►
From YouTube: Filecoin Core Devs #41 (Meeting 2)
Description
Recording for:https://github.com/filecoin-project/tpm/issues/98
For more information on Filecoin
- visit the project website: https://filecoin.io/
- or follow Filecoin on Twitter: https://twitter.com/Filecoin
Get Filecoin community news and announcements in your inbox, monthly: http://eepurl.com/gbfn1n
A
Recording
all
right
welcome
everyone.
This
is
file
coin
core
devs
meeting
number
41.
This
is
meeting
two
today
thursday,
the
21st
of
april
2022.
A
on
today's
agenda.
As
always,
we
have
brief
updates
from
all
of
the
implementation
teams,
as
well
as
from
the
file
queen
foundation
on
both
fips
and
security,
and
today
we
have
two
discussion
topics
which
we'll
discuss
in
this
meeting.
The
first
one
is
update
to
the
timeline
for
the
network.
V16
upgrade
the
second
one
is
the
index
or
advertisement
protocol
fit
that
was
just
pushed
into
the
repo
and
hong
hao
who's
in
the
meeting
with
us
is
going
to
be
presenting
that
to
us.
A
The
final
discussion
topic
is
the
node
state
compaction
proposal
that
the
fujan
team
has
put
together,
but
I'd
ask
that
you
check
the
meeting
notes
and
the
reporting
from
meeting
one,
so
you
can
see
their
proposal.
I
didn't
get
their
slides
in
time
to
be
able
to
present
them
to
you
on
this
call.
Today.
A
A
I
believe
and
have
updated
rest
actors
through
v6
and
network
through
v14
security
audits
are
ongoing,
as
mentioned
it's,
where
they're
spending
most
of
their
time
at
the
moment,
but
they
are
working
to
prepare
v15
for
mainnet
and
calibration
sync,
and
they
also
have
a
fantastic
new
wiki
where
they
track
all
of
their
team
updates
in
real
time.
A
So
do
recommend
you
taking
a
look
at
those
for
the
fuhan
team,
they've
continued
to
work
on
bug,
fixes
and
improvements,
as
well
as
shipping
their
cli
enhancements,
and
are
currently
working
on
implementing
an
advanced
scheduler
for
the
fujan
miner.
Any
questions
that
you'd
like
me
to
relay
to
either
the
forest
or
the
fujong
huntings.
A
A
All
right,
yep
and
andrew
you
asked
for
the
the
repo
link
I'd
be
happy
to
give
it
to
you.
Are
you
looking
for
forest.
A
D
Hello,
this
tim,
yes,
our,
like
monthly
feature
release,
is
still
ongoing.
This
is
again
a
very
exciting
one.
We
have
like
post
worker
introduced
in
this
release.
We
have
a
lot
of
like
scheduler
enhancement
for
the
ceiling
pipeline.
Basically,
it's
like
storage
rather
used
to
have
you
know
a
lot
of
resources.
D
However,
tasks
are
only
assigned
to
one
specific
resource
and
the
other
continues
like
to
stay
idle,
and
this
one
fixed
that
bug
for
the
post
worker
we
have
both
like
window
post
and
winning
post
worker
and
like
one
sp,
can
run
multiple
post
workers
as
they
need
it.
So
that's
all
like
very
exciting.
D
Of
course,
the
lotus
team
has
been
working
with
fvm
team
to
make
sure
that
we
deliver
like
fvmv
one
at
the
moment.
Flip
31
is
completely
implemented,
including
the
execution
trace
as
well.
Right
now,
the
team
is
working
on
getting
the
guests
like
remodeling,
like
the
parameters
right
now
we're
expecting
an
answer
by
the
end
of
week.
The
team
is
like
here
like
in
amsterdam
like
working
on
that,
and
then
we
will
be
starting
our
5th
31,
like
the
test
flow
proposed
in
the
fifth
city,
one.
D
D
Sorry,
one
more
thing:
yes,
because
I
want
to
give
a
shout
out
to
boost-
is
like
the
market.
V2
implementation
that
is
supposed
to
be
like
a
replacing
or
like
an
alternative
for
the
lotus
like
minor
market
or
gold
fuel
market
in
general,
is
like
launching
alpha
and
testing.
With
a
couple
of
stored
feathers.
We
are
seeing
very
promising
result
it
ships
with
a
very
nice
ui.
D
We
did
a
ama
yesterday
and
there's
a
recording
on
falcon
youtube.
We
would
encourage
everyone
to
like
check
it
out,
especially
storage
provider,
who
are
making
deals,
giving
the
team
any.
You
know
feedbacks
in
the
booth
help
channel
on
the
foul
queen
slack,
and
if
you
guys
have
any
more
like
market
related
requests,
you
can
still
report
your
loaders.
However,
you
will
see
again
again
like
trans
and
try
like
transfer
to
boost
channels
the
boost
repository,
because
that's
gonna
be
our
go
to
market
implementation.
A
And
then
stephen,
do
you
want
to
offer
venus's
updates
for
the
week.
B
Okay,
yeah,
thank
you.
Yeah
very
last
is
going
on
on
the
implementation
and
yeah
yeah.
Many
on
two
parts.
One
is
about
the
venus
cluster.
We
actually
yeah
combined
about
the
yeah,
the
ceiling
and
work
and
some
some
yeah
some
see
some
secret
managements.
Then
yeah
together
to
come
to
your
comp.
You
come
up
with
a
component
yeah
with
a
diverse
cloud
breast
cluster,
to
make
it
much
easier
for
our
user
to
use
to
use
venus.
B
So
currently
we
have
done
the
videos
classes
to
support
slot
slapdo
and
the
testing
is
ongoing
and
we
think
that
we
will
have
this
be
yeah
deployed
in
some
and
yeah
for
some
notes
perhaps
next
week.
B
Another
part
is
that
your
business
market,
so
virus
max,
is
another
very
important
component
for
yeah
for
the
mining
pool
yeah,
because
the
various
market
we
we
are
in
the
phase
two
to
support
multiple
storage
providers,
so
that
yeah,
which
means
that
we
could
have
this
kind
of
service
or
yeah
and
anyone
can
use
it
as
market
as
a
yeah.
B
B
B
And
also
we
are
supporting
one
important
customer.
The
file
drive
yeah
well
of
the
file
drive,
is
doing
yeah
zero
data
storage
service
as
with
tables
service
yeah
for
their
users,
and
now
they
are
yeah.
They
want
to
support
yeah
to
have
velas
and
the
speaker
to
have
villas
market
and
in
their
system
to
serve
their
customers
and
and
their
storage
providers.
B
So
we
have
some
discussion
with
file
drive
team
and
to
see
how
to
support
and
integrate
and
yeah
there
an
app
there
yeah
service
software
and
yeah.
I
wasn't
yeah
so
far,
so
good
and
everything
works.
Will
you
test
him
and
overthink
that
the
file
job
might
be
deployed
this
one
in
early
may
yeah?
That
is
a
very
good
milestone
yeah
for
villas.
I
would
think
it's
basically
for
your
real
data
and
your
storage
yeah.
It's
all.
A
Also
steven
I'm
finalizing
the
roadmap
updates
that
we
do
on
a
quarterly
basis
and
I
think
file
drive
file
drives
project
suite
is
going
to
be
included.
Also
so
it'll
be
a
nice
little
linkage
for
the
work
you're
doing
across
the
ecosystem
too,
so
be
on
the
lookout.
B
A
So
from
the
foundation
last
week
we
accepted
fifth
27,
which
is
the
deal
proposal
label
type
which
we're
going
to
be
implementing
as
part
of
the
v16
upgrade.
So
it's
great
that
this
is
squared
away
and,
as
we
all
know,
the
fib
authored
by
stephen
29
for
beneficiary
address
provisioning
for
storage
providers.
It's
currently
in
last
call
through
next
friday.
A
We've
had
nothing,
but
an
overwhelming
amount
of
support
for
this
change,
especially
from
storage
providers.
So
I'll
be
going
to
all
of
the
working
group
next
weeks
to
triage
and
answer
any
questions
they
have
specifically.
A
But
we
expect
this
one
to
pass
and
hopefully
be
implemented
in
a
very
quickly
upcoming
network,
upgrade
in
the
near
future
other
than
that
the
list
of
open
fit
drafts
remains
unchanged,
except
for
this
last
one,
which
is
the
unnumbered
fip
index
or
advertisement
protocol
which
again,
hong
kong,
will
walk
us
through
in
just
a
moment
and
otherwise,
as
before,
we
do
have
our
weekly
governance
updates
being
pushed
into
the
fibs
discussion
forum.
A
So
you're
always
welcome
to
check
even
more
detailed
and
updated
information
related
to
this
list,
of
where
each
fifth
is
in
its
process
there
on
a
weekly
basis
on
the
security
side.
This
morning
we
welcomed
parth.
More
formally
parth
has
joined
the
foundation.
I
think
he
started
in
early
february
and
he's
currently
working
with
jennifer
to
triage.
Some
of
the
ongoing
audit
needs
that
we
are
currently
covering
since
dudley
has
left
the
foundation.
A
A
D
As
we
mentioned,
like
multiple
times
like
if
vm
is
like
a
huge
protocol,
change
like
to
the
internet
like
network-
and
we
are
taking,
like
you-
know-
additional
security
consideration
during
all
the
implementation
to
make
sure
this
landed
in
the
network
securely
and
plus
that
we
are
doing
like
extensive,
like
testing
these
days.
So
that
means
like
you
know,
the
timeline
is
not
as
certain
as
the
past.
D
A
couple
network
upgrades
because,
like
the
work
item,
is
like
very
well
defined,
but
with
fvm,
it's
like
every
time
we're
implementing
something
we're
doing
a
lot
of
audio
and
testing,
there's
new
area
being
discovered
and
so
like
we,
we
hope,
the
community
and
all
the
implementation
team
to
you
know
thank
you
for
your
patience
and
like,
and
you
know,
being
like,
responsive
and
flexible
with
the
timeline
we're
having
right
now,
we
do
commit
to
give
the
community
the
latest
timeline,
as
we
can
so
right
now,
flip
27
is
already
like
in
last
call,
I
believe,
and
it's
like
fully
implemented
and
being
tested
like
the
migration.
D
Everything
is
like
all
good.
As
we
all
know,
flip
30
is
just
like
a
conceptual
like
flip,
so
there's
no
implementation
work
flip
31
the
implementation
wise.
It's
done
right
now
in
the
lotus
of
the
implementation.
D
Right
now
we
are
trying
to
figure
out
because,
like
the
cid
before
it's
just
like
a
name
string,
however,
now
the
accuracy
id
is
actually
content
addressing
so
a
lot
of
the
things
like
kind
of
change,
because
before
it's
like
as
long
as
you
for
each
system
actor,
you
have
a
name
string
now,
every
time
you
deploy
any
change
in
actor,
the
cid
will
change
and
it's
not
like
you
know,
aesthetic
anymore.
So
we
have
some
like
implementation
challenge
and
refactoring.
D
We
have
to
do
so
that
all
the
actors
are
loaded
in
implementations
like
pro
play.
That's
why
there
are
some
like
implementation
delay
there.
Another
reason
why,
as
everyone
knows,
expect
actually
it's
really
in
gold,
venus
and
lotus
are
developing
go.
However,
now
the
built-in
actually
using
rust,
so
we
kind
of
have
to
create
a
new
wrapper
so
that,
like
you
know,
the
gold
implementation
can
actually
talk
to
the
built-in
actor.
D
So
we
are
working
on
that
and
that
work
is
also
done
and
we
will
be
testing
that
we
are
actually
deploying
that
in
caterpillar
nets,
which
is
a
test
net
for
implementation
team
like
today
and
and
hopefully
going
through
like
a
week
to
two
weeks
of
like
upgrades
like
testing
before
we
deploy
to
the
community
test
net,
which
is
butterfly,
I
will
get
to
that
in
a
second
on
fifth
32.
As
I
mentioned
earlier
on
cuba,
ralph
stab
and
stephen
are
still
working
on
like
the
final
gas
parameter.
The
implementation
is
already
done.
D
We
are
just
trying
to
figure
out
the
right
parameters
to
make
sure
the
network
is
like
secure
the
chain
capacity
and
like
things
are,
like
cans
can
be
like
you
know,
safe
in
a
sense,
you
can
see
more
details
in
532
and
the
final
parameters
will
be
updated
within
a
week
before
we
can
move
it
to
last
call.
So
that
being
said
on
the
summary,
the
test
plan,
we
are
still
going
going
to
do
that.
However,
we're
gonna
eliminate
because
of
all
the
testing
we
are
doing
during
the
implementation
phase.
D
We
are
eliminating
one
test
phase.
It's
like,
I
believe,
is
test
phase
four.
So
before
we
are
saying
for
the
butterfly,
we
are
gonna.
Do
two
upgrades.
One
is
the
upgrade
q531
basically
switch
to
fvm
and
make
sure
the
butane
actually
is
properly
used.
Then
we
are
going
to
do
a
reset
and
upgrade
to
network
b16
which,
including
both
531
and
532,
we
decided
the
intermediate
like
the
first
upgrade
is
unnecessary,
given
all
the
testing
we
are
doing
on
devnet
and
the
localnet
and
the
caterpillar.
D
So
we
are
saying
the
butterfly
net
will
just
directly
upgrade
to
a
v16
around
may
sorry
around.
May
the
6th
so
about
like
two
weeks
a
day
from
the
previous
timeline,
and
then
we
will
be
testing
there
for
a
good
like
two
and
a
half
weeks
so
also
like
from,
like
you
know
two
weeks
ago,
until,
like
the
end
of
the
may,
we
will
be
doing
like
a
lot
of
external
auditing
and
avm
bug
bounty
program.
D
So,
in
this
testing
period
like
period,
we
are
hoping
to
address
all
the
findings,
especially
the
p0p1
fundings,
in
the
testing
phase.
Before
deploying
that
to
calibration,
assuming
that,
like
no
major
consensus,
breaking
findings
are
being
reported,
we
are
planning
to
upgrade
calibration
to
v16
network
v16
on
may
the
26th
and
then,
as
always,
we
will
be
testing
that
for
like
three
weeks
before
the
midnight
upgrade.
That
also
gives
like
an
implementation
team,
to
put
out
a
release
and
giving
their
user
time
to
upgrade.
So
right
now
we
are
looking
at.
D
I
mean
that
upgrade
time
around
like
june
16th
to
june
20s,
depending
on
because,
like
normally,
we
do
network
upgrade
on
like
tuesdays
like
early
tuesdays,
so
that,
like
people
have
a
week
to
respond
in
case
anything
happens.
That's
why
20s
is
possible.
However,
if
based
on
our
like
testing
flow
june,
16th
is
the
earliest
we
can
do,
but
that's
the
thursday.
D
So
we
as
called
out
we
should
discuss
like
you
know,
because,
like
thursday,
if
anything
happens,
we
only
have
like
two
days
otherwise
we're
heading
into
weekend
and
for
people
to
dealing
with
like
bugs
and
things.
So
it's
less
some
deal
but
like
but
like
it's
around
like
mid-june,
is
what
we
are
looking
at
for
the
midnight
upgrade
and
so
far
again,
I
want
to
reiterate:
we
are
we're
gonna,
like
all
the
implementers
are
gonna,
try
really
hard
to
commit
to
this
timeline.
However,
this
is
one
network
upgrade
we
don't
want
to
rush.
D
B
Okay,
so
yeah
and
as
jenny
information,
so
it
looks
at
the
plan.
Yeah
from
those
point
of
view
is
yeah
from
this
page,
and
you
know
it's
kind
of
two
or
three
weeks
yeah
the
delay
based
on
this
yesterday
in
this
life
yeah,
everything
is
okay
in
yeah
from
very
last
point
of
view.
Yes,
because
we
are,
and
actually
we
use
most
of
the
code
of
the
yeah
zwast
based
actors
and
also
the
afm
yeah.
B
B
Yes,
something
like
this,
so
it
should
be
no
problem
yeah
for
venus
to
to
catch
up
and
to
meet
this
timeline,
I
will
send
yeah
it's
the
one
thing
I
want
to
mention
and
another
one
is
that
I
said:
okay,
I
like
this
plan.
Actually
we
have
the
network
version
16.
We
only
have
the
a
frame
deleted
and
we
add
one
more-
the
yeah
yeah
the
27
yeah,
but
I
think
this
is
a
very
small
one
right,
yeah,
it's
kind
of
okay.
I
would.
B
B
That
is
my
expected,
my
expectation
yeah,
because
I
want
to
say
that
and
the
your
next
version
is
only
about
fm.
No
any
other
sense.
You
have
to
make
it
very
pure,
okay,.
A
A
In
jennifer
two
this
morning,
I
let
everyone
know
that
there
are
changes
and
that
it
is
still
quite
tentative
and
no
one
raised
any
concerns.
The
only
thing
they
ask
is
that
we
do
share
updated
thinking
around
the
test
plan
and
the
actual
dates
specifically
around
the
butterfly
net
upgrade.
So
do
you
have
any
concerns
with
us
sharing
the
information
you
just
shared
with
all
of
us
here
in
the.
D
Not
at
all,
I'm
actually
to
update
the
like
the
community
forum
like
network
announcement
as
well.
It's
just
yeah!
I
didn't
get
to
it
because,
like
all
the
channels
but
like
yes,
I
will
post
that
like
to
both
like
implementers
and
community
tomorrow
morning,
when
I
woke
up
perfect.
Okay.
Thank
you
so
much.
That
was
the
only
ask
from
this
morning.
B
So
yeah
there's
one
thing
I
want
to
discuss
here
so
after
the
yeah,
this
version
16
and
yes
in
our
yeah
yeah
and
our
plan
is
to
actually
to
support
the
user
programmable
actors
right
yeah.
That
will
be.
After
this
we
will
have
yeah,
for
example,
yeah
another
late
work
version
yeah
to
support
your
fully
user
programmable
actors
yeah.
B
But
I
would
think
that
before
that,
after
the
version
16,
we
might
be
have
another
one,
for
example,
version
17
to
support
some
other
fix,
for
example
yeah,
because
there
are
many,
many
other
and
yeah
some
faves.
For
example,
the
29
and
some
others
alex
proposed
to
support
yeah
support,
some
flexible
markets
and
other
things
that
kind
of
change
might
be
made
before
we
really
support
the
user
programmable
actors.
B
D
Agree,
I
think
I
believe
that
we
did
have
a
such
a
conversation
in
the
past
like
core
that
call
already
so
like
we
have
529.
We
have
like
basically
the
flips
that
you
enable
like
to
make
our
system
actor
more
flexible,
to
support
the
user
program,
both
like
actor
for
m2.
So
we
should
like
continue
like
discussing
that,
of
course,
but
like
one
concern
from
the
lotus
team
is
like,
because
the
lotus
team
is,
like
you
know,
supporting
the
fbm
team,
deploying
both
like
m1
and
m2,
a
lot
so
like.
D
If
we
do
another
network
upgrade.
That
means
like
lola's,
had
to
allocate
resources
to
support
all
the
fips
that,
like
all
implementers,
decided
to
accept
or
like
finalize
in
the
next
like
network
version,
and
that
means
there
will
be
less
resources
going
to
the
fvm
in,
like
you
know,
development.
D
So
for
us
like,
we
are
still
trying
to
decide
which
one
we
should
be
prioritizing.
After
that
you
know
we
can
share
and
discuss
with
the
rest
of
the
implementation
team
to
say
what
can
we
be
committed
to
in
the
next
quarter?
Basically,.
B
A
Once
once
m1
shifts,
we
need
to
have
yet
another
retro,
as
we
always
do,
but
I
think
this
one
needs
to
really
focus
on
timelining
and
resource
planning
so
that
when
we
begin
to
talk
about
jennifer's
right,
we'll
have
another
upgrade
in
between
the
launch
of
m1
and
m2.
But
we
need
to
have
a
conversation
as
a
group
about
expectations
around
the
scale
of
that
interim
upgrade,
as
well
as
the
projected
timeline
from
the
beginning.
A
Since
we've
now
pushed
out
the
original
original
plan
for
the
f
non
non-programmable
fvm
to
go
live
was
march
31st,
so
we've
obviously
changed
quite
a
bit
from
that
original,
and
we
want
to
like
be
able
to
keep
that
in
mind
when
we
look
at
planning
for
the
next
one
as
well.
E
Okay
sounds
good,
I'm
here
with
andrew
today
too,
and
I
walked
through
the
industry
project
and
then
for
we
can
do
q
a
and
also
andrew,
feel
free
to
chime
in
to
add
more
technical
details
too.
If
I
miss
anything
first
of
all
to
set
the
ground
like
high
level
context
is
of
why
we're
building
doing
investor
is
because
firecoin
has
a
lot
of
content
that
is
being
stored
and
we
want
to
make
them
useful
when
people
try
to
look
for
it.
E
E
But
falcon
has
a
lot
a
lot
more
large
amount
of
data
compared
to
that,
and
if
we
put
all
these
sids
into
the
dht,
it
will
blow
over
the
dht,
and
so
what
we
do
is
we
spin
up
the
industrial
node
to
store
these
kind
of
high
volume
indexes
and
to
support
when
customers
try
to
retrieve
the
data
they
can
hit
the
indexer
which
people
able
to
look
up
like
where's
the
content
being
stored
and
be
able
to
achieve
the
data
back.
E
So
it's
a
pretty
crucial
component
that
supports
the
retrieval
for
both
filecoin
and
in
the
future,
the
interrupt
in
durability
between
ipfs
and
falcon.
So
on
the
right
hand,
side
is
a
diagram
overall
about
how
things
works,
with
the
indexer
and
in
this
provider.
E
So
today,
like
storage
provider,
they
make
the
deal
and,
alongside
with
that
storage
provider
market
process,
they
also
run
the
index
provider
process
that
can
publish,
which
is
like
a
non
published
advertisement
messages
to
the
industrial
node
to
let
them
know
that
to
let
them
know
that
the
data
is
available
and
it's
published
over
the
gaza
subsystem
to
announce.
E
Those
advertisement
message
and
the
advertisement
message
are
contains
the
information
of
the
sit,
the
multihash
and
as
well
as
the
previous
advertisements
id,
and
so
that
together
they
forms
a
chain,
and
the
investor
is
basically
consuming
this
chain
from
the
initial
advertisement.
All
the
way
to
the
latest
one-
and
another
thing
to
know
is
that
this
industry
also
updates
over
grabsing,
so
every
24
hours
it
look
up
the
data
to
do
and
to
do
the
update.
E
If
multiple
storage
provider
trunks
have
the
same
sid,
then
industry
will
only
look
up
from
one
of
them,
which
saves
the
bandwidth
bandwidth
as
well
and
after
the
data
is
being
published,
which
is
two
way
right
now,
like
one
is
through
the
advertisement,
another
one
is
http,
and
once
the
data
and
lands
on
the
industrial
node
and
when
retrieval
client
trying
to
make
the
request
the
data,
then
it
will
look
up
the
industrial
node
first
to
know
that
for
this
cit,
which
I
look
up
for
the
story
provider
to
get
the
data
back
and
then
they'll
actually
make
a
query
to
the
storage
provider
and
retrieve
the
data
back.
E
So
that's
the
overall
like
how
industrial
works,
and
why
do
we
need
indexer
specifically
for
this
fib?
It's
it's
worth
noting
that
we're
still
a
jeff
in
progress.
We
expect
to
finalize
the
details
in
the
fib
next
week,
but
specifically,
what
we
want
to
standardize
as
part
of
the
standard
is
about
not
about
like
how
do
we
run
intestinal,
but
more
about
how
do
we
publish
content
to
the
industry?
E
And
specifically,
those
are
one
advertisement
message
like
using
the
schema
of
advertisement
and
using
the
mechanism
of
everyone
has
a
message
to
publish
the
data
and
then
leveraging
the
gaza
sub
mechanisms
to
announce
these
advertisement
message.
So
those
are
the
two
things
that
we
really
want
to
standardize
as
part
of
this
industrial
effort.
E
One
more
thing
to
know
is:
we
industry
is
currently
live
in
production.
We
are
with
our
pl's
implementation.
We
are
working
with
more
partners
to
run
multiple
instances
of
investors
as
well.
It's
like
with
lotus
1.15.1,
and
we
have
currently
invest
a
little
bit
more
than
2.5
billion
indexes
and
for
more
information
about
industry
itself.
You
can
find
it
on
sid.contact
and
over
there
we
we
periodically
update
the
latest,
and
you
can
also
use
that
to
look
up
the
content
as
well.
E
Given
look
up
where
to
find
the
content,
given
this
yeah,
that's
that's
pretty
much
it,
and
I
don't
know
andrew,
is
the
expert
here
and
the
hero
here
for
the
project
and
feel
free
to
add
anything
as
well.
If
I
miss
any
like
things
that
we
should
call.
C
I
think
that
was
an
excellent
overview
and
I
just
as
was
stated
earlier,
the
fip
is
sort
of
a
work
in
progress,
so
there
are
going
to
be
some
updates,
so
don't
don't
start
coding
off
of
it
just
yet,
but
yeah
we
want
to
finalize
that
protocol
and
make
it
very
clear
to
all
implementers
exactly
what
it
allows
them
to
do.
What
it
doesn't
allow
them
to
do.
I
think
also
you'll
see
some
changes
in
the
very
near
future
with
the
infrastructure.
C
So
if,
while
the
indexer
is
live,
we're
constantly
having
to
expand
our
infrastructure
as
it's
growing
greatly
in
popularity,
so
we're
doing
some
moving
around
of
the
where
actual
nodes
are
are
being
hosted
from
at
this
point,
but
otherwise,
yes,
it
is
live.
You
can
use
it.
You
can
index
content.
You
can
see
where
that
content
is
residing.
D
Thank
you.
I
do
have
like
one
question
or
like
not
really
question
or
like
comment
so
like
lola's.
Just
like
you
know,
lucky
enough
that
we
have
the
store
the
index
team
to
integrate
the
index
provider,
which
is
the
you
know,
working
with
the
storage
providers
market
notes
to
actually
announcing
the
indexing
to
the
indexer.
D
We
we
are
lucky
enough
to
have
the
team
to
integrate
that
into
low
levels.
For
us
also
like
that
they've
seen
boost
the
other
market
implementation.
We
have
right
now.
My
question
to
I
guess
like
andrew
and
hong.
How
is
like
is
this
like
fit
a
protocol
standardized
like
the
indexing
advice,
like
you
know
the
advice
constructor
as
well,
so
that,
like
implementation
like
venus
market
and
cities,
they
can
all
be
the
index
provider
to
the
indexer
network.
So
is
that
a
part
of
the
fifth?
D
C
The
fip
should
cover
two
aspects
of
integration:
one
is
as
a
library
and
we
wanna
document
the
actual
protocol
so
that
if
they
want
to
write
their
own
implementation
of
that
protocol,
they
can
do
so,
but
also
to
document
where
existing
libraries
for
and
reference
implementations
are
so
that
they
can
also
reuse
the
existing
software
as
components
in
their
own,
their
own
applications,
libraries
etc.
C
So
we
can
also
provide
what
fip
should
also
specify
the
software
apis
that
are
available
too.
Not
just
the
protocol.
A
Nice,
sorry,
just
to
make
sure
I
understand
we
had
this
conversation
this
morning
too,
because
of
those
two
separate
components,
though
you
wouldn't
require.
There's
no
like
protocol
code
base
change,
it's
more
of
a
standardization
across
implementations,
correct,
correct!
That's
right,
yeah,
okay,
okay
cool!
So
I
mean
in
terms
of
the
fit
process.
Then,
as
soon
as
you
deem
the
current
fifth
draft
complete,
we
can
work
together
to
put
it
in
last,
call
and
hopefully
get
it
accepted,
never,
and
we
can
make
sure
that
implementers
incorporate
it
but
yeah.
C
This
just
describes
what
the
protocol
is,
and
then
that
gives
the
implementers
the
option
of
writing
their
own
implementation
of
that
protocol
in
whatever
language
they
want,
or
also
using
the
software
that
we
provide
primarily
in
go,
but
we
may
also
have
other
implementations
in
other
languages
as
well,
so
it's
either.
So
it
should.
I
think
it
should
cover
both,
but
the
fip
being
a
fifth
should
primarily
cover
the
protocol
waiting.
What
the
protocol
is
on
the
wire
so
anyway,
any
implementer
can
implement
that.
B
Yeah,
okay,
this
one
is
very
great.
I
would
think
yeah,
it's
a
very
good
start
for
the
yeah
for
the
youtuber
market,
and
I
like
this
design.
I
I
had
a
quick
review
about
the
yeah,
the
vape
and
it's
good,
and
I
would
like
to
update
you
about
the
the
builders
team-
and
I
think
I
missed
this
in
in
the
videos
and
upgrade
your
your
previous
sled.
B
We
are
we.
We
have
start
started
to
to
study
this
and
we're
working
on
to
integrate
the
index
provider
with
yeah
with
us
market
yeah.
We're
yeah,
we're
working
on
this,
and
we
may
have
some
questions
and
I
will
yeah
reach
you
about
this.
We
yeah
there
might
be
some
change
actually
based
on
the
currently
the
index
provider.
B
Yeah
software
is
because
we
need
to
integrate
them
with
the
business
market
yeah,
but
I
would
think
that
the
most
important
part
is
about
the
product,
as
you
mentioned,
is
about
the
communication,
the
standardization
between
the
components,
right,
the
storage
provider
and
the
index
load
and
how
the
announcements,
the
message
and
the
protocols
hoses
yeah.
C
B
Would
think
was
all
of
this
are
well
defined
and
accepted
by
the
yeah
community
yeah.
That
will
be
very
good,
and
I
will
yeah
follow
this
and
to
put
it
in
and
the
importations.
D
C
Yes,
I
second
that
thank
you
for
all
of
the
support
you've
given
as
we've.
Both
both
sides
of
this
team
have
worked
through
the
process
of
integrating
this,
and,
as
always,
if
there
are
questions
or
concerns,
or
even
innovations
feel
free
to
contact
my
team,
hong
hao
and
that
make
sure
that
we
can
communicate
or
integrate
whatever
it
is.
That
is
needed.
D
A
All
right,
then,
if
there's
no
more
questions
on
the
indexer
advertisement
protocol,
just
a
reminder
that
this
morning
the
the
fujan
team
also
had
a
separate
presentation
on
their
note
today,
compaction
proposal.
It
was
a
pretty
detailed
description
of
what
they're
proposing.
So
I
would
recommend
you
take
a
look
at
the
meeting
notes
when
they're
available
early
next
week,
but
other
than
that
we
are
almost
at
time
and
at
the
end
of
our
meeting.
So
thank
you
all
for
joining
us
today
and
we
will
see
you
at
a
minimum
in
two
weeks.