►
Description
Join us for a discussion about the Filecoin roadmap, including product updates for Lotus, Boost, Indexer and Saturn. We end with an open discussion for Boost.
Watch the recap of the Storage Provider meetup in Austin, hosted by the Filecoin Foundation!
Meetups are an opportunity to meet the amazing people and teams participating in the Filecoin community's decentralized storage network. Learn more at https://sp.filecoin.io/
A
I
wanted
to
talk
about
some
of
the
major
themes
that
we're
kind
of
seeing
across
core
improvements
in
the
file
coin
project
protocol
ecosystem
over
the
past
two
quarters,
like
really
over
all
of
2022,
the
the
things
that
we've
been
seeing:
land
in
q1
and
q2,
and
the
really
exciting
breakthroughs
that
are
coming
in
q3
and
q4
that
everyone
is
probably
already
sick
of
hearing
about,
because
we're
all
so
excited.
We
can't
shut
up.
A
I
kind
of
grouped
the
set
of
milestones
that
I
see
coming
across
kind
of
like
core
improvements
at
a
protocol
level
into
three
main
categories.
First
and
foremost,
where
a
ton
of
well,
maybe
not
first
and
foremost,
first
and
foremost,
is
honestly
capacity
and
data
onboarding.
This
has
been
a
massive
focus.
A
There's
been
a
ton
of
improvements
happening
there
if
you
haven't
yet
updated
to
lotus
15.2,
there's
a
lot
of
new
improvements
in
there
that
you
should
be
taking
advantage
of
all
of
the
work
that
we
did
kind
of
starting
in
august
of
last
year
and
going
through
march
on
snap
deals,
was
to
help
accelerate
data
onboarding
and
create
the
ability
for
new
groups
to
come
into
the
network
and
just
focus
on
upgrading
existing
sectors.
With
data.
A
There's
been
a
ton
of
work
that
the
slingshot
and
other
kind
of
related
programs
have
been
doing
to
help
onboard
public,
socially
good
data
and
preserve
it
over
time
and
a
ton
of
new
products
that
are
coming
fillmine
boost,
which
you'll
hear
a
lot
more
about
soon.
That
really
help
with
both
capacity
onboarding
and
data
onboarding.
So
that's
really,
first
and
foremost,
it's
falcon's
bread
and
butter.
A
Personally,
this
is
the
thing
that
I'm
super
excited
about:
bringing
programmability
and
compute
into
the
falcoid
network.
We
have
been
hard
at
work
on
this
for
a
long
time.
It
is
we're
approaching
fvm
milestone,
one
which
is,
we
are
going
to
be
re-platforming
the
entire
file
coin
network,
from
kind
of
our
existing
actors
model
and
our
existing
vm
structure
onto
a
wasm,
ipld,
vm
and
switching
over
to
rust
proof
rust
actors
as
like
the
main
kind
of
like
consensus
securing
so
it's
very
exciting.
A
A
That's
our
top
priority,
even
though
we
want
to
bring
these
things
to
you
like
two
years
ago,
and
so
there's
been
a
ton
of
work
happening
over
time
with
the
early
builders
program
to
also
start
today,
building
the
smart
contracts
and
loan
markets
and
new
storage
markets
and
and
compute
systems
that
are
going
to
be
running
on
top
of
file
coin
in
the
future.
And
so
there's
some
really
exciting
work
happening
in
the
compute
over
data
working
group.
Where
a
lot
of
different
groups,
phil
swann,
I
think
I
said
charles
cowell
over
here.
A
Some
of
the
groups
on
bakalao
a
number
of
other
groups-
are
getting
involved
on
building
computation
networks.
So
you
don't
just
compute
over
the
state
and
run
programs
over
the
state
in
file
coin.
But
you
can
then
also
run
programs
that
access
data,
ipl
data
stored
in
file
coin
sectors
and
do
computation
restore
their
data
in
new
file,
coin
sectors
and
iterate
and
like
this
is
so
exciting,
so
cool
not
possible
in
in
blockchain
systems
to
date,
and
so
we're
really
excited
to
be
kind
of
like
breaking
new
ground
here
and
then.
A
Finally,
the
the
last
major
thread
that
is
getting
you
know
more
and
more
of
a
priority.
Many
of
the
people
in,
like
the
spx
working
group,
have
been
doing
a
lot
of
work
around
retrievals
as
part
of
slingshot
evergreen,
where
we
are
retrieving
data
from
you
know
existing
deals
on
the
falcon
network
and
making
sure
that
we
can
up
their
replication
or
we
can
increase
their
geographic
distribution
so
that
data
is
really
effectively
preserved.
A
I'm
sure
many
of
you
have
already
heard
about
retrieval
markets.
If
not
go
talk
to
patrick,
he
will
tell
you
an
onscar
amazing
yeah.
They
will
tell
you
all
about
saturn
all
about
the
amazing
work.
That's
happening
across
mile
there's
a
ton
of
groups
that
I
don't
even
know
like
titan
and,
like
other
amazing
things,
so
go
talk
to
them.
This
is
also
super
super
cool.
A
You
guys
will
probably
all
get
to
run
the
first
retrieval
nodes,
which
makes
me
sad
because
I
want
to
run
one
on
my
phone
all
the
time,
but
really
excited
to
bring
a
whole
new
community
of
folks
into
the
file
coin
network
and
maybe
they're
not
storage
providers,
but
we're
going
to
love
our
retrieval
providers
as
well
and
and
bring
them
into
the
to
the
networking
community
and
then
there's
some
really
cool
stuff.
That's
happening
on
the
crypto
net
lab
team
that
you
should
definitely
go
read
their
notion.
It's
open!
It's
accessible!
A
You
can
read
all
about
these
awesome
new
projects
that
they're
creating
that
will
be
launched
on
top
of
fvm.
There's
a
data
retrievability
oracle
that
there's
already
an
mvp
up
and
running
on
ethereum
testnet.
So
you
can
go
and
play
with
it.
We
may
start
using
this
for
various
things.
I
have
some
ideas.
If
anyone
wants
to
talk
to
me
about
how
we
can
add
additional
guarantees
for
fall
coin
plus
data
to
be
retrievable,
I
would
be
very
interested
in
talking
to
people
about
that.
A
I
think
could
be
cool
and
then
there's
also
some
really
exciting
new
work.
That's
you
know
coming
more
in
the
q1
time
frame,
but
research
and
development
and
prototyping
is
starting
today,
because
we
have
a
long
research
development
pipeline.
We
don't
just
think
about
tomorrow
or
the
next
day
we're
thinking
about
next
year
years
after
that
and
the
amazing
improvements
you
want
to
bring
that
can
enable
you
to
do
data
retrievability
controls
or
data
access
controls
on
top
of
smart
contracts
in
fem.
A
So
we
can
have
all
of
the
programmability
and
compute
and
fill
the
chain
full
of
tons
of
amazing
growth
and
progress,
and
then
we
have
the
the
space
sharding
sub-child
chains
to
make
to
make
room
for
all
of
that,
and
so
that's
just
a
big
picture
thing.
Daniel
and
and
brenda
are
going
to
dive
more,
especially
into
some
of
the
the
boost
and
and
deeper
kind
of
capacity
and
data
onboarding
improvements
that
have
happened
over
the
past
couple
of
months,
but
I
wanted
to
give
you
a
big
picture
and
request
your
feedback
like.
A
I
know
this
needs
to
improve
and
we'll
be
working
on
putting
in
a
place
where
you
can
add
lots
of
stuff
to
this
as
well,
but
I
also
want
to
know
about
all
the
cool
stuff.
That's
happening
so
come
and
tell
me
about
it.
If
it's
not
my
list.
Thank
you,
oh
yeah.
I
have
time
to
take
questions
I
mean.
Actually,
I
think
maybe
let
them
talk
and
then
we'll
all
take
questions
together.
Sound
good.
B
All
right
so
I'll
talk
about
lotus
and
then
I'll
hand
it
over
to
brenda
talk
about
boost
the
indexer
and
saturn
great.
So
last
time
we're
in
austin,
we
gathered
a
lot
of
great
feedback
from
storage
providers.
We
took
that
feedback
and
consolidated
into
a
priority
list
and
that
priority
list
was
passed
along.
The
lotus
team,
the
lowest
team,
was
able
to
use
that
list
to
actually
prioritize
the
work
that
was
going
into
the
next
quarter.
B
B
As
for
the
ccs
core
scheduling
just
significantly
decreasing
the
ceiling
time,
both
those
went
straight
to
the
lotus
team,
they
were
able
to
take
account
that
this
was
a
big
pain
point
and
prioritize
these
issues.
Both
these
have
been
resolved
or
are
in
testing.
The
spx
team
worked
with
the
lotus
team
to
get
some
of
these
a
fix
in
place.
That
fix
is
currently
in
testing
and
is
actually
going
really
well.
So
it's
on
track
to
getting
resolved
as
well.
B
A
few
other
issues
here
related
to
the
lowest
wallet
and
related
to
the
multiple
markets,
node
that
was
talked
about
earlier.
All
of
these,
thanks
to
your
feedback,
has
come
to
the
front
and
center
right
now
is
being
worked
on
by
the
lotus
team
by
the
boost
team
to
get
resolved.
This
quarter,
as
well
as
far
as
upcoming
projects
go
all
this
is
now
public.
This
was
a
private
page
before
this
lotus
roadmap
page,
you
can
see
upcoming
projects.
B
Molly
mentioned
some
upgrades
that
are
happening.
The
lotus
release
timelines
are
now
on
that
page
in
public
as
well.
Network
v16
is
currently
testing
in
butterflynet
it'll
be
testing
calibration
as
early
as
next
week,
and
then
the
full
upgrade
is
expected
to
happen
at
the
end
of
june
and
early
july.
In
the
meantime,
network
v17
is
also
being
scoped,
and
you
can
see
all
those
large
release
timelines
on
that
same
page.
B
B
B
C
Thanks
again
for
coming,
I'm
going
to
cover
a
bit
about
boost,
also
indexer
and
saturn.
Actually,
there's
also
folks
from
the
indexer
team
and
sat
our
team
here.
So
we
can
ask
questions
to
them
later,
but
I'll
just
be
presenting
on
their
behalf,
because
I
know
we're
a
little
short
on
time
so,
really
quickly.
What
is
boost,
I
think
most
of
you
guys
have
actually
given
us
great
feedback
already,
but
it's
essentially
a
new
markets
module
that's
going
to
eventually
replace
replace
the
lotus
market
subsystem.
C
So
it
comes
with
a
number
of
new
features
thanks
to
feedback
from
you
guys
actually
later
earlier
in
the
fall.
So
we've
incorporated
a
number
of
your
feedback
into
into
boost.
So
there
is
a
new
web
ui.
There
is
more
data,
transfer
options,
there's
more
tooling
support
and
if
any
of
you
are
using
bidbar
or
esri
today
to
receive
deals,
it
should
be
working.
So,
if
you're
running
into
problems,
please
report
it
to
us.
C
C
So
if
you
run
into
problems,
please
let
us
know
we
want
to
address
them
immediately
and
quickly,
so
that
we
have
a
very
stable
release
for
everyone
and
really
quickly
to
share
roadmap
as
daniel
and
molly
both
mentioned
were
looking
to
scale
boost,
so
you
can
run
multiple
markets
nodes
to
work
with
your
miners.
So
we'd
really
like
to
better
understand
what
the
main
limitations
are
with
with
when
you're
onboarding
data,
so
we
can
better
scale
out
boost.
C
We
also
are
looking
to
do
some
of
the
retrieval
capabilities
that
we
haven't
been
able
to
focus
on
yet.
So,
if
you
have
any
information
about
how
you
want
to
operationalize
your
retrieval
deals
and
how
to
run
all
of
your,
like,
I
guess,
interact
with
clients
to
work
on
retrievals.
Please
let
us
know
we
that
way
we
can
help.
It
will
help
us
prioritize
how
we're
going
to
work
on
whether
it's
enabling
http
retrieval
or
like
bitswap
retrieval
for
ipfs
retrievals.
That
would
be
really
helpful
to
us
yeah.
C
So
that's
really
awesome.
Thanks
to
the
great
team
that's
been
working
on
this
hong
kong
from
the
indexer
team
is
also
here,
so
feel
free
to
ask
him
questions
later
about
indexer.
If
you
either
want
to
run
it,
we're
looking
for
people
to
run
indexer
nodes
as
well.
So
if
you're
interested
in
doing
that,
let
us
know
we'll
help.
You
get
started
all
right,
yeah
updates.
C
Our
team
is
running
an
indexer
node
in
production.
So
for
those
of
you
who
have
enabled
this,
you
were
actively
indexing.
Your
data
there's
also
other
teams
that
are
running
index
or
nodes.
So
it's
really
awesome
to
see
how
much
content's
been
indexed
and
enable
more
content
discovery.
So
a
couple
next
steps
here
you
can
read
about
it
here,
but
essentially
we
want
to
scale
it
to
pebby
byte
scale.
We
want
to
also
decentralize
the
index
or
nodes.
C
Some
resources
here
feel
free
to
yeah
people
to
read
at
your
leisure
saturn,
alright,
so
I'm
also
presenting
this
on
behalf
of
the
saturn
team.
So
please
talk
to
patrick
or
onskar,
who
are
there
in
the
back?
If
you
have
additional
questions
later
on
so
saturn,
I
think
if
you
guys
haven't
heard
of
it
yet
is
actually
all
the
retrieval
markets.
Work
that's
been
happening,
but
it's
going
to
be
at
a
centralized
cdn
for
retrieval
on
file
coin.
So
to
give
a
really
quick
overview.
C
There's
kind
of
like
three
layers
of
this
cdn
that
that's
being
worked
on
right
now:
one
is
kind
of
the
network
edge
nodes
which
are
called
the
l1
nodes,
there's
also
the
intermediate
cache
nodes
called
l2
nodes
and
then
there's
a
storage
writers.
You
guys
who
are
actually
holding
all
the
content,
and
this
is
the
content-
that's
going
to
be
served
on
the
cdn,
so
the
way
that
the
network
flow
is
going
to
go
is
that
you
know
you
have
these
l1
nodes
who
are
going
to
be
doing
most
of
the.
C
C
Right
now,
the
l1
network
is
live
and
private
testing
and
you
can
see
the
roadmap
here
in
q2.
We
want
to
hopefully
beat
the
ipf's
gateway
and
performance.
We
want
to
launch
the
l1s
publicly
in
q3,
so
people
can
run
their
own
l1
nodes
and
earn
filecoin
for
doing
so
and
yeah
in
q4
more
launches
for
l2
l2
nodes,
so
yeah.
If
you
want
to
get
involved,
please
reach
out
to
onscore
or
patrick
there's
lots
of
work
going
on
there.
D
A
So
I
think,
there's
a
couple
of
different
fvm
launches,
there's
fem
m1,
which
is
coming
in
the
next
network,
upgrade
that
folks
mentioned.
I
think
rc
going
out
next
week
and
calibration
net
upgrades
you'll
be
able
to
test
for
yourself
in
calibration
net
how
how
gas
changes
there.
There
was
an
estimate
that
I
believe
was
put
in
the
fip
that
I
don't
remember
off
the
top
of
my
head
for
the
the
sort
of
increases
that
we
expect.
A
I
think
there
is
going
to
be
a
slight
effect,
but
we
think
is
pretty
minimal,
especially
because
we
have
all
of
the
space
of
of
aggregations
for
pre-commit
post
command,
all
of
that
stuff,
all
the
batching
capabilities
of
the
network
that
are
currently
not
being
utilized.
So
we
have
lots
of
space
to
to
move
into
that
before
we
could
should
expect
any
significant
impacts
on
like
base
fee
or
anything
like
that.
A
We
don't
have
enough
data
yet
for
what
it'll
look
like
once,
we
have
a
lot
of
smart
contracts
on
filecoin,
that's
an
unsolved
problem.
Yet
we
don't
know,
but
we'll
be
doing
a
lot
of
benchmarking
and
testing
ahead
of
time
to
see
how
things
go,
and
we
also
that's
one
of
the
reasons
we're
working
on
all
of
the
consensus,
starting
in
parallel,
so
that
we
have
the
increased
chain
space
so
that,
as
we
hopefully
have
many
amazing
smart
contracts
running
on
top
of
file
coin,
doing
all
sorts
of
useful
things.
A
We
also
have
the
space
to
expand
into
again
we're
not
currently
utilizing
anything
close
to
the
capacity
that
falcon
has
available,
and
we
can
always
increase
our
ability
to
batch
proof
commits
and
pre-commits,
even
more
than
what's
currently
allowed
in
the
network.
If
we
were
to
reach
those
limits,
and
so
we
actually
do
have
a
lot
of
space
to
expand
into
before.
We
should
expect
any
like
massive
changes.
E
A
Deals
that
like
no
one
can
get
out
of
without
getting
slashed
massively
so
like.
We
need
that
capability,
like
obviously,
if
we
get
to
five
years
from
now
and
we're
like
nope
everything's
cool,
we're
all
good.
It's
chill
then,
like
you
know,
those
are
those
are
numbers
that
we
as
a
network,
can
file
fips
and
talk
about
changing
back
in
whatever
it
was.
A
When
the
the
v1
proofs,
I
think,
a
year
ago
february,
when
the
v1
proofs
were
expiring
and
there's
a
set
of
people
who
had
done
six
month,
v1
proofs,
they
were
like
okay,
like
what,
if
we
can
just
extend
those
up
to
the
length
of
all
of
the
other
proofs,
we
decided
as
a
network.
You
know,
I
think
this
was
initiated
by
either
someone
on
the
ipf's
main
team
or
someone
on
the
ipfs
force
team,
one
of
those
that
that
okay,
there
was
a
vote,
a
poll
of
all
storage
providers.
A
We
decided
we
wanted
to
to
take
that
on
and
we
did,
and
so
these
are
things
that
that
we
can't
adjust.
Personally,
I
would
I
would
love
to
see
the
the
length
of
a
deal
that
you
can
initially
commit
to
the
network
be
longer
again.
It
is
a
security
parameter,
and
so
we
need
to
do
that
with
kind
of
a
thoughtful
modeling
of
like
well.
F
Oh,
I
was
just
gonna,
also
throw
in
one
comment
on
that,
like
I
think
it's
also
probably
important
to
distinguish
between
like
what
do
we
need
at,
like
the
proof
level
and
what
are
things
that
could
be
smart
contracts
on
top.
If
we're
talking
about
like
from
the
client
side,
how
do
we
get
like
an
arbitrary
length
deal
that
could
easily
be
a
smart
contract
which
is
like
cid,
some
number
of
copies,
some
funds
that
are
just
set
as
a
bounty
and
then
building
logic
in
between
that
just
says.
Like?
F
Do
we
always
see
some
number
of
proofs
within
some
number
of
hours
so
like
from
the
perspective
of
like
the
minor,
where
it's
like,
we
want
to
minimize
the
amount
of
times
that
you
have
to
like
over
a
ten
year
period
like
reseal,
do
whatever
and
interact
with
some
deal
making
thing.
I
think
that's
like
a
good
ux
thing
to
like
solve
for,
but
from
the
perspective
of
the
client,
hopefully
we'll
have
like
smart
contracts.
That
sort
of
like
hide
a
good
chunk
of
that
bit,
so
we
can
actually
get
to
something.
G
A
Retrieval
markets
that
are
being
produced
so
there's
like
the
mile
pop
network,
which
is
you
know,
retriever
or
client,
pays
for
fetching
data.
I
think
they're
working
on
one,
though
that
is
provider
pays
as
well.
You
can
like
prepay
for
for
your
retrievals
and
there's
saturn,
which
is
like
provider,
pays
and
there's
like
a
couple
of
others.
So
every
flavor
of
payment
scheme
that
you
want
there'll,
probably
be
a
retrieval
market
for
it.
G
Also,
the
survey
you're
gonna
get
like
a
link
to
the
survey
right
that
daniel
mentioned.
I
just
want
to
re-emphasize
that
because
the
great
like
what
you're
seeing
here
today
is
an
outcome
of
the
previous
storage
meetup
and
really
appreciate
all
your
your
feedback.
So
please
use
the
survey
as
well
to
provide
your
feedback
into
the
engineering
thing,
because
the
engineering
team
wants
to
hear
from
you.
So
please
help
us,
like
you
know,
whatever
features
you're
looking
for
help
us
prioritize
it.
It's.
A
D
C
Actually,
we've
literally
just
talked
about
this.
The
boost
team
actually
was
just
talking
about
this
feedback
yesterday
during
our
stand-up,
and
I
think,
if
you
have
specific
items
that
you're
looking
for
like
if
there's
specific
commands
that
you
want
to
us
to
support
or
things
that
are
in
the
gui
that
you
actually
would
prefer
on
command
line
like.
I
would
encourage
you
to
like
create
an
issue,
because
we,
I
think
it's
hard
to
track
all
the
things
that
are
happening
in
different
threads
and
different
discussions.
C
So
if
you
have
it
on
github,
it's
really
easy
for
the
dev
team
and
me
to
look
at
it
and
go
through
and
actually
like,
take
action.
So
that
would
be
really
great,
because
we
were
looking
for
additional
details
and
it
was
like.
Where
are
these
details
living
we
don't
quite
know,
and
so
so
yeah?
That
would
be
great
if
we
could
hear
that
from
you
yeah,
and
I
think
also
one
thing
I
did
want
to
call
out
for
everyone
here.
C
I
know
you've
been
super
helpful,
stu,
giving
us
feedback
as
well,
but
for
anyone
who
has
specific
feedback
on
scaling
markets
nodes,
there
is
a
discussion
link.
That's
in
this
deck
for
boost
specifically,
please
go
and
write
your
feedback
there,
because
we're
brainstorming
ideas.
How
we
want
to
do
the
scaling,
how
I
want
to
make
boosts
more
stateless,
and
so
it
would
be
really
helpful
if
we
can
get
information
around
like
how
your
mining
operation
is
set
up.
So
we
can
better
optimize
for
the
folks
here.
C
So
that's
one
thing
and
then
one
other
thing:
I'm
just
calling
out
all
the
boost
items
right
now,
if
you
have
also
input
on
how
you
want
to
do
your
retrievals
like
if
you
want
to
do
hdb
http
retrieval,
I
think,
there's
some
overhead
that
might
come
with
like
running
certain
operations
for
that
or
if
you
are
interested
in
having
like
a
way
to
do
bit,
swap
data
transfer,
which
is
would
be
compatible
with
ipfs
like.
Is
that
something
that
you
want
and
are
willing
to
run?
C
D
And
the
other
question
was
around
the
indexer
now
as
storage
providers
rather
than
running
an
index
node,
but
what
I've
seen
is
there's
a
percentage
like
15
of
my
you
know.
Pcids
are
not
in
a
good
state
right,
so
part
of
it
is
trying
to
figure
out
how
to
how
to
diagnose
and
test
that.
But
on
the
other
side
of
things,
what
we're
seeing
is
people
actually
publish
something
gets
on
chain?
It's
it's
proving
yeah.
H
D
C
Okay,
so
I
don't
know
like
home,
call
feel
free
to
jump
in
as
well,
but
based
on
my
understanding,
the
indexer
notes
actually
will
go
and
pull
a
delta
of
like
from
directly
from
you
as
a
storage
writer,
in
terms
of
like
the
sids
that
you
have.
So
I
think,
if
you
are
like
you
know,
fixing
something
based
on
whatever
it
is
that
you're
storing
like
it
should
be
able
to
get
that
delta
on
a
like
daily
or
maybe
like
three
or.
D
H
D
H
D
I
Think
previously
there
was
some
issue
with
the
with
the
uptime
of
the
investor
and
we
like
to
solve
that.
We
have
basically
around
another
instance
of
the
investor
now
like
to
keep
the
uptime
high.
I
suspect
that
might
be
related
to
that,
but
we
I'm
happy
to
follow
up
with
you
offline
and
we
can
kind
of
open
the
issue.