►
Description
Moderated by Director of The Graph Foundation Eva Beylin, this Graph Day 2022 panel shone a light on how each core developer contributes to The Graph ecosystem. Joseph Al-Chami of Figment, Jannis Pohlmann of Edge & Node, Sam Green of Semiotic Labs, Uri Goldshtein of The Guild, and Chris Wessels of GraphOps engaged in a dynamic conversation regarding what it means to be a core dev. During the panel, Al-Chami announced The Graph’s beta hosted service integration of Cosmos Hub, marking the first Cosmos chain to be indexed by The Graph. Upcoming support for Osmosis was announced as well. Additionally, Beylin prompted panelists to consider how developers have been using The Graph Client, a tool built by The Guild, and what benefits it provides.
Read more about The Graph’s Cosmos Hub integration: https://thegraph.com/blog/the-graph-cosmos-hub-integration
Discover The Graph Client on Github: https://github.com/graphprotocol/graph-client
Watch more Graph Day 2022 talks, panels, and announcements here: https://www.youtube.com/playlist?list=PLTqyKgxaGF3SWt8ipQw-Ul9HQvmxulbUu
Time Stamps
0:00 Introductions
10:25 Figment and Multichain
14:24 Graph Node Team
17:11 Snark Force
21:30 What is it like working on each team
30:10 What can end users understand about the roadmap?
31:45 The hosted service
34:35 Advice to indexers
41:55 How to get more web2 developers onto web3
Keep up with The Graph on social media!
Twitter: https://twitter.com/graphprotocol
Instagram: https://instagram.com/graphprotocol
LinkedIn: https://www.linkedin.com/company/thegraph
GitHub: https://github.com/graphprotocol
Website: https://thegraph.com
A
I
will
say:
I'm
surprised
the
stage
isn't
being
weighed
down
from
all
the
big
brains
sitting.
Next
to
me,
I
want
to
first
get
started
with
just
a
quick
intro.
I
know
I
gave
a
little
bit
of
an
overview
today,
but
if
everyone
can
go
around
say
your
name,
maybe
what
you
do
at
your
core
developer
team
and
where
you
work.
B
Yeah
sure
should
I
start
yeah:
okay,
hello,
everyone.
My
name
is
joseph
hashemi
and
I
am
product
manager
at
figment
and
that
figment.
We
recently
joined
like
almost
a
year
ago
as
a
core
dev
team
with
the
graph,
and
our
focus
has
been
majorly
on
integrating
new
chains
and
working
on
the
multi-chain
vision.
C
E
Hi,
I'm
molly,
I'm
one
of
the
founders
of
a
group
called
the
guild.
We've
been
doing
a
lot
of
open
source
around
graphql
and
we've
joined
the
graph
around
six
months
ago
to
help
with
basically
everything
graphql
and
also
developer
toolings,
and
it's
the
first
time
in
history
of
the
guild
that
everyone
is
here
in
person.
So
thanks.
F
F
A
A
So
I
want
to
set
the
stage
a
little
bit
and
remind
everyone
that
we
have
five
working
groups
within
the
graph
core
development,
indexer
experience,
protocol
economics,
snark
force,
protocol
operations
and
network
operations,
and
then
data
and
apis,
and
I
want
to
start
actually
with
yuri
to
speak
a
little
more
about.
You
know
what
you've
been
doing
with
the
graph
client
and
how
that
ties
into
these
working
groups.
E
Yeah
so
today
we
announced
the
graphic
line
1.0.
Finally,
so
thanks
for
everyone,
who's
been
involved,
all
of
the
guild
is
sitting
here.
I
mentioned
like
it's
the
first
time
we
all
actually
met
in
person
after
five
years
of
working
together
and
the
idea.
So
when
we
joined
the
graph
about
six
months
ago,
we
looked
at
like
what
is
the
developer
experience
of
people
using
the
graph
and
using
graphql
with
the
graph
and
the
graph
client.
I
think
how
we
see
the
graph
client
is.
E
You
know
there
was
a
lot
of
talk
during
the
conference
about
you
know
all
the
amazing
things
that
the
graph
and
everything
that
everyone's
building
is
doing
and
I
think
for
us.
The
graph
client
comes
from
the
idea
that
doing
the
best
thing
should
be
the
easiest,
so
the
graphic
line
for
us,
it's
not
just
the
best
way
to
interact
with
the
graph.
E
It's
the
best
way
to
integrate,
to
interact
with
graphql
in
general,
like
we
believe
that
this
release
of
the
graph
client,
and
also
the
next
releases
after
that,
even
for
let's
say
web
2
developers
that
use
graphql,
they
should
look
at
the
graph
client
like
we
integrated
a
lot
of
ideas
that
we
learned
throughout
all
the
years
in
graphql
and
throughout
the
six
months
with
the
graph
into
this.
Let's
call
it
a
framework
of
basically
using
graphql
in
your
applications,
so
there's
many
things
in
it.
I'm
not
gonna
dive
into
all
the
details.
E
A
E
So
there's
many,
I
think,
first
of
all,
a
lot
of
when
we
started
the
graphic
working
on
the
graph
client.
We
basically
interviewed
a
lot
of
sub
graph
developers
and
app
developers,
and
we
just
saw
the
patterns
that
people
keep
doing
manually
and
we
tried
to
put
that
into
a
framework.
So,
for
example,
if
you
now
want
to
query
multiple
sub
graphs
now
you
can
basically
automatically
automatically
with
small
configurations.
E
I
can
basically
like
basically
mention
all
those
different
sub
graphs
and
get
one
graphql
unified,
graphql
api.
That
runs
on
your
client
that
calls
all
these
different
subgraphs.
So
that's
where
it
started,
but
then
we
saw
more
and
more
patterns
like
you
know
how
people
today
do
pagination,
for
example,
so
the
framework
actually
now
takes
care
of
that
for
you
automatically,
so
it
does
automatic
pagination
automatic
block
tracking,
like
all
kinds
of
different
things.
E
A
I
don't
know
how
much
time
yeah
that's
great.
I
wanted
to
switch
over
to
chris
first
off.
Congratulations
again
on
your
grant
really
exciting.
I
know
your
team
is
working
closely
with
the
core
subgraph
for
the
graph,
but
wanted
to
hear
more
about
your
vision
for
the
team
and
graph
ops
joining
the
mission.
F
F
I
think
it's
it's
worth
talking,
maybe
about
two
of
the
things
that
we'll
be
focusing
on
in
the
near
term.
The
first
is
trying
to
create
orchestration
and
automation,
tooling
for
indexers
that
are
kubernetes
centric
and
so
generally.
We
believe
that
kubernetes
is
the
right
platform
that
indexers
should
be
targeting
when
they
deploy
their
indexing
operation
across
whatever
infrastructure
that
they
leverage,
whether
that's
public
cloud
or
private,
bare
metal
machines.
F
And
today
the
state
of
of
the
graph
stack
on
kubernetes
is
is
okay.
Certainly
some
indexers
run
it
that
way,
but
we
would
like
to
make
that
the
de
facto
choice
for
indexing,
because
we
think
that
it
is
the
right
framework
for
indexers
to
reason
about
their
infrastructure,
and
it
is
far
more
amenable
to
the
scaling
requirements
that
we
know
are
coming
for.
Indexers
as
the
graph
expands
across
many
blockchains
and
also
as
the
fire
hose
becomes
more
of
a
core
component
of
the
indexing
stack.
F
I
effectively
need
to
do
that
on
chain
and
the
costs
of
doing
so
are
obviously
quite
substantial,
and
we
hope
that
l2
and
we
know
that
l2
will
lower
that
substantially,
but
having
the
protocol
intermediate.
All
of
those
interactions
somewhat
creates
like
a
a
floor
for
like
the
minimum
value
of
a
message
or
a
signal
that
any
participant
would
need
to
send
to
any
other
participant
right.
F
It
needs
to
be
worth
the
gas
and
so
by
introducing
a
gossip
network
and
allowing
initially
indexers,
but
we
do
think
that
this
could
scale
quite
nicely
across
other
participants
too,
to
effectively
gossip
with
one
another
over
a
p2p
network.
This
really
unlocks
a
new
design
space
for
coordination,
and
I
would
say
that
that
that
doesn't
replace
the
coordination
that
the
protocol
does
today.
F
A
Incredible
really
looking
forward
to
that
as
a
segway,
you
mentioned
proofs
of
indexing,
and
this
is
highly
relevant
to
new
chains
and
supporting
multi-chain.
I
wanted
to
speak
with
joseph
here,
learn
a
little
more
about
your
role
and
and
with
figment
and
what
you
guys
have
been
working
on
in
regards
to
multi-chain.
B
B
Looking
at,
you
always
want
to
look
at
the
people
at
the
frontier
who's
pushing
this
front
here,
and
you
want
to
see
like
where
they're
headed
and
when
we
looked
at
web
3
and
we
we
try
to
make
the
same
analogy
in
warp
3.
We
start
looking
at
the
builders,
because
the
builders
are
the
one
who's
pushing
the
frontier
of
f3
and
we're
trying
to
look
where.
Where
are
these
builders
going
to
and
which
front
here?
Are
they
pushing
towards
and
it
became
natural
to
think
about
to?
B
So
if
you're
a
builder,
you
know
solid
solidity
and
you
want
to
like
leverage
the
highest
security
blockchain,
you
will
build
your
applications
on
top
of
ethereum.
But
if
you
want
a
different
language,
you
wanted
something:
that's
more
flexible,
less
transaction,
because
fees,
you
might
let's
say,
go
into
cosmos,
and
so,
when
we
start
looking
at
the
space
is
moving
into
this
multi-chain
future.
We
start
being
a
lot
as
well
as
your
company
agnostic
chain.
So
we
we
believe
in
all
chains.
B
And
when
we
talk
to
the
builders,
a
lot
of
them
are
building
like
many
cool
awesome
applications.
But
their
main
problem
is
how
they're
going
to
get
the
data
out
of
the
blockchain,
and
so
a
lot
of
them
are
wasting
their
talent,
wasting
their
effort,
trying
to
build
indexers
trying
to
run
nodes,
and
it's
not
working
for
them.
They're,
wasting
resources
to
figure
it
out,
and
so
they're
losing
a
lot
of
advantage
in
the
space.
B
B
So
if
we
identify
builders
in
many
different
chains
building-
and
they
are
using
essentially
service,
it's
our
job
and
our
duty
to
bring
this
naturalization
to
them,
and
so
that's
what
we've
been
trying
to
do
and
the
first
network
we
started
with
is
cosmos
and
nine
months
ago
we
announced
that
we're
going
to
be
integrating
cosmos.
So
we
started
researching
cosmos
and
cosmos
is
like
a
really
innovative
iso
idea
like
the
way
it
works
in
the
cosmosic
system.
B
Everyone
could
just
come
and
use
it
to
build
applications
on
top
of
it,
and
so,
when
we
start
researching
it,
we
we
wanted
to
know
like
how
could
the
graph
make
the
most
impact,
and
so,
instead
of
starting
integrating
the
application
layers,
we
went
all
the
way
down
and
we
integrated
tendermint.
So
we
can
have
all
the
applications
later
built
on
top
of
tendermint
easily
and
rapidly
enough,
and
I'm
so
excited
today
to
share
the
news
that
we
finally
completed
the
integration
of
cosmos
hub
network.
B
And
now
all
the
cosmos
developers
they
can
use
the
graph
to
get
their
decentralized
blockchain
data
using
the
graph
and
thanks
to
the
beautiful
team
that
we
have
over
here
to
your
efforts
to
our
partnerships
with
edunote
and
streaming
tests.
We
made
this
possible
oh
and
one
more
announcement
five
minutes
ago
or
something
like
that.
We
released
the
osmosis
gip
and
we
have
a
better
version
of
the
integration
of
osmosis.
A
Wow
amazing
really
excited
to
hear,
I
would
say:
cosmos
has
been
one
of
the
more
more
requested
chains.
I
wanted
to
flip
it
over
to
giannis
and
hear
a
little
more
about
the
graph
node
team
and
what
you've
been
doing
to
improve
indexing
performance.
This
is
one
of
the
things
dapps
look
out
for
the
most
and
you
know
substreams
was
mentioned,
but
we'd
love
to
hear
a
little
more
about
that
and
how
that
impacts.
App
developers.
C
Yeah,
I'm
going
to
try
not
to
spill
the
beans
too
much,
because
we
have
sebastian's
talk
still
coming
up
and
I
think
a
couple
hours
was
meant
to
be
earlier,
but
he
he
missed
one
of
his
planes
so
yeah.
The
graph
now
team
has
been
working
on
a
number
of
performance
improvements.
I
do
so
like
that
goes
all
the
way
to
making
making
changes
to
postgres
itself
and
trying
to
upstream
index
so
like
database
index
performance
improvements
into
postgres,
which
would
be
pretty
huge.
C
I
think
we
have
reached
sort
of
the
limits
of
what
sequential
processing
can
can
provide
in
terms
of
performance,
which
is
why
the
whole.
You
know
part
of
the
reason
why
we're
having
the
substrings
conversation
where
substreams
have
been
worked
on
and
the
key
to
kind
of
pushing
beyond
that,
and
not
just
you
know
achieving
incremental
performance.
Improvements
like
we
have
been
the
last
few
years,
I
think,
is
to
parallelize
processing
a
lot,
and
this
is
part
of
what
what
substreams
is
about.
C
It's
about
parallelizing
transformations
that
don't
have
data
dependencies,
whereas
in
subgraphs,
unless
users
draw
specific
boundaries
for
data
dependencies
which
they
currently
can't
do
it's
hard
to
do
anything
in
parallel
and
substreams
yeah.
The
parallel
processing
is
one
aspect.
Composition
at
indexing
time
is
something
we've
wanted
to
do
in
graph,
node
and
subgraphs
for
a
while,
but
kind
of
the
flexibility
modularity
of
substreams
make
that
a
little
easier
and
yeah
there's
a
lot
more.
C
The
essentially
yeah
substreams
do
allow
use
cases
that
are
not
currently
possible
with
the
graphql
api
that
that
graph
node
is
providing,
I
mean
graphql
is
very
well
suited,
for
you
know:
web
user
web
applications,
mobile
applications
less
so
for
analytics
use
cases
or
consuming
subgraph
data
in
yeah
in
and
kind
of
pulling
it
into.
Other
databases
usually
requires
kind
of
scraping
paginating
through
data
in
subgraphs,
which
a
lot
of
teams
are
doing.
That's
not
fun
and
not
very
efficient,
so
yeah
we'll
prove
a
lot
on
that.
C
A
Amazing
and
kind
of
on
the
same
track.
You
know
one
of
the
the
components
of
the
network
and
improving
indexing
performance
is
verifiability,
so
I
wanted
to
pass
it
on
to
sam
hear
a
little
more
about.
You
know
what
is
the
snark
force?
What
what
is
the
path
to
verifiability,
and
also
you
know
why
of
all
problems?
Did
you
guys
choose
to
solve
the
graphs
verifiability
problems.
D
So
what
this
team
is
doing
is
they
had
to
they?
These
guys
basically
had
to
pre
comb,
the
previous
literature
and
all
the
previous
available
solutions
for
four
techniques
for
zero
knowledge,
proofs
and
verifiable
computing,
and
there
were
no
techniques
that
satisfied
all
the
needs
of
the
graph
and
then
they
basically
had
to
come
up.
Of
course,
building
on
top
of
existing
techniques.
D
All
requirements
for
credible
neutrality
within
the
graph,
and
we
want
cryptographic
proven
we
want
cryptographic
security
for
everything
we
do
within
the
protocol.
We
don't
want
any
any.
You
may
not
have
heard
this
word
today,
but
we
don't
want
any
human
arbitration
when
there
are
problems
with
with
the
various
participants
in
the
protocol.
We
want
everything
to
be
provable
on
chain
and
it
needs
to
be
provable
efficiently
on
chain
and
that's
what
snarks?
Let
us?
D
This
is
the
history
expiry
problem.
This
will
let
it
this
will
make
it
possible
for
people
to
run
full
nodes
without
having
to
have
ssds
that
can
have
that
can
be
10
plus
terabytes
in
size.
It
will
allow
people
to
run
full
okay,
so
full
nodes
are
the
things
that
actually
are
providing
security
within
ethereum.
And
how
can
we
get
rid
of
this
requirement?
D
Well
with
this
with
this
state
expiry
problem,
it
can
let
the
graph
potentially
be
used
for
users
to
query
historical
data
which
is
currently
done
by
full
nodes,
so
yeah,
so
our
work
for
verifiable
queries
could
lead
to
a
solution
for
ethereum
itself
and
the
question
of
why.
Why
are
we
working
on
this?
So
if
you
look
across
the
snark
team,
there
are
people
that
have
been
in
our
team.
There
have
been.
D
We
were
currently
experiencing
another
explosion,
which
is
being
motivated
by
the
need
for
offloading
computing
from
the
evm
to
a
verifiable
method
like
what
snarks
provide,
and
this
is
going
to
this
is
going
to
give
us
this.
We
don't
really
know
how
it's
going
to
be
huge.
I
think,
just
like
e-commerce
was
huge
and
we're
going
to
see
something
else
huge
starting
within
the
graph
from
these
techniques.
A
E
I
mean
I
can
say
from
our
side
we
kind
of
like
to
we're
like
we
just
in
our
own
private
slack.
We
just
called
ourselves
like
the
web
two
boomers
around
here
like
and,
and
the
thing
is
we're
like
we
came
from
like
you
know,
maintaining
and
building
a
lot
of
tools
for
the
web
to
world
and
then
came
here
and
for
us
actually
like
this
transition.
E
C
C
But
we
don't
have
like
the
level
of
experience
that
the
guild
has.
For
instance,
we
don't
have
level
of
cryptography
experience
and
an
ai
experience
that
somebody
have,
and
you
know
not
the
multi
blockchain
experience.
The
graph
has
been
ethereum
only
for
a
long
time.
What
hasn't
been
for
a
while,
thanks
to
the
work
that
you've
been
doing
and
the
work
that
streaming
fire
has
been
doing,
so
I
think
it's
a
necessity
and
it's
very
very
good.
C
A
C
Yeah,
I
mean
obviously
brings
its
own
challenges.
You
know,
especially
in
terms
of
governance
and
coordination.
To
me
personally,
I
mean
for
me
kind
of
working
in
the
opens
very
personal
thing.
I've
been
doing
open
source
for
almost
20
years
now,
since,
like
20
2004
or
something,
and
whenever
I
didn't,
I
wasn't
happy
so
to
me
like
working
in
the
open
working
software
that
other
people
can
contribute
to.
C
A
F
I'd
like
to
add
something
it
touches
on
the
prior
question
as
well,
and
I
I
just
want
to
say
something
about
the
culture
of
the
graph,
because
I
think
that
is
so
seminal
to
this
being
possible
and
in
in
great
companies
you
tend
to
see
a
culture
where
the
best
idea
wins.
It
doesn't
matter
where
it
comes
from.
F
What
we
see
here
would
not
be
possible
so
that
that's
one
really
amazing
thing
about
working
in
this
environment,
and
I
think
you
also
have
this
cross-pollination
of
ideas
right,
and
that
is
great
for
coming
to
the
kind
of
best
technical
solution
which
I
feel
is
like
a
lot
of
what
we've
spoken
about
so
far.
But
I
think
that
you
know
we
across
the
core
developer
group-
and
you
know
the
core
developer
group
is
certainly
not
the
only
contributors
to
the
protocol.
F
F
The
life
experiences
the
relationships
that
these
individuals
have
with
their
governments,
their
institutions
they
help
perhaps
less
in
a
technical
context,
but
they're
enormously
valuable
in
a
product
context,
because
ultimately
we're
building
products
for
people,
and
I
think
that's
yeah,
just
something.
That's
really
struck
me
working
in
the
ecosystem,
such
a
broad
diversity
of
personal
experience,
really
brings
something
special
to
our
prioritization
processes
and
what
we
decide
to
build.
B
If
I
jump
jump
in
and
for
me
just
to
provide
like
a
different
angle,
it's
very
challenging
and
very
innovative
like
it's
something
that
you
really
want
to
look
at
from,
like
from.
Like
close
perspective
like,
I
think,
a
lot
of
people
here
today
are
building
their
own
startups
and
they
want
to
scale
their
own
startups,
like,
as
you
start,
hiring
a
lot
of
people.
You
start
having
multiple
departments
and
the
challenge
becomes
a
lot
like.
B
How
would
you
how
should
the
marketing
team
talk
and
prioritize
with
the
sales
team
with
the
product
team
with
the
engineering
team,
and
you
start
seeing
a
lot
of
this
type
of
communications
that
becomes
like
one
of
the
biggest
challenges?
As
for
you
to
scale
your
own
startup,
but
in
startups
it's
easy
because,
like
it
always
funds
falls
under
one
leadership.
You
have
like
our
leaders,
the
exact
they
are
thinking
they
are
giving
us
like
objectives
roadmaps.
B
If
they
have
like
like
a
weak
mission,
the
startup
is
going
to
fall
unless,
like
they
really
have
something
that
they
look
forward
to.
I
think
here
it's
what's
so
easy
with
all
the
other
car
depths
is
because
we
were
looking
at
the
graph,
and
we
know
how
much
the
graph
is
important
for
the
entire
ecosystem,
and
I
think
that
brought
the
entire
core
death
together,
and
it
was
easy
actually
to
start
working
with
all
these
great
minds
over
here.
A
So
almost
like
the
road
map,
united,
all
these
teams-
yeah
yeah.
I
know
we've
been
pretty
public
and
transparent
about
our
roadmap,
but
it's
also
a
little
convoluted.
Sometimes
so,
sam
I'd
love
to
hear
from
you.
You
know,
what's
one
thing
that
you
wish
dapps
or
end
users
better
understood
about
the
roadmap,
and
you
know
some
something:
that's
really
impactful
to
them
that
maybe
they're
not
grasping.
D
D
This
is
going
to
make
it
easier
for
indexers
to
choose
which
subgraphs
subgraphs
to
index
in
order
to
maximize
their
profits,
and
we
also
released
a
tool
for
indexers
to
is
called
auto
agora.
It's
a
tool
for
indexers
to
automatically
adjust
prices
to
be
competitive
in
the
market,
and
we
see
this
is
just.
A
Sounds
good,
I
want
to
switch
gears
just
a
little
bit
back
to
the
announcement
earlier
today
that
the
hosted
service
is
sun,
setting
in
q1
2023
very
exciting
and
wanted
to
hear
from
giannis.
You
know
this
milestone
is
massive
you've
been
here
since
the
hosted
service
started.
What
does
this
mean
to
you
and
what
are
you
most
excited
about
on
this
journey?.
C
C
C
The
hosted
service
was
then
the
quickest
way
to
then
get
that
to
have
a
hosted
experience
where
people
wouldn't
have
to
run
their
own
infrastructure
for
consumers
for
developers
much
very,
very
similar
to
the
network
now
where
they
don't
have
to
worry
about
the
infrastructure
themselves,
but
it
was
never
gonna,
be
the
end
game,
and
so
for
me,
it's
a
natural
next
step,
I'm
glad
we're
making
it.
I
think
the
the
team
that's
been
running.
C
The
hosted
service
has
been
amazing
at
keeping
it
up
and
you've
seen
the
numbers
of
kind
of
the
traffic
it's
seen.
So
I
would
like
to
all
would
like
you
all
to
give
them
a
round
of
applause.
Don't
see
that
where
they
are
sitting
at
the
moment
I
see
leo
over
there,
I'm
not
sure
what
david
is.
Perhaps
we're
gonna
give
them
a
round
of
applause.
C
One
thing
that
I'm
really
excited
about
for
the
next
next
few
months
is
building
out
the
network
kind
of
especially
the
the
number
of
chains
that
the
network
supports.
Like
we've,
seen
the
hosted
service
supporting
something
like
34
different
chains
right
now,
whereas
the
network
supports
ethereum
mainer
and
that's
it.
C
I'm
yeah
very
much
looking
forward
to
the
team
to
the
work
that's
already
under
underway,
which
will
enable
indexing
rewards
for
other
chains,
and
then
that
will
enable
you
know
then
I'll
incentivize
indexers
to
support
subgraphs
for
these
other
chains
as
well,
and
then
we'll
slowly.
Move
things
over
from
those
servers,
provide
more
and
more
networks,
more
and
more
chains
on
the
network,
and
that's
something
really
looking
forward
to.
A
Yeah
really
incredible,
you
know
that
we're.
Finally,
at
the
point
of
adding
new
chains,
I
know
joseph
this
is
very
close
to
your
heart.
You
know
what
advice,
maybe
can
you
share
with
indexers,
who
not
only
need
to
now
run
ethereum
mainnet
to
support
the
network,
but
will
now
start
to
run
that
of
different
chains
and
they'll
even
choose
which
chains
they
want
to
run
so
anything
you
could
share.
B
Yeah,
so
I
would
say
like
one
of
the
most
things
when,
when
you
want
to
start
supporting
like
new
networks,
familiarize
yourself
with
these
ecosystems,
so
let's
say
like
right
now:
we're
going
to
be
integrating
cosmos
hub
chain
and
next
osmosis
and
maybe
like
other
few
chains
in
the
cosmosic
system
so
like.
If
you
were
already
someone
building,
indexers
and
developer
in
this
ecosystem,
it's
gonna
be
easier
for
you
to
start
building
it.
But
if
not
and
you're
interested
like,
I
really
suggest
that.
First,
you
start
familiarizing
yourself
with
the
entire
ecosystem.
B
The
mission
see,
if
that's
something
that
you
would
love
to
be
a
part
of
or
not.
It
aligns
with
objectives,
principles
and
all
these
type
of
things,
and
then
we
on
our
side,
we're
gonna,
be
like
when
we're
integrating
new
chains,
we're
going
to
do
our
best
to
create
as
much
possible
as
learning
materials
so
like
as
an
indexer.
You
want
to
be
in
this
new
system
expect
for
us
that
we're
going
to
be
sharing
a
lot
of
learning
material
and
one,
I
think
the
most
important
advice
is
don't
be
shy.
B
Reach
out
you
there's
a
lot
of
people
ready
to
help
and
we're
going
to
be
really
like.
I
think
today,
announcement
from
missouri
that
they're
going
to
be
building
subgraphs
and
that's
going
to
be
something
huge.
So
every
time
we
integrate
the
chain
now
we
have
at
least
someone
who's
gonna,
be
building
some
grass
for
it
and
and
creating
this
schema
and
set
the
right
way
so
reach
out
to
them
reach
out
to
their
core
devs.
And
let's
build
this
stuff
together.
Amazing.
C
What's
great
is
that
we
now
have
sort
of
a
repeated
and
well
repeatedly
proven
process
for
adding
support
for
new
chains
and
not
evm,
for
instance,
and
now
it's
just
a
question
of
enabling
the
the
rewards
to
have
the
incentivization
in
place
these
network
for
these
chains
to
be
supported
on
the
network
and
then
just
need
to
do
that.
Work
repeatedly
to
support
these
new
chains
in
general.
A
Now
a
big
part
of
this
of
supporting
new
chains
and
indexing
performance
is
fire
hose.
We
don't
have
the
streaming
fast
team
here,
but
chris
I'm
looking
at
you.
I
know
that
graphops
has
been
getting
dirty
with
firehose
we'd
love
to
hear
from
you
a
little
more
about
firehouse,
maybe
for
the
group
that
doesn't
know
and
what
you
think
the
impact
is
of
fire
hose
on
sunsetting
the
hosted
service.
F
F
F
F
F
It's
possible
to
share
that
data
between
participants
and,
in
this
case
indexers.
So
what
we
will
see
in
the
future
is
the
ability
for
new
indexers
to
rapidly
bootstrap
their
own
operations
by
synchronizing,
the
flat
files
underlying
the
fire
hose
from
other
indexers,
and
so
just
in
terms
of
the
ability
for
indexers
to
get
up
to
speed
with
new
chains
or
data
sets.
This
is
a
real
game.
Changer.
A
Thank
you
for
that
thorough
explanation.
We
have
time
for
one
more
question
and
I
think
this
is
for
all,
but
I
want
to
start
off
with
yuri,
because
you
are
the
web
2
boomer
on
stage
and
sun
setting
the
hosted
service
is
really
the
final
you
know
match
for
us
with
web
2.
and-
and
I
I
want
to
know
everyone's
vision,
you
know,
for
how
do
we
actually
get
more
of
web
2
developers
into
web3?
What
does
that
process
look
like,
and
maybe
here
you
can
go
first.
E
Yeah,
I
think
it's
related
to
what
I
said
at
the
beginning.
I
think
for
us.
You
know
we
came
in
and
we
looked
first
of
all
from
both
sides,
we
started
like
first
of
all,
contributing
to
the
graph
node
itself
and
exploring
the
code
base
and
exploring
the
api
and
then
looking
when
we
started
into
graph
client.
E
We
we
did
what
we
usually
do,
which
is
just
interviewing
developers
and-
and
we
interview
you
know-
web
2
developers
and
web
3
developers,
and
I
think
what
we
came
to
realize
is
that
the
best
way
to
make
people
use
a
new
thing
and
then
and
use
the
right
thing.
Let's
say
the
this
decentralized
network
is
to
make
it
just
easier
and
better
for
them.
E
You
know
it's
like
not
because
it's
you
know
not
because
it's
good
for
the
society
or
anything
like
that,
but
because
it's
just
better-
and
I
think
you
know
you
asked
one
of
your
previous
questions-
was
you
know?
What
would
you
wish
everyone
to
know
about
the
roadmap
and
everything
I
think
we're
in
some
ways
we're
already
there
like
in
some
ways
today
we're
getting
into
the
point
where
it's
easier.
E
If
you
want
to
start
an
app
with
all
the
tools
that
you
know
we
know
were
announced
today
like
it
might
be
already,
in
some
cases,
easier
to
start
doing
a
decentralized
app
using
all
these
tools
than
you
know,
setting
up
the
regular
hosted
web
tools
services.
So
we
are
for
sure
very
excited
about
everything,
and
it
has
to
do
with
everything
that
was
being
talked
about
here.
E
Also,
the
way
that
all
of
the
teams
here
are
working
together
like
we're
part
of
many
foundations,
the
linux
foundations,
the
graphql
foundations,
many
other
others,
and
the
way
that
things
progress
here
and
collaborate
here
and
like
someone's,
I
think
you
said
like
the
best
idea
wins.
I
think
that's
what
brought
us
here
so
yeah.
E
And
also,
if
not
like
talk
to
us
like
that's,
why
we're
all
here
like
we're
just
waiting
for
feedback,
and
I
think
you
know
just
on
the
graph
client,
every
feedback
that
we
got
just
in
the
last
month.
Two
to
two
months,
we
integrated
into
the
framework
so
tried
out,
and
just
you
know
everything
that
was
being
revealed
here
and
just
let
us
know
how
could
we
do
better
because
for
sure
we
can,
but
we
also
can
execute
on
that.
B
Yeah
sure
so
I
guess,
as
a
lot
of
us
know
here
in
web,
3
incentives
shapes
behavior.
So
I
think
we
I
think
like
something
interesting
is
the
more.
We
are
creating
incentives
for
web
developers
to
move
into
up3
the
more
we
are
getting
out
of
them,
and
so
I
think
we
at
figment
at
some
point
we
launched
something
called
figment,
learned
and
figment.
B
Learn
was
actually
free,
tutorials
free
courses
for
with
developers
to
start
learning
how
to
build
in
web3,
and
we
started
seeing
the
big
migration
of
web
two
developers
into
f3,
and
that
also
was
coupled
with
with
incentives,
and
I
think,
like
we.
The
grants,
like
the
total
grants,
was
something
around
half
million
dollars
like
in
total.
That
was
given
to
developers
and
it's
not
just
about
incentives.
It's
like
something
that
are
attracting
webtoon
developers,
we're
telling
them
hey.
Just
learn
and
you're
gonna
get
the
rewards
for
learning,
and
this
small
reward
might
be.
B
You
know
an
opportunity
for
you
to
start
building
something
you
have
like
small
funding
and
then
at
the
end,
it's
about
the
community
when
they
get
when
they
start
asking
questions
and
they
get
involved
in
the
communities.
This
is
where
they
start
meeting
other
people
and
they're,
starting
like
realizing
how
interesting
this
space
is
like
something,
a
lot
that
I
do
like
when
I
go
to
our
three
conferences,
meetups,
and
when
we
do
some
interviews,
I
talked
to
a
lot
of
people.
I
asked
them.
B
Oh,
that's
like
something
interesting
I
want
to
buy
to
just
like
kind
of
like
they
thought
the
talk
is
going
to
go
up
and
once
they
start
learning
about
it,
they
sat
realizing
it's
not
about
just
the
token
about
the
price,
there's
something
much
bigger
behind
it
right
and
I
think
it's
almost
the
same
idea
or
the
same
vision
with
developers.
Maybe
they
are
coming
for
the
incentives
to
start
learning
to
start
building,
but
soon
enough.
They
are
realizing
that,
what's
behind
it's
just
like
much
much
much
bigger
than
what
they
saw
at
the
beginning.
C
Yeah,
I
would
split
it
up
in
sort
of
consumers
and
developers
so
from
a
consumer
perspective.
I
think
both
both
cases,
like
yuri,
said
in
both
cases
you
need
to
provide
something.
That's
better
better
can
mean
a
lot
of
different
things
for
different
people.
Some
people
may
care
about
privacy.
Some
people
may
care
about
censorship,
resistance,
openness,
collaboration
and,
like
we've,
seen
a
good
example
with
geo,
where
collaboration
around
like
global
information,
you
know
factual
information,
that's
that's
that's
curated,
together
and
is
is
more.
C
You
know
is
true,
rather
than
from
questionable
sources
that
are
hard
to
trace
back
to
to
anything,
I
think.
That's!
That's
one
aspect
provide
applications
that
that
people
need
that
provide
that
offer
something
that
the
traditional
web
doesn't
offer
censorship
resistance,
for
instance,
for
developers.
I
think
it's
yeah.
I
think
a
lot
has
to
do
with
okay.
What
do
I
need
to
to
build?
What
I
want
to
build
and
sure
it's
it's
definitely
easier.
C
I
think,
if
you
don't
need
to
spin
up,
you
know
find
familiarize
yourself
with,
like
a
cloud
provider,
learn
how
to
write
a
server.
You
know
application
host
that
somewhere
learn
how
to
scale
it.
All
of
these
things
become
unnecessary
through
the
graph,
especially
so
I
think
that's
that's
a
big
leap
already,
but
I
think,
there's
more,
I
don't
think
it's
easier
to
write
a
smart
contract
compared
to.
C
A
D
Yeah,
so,
in
my
opinion,
I
think
we
need
to
continue
what
we're
doing
the
the
snark
force
is
going
to
be
so
I'm
going
to
be
coming
at
this
question
from
a
research
and
development
perspective,
snark
snarkforce
is
going
to
be
publishing
multiple
papers,
academic,
peer-reviewed
cryptography
papers
in
the
coming
year
in
the
ai
work.
As
far
as
I
know,
we
in
the
graph
protocol
have
released
the
first
reinforcement
learning
agent
in
web3.
If
anybody
knows
different,
please
let
me
know,
but
these
are.
D
These
are
pretty
big,
pretty
big
deals
and
I'm
starting
to
see
some.
I
think
that
we're
going
to
be
attracting
a
lot
of
top
talent.
I've
had
I'm
having
people
cold,
cold
contact
me
that
are
in
in
like
in
web
2
or
in
industry.
People
that
you
know
went
to
mit
oxford
to
darmstadt,
like
people
from
these
that
are
like
doing
these
great
things
that
are
in
more
traditional
roles
right
now
and
the
stuff
that
we're
doing
the
we're,
really
we're
really
pushing
the
limits
of
these
technologies.
D
We're
working
on
not
not
just
in
the
not
just
the
ai
and
r
d,
but
just
the
whole.
The
whole
way
we're
building
it,
we're
really
pushing
the
limits
and
that's
going
to
attract
top
talent,
and
you
know,
just
speaking
from
semiotics
perspective,
we're
getting
top
talent
coming
to
us.
Just
saying:
hey,
can
you
it's
my
really?
D
What
I
want
to
do
is
web
3
plus,
say
hi
and
we're
gonna
get
we're
gonna
get
some
of
these
people
and
then
then
we're
gonna
also
we'll
be
able
to
be
in
a
better
position
to
attract
really
good
engineers
that
want
to
also
see
these
r
d
ideas
instantiated
in
real.
You
know
in
in
real
systems,
so
I
think
we
just
need
to
keep
keep
doing
what
we're
doing.
F
Sure
I
certainly
want
to
echo
everything
that's
been
said
so
far.
I'd
definitely
agree.
I
think
another
major
driver
will
be
business
efficiencies
and
I
think,
for
me,
a
lot
of
what
web
3
is
about
is
less
about
the
technology
and
more
about
the
business
models
that
the
technology
unlock,
and
I
think
it
is,
I
think,
defy-
is
a
particularly
great
example
of
this.
Where
you
know
we
see
that
companies
like
uni
swap
who
you
know
have
relatively
speaking,
small
teams
are
rivaling.
A
Yeah,
I
would
echo
that
you
know
what
really
takes
it
home
for
me
as
a
non-developer
is
you
know
we
should
not
have
dap
developers
focusing
on
their
back
end
really,
they
should
focus
on
their
core
competencies.
Building
good
products
front
end
development
and
let
the
hard
stuff
be
taken
care
of
by
a
network.