►
From YouTube: GraphQL.js Working Group - 2020-09-16
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
B
Oh
good
question:
I'm
I'm
good.
A
So,
in
a
sense
like
the
planet,
for
I
think
like
more
than
a
month,
not
from
the
last
working
group,
but
a
working
group
before
that-
and
we
don't
have
quite
official
agenda
because
I
didn't
set
up
a
branch
and
nobody
create
pr
for
it
so
like
at
some
point.
We
actually
it's
good
that
we
have
recording.
A
So
I'm
not
sure
if
we
need
the
notes,
I
think,
like
second
one
would
be
more
official
and
we
will
have
like
a
gem
there
and
all
the
stuff.
So
it's
better
to
to
start
doing
something,
then
for
it
to
be
ideal.
So
one
thing
is
that
we
discuss
on
what's
working
group
is
that
working
group
should
have
a
separate
repos
on
the
graphical
foundation.
A
A
I
will
and
let's
use
the
same
issue
to
coordinate
I
think
like
for
now,
until
we
have
a
ripple
issue
that
we
use
for
organizing
with
always
good.
So
just.
B
C
A
Especially
somebody
I
always
try
to
take
a
note,
because
I'm
not
a
native
speaker,
so
it's
hard
for
harder
for
me,
or
at
least
it's
my
excuse
for
not
doing
that
so
yeah
yeah
another
before
we
go
around
and
introduce
ourselves
and
mention
what
we
want
to
discuss
today,
because
we
don't
have
agenda
one
one
note
is,
I
only
have
one
hour
maybe
an
hour
and
a
half,
but
since
it's
the
first
one
and
it
bootstrapping-
that
probably
would
be
enough
so
like,
let's,
let's
target
one
hour
from
now
and
if
we
overflow
a
little
bit
as
you
have
a
time
afterwards
yeah.
A
So
let's,
let's
start
from
me,
basically
I'm
I
would
say
how
to
say
I'm
a
contractor
for
graphql
foundation,
part
time.
So
it's
like
we
discuss
it.
It's
around
40
hours
and
I
hope
to
maintain
graphql
js
for
the
last
three
years.
A
A
C
B
Okay,
okay,
so
I'm
oli
and
I'm
from
the
guild
where
a
couple
of
people
here
from
the
guild
we
do.
We
maintain
a
couple
of
like
graphql,
related,
open
source
libraries
and
you
know
all
of
them
are
using
graphqlgs
and
it's
been
amazing
for
us,
and
I
think
so
I
guess
the
main
thing
that
I
would
love
would
be
to
have
kind
of,
like
a
let's
say,
a
public
roadmap.
B
B
You
know,
because
I
think
once
you
have
a
plan,
then
you
know
it's
a
good
thing
for
all
the
community
to
see
where
we're
going
and
also
help
people
know
where
and
how
to
contribute
and
also
maybe
suggest
other
things
so
have
a
public
romance.
I
think
it's
like
probably
like
a
good
idea.
B
Another
thing
I
would
want
to
talk
about
is
how
can
we
like?
What
did
what
this
mean?
What
does
it
mean
to
become
a
maintainer
with
graphql
gs
like
what
are
the
let's
say,
characteristics
of
a
maintainer
like
what
does
it
mean
to
be
a
maintainer?
And
how
can
we
find
and
because
I
think
that
will
maybe
help
us?
B
You
know
always
it's
always
good
to
get
more
maintainers,
but
then
there's
always
it's
very
hard
to
say
well,
who's
good
enough
or
who
is
should
we
or
shouldn't
be,
and
I
think,
having
the
like
like
clear
criterias
would
be
great
because
then
you
know,
I
think
it
will
open
the
doors
for
others
as
well.
We
would
love
to
spend
more
time
and
and
help
as
much
as
we
can,
what
else
yeah,
and
I
think
those
two
are
the
main
things
that
I
would.
C
Oh,
you
mean
camille
right
yeah,
so
I'm
the
member
of
the
guild
pretty
much
we
work
together
for,
like
I
don't
know,
I
don't
remember
since
I
was
a
kid
probably
yeah,
yeah
most
interesting
in
the
typescript
integration,
I
mean
just
to
rewrite
the
coding,
typescript
and
just
get
the
you
know
the
native
typings,
so
it's
always
in
sync
and
maybe
the
yeah
for
sure
the
roadmap.
C
So
it's
clear
you
know
what's
coming
on
and
how
can
we
help
etc
and
yeah
like
what
is
the
basically,
what
woody
said,
which
is?
How
can
we
contribute?
What
are
the
rules?
You
know
this
kind
of
stuff.
A
Okay,
great
marcel
user
next.
D
Right,
welcome
sorry,
hi
everybody.
My
name
is
marcelo.
I
use
graphql.js
on
about
20
projects
that
I
have
running
in
production.
We
have
a
internal
framework
that
we
developed
specifically
for
subscriptions
in
serverless
environments
and
we
are
targeting
to
open
source
this
framework
and
I'm
especially
interested
in
the
roadmap
for.
E
Yeah
rob
from
first
dibs,
been
working
on
deferred
stream,
wanted
to
talk
about
the
pr
for
async
intervals
and
deferred.
E
F
Yeah,
it's
okay,
it's
not
done
so.
I'm
hi
everyone,
I'm
the
tan,
I'm
also
a
member
of
the
guild,
mostly
working
in
like
the
areas
for
open
source,
I'm
mostly
working
on
graphical
coding.
F
I
would
love,
if
we'll
have
some
time
today,
to
talk
about
what
already
suggested
specifically
on
what
is
the
path
or
the
process
of
becoming
a
maintainer
or
a
collaborator
on
the
graphql
gs,
repo
and
the
other
topic
is
typescript
migration.
I
brought
this
up
a
few
times
yeah
and
just
because
we
we
love
typescript,
and
I
know
this
is
planned
and
we
would
love
to
to
help
and
do
whatever
it
needs.
B
If
there's
no
one,
then
I
would
just
I
forgot.
I
also
wanted
to
add
the
point
into
there,
which
is
the
like
esm
support.
A
Okay,
it's
actually
related
to
typescript
and
sure
so
yeah,
let's
discuss
it
in
in
order
like
from
from
much
more
generic
to
more
specific,
I
would
suggest-
and
let's
let's
time,
time
frame
on
items
so
like
let's
discuss
about
contribution,
how
to
become
contributor
when
typescript
and
when
what
what
is
blocked
in
different
stream
and
subscription,
because
I
actually
like
second
and
third
related
to
to
first
to
like
how
community
can
be
involved
and
yes
and
like
after
typescript
ds7,
then
stream
d4
and
have
like
15
minutes
each
and
we
will
still
fit
yeah
so
about
maintenance.
A
A
Yeah
I
tried
couple
of
times.
One
of
the
issue
is
like
I
learn.
You
need
to
communicate
with
a
person
because
you
never
know.
Is
it
wait?
Sometimes
person
just
want
this
change,
so
he
don't
want
to
learn
stuff
about
your
project
or
how
to
contribute
or
how
to
adhere
like
the
standards.
He
just
want.
This
change
to
be
pushed,
and
if
you
try
to
teach
him,
it's
actually
bad
for
both
parties
yeah.
A
A
A
A
I
just
think
it's
better
to
do
them
at
the
same
time
on
boarding
person
and
changing
how
we
do
things-
and
it's
related
to
my
biggest
struggle
right
now
with
like
prs,
is
that
we
have
a
quality
level
and
you
probably
saw
that
we
don't
have
bugs
most
of
the
feature,
requests
or
mission
feature
or
like
some
problem
with
spec
or
some
somebody
not
to
permit
something.
We
almost
don't
have
bugs
and
quality
of
always
good,
and
I
don't
want
to
over
it.
A
So
from
my
and
it's
keep
get
history
clean
because
I'm
like,
for
example,
I'm
not
a
facebook
employee-
and
I
don't
have
like
a
lot
of
access
to
facebook,
engineers
and
plus
graphql-
was
a
cod
dump
at
some
point
in
2016.
They
just
work
numb
code
on
github.
So
it's
very
valuable.
For
me
to
see,
give
history,
and
I
want
to
preserve
it
for
other
as
a
future
maintainers
and
core
contributors,
so
idea
is
that
pr
should
be
topic
and
as
small
as
possible
and
changes
should
be
gradual
and
reviewable.
A
A
I
need
to
ensure
that
it's
done
according
to
like
current
architecture.
We
can
change
it
to
either,
but
so
what
I'm
trying
to
communicate
is
that
I
would
suggest
to
stick
with
current
process
until
we
form
a
like
a
core
team
like
at
least
we're
on
board
like
couple
of
people
and
when
we
can
discuss
how
to
change
processes,
and
we
can
actually
change
process
during
this
thing,
but
it
should
be
graduates
not
only,
and
my
like,
my
strongest
requirement
is.
That
too,
is
that
pair
should
be
reviewable,
meaning
like
if
it's
not
the
same.
A
It's
the
same
thing
if
by
command
or
some
changes
and
ask
to
do
the
same
changes
write
down
the
files,
so
I
basically
limit
because,
as
we
discussed
with
brian
previously,
it's
not
my
full-time
job.
It's
like
basically
a
foundation,
try
to
support
me
for
like
some
number
of
hours
and
we
discuss
it
like
around
40
hours
per
month,
and
I
maintain
like
some
other
person,
so
I
I
cannot
dedicate
like
huge
amount
of
time
to
reviewing.
A
So
I
would
ask
for
people
who
want
to
contribute
to
make
the
api
easy
for
review
and
basically
it's
like
my
requirement
and
like
to
self
review.
First
thing
is
to
self
review
and
from
my
side
I
think,
is
bad
to
demand
unwritten
rules,
so
I
know
that
we
discussed
it
previously
and
I
failed
to
deliver
on
that.
I
need
to
write
contribution
guide.
I
have
an
issue
yeah.
I
will
send
the
link
to
it
when
I
have
like
some.
A
B
I
think,
by
the
way,
I'm
sorry
if
there's
like
a
noise
and
stuff,
but
I
I'm
I'm.
I
think
that
yeah,
I
very
much
agreed
everything
you
said.
I
think
you
know
we
really
like
all
of
us.
I
think,
as
some
people
that
work
a
lot
a
lot
of
time
with
graphql
in
javascript,
and
we
are
ready
to
dedicate
a
lot
of
our
time
for
free
to
help
you.
So
I
think
you
know
maybe
we
could
just
so.
B
The
question
is:
let's
say
we
nominate
at
the
beginning
just
one
person
from
the
guild,
because
you
know
too
many
people
at
the
same
time
will
probably
take
a
lot
of
time
from
you,
but
and
I'm
saying
from
the
guild,
but
it
might
be
someone
else.
Just
I
think,
let's
pick
up
someone
and
let's
debate
like
what's
the
best
use
of
your
time
and
who's,
this
person
should
be,
and
then
you
know
I
think,
let's
start
with
thinking
together.
B
How
can
we
onboard
that
specific
person?
I
think
that
would
mean
then
we'll
get
an
actual
action
plan
that
we
can
even
like,
I
don't
know,
execute
until
the
next
meeting.
So
I
think
the
main
question,
maybe
would
start
is
like
let's
say
a
lot
of
people
are.
I
know
a
lot
of
people
that
a
bunch
of
us
joining
into
that
meeting
want
to
help
you.
So
what
and
let's
say
we
pick
one
person
first
of
all,
how
should
we
choose?
Who
that
one
person
should
be?
B
Second
of
all,
what
would
be
the
first
task
for
that
person
to
prove
themselves
or
to
make
sure
that
they
learn
what
they
need
to
learn.
A
So
I
don't
want
to
wait.
You
know.
One
of
the
problem
of
working
groups
is
to
not
to
turn
it
in
dialogue
and
between
like
two
people.
So
if
anybody
else
have
like
some
ideas
or
some
some
stuff
to
say,
let's
go
around
and
everybody
will
talk
and
I
will
respond
and
we
will
continue
anybody
else
want
to
like
suggest.
D
Well,
just
to
make
clear
when
you
were
talking
about
the
size
of
the
prs
and
that
kind
of
things
that
bothers
you,
I
totally
feel
the
same
as
you
on
that
matter.
I
agree
with
that,
but
that
was
when
pr's.
D
That's
that
usually
happens
when
you
don't
have
a
clear
road
map
and
like
tasks
and
deliverables
define
it
so
people
are
mostly
at
the
hawking,
adding
features
and
trying
to
do
stuff.
I
think
that
that
will
be
solved
when
we
have
a
clear
roadmap
and
deliverables,
and
then
you
can
get
a
small
task
and
iterate.
On
top
of
that,
and
as
yuri
mentioned
that
I
have,
I
have
a
big
time
available
for
working
on
this.
Specifically,
this
is
very
important
project
for
me
and
I'm
available
for
anything
that
I
can
help.
F
I
just
want
to
to
say
that
I
completely
agree
with
you
and
from
I
don't
know
our
point
of
view.
We
feel
the
same
with
a
lot
of
like
other
libraries
that
are
open
source
that
we
maintain
small
pros
are
usually
easy
to
review
and
it's
much
much
simpler
to
get
this
merged
and
released
and
available
for
everyone.
F
A
So
if
everybody,
okay
by
the
way,
it's
okay
to
interrupt
me,
we
have
like
only
six
people
so
with
small
sizes,
actually,
okay
like
if
we
have
like
20
people
and
everybody
try
to
speak
so
yeah.
A
Let's
use
this
opportunity
so
to
answer
like
first
uri
and
like
correct
me,
if
I
like
miss
something
or
misinterpreted,
so
I
would,
I
would
say
in
a
sense,
it's
it's
kind
of
not
fair
to
to
designate
one
particular
person
and
and
and
let's
put
the
pressure
on
this
person-
maybe
it's
like
in
the
end
he
will
learn
because
it's
open
source.
So
it's
not
it's
not
like.
A
A
I
always
try
to
respond
in
24
hours,
except
like
cases
I
have
like
some
some
health
issues
or
something
else.
So
it's
like
my
commitment
to
answer
in
24
hours.
If
you
ask
questions,
if
I
don't
have
a
time
for
review,
or
I
would
answer
if
you
one
thing,
I
I
had
like
some
some
surgery,
so
my
did
and
inbox
is
like
huge.
A
So
if
you
want
from
me
to
answer
just
like
tag
me
so
answer
to
ever,
you
know,
like
github
inbox,
have
a
special
category
mentioned.
A
So
if
you
want
something
to
be
like
reviewed
or
stuff
mention
user,
because
I
actually
like
go
every
day,
I
go
for
it
as
if
you
just
create
a
pr
without
mentioning
it's
like
basically
mean
like
next
time.
I
want
to
open
pandora
box
of
issues
and
pr's
like
yeah.
It's
one
thing
and
a
second
thing,
partial
partial
reviews,
so
I
learn
is
actually
helpful
from
both
it's
like
it
can
be.
Frustrating
like
it
can
be
frustrating
if
you
don't
know
why.
A
A
When
I
saw
something,
I
just
ask
you
to
change
it
in
other
places,
so
if
I
say
see,
one
one
place
that
you
did
something
not
according
to
standard
or
something
problem
with
it,
I
ask
you
to
self
review
the
rest
of
yeah
and
change
it.
So
I
would
I'm
okay
with
reviewing
like
honestly,
if,
like
two
three
four
people
working
full
time
you
can
attribute,
but
I
don't
expect
this
because
you
also
have
in
the
guild
and
other
contributors
have
eq
projects.
A
So
let's
do
it
like
that,
like
let's
start
from
my
people
with
hope
and
write
something
in
the
issue.
So
I
remember
your
github
accounts
and
if
you
mention
me
like
I'll
answer,
answer
to
everybody
who
mentions
me
at
least
I
try
and
in
your
aprs,
I
will
try
to
give
feedback
instead
of
like
reimplementing
it
or
fixing
it
myself,
because
yeah.
If
it's
not
like
super
complex
stuff,
to
explain
I
give
feedback.
Is
it
like
really
complicated
or
is
it
like?
A
Last
final
touches
I
can
like
commit
on
top,
but
I
will
like
try
to
give
feedback,
but
at
the
same
time,
first
self
review
and
second
thing
when
I
ask
in
comments
to
change
it
to
self
review
and
change
it
to
other
places.
If
it's
big
pr,
it's
not
because
I'm
like
my
my
personal
goal
is
not
to
demand
anything
from
anybody
that
I
I
wouldn't
do
myself.
I
don't
do
myself,
so
it's
the
same
about
small
fps.
I
always
break
my
work.
A
So
when
I
like
ask
you
to
do
something,
it's
basically
the
same
thing,
I'm
doing
personally,
as
I
said
like
a
week
later,
we
can
discuss
how
to
how
to
change
our
workforce,
but
at
the
moment
it's
like
that
so
self
review
and
and
like
reacting
partial
reviews
and
reacting
on
reviews
and
changing
to
other
places.
So
this
must
address
and
yeah,
let's
limit
batch
to
people.
This
call
and
next
call
maybe
some
somebody
else
want
to
join
and
and
by.
A
And
feel
free,
if
I
didn't
answer
like
because
I
didn't
remember
the
oh
questions
and
I
I
need
to
duplicate
with
in
contribution,
I
will
basically
duplicate
the
thing
about,
like
partial
pr
self,
reviewing
and
other
things,
yeah.
B
Okay,
so
I
think
well,
there's
two
separate
things
here:
one
is,
how
can
we
get
more
prs
in
which
you
know
or
like
increase
the
throughput,
which
is
what
you
said
is
I
think
okay
I
mean
we
would
love
to
like
we
would
from
even
tomorrow,
we'll
start
breaking
prs
into
smaller
chunks,
and
we
would
you
know
we
will
tag
you
and
you
know
we
will
do
those
things,
and
I
hope
that
you
know
that
will
help
getting
more
things
in.
B
I
think
the
other
thing
that
I
would
I
think
would
really
be
great
like
was
suggested,
would
be
to
have
a
road
map.
Then
we
can,
like
you
know
we
can.
C
B
Use
that
from
it
like
put
each
task
on
like
as
a
github
issue
or
pr
and
have
like
a
you
know,
maybe
like
a
zen
hub
or
some
board,
we
can
all
see
what's
the
progress
and
everything
I
think
that
also
could
be
nice,
but
I
think
all
of
that
is
you
know,
that's
all
those
things
are
about.
How
can
we
get
more
prs
in
and
increase
the
throughput
there's
another
subject
that
I
think
is
as
important,
which
is:
how
can
we
increase
the
bandwidth
meaning?
How
can
we
get
more
like
this?
This
whole
means.
B
How
can
we
everything
you
said
was:
how
can
we
optimize
the
time
of
ivan
as
the
main
maintainer
there's
another
subject,
which
is:
how
can
we
get
more
events
in.
A
A
A
If,
if
person,
if
person
like
normally
react
to
feedback
and
correct
his
things,
and
we
he
start
understanding,
what's
happened
to
a
project
yeah,
I'm
okay
with
after
one
thing,
I
want
to
be
sure
that
this
person
a
person's
commitment
and
he
is
like
so
because
we
don't
want
to
put
barrier
to
war
in
a
sense.
So
I
think,
like
like
two
months
of
contribution,
political
contribution
to
any
person
like
iterate
and
buy
some
feedback
and
like
basically
like
learning
ways
of
a
project.
A
When
we
have
like
core
group,
we
can
review
each
other
and
if
you
so,
I
basically
like
try
to
adhere
to
standards
and
actively
can
you
commit
for
actually
create
pr
for
a
couple
of
months?
A
Here's
my
like
personal
criteria,
I
feel
like
and
brian
approached
me,
brian
from
graphical
foundation.
Actually,
I
have
called
with
him
today
and
he
said
about
some
way,
how
we
communicate
this
foundation
and
it's
always
thing,
but
I
think
we
need
some
input
and
agreement
from
foundation,
so
I'm
not
ready
to
put
official
criteria,
but
not
an
official
criteria.
From
events.
A
Point
of
view
is
to
actually
contribute
and
to
like,
even
if
you
disagree
with
how
it's
done
right
now
to
adhere
to
standards
and
later
we
can
change
them
and
I'm
like
yeah,
so
onboarding
period
is
for
me
it's
my
personal
criteria.
Yes,
so
before
I
speak
with
like
foundation
or
like
we
have
some
discussion,
I
cannot
say
it's
official,
but
it's
my
position.
I
will
advocate
for
that.
B
So
can
you
just
to
summarize
your
criteria
like
and
again
it's
just
your
point
of
view
and
not
the
official
point
of
view.
It's
you
want
someone
that
is,
first
of
all
like
a
person
that
learns
from
comments
and
not
repeating
mistakes.
B
Like
you
said,
second
of
all,
someone
that
is
like
invested
in
the
project
in
graphql
long
term
like
we
need
to
define
what
is
long
term
right
like
is
it
a
month?
Is
it
six
months?
Is
it
a
year.
A
I
would
say
like
two
months:
is
it's
a
good
criteria
because,
like
in
one
month
it's
like
a
lot
of
and
if
something
happens,
it's
also
okay,
so
it's
totally
normal.
For
example,
like
I
try
to
onboard
daniel-
and
I
know
daniel
actually
worked
for
you
and
we
have
a
good
progress
and
good
process,
but
he
changed
his
job,
so
I
spoke
with
him
and
he's
still
interested
in
being
good
contributor,
but
on
a
new
job.
He
highlights
some
other
stuff
to
do
and
it's
totally
normal.
It's
unpredictable.
A
It's
like.
I
just
want
to
know
a
person
that
he's
not
losing
interest
or
finding
the
new
fancy
thing
in
like
maybe
there
is
something
fancier,
some
graphql
and
ipi
space
I
don't
know
or
whereby
somebody
or
something
yeah
like.
B
B
Shifts
their,
I
don't
know
their
interest
or
they're
losing
they're,
not
spending
more
time.
I
think
it's
okay
to
then
just
drop
that
maintainer,
meaning
like
yeah
and
that's
okay.
I
think
this
is
a
process
that
happens
a
lot
in
projects,
so
I
think
it's
more
less
about
anticipating.
What's
going
to
happen
and
more
about
just
keeping
examining
the
people
involved
and
seeing
you
know.
C
B
Then
you
keep
the
barrier
possible
to
get
in,
but
also
it
means
that
you
know
when
people
stop
being
like
that,
we
can
always
drop
them
and
it's
not
such
a
big
deal.
They
can
always
come
back,
but
yeah.
A
Yeah
yeah,
on
the
technical
side,
we
have
like
60
people
from
facebook,
more
than
60
people.
Facebook
is
commit
rights
and
it's
an
easy
question.
But
at
some
point
you
actually
need
to
ask
what
to
do
with
these
people.
So,
and
I'm
totally
agree,
I
would
say,
like
I
would
say
right
now.
What
I
did
with
daniel
is
to
contact
him
and
ask:
do
you
want
to
contribute
in
future,
and
he
said
yes,
just
like
I
have
like
I'm
boring
experience
with
this
work,
so
I
would
expect
criteria
to
be
the
same.
A
A
Yeah,
but
I
think
like
right
now,
we
have
like
a
good
thing.
It
can
be
like
a
lot
of
things
can
can
be
manual
because
we're
like
in
bootstrap
mode.
You
know
like
this
article
from
from
graham
the
guy
from
by
combinator
boys
articles,
say
titles
say:
do
things
that
don't
scale
so
because
we're
bootstrapping
our
processes
can
be
different
from
at
times
that
we
fully
bootstrap,
we
can
cut
a
corner
and
like
simplify
stuff
and
be
less
less
official.
A
A
Yeah
actually
another
thing
we're
running
on
time.
I
think,
like
it's
yeah,
it's
basically
most
important
question.
So
if
somebody
like,
I
will
resume
and
if
somebody
disagree,
we
still
have
a
time,
because
I
think
it's
like
most
important
question.
If
not,
we
will
continue
to
type
scripts
and
to
other
issues.
A
So
for
me,
action
items
because
action
items
is
always
helpful.
To
put
it
official
is
to
create
a
repo,
and
I
would
prefer
somebody
actually
like
a
first
first
like
a
way
to
help
me.
I
set
up
a
repo
and
you
just
copy
template
and
stuff
from
from
main
working
group
and
replace
graphql
spec
with
graphql
js.
A
A
A
So
next
thing
is,
I
actually
will
put
like
in
contribution
guide.
I
will
put
the
stuff
that
we
that
I
discuss
so,
if
you
mention
me
or
any
other
montana,
try
to
respond
and
like
yes,
they
said
lesser
one
and
more
like
maintainers,
so
contribution
guide
is
like
maternity
should.
If
you
mentioned,
he
should
try
to
respond
in
24
hours
respond.
A
Not
he
can
say,
I
don't
have
a
time
to
review,
but
he
can
should
respond
and
same
thing
about
like
partial
reviews
and
like
asking
like
basically
to
put
more
work
on
on
a
person
who
submitter
and
it's
a
it's
always
helpful,
because
people
will
burn
out
not
only
in
this
particular
situation
when
we
have
like,
and
another
thing
is,
if
somebody
interested
in
being
reviewer
for
other
peers.
D
A
Quite
feature
requests
put
a
lot
of
pressure
on
me
to
to
my
decision
and
to
choose
that's.
Why
actually
like
more
maintenance
is
helpful
because
you
you
like
discuss
and
brainstorm
and
people
have
a
context
about
project,
so
I
I
would
suggest,
let's,
let's
start
from
like
obvious
features
like
yes,
I
am
type
script
features
that
will
agree
and
features
that
we
don't
need
to
design.
A
Actually
it
will
be
helpful
if
somebody
volunteered
to
classify
like
new
stuff
from
like
improvement
and
problems
like
dino
support
is
obvious,
and
I
spoke
with
dynamic,
guys
and
they're
interested
and
something
like
set
up
test
on
browsers
like
run
or
we
have
like
integration
tests,
support
more
integration.
Tests
like
webpack
is
not
controversial
and
it's
helpful
that
we
don't
break
back
or
don't
break
flow
by
switching
to
typescript.
A
D
A
Yeah,
so
I
I
create
issues
for
that.
I
will
bring
some
some
ideas
and
one
of
the
issues
that
I
would
use
help
on
is
to
find
issues
or
prs
that
don't
change
api
or
don't
introduce
new
features.
A
Is
that
just
quality
of
life
improvements
and
we
can
start
from
that
because
for
new
features
I
don't
want
to
be
one
of
the
problems
like
personal
spend
a
lot
of
time
and
I
kind
of
pressured
into
giving
him
feedback,
and
I
don't
have
a
right
answer
at
the
moment.
I
need
to
think
or
consult
with
somebody,
or
I
just
don't
know
the
right
answer,
so
I
would
yeah
I
create
issue
for
somebody
to
go
through
and
by
the
way.
A
If,
if
somebody
interested
we
can,
we
can
have
a
model
github
of
all
trash
thresholds,
so
you
can
help
by
reviewing
and
by
answering
issues
from
time
to
time,
people
ask
random
questions
or
somebody
something
is
crushing.
You
can
ask
him,
ask
personal,
provide
like
detailed
bug,
report
or
some
details,
or
so,
if
you,
it
would
be
also
helpful,
and
I
can
quickly
promote
you
to
trash
raw.
A
A
Because
I
actually
like
it
not
just
foundation,
I
this
idea
that
it's
very
easy
to
become
to
have
treasure
and
I
think
it's
actually
test
people,
but
at
the
same
time
it's
motivation.
You
don't
need
to
wait
like
couple
of
months.
A
A
Is
a
sense
of
like
let's
so
if
it's
include,
if
it's
included
like
non-controversial
no
like
non-design
stuff,
let's
say
it's
like
non-design
stuff
like
yes,
I
am
typescript
and
everything
I
think
like.
A
If
you're
interested
in
that
open
an
issue
for
that
and
like
just
claim
it
right
now
after
call
tomorrow
say
I
would
work
on
that
by
the
way.
If
somebody
wants
to
start
working
on
something
starting
from
issue
is
a
good
thing
or
I
commend
an
existing
issue
so
yeah
it
would.
It
would
be
great,
and
you
provide
the
list
of
stuff
that
you
want
to
include
and
yeah
so
yeah,
I'm
like
totally
for
it.
A
B
Yeah,
I
will
just
create
it
and
suggest
it.
It
doesn't
mean
that
it's
there
just
like
it's
for
your
review
and
then
you
know
you
can
change
whatever
you
want
or
comment
on
it.
Just
I
can
start
taking
on
myself
just
to
create
that
initial
issue.
I
can
write
a
bunch
of
down
and
then
you
can
like
delete
all
of
it.
Just
like
yeah.
A
If
you're
interested
what
we
can
experiment
with
projects
on
github,
I
can
give
you
trudge
raw
and
we
can
experiment
with.
If,
if
you
yeah
one
thing,
the
same
thing
applied,
if
you
volunteer
to
do
that,
you
need
to
do
it
in
reasonable
time
so
like
like
couple
weeks
or
until
the
next
until
the
next
working
group
called
basically
like.
A
Let's
time
it
like
that,
yeah
so
action
item.
For
me
after
call,
I
will
add
us
trash
roll
and
you
create
like
project
modes,
think
like
project
boards.
I
don't
know
how
to
discuss
stuff
on
project
boards
or
should
it
be
move,
but
we
can
figure
that
out.
I
think
like
project
box
is
dedicated
for
road
maps,
github
use
it
as
a
public
roadmap,
so
you
can
just
fall.
A
Yeah
yeah.
B
I
think
it's
a
good
idea,
and
I
think
I
mean
we
can
always
fall
back
from
projects
to
zenhub,
which
we
also
very
like
a
lot
like.
Then
you
can
add
some
more
data,
but
I
think
starting
with
projects
will
be
good
and
just
to
add
on
the
adding
maintainer.
So
you
said,
like
you
suggested
you
are
needed
for
characteristics
and,
of
course
it's
not
official.
It's
like
your
like
your
thing,
but
the
good
thing
is
that
I'm
also
part
of
the
foundation.
B
So
what
we
can?
I
will
make
sure
that
tomorrow,
on
the
foundation
meeting,
I
will
present
your
characteristics.
Guys
will
ask
them
if
they
like
can
officially
like
officially
agree
to
those
characteristics.
I
think
it
will
be
easier
if,
until
tomorrow
you
can
create
an
issue
and
write
those
things
down,
and
then
you
know
we
can
actually.
I
can
actually
present
them,
something
that
you
wrote
and
not
that
I
say
what
you
say.
So
I
think
it's
it's
better.
It
makes.
A
A
A
So
in
foundation
documents
there
is
it's
officially
described,
there
is
a
governing
board
and
government
board
is
like
general
issues
around
projects
and
like
finance
allocation,
but
there
is
a
technical
steering
committee,
which
is
like
same
tasty
as
like
not
just
projects
or
other
projects.
A
So
it
not
just
us
understand
like
we
think
with
like
work
of
policy
or
how
we
probably
need
to
contribute
done
by
this
team.
So
I
might
I'm
not
sure
what
one
thing
I'd
like
basically.
G
A
A
So
yeah,
let's
communicate.
Let's
not
took
time
because
it's
like
open
me
on
on
truck.
So
it's
basically.
B
D
A
A
A
B
A
I
think
foundation
approved
it,
it
was
like
they
have
a
void
and
can
they
approve
the
list
like
in
last
year?
So
maybe
it
was
before
we
joined
it.
So
I
I
will
just
see
them
in
the
issue.
I'm
not
feel
like
creating
like
official
press
release
or
something
is,
is
beyond
my
my
throw
or
something
that's.
A
Yeah,
so
I
think
we
discuss
like
in
this
topic.
We
spend
but
yeah
it's
most
important
topic
and
it's
unblock
a
lot
of
other
stuff.
So
next
thing
is
like
typescript.
A
So
it's
a
couple
of
couple
of
things
about
typescript.
It's
the
same
thing
so
I'm
like.
If
we
want
one
conversion,
I'm
like
totally
okay,
if
we
like
add
standard
typescript,
it's
also
totally.
Okay,
if
we
add
some
typescript
meta
programming,
so
like
first
step
would
be
to
just
want
one
conversion
for
four
to
ten
strip
without
adding
new
new
stuff,
like
whatever
we
can
discuss
an
exam.
A
But
if
it's
like
core
code,
I
would
stick
with
like
standard
typescript
without
a
lot
of
meta.
Parkan
is
my
position
because
it
should
be
readable
by
like
a
person
from
graphql
java
or
graphical
python,
maintainer
or
php,
and
not
always
obviously
like
not.
Every
language
have
a
types
and
we
have
a
type
so
yeah,
but
in
a
sense
don't
we
should
not
illustrate
it.
So
a
person
should
learn
like
some
javascript
to
be
able
to
fully
read
it,
but
he
shouldn't
be
proficient
in
in
that's.
A
Why
and
it's
also
up
to
discussion
after
we
migrate
as
a
separate
step,
we
can
discuss
like
more
complex
usage
scenario
for
typescript,
but
beforehand.
A
F
Can
I
just
say
something
about
that?
I
mean
this
is
a
bit
hard
for
me.
I
mean
because
think
of
I
mean
it's
very
hard
to
do
reference
implementation
with
the
language
that
isn't
like
pseudocode,
because
let's
say
that
someone
from
go,
which
is
very
popular,
would
like
to
implement
graphql.
He
won't
be
able
to
use
gener
generics,
and
this
is
something
that
we
use
in
even
now
with
flow.
F
So
it's
it's
kind
of
like
it's
endless,
so
I
think,
in
terms
of
like
logic,
I
do
agree
with
you
that
it
needs
to
be
a
reference
implementation
in
terms
of
the
logic
but
in
terms
of
just
a
second,
but
in
terms
of
like
the
typing
system
and
the
features
around
it.
I
I
think
it's
like
it's
endless.
I
mean
if
we
start
to
limit
our
like
limit
the
way
we
implement
graphql
gs,
we'll
eventually
want
to
be
able
to
use
any
feature-rich
language.
F
A
So
we
can
better
discuss
this
policy
and
agree
on
that
and
define
like
a
border.
I
have
some
ideas
about
that.
Just
don't
want
to
you
know
like
me,
okay,
so
yeah,
it's
a
first
point,
so
I
kind
of
feature
freeze
on
typescript:
don't
do
more
than
you're
already
using
for
or
more
than
necessary,
and
second
thing
is-
and
it's
back
to
my
point-
that
I'm
I'm
don't
ask
to
do
anything
that
I'm
not
willing
to
do
myself
is
to
try
to
do
as
much
conversion.
A
So,
for
example,
I
like
ported
titans,
so
I
typed
every
function
in
full
right
now,
so
so
like
typescript
change,
that
clip
should
not
contain
like
new
typings.
It
just
should
contain
like
some
changes
in
titans.
So
it's
like
the
stuff
that
I
will
ask
to
do
during
conversion.
This
like
to
do
as
much
as
possible
stuff
in
full
and
so
like
texture
conversion
gives
more
in
a
sense.
A
So,
for
example,
if
benchmark
is
broken
during
conversion
and
if
we
see
a
way
how
what's
what
I'm
basically
working
right
now
so
like
migrating
benchmark
to
be
separate
from
from
some
basic
source
tree
so
like
adapted
to
future
conversion
before
we
do
conversion
so
not
to
have
broken
thing
during
conversion
effects
afterwards,
it's
so
what
I
ask
is
that,
like
in
a
sense?
A
A
A
Okay,
yeah
so
yeah.
G
A
Another
thing
I
need
to
buy
in
from
you
and
it's
like
is
that,
where
lines
is
that.
A
If
we
do,
we
have
some
problem
with
step
typings,
because
we
need
to
support
some
old
typescale
version,
I
think
like
2.7
and
this
tube
support,
because
when
they
had
titans
from
deftly
typed,
I
didn't
want
to
create
release
that
break
immediately.
I
wanted
for
entire
system
to
switch
on
using
graphql,
js
and
later
to
do
breaking
changes,
so
I
feel
if
we
introduced
a
bunch
of
pennies
in
our
test
and
bunch
of
other
like
compromises
from
2.7,
so
I
want
to
do
new.
A
Braking
change,
0.0
and
I
think,
like
community,
will
accept
that
if,
if
we
say
it's
for
typescript
conversion
and
for
doing
it
cleanly,
instead
of
doing
bunch
of
work
around
to
support
all
typescript
version
or
to
to
support
like
some
other
stuff,
so
some
preparation
work
pass.
It
allow
us
to
remove
duplicated
fields,
duplicated
features
instead
of
migrating
them.
So
likewise,
migration
is
like
everybody.
Okay,
that
we
do
typescript
migration
is
breaking
change.
That's
a
new
release,
basically
1600.
A
F
And
one
of
the
things
that
I
want
to
ask
you
is
so
I
I
had
this
big
pr
for
typescript
migration
and
one
of
the
things
that
I
noticed
is
that
there
are
some
bugs
in
the
typescript
typings
with
the
current
ones
that
we
have
like
on
the
dps
files.
F
So
maybe
I
can
start
with
small
prs
just
to
fix
those,
because
those
are
bugs
and
could
be,
I
mean
potentially,
this
could
be
in
the
next
release
of
graphql
gs,
because
it's
a
bug
fix
on
types
and
it's
just
aligning
the
types
to
the
types
that
we
have
on
the
flow
types.
So
I
feel
like
so.
A
Yeah
so
yeah,
let's
discuss
actual
approach,
so
actual
approaches.
Actually
I
think
I
mentioned
it
in
issue
the
only
thing
I
suggested
to
do
it
on
branch
and
right
now
I
learned
that
we
need
to
drop
all
typescript
version
and
do
break
and
release.
So
I
would
suggest
approach
when
we
do
as
much
preparation
work
in
full
and
later
we
we
do
a
set
of
commits
that
first
migrate,
stuff
to
syntactically
so
like
we
will
use
prettier
or
probably
prettier
to
valid
as
validation
tool.
A
A
If
I
also
use
them
as
a
reference
and
you
repo
for
entire
repo,
just
will
create
syntax,
first
and
latch,
enable
slowly
enable
typings
and
because
it's
breaking
change
and
we
will
have
separate
branch,
call
1500
line
and
support.
We
can
allow
ourselves
to
have
like
a
week
or
something
when
when
mastery
is
broken,
and
I
think
we
need
that
to
do
quality
results,
so
I
will
so
part
of
a
plan
is
to
migrate
javascript
files
as
a
whole,
so
at
least
like
for
syntax
changes
and
for
basic
typescript
compilation.
A
A
A
With
preparation
steps
that
we
need
to
do
a
separate,
separate
issues,
and
I
have
a
branch
for
breaking
release-
I
basically
remove
stuff,
so
one
is
the
same
as
liking.
This
issue
about
row,
two
typescript
is
just
it
should.
A
A
To
yeah
so
so
as
much
preparation
work
as
possible
and
release
1600
alpha,
one
that
doesn't
have
typescript
conversion
but
have
all
breaking
changes
and
preparation
work.
We
anticipate,
and
so
we
have
a
point
of
reference,
so
people
from
community
know
what
what
broke
the
typescript
conversion
or
removing
duplicated
stuff.
A
One
thing:
I
wanted
to
discuss
that
sim
that
will
actually
simplify
conversion
and
which
is
like
separate
projects
that
you
can
really
help
me
with
it's
like
four
typings,
I
feel
like.
We
need
to
figure
out
the
way
not
to
pressure
people
into
updating
flow,
at
least
like
for
most
things.
So
the
dot
only
suggests
that
on
working
group
calls
yesterday
too,
it's
written
by
orta,
so
you
can
and
north
is
also
interested
in
graphql.
So
we
can
bring
him
too
much
stuff.
A
I
don't
expect
it
to
be
able
to
convert
everything,
so
we
can
maybe
have
some
gifts
or
patch
to
apply
on
top
of
it,
but
I
feel
like
if
we
pressure
every
contributor,
learn
typescript
and
learn
flow
at
the
same
time,
and
it's
like
and
another
thing
I
don't
want
to
be-
I
don't
want
to
be
person
responsible
for
right
and
four
times,
and
I
don't
want
anybody
else
to
be
responsible
just
for
full
time.
Yes
yeah.
So
let's
try.
Let's
try
a
technical
solution
and
if
it's
not
work,
we'll
think
something.
A
But
if
it's
work
it
will
be
great
another
thing
in
section:
zero:
zero.
I
want
to
drop
polyfills
because
they
have
to
migrate
and
the
bunch
of
hacks.
A
A
So
I
feel
like
if
we
polish
you
or
transpire
a
lot
of
things.
It's
great
problem
for
debugging
and
that
and
stuff.
So
I
think,
like
we
can
figure
out
something
out
like
I
was
publish
another
package
transpiled
one
or
have
like
a
polyfill
package
like
react
has,
but
it's
pretty
hard
and
it's
reduced
bunch
of
hacks
hacksawing
titles
like
yeah.
So
if,
if
you're,
okay,
that
we.
A
So
we
would
not
introduce
very
complicated
the
same
correct
stuff
that
is
hard
to
polyfill,
but
we
need
to
use
like
object
values,
because
if
we
use
polyfill
for
object
wireless,
it's
like
we
need
to
type.
It
is
not
working
good
with
4,
it's
not
working
with
typescript
one
giant
hug.
What
do
you
think
about
that?.
A
It
sounds
like
like
a
good
one.
Okay,
so
I
it's
so
section
zero
zero
alpha
one
will
contain
polyfills,
removing
perfumes,
removing
duplicated
stuff,
removing
and
try
to
remove
four
hacks.
So
if
we
go
to
typescript,
we
don't
need
for,
like
I
said
we
don't
need
to
take
four
weeks
into
typescript
would
because
for
typescript
will
introduce
its
own
set
of
workarounds.
A
F
Yeah,
I
think
in
typescript
four
we
can
do
some
tricks,
but
I
guess
that
the
backward
compatibility
will
eventually
force
us
to
yeah,
but
is
there
a
plan
to
drop
support
for
old
versions
of
typescript
at
some
point,
because
I
I
using
version
two
of
that
clip
right
now.
A
Yeah
so
idea
is
to
support
3.7
have
recursive
types
and
they
are
pretty
important
for
code
bases
for
cash
support
for
recursive
type,
so
we
will
lose
a
bunch
of
type
checkings.
If
you
drop
the
question
type
in
graphql,
a
lot
of
things
are
recursive.
We
can
use
download
dts.
There
is
a
special
tool
from
attached
team
to
down
level
titans.
A
A
3.7
I
checked
like
3.7.
It
was
released
more
way
almost
like
two
years
ago,
so
it
was
plenty
of
time
to
great.
I
don't.
I
feel
like
over
over
supporting
too
much
work
as
it
talks
about
download
projects
and
create
a
lot
of
frustration,
especially
for
new
contributors.
You,
you
know
why
it's
not
possible,
but
every
time
any
npr,
you
need
to
explain
a
person.
You
cannot
use
this
new
types,
new
js
virtual
typescript
feature,
so
it's
free,
so
yeah.
A
A
A
So
yeah
yeah,
so
anybody
else
have
some
suggestion
about
typescript
and
we
can
continue
inside
the
issue.
I
will
like
cruising
my
and-
and
I
will
publish
a
timestamp
in
this
video
in
touchscript
issues,
so
we
can.
If
something
is
missing,
we
can
discuss
just
like
we
promised
to
also
discuss
like
streaming
before
and
how
it's
blocked.
B
I'll
just
add
that
I
think,
basically
repeat
what
you're
saying
I
think
everything
is
is
fine.
I
just
I
wrote
I
added
to
the
task
where
you
know
you
could
or
update
the
existing
typescript
issue
or
break
into
a
couple
of
like
smaller
issues.
Just
so
we
can
track
and
it
will
be
easier
for
us
to
see.
You
know
where
we
stand
and
what
what
we
need
to
do
and
what
we
need
to
take
care
of,
and
things
are.
A
Yeah
yeah
one
of
the
to
explain
one
of
the
things
like
it's
not
mentally
hard
when
you're,
like
in
in
a
mode
that
you
failed
to
deliver
and
to
write,
is
the
same
issue
from
year
ago.
So
but
like
actually,
this
group
is
helpful
because
I
have
a
real
feedback
from
people
and
understand
that
it's
not
blaming.
It's
like
community
wants
to
help.
It's
not
like
yeah,
so
yeah,
so
switch
to
to
stream
and
defer
so.
E
Yeah
like
so
I
I
guess
I
I
have
a
it's
a
pretty
large
pr
for
a
deferred
stream
that
definitely
needs
to
get
broken
up.
The
the
first
part
of
it
was
doing
the
async
intervals
and
where,
where
I'm
at
with
that
is,
you
had
commented
that
there
may
be
a
memory
leak
because
it
was
a
similar
pattern.
E
A
A
So
what's
your
motivation
for
streaming
g4,
like,
obviously
you
want
to
want
to
feature
to
be
merged,
but
do
you
interested
in
becoming
core
contributors
is
like
why
I'm
asking
so
to
explain
why
I'm
asking
is
that
some,
some
small
things
it's
easier
to
do
than
to
explain
so
it's
totally
okay,
if
you
interested
in
this
picture
to
be
added,
it's
mean
like
I
can
ask
you
to
do
some
stuff
and
some
some
stuff.
A
I
can
like,
for
example,
in
your
pr's
I
can
commit
on
top
of
so
when
you
review
and
stuff
it's
easier
to
just
like
fix
it
yourself
and
commit
on
top
and
ask
person
and
ask
you
to
review
it
in
case
of
missing
something
or
you're
interested
in
long-term
contribution.
In
that
case,
I
will
try
to
provide
feedback
and
explain
like
especially
on
small,
like
obviously
you
more
knowledgeable
about
stream
and
defer,
but
about,
like
you
know,
needs
and
pieces
and
how
we
create
like
tests
and
stuff.
So
what's
what's
your
interest
in
that.
E
I
I
I'm
more,
I'm
more
interested
in
the
feature
than
being
a
maintainer
long
term.
Maybe
after
this
feature
is
done,
I
could
I'll
consider
it
but
like
right
now
I
I'm
kind
of
using
all
of
the
free
time
I
have
towards
for
this,
and
this
back
and
the
other
parts
of
it
and
to
focus
more
on
graphql
js.
If
that's
okay,.
A
Yeah
yeah
yeah.
It
is
definitely
it's
like
open
source.
It's
like
the
fact
that
you're
contributing
it's
you
you're
doing
rainbows
and
all
other
most
of
other
programmers.
It's
like
it's
like
everybody,
choose
a
way
to
contribute
that
he's
comfortable
with
it.
Just
like
we
need
to
align
expectations.
So
I'm
I'm
not
feel
bad.
When
I
like
commit
on
top
of
your
pr.
E
Yeah,
no,
not
at
all
that
that
would
be
very
helpful,
like
whatever
is,
is
easiest
for
you.
I
I
want
to
do
so.
If
you
would
like
me
to
fix
something,
I'm
happy
to
do
that
or
if
it's
easier
for
you
to
just
push,
I
could
give
you
access
to
the
branches
that
we're
working
on
too.
If
that's
easier
or
I
feel
like.
A
A
We
discussed
this
idea
for
experimental
folder,
but
it's
at
some
stuff
that
it's
really
hard
to
isolate
without
basically
copying
the
code
right
yeah
and
because
we
will
do
texture
conversion,
wait.
I'm
actually
like
for
me
it's
easier
to
help.
You
convert
your
full
code
into
typescript
after
conversion,
then
to
merge
it
and
like
it's
too
much
uncertainty.
A
So
I'm
like
okay,
this,
I
think
iteratible
like
module
like
if
we
figure
out
all
the
stuff
but
general.
At
the
same
time,
I
see
how
you
want
to
get
the
momentum
attraction.
So
what
what
we
can
do
and
is
just
the
first
step-
is
to
give
it
up
there.
So
you
can
write
a
blog
post
and
advocate
and
get
some
feedback
on.
It
is
to
release
basically
what
we
have
right
now.
14
points.
A
A
From
a
branch
so
like
so
action
items
for
me,
I
set
up
a
branch
and
basically
like
much
stream
and
default
vr
that
you
have
right
now
into
that
branch
and
manually
create
a
release
on
top
of
it.
So
I
need
to
create
release
what
we
have
on
master
and
release
what
we
have
on
this
branch.
So
we
would
have
like
14
15
point
something
point
something
and
the
same
version,
but
dash
stream
before
experimental
stream
default,
and
that
way
we
partly
unbox
you,
but
don't
introduce
a
lot
of
uncertainty
like
this.
A
E
Yeah,
the
the
only
thing
is
keeping
at
the
branch
up
to
date,
because.
G
Yeah,
I
can
help
with
that.
So.
A
A
If,
if
somebody
else
create
pair,
I
will
try
to
replace
it
and
if
it's
too
complex,
I
will
ask
a
person
to
create
pr
tools
so
not
not
to
put
too
much
pressure
because
I
know
like
we
switched
to
that
trip
and
bunch
of
other
changes
and
basically
like
it,
it
would
be
impossible
to
yeah.
So
I
will
take
up
myself
at
least
like
changes
that
I
made
myself
or
make
myself
or
is
easy
easily
replaceable
and
for
a
rest,
I
will
ask.
E
Okay,
yeah
that
that's
fine,
it's
just
that
I
I've
been
rebasing
it
now
every
couple
weeks
and
if
it's
on
the
graph,
if
it's
on
the
graphql
js
main
branch,
then
I
won't
have
access
to
do
that.
So
I
don't
want
to
make
more
work
for
you.
A
Okay,
main
branch
I
feel
like
I
feel
like
for
me-
zero.
It
would
be
easier
to
rebase
yours,
your
stuff
or,
if
you
yeah,
if
you
want
to
cherish
work
and
to
help,
if
you
have
capacity,
I
can
give
you
commit
rights
to
do
that,
particularly
if
you
promise
not
yeah.
Of
course,
anything
else
I
can
do.
One
thing
I'm
like
not
comfortable
is
like
releasing
somebody
else's
branch
is
official
package
yeah.
A
Okay,
yeah
and
I
actually
like
like
to
avoid
as
much
work
as
possible,
so
yeah
yeah,
so
for
me
to
step
up
branch
release,
master
and
release
your
version,
so
idea
is
that
we
will
have
parallel
package
packages
version
so
yeah
every
time
there
is
from
a
master
we
release
from
your
branch.
A
So
ideally
we
would
release
one
last
release
from
master
and
your
package
and
next
it
would
be
1600
alpha
one
with
all
the
braking
changes
and
your
version
of
package,
the
same
like
just
with
like
attack,
wait
name
different
name,
and
it's
the
same.
By
the
way
I
can
do
the
same
for
express
but
okay
change,
because
you
also
depend
on
it.
So
we
can
have
like
two
packages
express
graph
here
and
graphical
js.
Is
experimental
stream
detail.
E
Okay,
okay,
so
you
are
planning
to
make
version
16
soon.
That's
right!.
G
A
It
would
be
before
typescript
conversion
so
to
do
all
the
breaking
changes
in
flow
to
clean
up
like
some
workarounds
to
remove
polyfill
and
polyfill.
Some
batch
of
other
stuff
release.
A
A
A
A
A
E
Yeah
also
I'm
on
the
graphql
slack
all
day.
If,
if
I
it's
easier
to
talk
over
there
yeah
and
we.
D
To
do
it
in
best
way,
sorry
to
interrupt
what
is
the
graph
curves
like
just
true.
A
Yeah,
so
maybe
we
would
switch
to
discord
in
the
future,
just
like
people
discussing
that.
But
right
now
we
have
a
swag,
it's
managed
by
foundation
and
it's
official.
Basically
one
thing:
we
don't
have
a
history,
so
history
disappears
yeah,
but
that's
useful
and
you
can
actually
reach
me
that
if
you
have
something
quite
important
or
discuss
something
you
can
reach
me,
I'm
I
one
g
here:
okay
yeah.
I
will
write
this.
A
Yeah,
so
just
like
that
and
yeah
you
can
use
the
same
service
level
agreement
apply.
I
will
try
to
reply
24
hours
because
sometimes
like
like
different
time
zones.
E
Oh,
so,
for
the
the
request
for
async
iterables,
what.
A
Yeah,
I
think
it
still
should
be
there,
because
it's
a
stepping
stone
and
some
work
is
done.
I'm
much
like
part
of
it
is
is
a
simple
turretable,
so
we
need
to
rebase
it,
but
like
most
of
the
work
like,
I
still
feel
like
the
release.
So
like
a
long-term
one,
it's
my
suggestion.
We
can
discuss
it.
I
just
don't
want
to
stretch
with
it
and
definitely
so
because
actually
losing
people.
Initially,
it
was
six
people
and
right
now,
so
one
is
type
script.
A
So,
like
this
intermediate
release
with
breaking
changes,
typescript
conversion,
we
release
it
and
merge.
I
think
iteratively
into
core
see
like
every
release,
will
still
release
this
experiment
or
something
package
for
every
version,
very
least
identical
version
just
with
streaming
d4
and
at
what
point
you
have
feedback
on
and
d4,
and
if
feedback
is
positive
with
people
using
it
and
like
everything
more
was
working,
we
just
merged
in
that
core
it.
A
It
would
be
also
hard
process
because
it
build
big
pr
and
execute
is
like
very
important
function,
especially
for
performance
and
asynchronously,
and
we
have
a
strange
arrangement
that
it's
like
if
every
resolver
synchronous
synchronous,
if
at
least
one
resolver
is
asynchronous,
it's
asynchronous.
So
it's
hard
to.
You
cannot
ensure
that
with
linters
or
anything
and
like
it's
pretty
complex,
so
I
think
I
feel
like
I
think
it
ratable
definitely
coming
after
touchdown
conversion.
A
A
So
yeah
so
so
basic
idea
is
to
invoke
you
in
gasoline
feedback
and
if
you
can
fix,
obviously
peer
dependency
would
be
an
issue
for
you,
because
if
you
try
to
use
some
like
a
power
server
is
depends
my
understanding
of
some
version
and
how
packages
and
will
work.
It
will
require
version
without
the
experiment
of
streaming
d4.
A
But
I
feel
like,
if
you
pass
all
the
tests
apollo
guys
can
measure
it
and
you
can
try
to
communicate
them
if
it's
hard,
I
can
help
you
getting
in
touch
with
them,
because
it's
like
features
that
community
wants.
So
I
feel
like
they
can
add,
like
one
line
and
packages,
so
I'm
saying
experiment
of
streaming
default
can
be
used
as
independency.
E
E
So,
just
just
make
sure
I
heard
you
so
after
the
typescript
migration.
That's
when
we'll
work
on
merging
the
async
intervals
from
resolvers,
okay,
yeah.
A
Yeah,
it's
it's
the
next
step
and
it's
like
something
that
I'm
like
totally
agree
with
it's
useful,
even
without
the
stream
and
d4.
It's
not
like.
We
imagine
something
that
is
not
useful.
I
think
a
terrible
is
part
of
javascript
and
stream
proposal
from
from
browser
stream
proposal
is
actually
user.
Sync
iteratable,
so
it's
not
only
a
secret
turbo
itself,
but
the
source
streams,
browser
streams
and
you
know-
support,
for
example,
streams.
So,
like
I
think
it's
useful,
and
so
it's
just
like
executor
execution
is
like
hardest
part
of
kraken.
A
Yes
and
like
it's
not
real
thing
about
like
yes,
as
I
said,
I
see
it's
like
really
hard.
So,
let's,
let's
let
it
be
next
step.
So
first
step
is
to
unbox
you
and
second
step,
as
is
a
simple
terrible,
but
for
merchant
stream
and
d4.
A
E
A
Yeah
but
we
will
have
like
a
new
directive
so
theoretically,
if
somebody
have
stream
id4,
they
can
get
like
strange
behavior
with
because
they
stream
in
d4
exit
and
premiere
4
have
different
arguments.
I'm
okay
with
right,
I'm
okay
with.
If
we
decide
to
edit
to
spec,
I'm
okay
with
breakage
and
stuff,
I'm
like,
I
don't
want
to
add
to
eks
and
break
an
experimental
feature,
so
experimental
features
that
brexit
people
and
later
would
change
it.
And
let's
break
like
people
in
different
way,.
G
A
E
A
E
A
E
A
So
I
think
it's
on
the
surface.
If
we
go
deeper,
it
would
be
like,
but
I'm
totally,
okay
with
them,
and
I
think
community
is
okay
with
them.
If
we
deliver
them
like
call
new
future
like
extremely
different
that
people
not
only
could,
but
people
actually
need
that.
It's
like
it's
like
problem
of
of
graphic
white
people
started.
It's
like
you,
have
one
huge
request,
but
that's
like
go
on.
You
need
to
wait
for
one
time,
so
I
think,
like
a
community
would
be
okay.
If
you
break
stuff
by
adding
it
I'm
like
not.
E
A
E
Okay,
I
think
I'll
reach
out
maybe
next
week
and
let
you
know
what
I'm
working
on
with.
As
far
as
the
memory
leak
yeah,
I
haven't
been
able
to
write
a
good
test
for
it,
but
I'll.
Let
you
know
what
I
did.
A
Yeah,
so
if
anybody
else
have
wait,
anything
that
cannot
wait
for
a
next
call,
because
we
will
definitely
have
a
call
next
month
and
we
can
discuss
stuff
and,
as
I
said,
you
can
reach
me
as
work
on
the
issue
or
like
in
other
ways.
But
if
you
have
something
great
that
you
want
to
discuss
right
now,
you
can
still
do
that.
D
Hold
on,
thank
you
very
much
for
your
time
and
all
the
effort
you
put
on
that.
E
A
A
Yeah,
it's
a
learning
experience
for
me
also,
I
like
yeah,
but
just
learning
how
to
do
community
management
stuff.
So
if
you
see
some
issues
that
or
suggestion
you
can
reach
me
like
publicly
or
privately
what
what
must
work
for
you
and
definitely
doing
call
next
month-
and
I
will
thank
you
all
in
initial-
and
hopefully
we'll
have
agenda
and
no
official
stuff
place
that
time.
So
you
can
add
your
items
and
agenda
and
user
basically
the
same
workflow
as.