
these
will
be
backed
by
flashlights,
and
this
is
just
the
start.
There
are
other
use
cases
that
we
should
see
in
the
future.
This
morning,
fiona
kobayashi,
which
is
in
our
erc
group,
released
something
where
you
can
actually
delegate
the
paying
bit
of
a
flashlight.
You
can
you
can
build
financial
instruments
with
this?
B
It
will
be
useful
for
sure,
and
we
cannot
wait
until
all
the
use
cases
are
known
really
before
we
do
the
standards,
because
now
that
it's
it's
in
it
it's
in
its
infancy
is
more,
is
easier
to
actually
manage
to
get
a
consistent
approach
between
all
lenders
later
when
the
pledges
are
very
very
established,
it
will
be
more
difficult.
So
now
is
the
time
that's,
I
think,
a
good
introduction.
As
for
flash
loans.
B
Well
last
thing
I
wanted
to
say
about
flash
launch
is
that
there
are
two
two
kinds
which
will
be
more
more
important
before,
but
I
would
like
to
explain:
there
are
the
flash
loans
themselves,
which
is
when
a
platform
like
avi
or
compound
has
some
asset
like,
for
example,
die
and
flash
lens
that
die.
That's
a
flash
alone.
There
is
also
flash
mint,
which
is
when
a
contract
itself
can
mint
the
token,
for
example
with,
and
it
can
actually
flash
lend
that
token
that
it
can
meet.
B
So
I
am
the
wrapped
ether
contract
and
I
offer
a
flash
lending
solution.
I
don't
need
anyone
to
have
given
me
a
width
for
me
to
land,
because
I
create
it
from
nothing.
So
if
I
want
to
give
you
a
flash
loan
of
a
trillion
billion
with,
I
can
because
I
just
create
them,
and
that
opens
a
certain
amount
of
fun
possibilities
and
risks
much
greater,
but
anyway,
those
are
the
two
terms:
flash
loan
flash
mint
from
the
point
of
view
of
the
borrower.
They
are
the
same
unless
you
are
really
really
interested
in
borrowing.
C
C
What's
the
protocol
need
to
have
as
a
prerequisite
to
be
able
to
offer
a
flash
loan
and
I'm
assuming
that
there
should
be
a
massive
difference
here
between
a
flash
loan
and
a
flash
men,
because,
obviously,
by
a
flash
loan,
you
actually
have
to
have
those
assets
in
reserve
to
be
able
to
lend
them
out,
whereas
by
flash
minting.
Theoretically
I
mean.
Is
there
any
prerequisite
for
a
platform
to
be
able
to
be
able
to
offer
a
flash
man.
B
C
B
Plus
that
might
work
great
yeah
good
stuff.
Thank
you.
So
if
I'm
a
basic
erc20
and
I
want
to
fly
to
offer
a
flash
launch
of
myself,
only
thing
I
have
to
do
is
to
implement
these
two
functions.
That
say,
how
much
can
I
lend,
which
might
be
the
maximum
there's
a
error
there,
but
it's
fine
and
how
much
am
I
going
to
charge
which
in
this
case
is
zero?
B
B
B
It
means
the
amount
on
the
receiver,
then
it
calls
on
the
receiver
this
callback,
which
is
basically
saying
I've,
sent
you
this
match
this
many
tokens
of
this
token.
You
need
to
repay
the
amount
plus
this
fee.
I
got
this
data
to
pass
you
along,
and
this
is
the
account
that
initiated
the
loan.
It's
just
calling
this
function.
D
B
The
the
lender
doesn't
doesn't
do
anything
with
that
data
is
because
the
borrower,
the
receiver
of
the
tokens,
which
often
is
the
same
person
as
the
borrower,
might
want
to
know
something
about
the
flashlight.
Since
this
has
a
pattern
where
the
portal
work
calls
this
function,
and
this
function,
that's
a
callback
on
the
receiver.
B
This
receiver
just
has
a
a
plain
public
function,
called
onflashlawn
and
might
not
know.
What's
going
on,
for
example,
the
the
borrower
might
be
getting
this
flash
loan
to
do
a
refinance
deal
on
a
certain
user
account
which
is
not
the
receiver.
The
receiver
is
probably
the
the
same
lending
platform
or
whatever.
So
probably
the
the.
B
B
B
Dydx
just
gives
you
the
gives
you
the
the
tokens
and
a
little
more
so
I
needed
to
send
some
data
to
the
receiver
of
the
flash
loan
so
that
he
would
know
what's
going
on,
because
otherwise
it
would
just
be
one
function
and
okay,
I've
I've
got
a
function
back.
Someone
has
called
this
function
now.
What
do
I
do?
So
you
need
that
data
parameter
often
to
basically
tell
the
receiver
of
the
of
the
assets.
What
to
do.
C
If
I'm
understanding
you
right
so
one
of
the
things
that
you
can
use
the
data
for
is
that
even
if
specific
platforms
have
flash
loans
implemented
differently,
you're
able
to
use
the
data
to
create
sort
of
a
standard
data
structure
that
can
then
be
passed
around
or,
like
you
know.
Well,
I
guess
standardly,
that's
the
point
of
the
standard.
I
suppose
that,
no
matter
what
data
comes
in
you're
able
to
format
that,
in
a
standard
in
a
standard
format,.
B
Yeah,
for
example,
the
the
receiver
of
the
flashlight
might
have
two
options
that
are
both
enabled
by
a
flash
loan.
Maybe
one
is
to
migrate
assets
from
a
to
b,
and
the
other
option
is
to
migrate
assets
from
b
to
a
on
the
other
direction,
since
there
is
only
one
callback
on
flash
loan,
the
receiver
has
no
way
of
knowing
if
it's
a
moving
asset
from
a
to
b
or
b
to
a
kind
of
option.
B
C
B
No,
no
that's
completely
chosen
by
the
by
the
user
of
the
flash
loans.
So
if
I'm
deaf
saver,
for
example,
and
I'm
taking
flash
loans
from
avi
to
do
stuff,
basically
he's
defy
saver
calling
flash
calling
avi
and
then
avi
calling
deffy
saver.
Because,
okay,
I
get
the
the
flashlight
from
you.
And
then
you
call
me
to
tell
me
that
I've
got
the
the
loan
in
my
in
my
contract.
But
when
you
call
me,
I
need
to
know
what
did
I
ask
you
the
flashlight?
B
C
Interesting
you
mentioned,
I
mean
you
mentioned
before
that
fubu
had
mentioned
to
you,
that
he
thought
this
should
have
a
standard.
I
think
we're
probably
touching
on
this
a
little
bit
as
we
talk
about
the
data,
but
what
what's
the?
What
would
you
say?
The
main
rationale
for
creating
a
standard
around
throughout
around
flash
loans
is
what
benefit
is
there
from
having
a
standardized
version
of
flashlights.
B
B
The
situation
until
now
has
been
that
avi
was
very,
very
easy
to
use,
but
it
charged
0.09,
which
is
not
much,
but
it's
something
you
could
also
get
flash
loans
for
free
from
dydx,
but
their
interface
was
hard,
so
very
few
people
use
them
even
if
they
were
free
and
you
can
also
get
flashlights
from
unisrap
that
are
much
more
expensive
than
avi,
but
they
are
also
quite
hard
to
do,
but
they
also
have
the
positive
that
you
have
a
really
large
amount
of
tokens.
I
need
token
that
you
can
trade
in
unisop.
B
You
can
take
a
flash
loan
from,
but
it's
not
easy
with
with
a
standard
just
telling
the
lenders.
Okay,
could
you
just
please
offer
a
common
function?
When
someone
wants
a
flash
flashlight,
you
can
actually
go
once
you
learn
the
standard.
You
don't
need
to
learn
separately.
How
to
flash
long
from
avi.
The
ydx
unit
swap
now
sushi
swap
now
make
it
out
with
ish
is
a
little
bit
hard
to
learn
how
to
do
something,
how
to
get
the
same
service
in
different
ways
from
all
these
providers.
C
B
C
B
B
B
I
I
have
a:
I
have
a
fun
story
actually
about
how
useful
this
is.
So
I'm
I'm
coding
now
an
mip
for
for
maker,
though
that
is
quite
unrelated
to
all
this,
and
I
found
at
some
point
of
coding
this
mip
that
hold
on
I
should
do.
I
could
do
this
easy
with
a
flashlight,
because
I
have
these
tokens
that
are
locked
here
and
I
need
to.
B
I
need
to
redeem
them
for
for
some
assets,
but
they
are
locked.
So
I
can't
so.
I
need
to
undo
my
position
somehow
and
I
think
okay,
I
cannot.
I
can
do
this
easily
with
a
flashlight,
and
I
know
that
sg
mcpherson
is
coding,
25
the
flash
lending
module
for
die,
so
I
thought
okay,
so
he's
in
my
group
he's
going
to
use
crc3156.
B
I
already
know
what's
the
standard,
I
can
just
implement
this,
knowing
that
he
will
actually
when
he
releases
he
will
release
with
this
standard.
I
can
start
working
now
and
if
he,
for
some
reason,
mip
25
doesn't
happen,
not
a
problem
either,
because
it's
the
standard,
I
can
just
change
one
address
and
borrow
my
assets
from
dydx
same
code.
Not
a
problem,
so
it's
useful
for
me
at
least.
C
Very
interesting
I
mean
yeah,
so
I
think
that
highlights
also
that
it
allows
for
anyone
who
wants
to
integrate
in
the
future
a
much
easier
format
for
doing
that.
Your
point
with
dydx
is
at
the
end
is
very
interesting
to
me.
Also,
I
mean
you're
talking
about
making
an
integration
to
maker,
but
then
it
would
be
flash
loading
off
of
dydx,
but
I'm
assuming
also
in
that
instance
that
probably
nobody
who's
actually
using.
It
would
even
realize
that
that's
what
happened
underneath
unless
they're
going.
D
B
No
then,
the
contract,
the
smart
contract
that
I
did
for
testing
flash
loans
is
actually
quite
fun
to
use
because
it
just
offers
a
single
interface.
That
says,
give
me
the
lender
address,
give
me
the
token
address
and
how
much
you
want
to
borrow,
and
you
can
say:
okay.
So,
let's
borrow
from
avi
this
match
died.
That's
the
whole
thing!
Then
you
go
to
the
interface.
Okay,
try
to
borrow
the
same
die,
but
this
time
from
units
to
a
plume.
That's
the
whole
thing,
so
it's
very
easy
to
switch
within
between
providers.
C
I
also
wanted
to
ask
you
if
you
see
a
specific
advantage
to
your
implementation.
You
were
talking
before
about
on
certain
platforms
having
flash
loans
that
were
harder
to
access
or
easier
to
access.
Do
you
see
an
advantage
to
your
implementation,
or
I
suppose,
alternatively,
what
makes
a
flash
loan
easier
and
what
makes
it
harder.
B
So
in
terms
of
safety,
I
know
that
the
erc
will
provide
higher
safety
than
almost
any
other
flash
launder
well
or
at
least
it
will
expose
what
are
the
risks
of
using
a
flash
loan?
And
how
do
you
need
to
act
in
order
to
trust
who
is
initiating
the
loan
and
to
avoid
being
exploited
by
someone
sending
you
fake
data,
as
I
think
it
happened
with
defeat
saber
quite
recently
after
they
managed
to
do
a
white
hat
saving
of
that
this
standard
is
being
designed.
So
that
is
easy
to
do
it
safely.
B
Okay-
and
it's
easy
to
do
it,
but
it's
not
the
end
of
all
flash
loans.
There
is
a
bunch
of
features
that
they
are
not
included
in
the
standard
because
they
make
the
standard
more
complicated,
and
maybe
they
are
not
features
that
everyone
is
going
to
use
with
this
standard.
I
don't
think
that
every
possible
flash
lending
feature
should
be
included
in
the
standard.
B
So
that's
not
in
the
standard
right
now.
Okay,
the
standard
offers
something
simple
that
can
be
used
by
everyone
safely.
The
lenders
can
offer
more
stuff,
so
watch.
Flash
notice
is
is
an
example.
I
think
that
uni
swap
also
offers
some
other
things
that
are
not
in
the
standard,
but
if
you
need
some
advice,
those
advanced
features
you
should
go
to
uni
swap
avi
also
offers
a
very
cool
feature
where
you
actually
don't
need
to
repay
the
flash
loan.
It
becomes
debt
at
the
end
of
the
flash
loan.
B
C
And
that's
definitely
the
pattern
that
we've
seen
out
of
I
mean
most
prominently.
Rc20
I
mean
anyone
who's
familiar
with
the
open,
zeppelin
suite
knows
that
you've
got
the
rc20
and
then
you've
got
like
15
different
variants,
adding
different,
like
you
know,
burn
ability
or
numerability
or
snapshotting
or
whatever
the
case
is
yeah.
D
C
B
Okay,
so,
for
example,
an
unsafe
flash
loan
and
then
save
flash
loan
for
the
lender
might
be,
for
example,
where
you
accidentally.
Let
someone
take
a
flash
loan
of
something
like
die.
For
example,
your
compound
and
your
flashlight
didn't
die,
and
then
the
borrower
takes
that
die
and
deposits
it
into
compound
back
into
compound.
B
The
flash
loan
function
thinks
that
you
have
returned
the
flash
loan,
but
what
you
actually
did
was
deposit
it
and
you
have
got
the
c
die
for
free
because
the
because
the
way
of
the
of
the
accounting
was
done
in
the
future.
That's
one
of
the
vulnerabilities
that
we
very
recently
designed
against,
because
there
is
a
way
of
actually
allowing
a
flash
lender
to
accept
the
assets
that
is
lending
back
and
do
the
accounting
properly.
That's
one
example
in
in
yield.
B
B
In
maker,
though,
we
are
backed
by
micro,
so
we
needed
to
we
needed
to
lock
the
redeem
function.
To
avoid
that
with
this
erc
that
wouldn't
have
happened,
because
now
the
we
know
better
how
to
code
these
things
so
that
you
can
actually
let
people
to
to
borrow
to
take
a
loan
on
the
on
the
currency
that
you
are
also
going
to
accept
as
another
part
of
your
of
your
platform.
B
The
callback
needs
by
nature
to
be
open
to
anyone,
so
in
defusaver
you
could
call
the
callback
with
data
that
you
make
up,
and
you
would
make
the
fee
saver
to
things
that
it
shouldn't
do,
and
you
could
trick
the
fee
saver
to
tell
to
make
it
think
that
you
are
heavy
and
that
you
and
that
the
fee
saver
should
send
tokens
to
the
proxy,
which
is
not
actually
the
proxy.
Is
your
own
wallet
address,
so
you
could
use
the
callback
to
just
grab
money
from
from
defi
saver.
C
I
thought
that
might
be
related
to
the
last
thing
you
said
I
would.
I
actually
wanted
to
circle
back
to
something
that
you
mentioned
before
about
refinancing
and
maybe
try
and
zoom
in
on
that
a
bit.
I
had
a
bit
of
trouble
understanding
myself
about
exactly
how
a
flash
loan
could
be
used
for
to
make
refinancing
easier
or
better
done.
Could
you
explain
it
a
bit
more.
B
Okay,
so
I
will
leak
some
alpha
now.
Congratulations
to
the
to
the
listeners.
So,
for
example,
imagine
that
you
have
put
ether
in
maker,
though,
and
you
have
taken
a
dialogue,
okay
and
then
that
diet,
you
have
used
it
for
something
else.
Maybe
you
bought
a
lambo
with
it.
You
don't
have
the
dye
anymore,
but
you
have
the
the
depth
in
maker,
though,
and
the
stability
fee
in
maker
starts
going
up.
B
If
there
is
a
collateral,
but
it
has
a
rate
of
two
percent,
so
you
think
I
would
prefer
to
have
my
ether
collateral
in
this
other
platform
and
my
depth
in
this
other
platform.
So
I
would
be
paying
a
two
percent
interest
rate
instead
of
a
five
percent.
Imagine
that
it's
just
a
mortgage,
you
took
your
mortgage
with
bank
a
and
bank
b
offers
you
a
better
interest
rate.
You
want
to
move
it
across.
B
You
cannot
just
take
your
ether
out
of
maker,
though,
because
first,
you
need
to
repay
your
debt,
and
you
don't
have
that
money,
because
you
bought
that
lamborghini
bad
luck
with
a
flash
loan.
You
can
take
the
day
the
die
from
avi,
for
example,
use
that
die
to
repay
your
debt
in
maker
dial.
Now
that
you
have
repaid
your
debt
in
maker,
though
you
can
get
your
ether
now,
you
can
get
your
ether
and
you
put
it
in
the
other
platform.
C
I
understand
that's
really
interesting.
There's
one
last
question
from
my
end,
I
think
puja
has
a
couple
about
the
eip
process
afterwards
on
a
slightly
unrelated,
but
defy
relevant
note
as
someone
who's
working
extensively
with
flash
loans.
What
would
you
say
the
the
purpose
place
and
meaning
of
audits
are
in
the
d5
space?
What
should
they
signify?
What
should
it
mean
to
a
user
when
they
see
a
platform
is
audited?
What
does
it
cover?
What
doesn't
it
cover.
B
In
generic
terms
or
in
flash
loan
terms,
I'd
be
interested
in
either
or
both
okay.
So,
as
as
the
subject
of
several
audits
myself,
what
you
get
as
a
developer
when
you
get
audited,
is
and
depends
a
lot
on
the
quality
of
the
auditing.
Not
all
audits
are
the
same.
You
get
a
fresh
pair
of
eyes,
often
with
a
very
specific
experience
to
go
through
your
code,
the
defy
products
most
of
them
are
done
by
very,
very
small
teams.
B
Does
that
mean
that
the
code
is
completely
safe?
No,
I
mean
an
auditing
is
just
that.
Another
thing
is
guaranteed
that
someone
else
under
and
who
that
that
someone
else
is
is
important
that
someone
else
has
gone
through
the
code
and
has
looked
through
it
is
it
if
the
code
were
had
maybe
a
50
50
chance
of
being
hacked
after
an
audit
by
a
by
a
firm
that
they
know
what
they
are
doing,
and
they
take
your
time.
Maybe
it's
90
safe.
B
C
I
think
that
I
think
that's
really
a
good
tick.
I
mean
in
general.
I
think
we
see
a
lot
of
people
in
the
space
who
understand
like
have
some
kind
of
understanding
that
as
soon
as
they
see
an
audit
stamp,
it
must
mean
that
something
is
absolutely
bulletproof
and
you
know
we'll
never
have
anything
happen
to
it
and
it's
in
general.
It's
just
not
the
reality.
I
mean.
If
you
ask
somebody
in
2005,
they
definitely
wouldn't
have
thought
the
layman
brothers
was
going
to
go
under
to
give
an
american
example,
but
yeah.
C
A
Thank
you
for
a
wonderful
question
answer
session,
both
to
william
and
alberto.
I
think
those
are
really
helpful,
so
one
more
question
I
have
with
regards
to
status.
Currently
I
see
that
this
is
in
the
review
status,
which
I
believe
is
the
most
appropriate
one
for
this
proposal
at
the
time
it
leaves
with
like
two
more
statuses
before
it
can
be
standard
like
class
call
and
final.
A
B
I
think
we
might
be
ready
pretty
soon
if
you
follow
the
conversation
in
ethereum
magicians,
the
guys
from
sushi
swap
difficult
to
pronounce.
I
made
a
further
review
on
them
on
the
standard
and
they
actually
raised
two
issues
that
were
raised
to
me
before,
one
of
them
by
mika
and
the
other
one
by
the
avi
guys,
which
are
the
ones
that
actually
led
me
to
think
more
about
what
needs
to
be
inside
the
standard
and
what
needs
to
be
outside
the
standard.
A
A
A
Great
great
so
yeah,
it
sounds
all
good
to
me
and
yeah.
That
was
my
only
question
related
to
process
part.
I
I
think
this.
This
was
a
very
wonderful
presentation
and
discussion
that
we
had
around
d5
projects.
I'm
sure
it
is
going
to
be
helping
dab
developers
and
other
projects
who
are
planning
to
implement
this
standard
in
their
projects.
A
So
generally,
security
consideration
is
is
important
for
any
upcoming
proposal,
but
when
it
comes
to
like
finance
and
decentralized
finance,
I
think
a
lot
more
eyes
revolve
around
the
security
consideration,
especially
after
the
recent
attack
on
on
one
of
the
85
projects,
glad
to
know
that
it
is
already
taken
care
of.
So
I
would
like
to
you
know
thank
you
for
your
time
and
explaining
these
things
for
our
viewers.
A
Great
so
for
viewers
who
are
new
to
our
show,
we
have
also
recorded
a
d5
governance
proposal.
That
is
the
ip2535
diamond
standard
that
is
not
in
a
final
state
yet,
but
if
people
are
interested
and
some
of
the
popular
d5
projects
are
already
implementing
that,
you
can
give
it
a
watch
that
also
contains
very
good
questions
around
decentralized
finance,
if
you
like,
this
episode,
do
like
it
and
share
button.
A
If
you
have
questions
that
could
not
be
answered
in
today's
episode
feel
free
to
leave
a
comment
or
you
can
directly
ask
from
the
author,
his
twitter
id
would
be
in
the
description
section
along
with
the
other
guests
learning.
A
new
proposal
is
always
interesting
and
I'm
sincerely
thankful
to
all
the
eap
authors
who
have
signed
up
to
join
us
for
going
an
extra
mile
to
make
it
even
simpler
for
ethereum
community
continuing
the
journey
we'll
be
joined
by
vitaly,
buterin
and
martin's.