►
From YouTube: Graphql.js Working Group - 2022-03-30
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).
A
C
B
B
It's
kind
of
hot:
do
you
hear
me
right.
B
B
No,
no
actually
yeah
yeah,
oh
yeah.
It's
it's
the
fifth
point
on
this
point.
So,
let's
start
from
point
number
one,
as
you
probably
know:
graphql
js,
hayaku
and
the
graphql
foundation,
and
so
as
any
project
on
the
graphql
foundation
to
take
part
in
it.
You
need
to
sign
membership
agreement,
participation,
code
of
conduct
and
probably
it's
only
two
documents.
B
So
if
anybody
watching
us
recording
this
meeting
and
want
to
join
next,
one,
please
sign
with
one
and
you
can
send
that
as
a
video
so
and
it's
like
five
minutes
yeah.
So
second
item
is
introduction
of
attendees
and
let's
start
from,
let's
go
in
the
order
of
in
agenda.
D
I'm
not
on
the
agenda,
I'm
about
to
be
I'm
submitting
a
pr
but
yeah,
I'm
alex
I'm
at
yelp
and
I'm
working
on
client
control,
nobility.
E
B
Yeah,
so
next
one
review
previous
meeting
action
items
no
update.
For
me
it
was
like
really
busy
month
for
totally
unrelated
reasons,
so
anybody
else
have
any
updates
on
action
items.
B
Action
item
first,
like
agenda
item
to
discuss
situation
because
I'm
basically
disappeared
for
for
a
month
before
last
time
was
a
gen
first
working
group
yeah.
So
I
canceled
last
working
group
for
like
officials
and
it's
coincide
like
next
day
was
start
of
the
war
in
ukraine
yeah.
So.
B
Like
not
at
the
top
of
my
list
of
priorities,
so
yeah
basically
disappeared
right
now,
I'm
back.
I
hope
I
I'm
right
now.
B
Sorry,
if
I
didn't
answer
any
of
my
tissue
or
anything,
because
I'm
like
this
end
of
last
week
and
beginning
of
this
week,
I
returned
back
to
work,
so
I'm
like
going
through
like
emails,
twitter
and
other
things
and
answering
everybody
so,
but
at
the
same
time
I
don't
know
what
happened
and
even
and
related
the
current
situation
of
in
ukraine
even
like,
without
I
think
we
need
contingency
so
and
I
want.
B
B
A
B
Everybody
except
jakob
from
from
the
review
team,
have
commit
access
and
yeah.
We
could
specifically
ask
not
to
give
it
to
him
so
other
than
that.
I
think
everybody
have
maybe
must
and
not.
But
if,
if
I
disappear,
it's
like
even
if
I
don't
disappear,
we
still
have
this
mechanism
and
you
can
use
it
right
and
you
should
definitely
use
it
to
disappear
one
more
time
for
some
reason.
D
Do
do
the
the
people
on
the
review
team?
Do
they
have
do
they
have
commit
access
like?
Can
they
merge
things
to
maine?
Can
they
do
releases
yeah.
B
Yeah,
I
think,
like
in
ben
and
they're
maxes
can
be
bottleneck,
so
yeah,
so
like
definitely
much
likes
how
much
access
I'm
not
sure
about
martian.
B
A
B
B
B
So
if
you
book
by
something,
if
like,
for
example
like
something
like
you
need
to
publish
and
brand
package,
please
reach
to
them
and
they
give
you
and
parent
keys
problem
within
penky.
Why?
I
don't
want,
because
always
you
know
how
profile.
B
Through
personal
accounts,
sadly,
I
can
paint
security,
it's
not
ideal.
You
cannot
restrict
tokens
to
particular
packages
and
other
effect
issues
like
yeah,
but
I
will
double
check
that
with
graphql
foundation,
what
to
do
with
keys
yeah.
So,
like
somebody
from
a
graphql
foundation,
I
teach
you
have
access
to
npm
guarantees.
B
And
by
the
way,
all
other
issues
like
graphql.org
by
documentation
website
like
domain
names,
everything
tc
should
either
already
have
access
to
it
or
they
can
reach
graphql
foundation
and
ask
for
access
to
it.
Refill
foundation
for
like
lastpass
account,
so
all
password
manager
is
there,
so
it
can
take
like
days
day
or
two
but
yeah.
I
think
it
can
download.
B
E
E
I
just
got
a
little
concerned
that
I
might
you
know,
there's
probably
a
better
way
to
do
it,
but
I
might
push
to
to
maine,
for
example,
in
graphql
gs
by
just
by
accident
instead
of
my
fork.
So
so
I
asked
just
to
be
removed
just
as
an
extra
layer
of
projection.
B
Yeah
well
totally
understandable,
it's
like
technical,
which
is
it's
important,
that
group
of
people
reliable
group
of
people
in
different
parts
of
the
world
have
access
to
it,
and
now
I
understand
that
these
people
should
be
geographically
distributed
because
anything
can
happen
to
any
country
at
any
moment
of
time
yeah.
So
if
you
at
some
point
want
to
get
like
right
access
this
rich
me
or
tc
or
like
anybody,
I
think
you've
done
enough
like
communities
that
you
totally
have
all
the
trust
for
for
stuff
like
that.
B
Since
we
have
agenda,
let's
move
to
another
topic
canary
releases,
such
viewers,
and-
and
can
you
please
tell
us
about
you.
A
Yeah,
so
for
the
kinder
releases
today
we
have
cannery
releases
on
every
pull
request.
So
if
you
open
something,
you'll
get
to
candy
on
every
pull
request,
but
there's
a
potential
security
concern
with
how
we
do
canneries
today
is
like
anyone
can
open.
A
pull
request
will
get
a
cannery,
which
means
any
bad
actor
could
open
a
pull
request
against
graphql
js
and
we'll
get
like
a
graphical
ds,
npm
version
for
something.
B
Yeah,
I
think,
like
it
was
a
part
of
what
we
discussed
so
people's
like.
I
did
some
measures.
First
thing
I
think,
like
two
days
ago,
I
watched
little
bit
different
version
of
what,
because
I
also
was
concerned
the
very
least
canary
for
every
for
every
girl.
B
B
I
didn't
use
any
like
server
or
anything.
It's
like
only
cia
action.
I
actually
discovered
some
issues
that
I
need
to
fix,
but
yeah
and
it's
working
right
now
so.
B
What
about
security
part?
I
I
think
I
took
necessary
so
like
a
problem
here
is.
A
B
B
Yeah
qualified
it'd,
be
since,
like
other
people
can
have
symbols.
I
I
personally
had
like
the
same
security
concern,
but
you
have
you
remember,
like
mining
scandal
like
a
year
ago,
when
people
tried
to
mine
on
ci
actions.
So
first
time,
people
who
opened
there
for
first
time
have
the
github
action
blocked,
so
the
same
mechanism
required
to
to
wet
action.
B
So
if
it's,
if
it's
an
unknown
contributor,
yeah
also
to
qualify
during
quite
with
canary
releases,
I
specifically
remove
all
scripts
and
I
had
duplication
message
saying
please
check
like
pr
before
you
before
you
use
this
package,
so
nothing
is
no
security
is
compromised
during
installation
phase,
because
there
is
no
scripts
and
it's
process
of
installing.
There
is
a
clear
warning,
you're
using
quite
something
experimental.
B
So
I
think
it's
not
ideal
scenario,
but
I
think
value
from
this
feature
is
bigger
than
some
potential
risk
unless
we
discover
like
actual
situation,
yeah
and
anything
else
on
this
topic
yeah
and
by
the
way
like
issue,
I
need
to
fix
it's
like.
B
It's
look
like
every
time
you
comment.
Now
your
command
gets
hidden
because
of
this
bot.
I
will
fix
it
if
you
see
with
behavior,
I
discovered
just
before
this
meeting
and
I
will
fix
it.
It's
like
really
weird
one
yeah.
So
anything
else
on
this
agenda
items
before
we
move
to
the
next.
D
One,
oh
thanks,
thanks
for
getting
this
out
by
the
way
it's
been
it
unblocked
me
and
it
was
super
helpful,
worked
great
yeah
and.
B
So
next
one
is
yeah
about
esm.
It's
also
use
a
hedge.
A
Yes,
so
we
have
this
full
request
for
setting
up.
Yes,
I'm
support
for
graphqlds,
so
I
will
send
the
link-
and
I
think
most
of
the
comments
are
being
addressed
on
it
and
it's
like
right
now.
It
gets
published
under
like
graphql-esm
package,
but
we
can
like.
I
think
the
discussion
is,
do
we
drop
cjs
completely
on
graphical
17,
or
do
we
like
how?
How
do
we
want
to
proceed
with
the
school
request,
because
this
is
already
state
to
be
merged?
A
Does
apollo
client
support
like
because
for
all
our
two
links
at
the
deal,
all
our
tools
will
work
with
the
asm
we
have
like
out
of
the
box.
He
has
some
support
for
stuff,
it's
more
like
who
else
will
be
impacted,
so
it's
like
is
apollo
client
ready
for
this
change,
because
I
think
that's
one
of
the
other
biggest
libraries
who
may
be
impacted
with
this.
A
Yeah
because
I
don't
think
we
would
be
able
to
get
esm
within
16.
Unlike
I
mean
there
is
sort
of
some
support
today,
but
not
complete,
so
it
would
require
a
breaking
change
in
the
line
to
get
complete.
Yes,
and
most
packages
have
dropped.
Yes,
most
packages
did
drop
like
many
are
dropping
today
with
cgs.
So
it's
like
it's
probably
the
right
thing
to
do.
I
don't
know
it
it.
It's
it's
up
to
discussion.
B
B
But
we
dropped
it
and,
like
I
didn't,
saw
any
anyway.
Nobody
reported
any
big
issue
with
the
people.
I
kind
of
understood
it
as
a
normal
evolution
of
things.
So
I
think
we
can
do
the
same
with
cgs,
so
we
can
drop
it
like
open
the
issue
about
it.
I
think,
is
a
good
thing
similar
to
for
one
and
drop
it,
and
if
anybody
concerned.
B
Figure
out
what
to
do
before,
like
before
release
1700
release
yeah.
So
I
I'm
like.
B
B
C
B
A
B
E
Issue,
sorry
is,
is
there
are
we
are
we
I
I
just
lost,
I'm
not
sure
I
got
100.
Are
we
talking
about
whether
to
drop
cg
cjs
support,
or
are
we
talking
about
whether
to
publish
like
publish
an
esm?
Only
version
like
that
I
saw
was
done
in
one
of
the
pr's.
E
You
know
under
under,
like
sort
of
like
a
deferred
stream.
Experimental
tag,
like
you
know,
science
system
are
we
talking
about
both
of
those
as
options
or
yeah.
B
I
think
ideas
here
is
the
same
as
we
had
like
four
versus
typescript.
We
can
support
both,
but
it's
harder,
let's
support
what
what
would
be
used
by
most
people
and
see
if,
if,
like
anybody,
have
like
big
issues
with
that
because
like
we
can
spend
efforts
on
supporting
both
and
nobody
will
actually
use
it.
E
Since
right,
but
but
I
mean
there
is
somewhat
of
a
difference
in
that
in
the
flow
typings
and
and
typescript,
you
know
sort
of
integrated
on
every
line
of
the
code
base,
and
you
know
I
I
think
we
have
a.
I
think
you
know,
whereas
here
we're
talking
about
sort
of
like
post,
you
know
the
build
scripts
that
can
just
sort
of
put
out
an
esm,
only
package
for
every
like
release.
E
B
Do
packages
with
peer
dependency
is
to
publish
codex
cjs
and
have
mg
ngs
wrappers,
but
in
that
case
it
was
tree
shaking
so
currently
we
have
like
three
shaking
and
but
if
we
go
official
route
we
lose
it.
There
is
this
approach
that
pablo
implemented
like
with
the
same
package
but
different
version,
which
is
which
can
work,
but
I
never
saw
anybody
actually
doing
that.
B
B
B
B
E
B
Especially
for
underweight,
I
think,
if
it's
okay,
to
experiment
like
with
graphql
stuff
with
it,
it's
the
main
concern
of
this
package,
but
for
for
distribution
or
esm
cjs
support,
I
think,
is
best
just
for
ecosystem,
but
in
case,
if
somebody
will
have
like.
B
Always
try
to
work
around
it
or
we
can
like
move
to
bubble
solution
just
like
other
way
around.
So
like
core
package
will
be
esm
and
if
you
want
to
use
cgs,
you
use
like
some
special
version,
so
yeah
anyway
wait
before
1700
release.
We
can
change
our
mind
based
on
feedback.
I
just
want
to
to
push
like
future
proof
solution
first
and
see
feedback.
B
B
B
Those
types
are
intended
to
be
for
result
of
getting
a
special
query.
It's
like
those
types
is
not
for
customer
transfection
queries.
B
Those
type
is
like
specifically
intended
for
for
response
of
quite
a
return
by
getting
transgression
quiet.
So
this
person
wrote
his
no,
I
think
no,
I
will
check
it
more
because
now
I'm
a
little
bit
concerned.
B
B
B
Yeah,
I
think
it's
like
design
issue
what
to
do
there
because,
basically
different
functions,
you
know
initially,
it
was
different
functions.
Initially,
bitcoin
schema
was
designed
to
build
schema
from
a
client
and
build
the
ic
schema
was
designed
to
build
your
own
schema
difference
here.
Is
you
want
useful
defaults
for
your
own
schema,
but
you
want
to
preserve
client
schema
in
the
same
way
and
for
example,
you
have
non-standard
non-standard
client
or
you
have
like.
B
Previous
version
of
like
spec,
have
little
bit
different
things
inside
the
introspection,
I'm
not
sure
here
it
gives
probably
yeah
specified
by
it's
like
new
directive,
so
you
don't
want
to
inject
specified
by
in
response
of
a
client.
B
You
want
to
check
if
client
have
specified
by
or
not
and
I'm
not
speaking,
even
about
74,
which
is
like
totally
different
issue,
but
even
like
for
specified,
buy
or
duplicate
like,
for
example,
duplicate
right
now
we
have
like
four
location,
but
previously
we
had
only
two
occasions
so,
but
I'm
like-
and
I
think
I
wrote
it
in
the
comment
of
this
issue-
that
initial
assumption
now
wrong
now,
because
people
save
their
introspection
as
a
sda
so
and
use
sdl
to
like
pop
away,
like
in
memory
schema
for
a
client.
B
B
B
B
A
Yeah
yeah,
so
17
would
be
like
having
a
some
some
like
some
state
of
the
asm.
Whatever
we
discuss
based
on
like
the
issue,
we
open
either
completely
drop
it
or
like
what
you
call
was
suggesting
right
now
and
then
what
would
be
some
other
things
that
we
are
planning
to
make?
Because
I
remember
if
it's
b16,
we
had
that
like
enum
issue,
like
the
enum
object,
like
the
typescript
enum
thing,
is
that
something
we
want
to
revert
back
for
v17,
because
that
will
be
a
breaking
change
again.
B
B
A
B
Yeah,
so
the
idea
was
that
I'm
briefly
interrupted
to
do
to
quickly
do
like
canary
release
things,
and
it
took
like
way
more
time
than
I
expected.
Basically
so
idea
right
now,
release
like
another
minor
version,
because
people
discovered
some
bugs
in
version
16
and
branch
it
out,
because
it's
how
correct
scripts
for
both.
B
B
B
Group
stream
like
what.
What
definitely
goes
into
the.
B
Yeah,
I
I
will
watch
it
and
unless
right
people
disagree,
the
idea
was
that
we
should
we
ship
something
so
stream
and
d4
idea
is
too
much
it
and
I
even
like
for,
for
things
like
alex
pr
or
of
no
ability,
is
also
measured
because
we
agreed
to
do
it
under
a
flag.
B
To
qualify,
let's
it's
about
syntax,
we
agreed
too
much
syntax,
not
semantics,
and
maybe
later
we
can
yeah,
but
I
think
you
mostly
book
one
syntax
similar
like
we
stream
a
d4,
I
think
like
d4,
is
more
agreed.
So
the
only
thing
we
can.
I
will
watch
working
group
and
I
also
asked
rope
of
current
status.
Might.
C
B
Stream
from
d4
yeah,
I
think
we
need
to
ask
current
status
and.
A
B
A
B
And
complex-
and
it's
actually
a
good
thing-
that
to
start
watching
version
17
because,
like
a
way
leading
the
pharma
correctly,
is
not
to
have
like
breaking
changes
at
all,
which
way
some
of
the
things
are.
But
what
what
they
are
chain
is
the
most
complex
blockchain.
I
ever
saw
in
my
life,
so
yeah.
B
A
B
B
B
A
This
changed,
I
know
for
roll
up
and
a
wheat.
It
works
because
I've
tried
that,
but
I
don't
know
about
a
backpack,
so
it
can.
I
think
it
should
work
fine
on
webpack,
because
that's
what
lauren
is
explaining
on
it.
So
I
think
it's
and
we
do
have
a
back.
That's
another
thing
somewhat
like
bundling
at
least.
B
Okay,
I
mean
like
a
removal
of
this
code
from
web
packing,
not
not
like
yeah.
In
initial,
I
will
write
weight.
B
B
E
Have
any,
I
think,
are
these
the
ones
about
the
foreign
stream.
A
E
Yeah
I
mean
they,
they
weren't
from
rob.
They
were
actually
they're
actually
were
from
me,
but
they
were
too
rough,
and
so
I'm
not
sure
if
we
can
get
get
very
far
on
it
without
him.
Actually,
so
you
know
I'll
leave
that
to
you
guys
at
the
side.
B
E
Yeah,
I
think
that
I
think
that
makes
sense
like
I
think
it's
more.
To
be
honest,
I'm
not
sure
it
might
even
belong
in
the
spec
meeting
as
well.
B
By
the
way,
same
goes
to
five
meetings.
Also,
if
discussion
starts
there,
we
can
always
create
a
separate,
separate
call.
We
had
this
for
one
off
and
for
some
other
topics
are
specified
by.
We
had
like
separate
course,
because
sometimes
people
can
go
really
deep
into
stuff,
and
this
requires
time
so
yeah
if
you're
interested
in
what
feel
free
to
suggested
during
work
in
gluco
or
reach
and
by
the
way
idea.
E
I
just
pasted
it
into
the
chat
I
mean
I
have
you
know
I
think
again,
if
rob's
not
here,
it's
not
the
right
context,
but
I
have
I
mean
basically
the
the
differences
that
you
know
I
could
set
up
and
do
you
know
short
discussion
about
in
terms
of
the
signatures
of
the
execute
functions,
and
you
know
how
things
would
look
a
little
differently
and
basically,
the
general
picture,
without
getting
into
the
specifics,
is
just
to
better
better
implement
the
ability
to
to
to
work
on
batches
of
results.
E
At
a
time
I
mean
right
now
we're
using
async
iterators
that
give
one
result
at
a
time
the
back
end,
but
if
they're
back
into
supplying
more
than
one
result
and
execute
is
able
to,
you
know,
push
more
than
one
result
you
know
sort
of
with
each
round
of
the
event
loop.
You
know
that
you
might
get
a
little
bit
better
performance
that
way,
and
I
just
I'm
not
100
sure,
on
whether
it's
actually
a
spec
change
or
only
an
implementation
detail.
B
Yeah,
I
think
I
think,
like
what
spark
have
was
algorithm
and
it
specifically
says
you
cannot
promote
it
differently.
B
If
you
get
the
same
observed
result,
so
it's
to
fall,
if
you
think
it's
valuable
for
all
implementation
and
if
you
think
it's
like
and
especially
wait,
and
if
it's
change
absolute
result,
it's
definitely
belong
to
to
main
working
group.
If
it's
like
one
of
the
way
to
implement
algorithm
when,
basically,
if,
if
you
want
to
send
multiply,
multiply
thing
in
the
same
http
package
kind
of
when
it's
it's
the
same,
multi-part
thing
how
cold
it's
like,
it
should
be
part
of
my
working
group.
E
B
Okay-
and
I
was
we
had
last
agenda
item
but
alex
said
he
need
to
go
next
time.
I
will
try
to
manage
time
better,
so
we
have
time
to
discuss
all
the
issues,
but
I
think
we
briefly
discussed
coin
control
now
ability
that
passing
part
syntax
part
is
getting
much
under
and
flag
and
semantic
part
is
blocked
by
discussion
in
working
group.
B
Anybody
else
anything
else.
Anybody
you
want
to
discuss.
Oh
like,
if
you
have
stuff
you
want
to
talk
about
control,
pronoun
ability
there,
you
also
can
discuss
it
anything
else.
Anybody
wanted.
A
A
B
B
I
start
reviewing
that,
because
what
one
is
the
simplest
one
and
I
think
warren
address
like
I
left
some
review
comment
and
it
was
addressed.
So
I'm
approached
forever.
B
Yeah,
so
it
was
very
productive
working
group
we
discussed
by
two
issues
and
agreed
on
next
steps,
so
see
you
in
a
month
and
hope
everyone
will
be
safe.