►
From YouTube: GraphQL.js Working Group - 2021-06-30
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
D
C
D
A
You,
if
you
are
up
higher
up
by
the
way
I
see
like
an
issue
with
my
video
sound.
We
had
the
storm,
oh
and.
B
A
So
there
is
no
electricity
and
I
use
mobile
internet.
So
if
you
experience
any
problem
with
my
audio
video,
I
will
try
to
change
something
so
like
just.
A
A
A
A
So
you
need
to
sign
it
only
once
so,
if
you're
watching
this
on
youtube
or
like
want
to
join
future
meetings,
please
sign
membership
agreements
and
second
item.
Let's
last
the
introduction
and
I'll
start
hi,
my
name
is
one
I
hope
to
maintain
graphqs
and
now
I
experience
problem
with
internet.
So,
let's
find
no
video
from
me,
hello,.
A
Yeah
so
let's
shift
to
not
taking
me
kind
of
don't
do
that,
but
at
the
same
time
we
have
video
recordings
on
youtube.
So
at
some
point
maybe
we
need
to
start
doing
notes,
but
now
I
think
we
can
skip
it,
especially
since
three
of
us
so
review
agenda
and
yeah.
A
It's
one
item
for
me
and
one
item
for
us
a
hit
any
anything
else
like
you
want
to
add.
Do
you
want
to
like
discuss
anything
from
your
side.
E
No
nothing
for
me.
I
I've
been
pretty
busy,
so
I
haven't
been
able
to
do
much
different
stream
stuff,
but
just
wanted
to
hear
what
the
latest
updates
were
for
version.
16..
Okay,.
A
So,
okay,
agenda
action
items,
action
items,
yeah
actually
like
about
security,
so
one
top
one.
If
you
I
was
hunted
in
canada.
So
if
you
open
it's
about
security
issues,
yeah
something
in
the
chat,
so
security
block
reports,
I
actually
need
to
fill
one
myself.
A
A
Add
fixed
to
1400
or
14x6,
while
so
yeah.
I
will
because
I
need
to
report
it
anyway,
so
I
claim
this
security
bug
reports,
setup,
discussion,
graphql.
I
think.
C
Yeah,
like
I
mean
people,
we
don't.
The
plan
was
to
have
habit,
I'm
still
down
to
like
help
out
with
it.
But
it's
like
we're
not
getting
many
like
the
questions
we
get
like
recently:
they're,
not
either
they're
too
technical,
they're
kind
of
like
they're,
not
sometimes
they're
like
feature
things,
and
then
others
are
like
example,
suggestion
which
shouldn't
even
like,
which
we
can
just
close
at
times,
because.
B
C
A
So
we
can
experiment
with
it
and
see
how
it's
going
if
it's
like
low
activity
or
it's
a
lot
of
containers
yeah.
Third
one
is
number
35
about
four
types
and
facebook
engineers.
I
will
close
it
because
it's
already
resolve
itself.
We
discuss
it
with
facebook
engineers,
so
they're,
okay
with
us,
dropping
four
types
as
they
said
they
will
basically
be
in
the
dependency
to
last
version.
A
A
Yeah,
I
think,
like
other
ones,
big
and
neat
for
me.
I
must
have.
There-
is
like
progress
on
on
that.
A
A
Yeah
status,
update
on
16x6
so
obviously
like
biggest
thing
that
was
done
in
like
preparation
for
16x6
1600
release
is
typescript
migration
and
it's
done
so
fro
from
start
from
a
ghost
panel
goes.
I
will
send
a
issue
here.
Sorry,
I
cannot
share
my
my
screen
or
I
like
the
video,
because
I
have
problem
with
internet.
So
I
just
posted.
A
A
D
A
A
Yeah
yeah,
oh
super
yeah,
and
maybe
I
will
have
a
decent
internet
in
like
a
couple
minutes,
yeah
yeah,
but
let's
continue
yeah
so
from
five
goals.
If
you
open
an
issue,
there
is
a
top
at
first
command.
There
is
like
a
list
of
goals
I
think
like
we
almost
converted
like
there
is
issues
like
inums
and
we
cannot
do
a
hundred
percent
ideal
conversion
anyway.
A
So
from
obvious
stuff,
we
missing,
like
probably
news,
is
the
biggest
thing
mission
from
from
media
magic
type
script,
and
there
is
parable
that
second
goal
is
address
fully.
We
remove
like
for
fully
remove
four
we
remove
polyfills
will
remove
like
yes,
five
clusters
like
tray
praise,
prototype
classes,
so
we
switch
to
property,
syntax.
A
Yeah
yeah
so
issue
my
current
right.
A
I
actually
think
like
for
them.
Support
is
critical,
because
then
community
is
not
using
npm
anyway,
and
they
don't
use
like
strict
version
control
in
a
sense
like
they
don't
use,
use
strict
releases,
preferred
ways
to
just
refer
to
attacks
or
master,
so
we
can
do
finish
like
dance
port
after
160
release.
A
As
for
esm,
I'm
I'm
kind
of
not
sure
what
to
do
there.
I
think,
but
let's
open
action
item
for
me
is
to
open,
separate
issue.
Discuss
problems
that
I
experienced
when
I
tried
to
edit
is
some
support.
Previously,
it's
kind
of
big
topic
so
other
than
that.
I
think
like
another
thing
that
can
be
kind
of
critical
is
yakov's
prs.
A
A
A
A
A
A
A
One
thing
I
can
do
and
it
will
probably
speed
up
stuff
so
like
we
don't
expect
a
huge
breaking
change
like
maybe
some
braking
change,
but
not
huge
braking
change,
so
I
think
we're
kind
of
ready
for
for
better
I'll
release
like
better
package
because,
like
from
alpha,
I
think
like
right
now,
alpha
4
or
alpha
5
or
during
the
release
kanji
that
or
like
during
the
release.
It's
not
logical,
it's
not
logic,
so
I
will
do
rc
today
tomorrow,
oh
no
thursday,
but
I
will
release
better
the
only
one
big
breaking
change.
A
I
would
potentially
anticipate
this.
Yes,
I'm
support
because
it's
like
really
hard
to
support
both
cjs
and
esm
in
the
same
package
and
like
for
it
to
be
enough
for
you
to
work
with,
like
3,
shake
and
everything
else,
but
I
think
it's
still
qualified
for
better
because,
like
we
switched
to
type
3,
we
did
like
huge
number
of
breaking
changes,
and
now
we
just
clean
up
this
smaller.
A
A
By
the
way,
if
you
want
to
hope
it's
a
good
thing
would
be
to
review
the
help
with
their
reviews,
so
it
will
take
me
some
time
to
review
all
them
and
if
you
like,
look
at
them
and
see
something,
please
comment
there,
so
you
know,
instead
of
waiting
for
me
to
give
feedback.
A
There
is
a
chance
like
you
can
discover
something
and
pr
after
can
iterate
on
it
by
the
way
like
big
thanks
for
like
say,
hi,
good
friends
for
bringing
a
new.
They
are
about
news
so
like
we
know
that
if
the
owner
wants
to
replace
it
or
not
so
big
thanks
to
that,
if
you
see
any
other.
A
Cars
quite
a
measurable
non-controversial
pr,
so
it's
not
a
big
change.
It's
not
a
key
feature
so
yeah.
It
would
be
great
if
you
do
the
same
because,
like
some
of
them
are
pretty
basic,
we
just
like
stuck
at
some
points.
A
A
A
Any
other
feedback
on
release
1600
anything
else,
you
think
like
worth,
including
or
like
stuff
that
you
think
is
maybe
not
important
for
sixteen
zero
zero.
Maybe
you
have
suggestions.
How
to
change
is
fun.
A
As
I
think
I
will
update
that
responds
like
more
details,
I
I
think
I
didn't
update
it
for
a
long
time,
yeah
for
like
half
a
year
or
more.
So,
let's
switch
to
another
topic
next
topic.
Next
topic
is,
I
think
it
was
yours
about
documentation.
C
C
So
that
will
help
us
and
like
docusaurus,
has
like
many
other
things
that
are
just
baked
and
we
don't
have
to
do
much
work,
so
the
goal
would
be
to
like,
hopefully,
by
next
meeting,
I
will
have
like
some
sort
of
mvp
ready
for
dark
side
that
has
all
the
dogs
pulled
in
from
our
code
comments,
and
then
I
will
also
merge
so
in
our.
We
did
talk
about
it
earlier
on.
C
C
So
this
will
allow
us
to
delete
the
docs
folder
and
then
have
our
new
site
up
in
the
docs
folder
place
and
yeah
then
after
that
is,
then
we
can
start
like
removing
from
the
main
graphql
site
and
direct
into
the
graphql
js
stock
site
and
and
then
soon
we
should
like
and
then
after
that,
we
can
also
have
like
some
example
code
references,
because
there
are
a
couple
issues.
As
there
are
a
couple
issues
I
try
as
they're
like
we
don't
have
like
good
doc.
C
We
don't
have
like
good
examples
for
someone
to
look
at
so
having
like
good
examples
where
someone
can
see
how
graphql
js
is
used,
because
most
of
the
examples
out
there
are
just
like
using
with
express
or
something
on
the
server
side,
but
like
using
it
with
something
like
using
in
browser
like
how
graphical
does
there
was
an
issue
on
it.
We
we
should
show
like
more.
A
I
think
maybe
we
can.
The
half
is
like
a
really
good
idea
or
not,
but
maybe
we
can
reuse
the
integration
test
either
integration
test
for
web
parts.
Maybe
we
can.
A
A
So
if
you
feel
free
to
like
iterate
on
it
like,
for
example,
compared
to
your
type
stripping
duration
or
typescript
migration
on
other
ripples,
it's
like
it
was
a
long
process
because,
basically,
we
wanted
to
maintain
the
same
moral
quality
and
it
was
big
with
four
and
I
wanted
to
to
keep
it
at
the
same
for
typescript.
A
A
Maybe
we
can
try
to
minimize
number
of
dependencies,
so
dependency
is
like
absolutely
necessary
because
it's
harder
to
maintain
dependency
on
updation
and
the
social
security
risk
other
than
that
right.
If
it's
don't
end
like
anything,
special
like
docosaurus
is
good
like
and
like
the
official
typescript
documentation
generation
is
also
good.
Some
popular
project
plugins
also
good
to
use.
A
Yeah,
just
because,
like
big
prs,
have
a
tendency
to
especially
things
that
that
stuff
went
to
packages
and
package
work.
So
I
I
encourage
you
when,
when
you
have
something
like
working,
even
if
it's
like
super
simple,
we
just
create
vr.
I
will
firstly
review
it
from
them,
I'm
not
expecting
from
them.
A
I
hope
like
in
front
and
yeah,
so
my
idea
is,
we
can
iterate
on
it.
So.
A
C
C
A
A
Is
too
much
graphql
railway.js
into
graphql
package,
because
now
it's
required
to
maintenance.
So
it's
like
a
lot
of
duplicated
work
on
both
so
by
merging
it
into
graphic
chess,
it
will
make
life
easier
for
maintenance
and.
C
So
question:
so
if
we
merge
it,
so
are
we
still
going
to
like?
So
if
I
do
npm
install
graphql
relay
right,
that's
what
I
download
so
are
we
still
gonna
publish
to
npm
as
graphql
relay
or
is
it
gonna
be
a
package
just
like
you
just
downloaded
and
graphql,
and
then
it
works
yeah
just.
A
I
think
like
for
beginning
just
because,
like
I
spoke
with
relative
and
facebook,
and
they
said,
like
some
things,
don't
make
sense
anymore,
like,
for
example,
client
id
mutations.
It's
not
really
one
in
real
way,
modern.
So,
like
some
some
functions,
some
stuff
is
not
really
one.
Some
stuff
need
to
be
updated.
A
A
We
don't
have
like
workspaces
or
other
things
that
complicates
things
especially
since,
like
both
packages
are
zero
dependency,
graphql
is
zero
dependency
and
the
way
depends
graphically
away
only
depends
on
graphql.
So
when
we
merge
them,
it
will
be
zero
dependency
and
it
will
be
also
possible
to
use
it
in
you
know.
A
A
C
A
A
Get
history
would
be
still
available
through
through
archived
repo,
so
it's
not
the
idea,
also
because
git
game
wouldn't
work,
but
out
of
box,
but
you'll
be
able
to
check
with
history.
So
I
think
one
one
thing
I
don't
want
to
book
6.00
release.
Basically
I
just
booked
you,
so
you
can
continue
with
attack
grip
conversion
because
it
will
take
some
time
things
that
need
to
be
done.
A
Okay,
no,
I
think
it's
like,
I
think,
maybe
it's
related
to
understand.
Okay,
thanks
so
did
I
answer
your
question.
C
C
A
So
yakov
it's
like
you
wanted
to
discuss.
I
think.
D
Yeah,
well,
I'm
not
sure.
Actually
it's
a
question
for
you
guys
and
for
rob
or
for
anyone.
I
I'm
not
sure
if
the
right
place
to
discuss
that
pr
is
this
working
group
or
the
like
the
spec
working
group,
basically
there's
a
bug
in
the
implementation
of
or
or
just
kind
of
bug.
D
You
know,
but
sort
of,
I'm
not
sure
if
this,
if
the
implementation
is
following
this
in
terms
of
whether
stream
payloads
are
allowed
to
be
sent
out
of
order
if
they
resolve
out
of
order
so
right
now
they
are
sent
out
of
order
if
they
resolve
out
of
order.
The
spec
is
not
a
hundred
percent
clear.
I
guess
it
depends
on
how
you
read
it
and
it
sounded
like
from
the
last
working
group
meeting
that
we
don't
want
that
to
happen,
but
the
spec
isn't
clear
and
the
implementation
lets
it
happen.
E
Way,
yeah,
I
think
the
intention
was
that
they
would
allow
be
allowed
to
be
out
of
order.
E
I
think
that
was
what
we
discussed
like
a
while
ago
when,
with
facebook-
I
I
don't
remember
100
but
but
yeah,
but
we-
I
didn't,
really
see
a
reason
like
why
they
should
we
should
hold
them
up
if
they're
ready
that
that's
kind
of
what
our
perspective
was,
I
mean
we
could
definitely
make.
We
definitely
should
make
it
more
clear
and
see.
D
Okay,
okay,
so
I
got
the
the
opposite
impression
from
the
working
group,
but
I'm
good
at
the
more
recent
one
though
just
the
one
I
attended
most
recently,
but
I'm
good
at
misunderstanding
things.
So
so
I
think
I
think
the
fix
then
is
is
not
a
code
fixed,
but
just
to
make
the
spec
more
clear
and
then
that's
that.
E
D
D
Okay,
well,
I
think
that
answers
the
question
from
this
end
and
I
guess
maybe
I'll
I'll
do
a
pull
request
on
the
spec
just
to
make
it
more
explicit
and
then
the
discussion
will
move
there,
but
I'm
happy
with
that.
That's
fine
with
me
thanks.
A
A
So,
if
you're
interested,
you
can
rewatch
it
and
maybe
found
when,
where
somebody
like
oppose
that
or
like,
I
really
want
basically
a
really
one
discussion.
I
just
wanted
to
mention
that
we
have
like
all
video
recordings
on
youtube
and
by
the
way,
if
you
need,
if
you
need
transcript,
I
have
access
to
zoom
transcript
for
working
group
meetings,
so
pin
pain
swag.
If
you
want
to
review
transcript
for
forward.
A
So
any
anybody
else
have
any
questions,
because
before
we
finish.
C
I
just
sent
an
issue
in
the
chat.
If
I
think
it's
something
you
have
to
answer
them,
maybe.
A
With
yakov,
in
a
sense,
so
in
other
issue,
another
pair
so
like
at
some
point,
we
need
to
add
middlewares
to
graphql
executor,
like
with
issue.
I
check
out
the
repo
that
people
posted
and
some
of
it.
I
don't
think
it
makes
sense,
a
lot
of
sense
that
you
include
in
graphql.js,
for
example,
like
wrapping
function
like
parse
or
validate.
A
A
A
But
what
cannot
be
done
outside
of
craftquest
is
the
performance
monitoring
of
execution
so
because
it
should
be
per
resolver
perfido
per
resolver.
So
I
think
we
need
at
some
point.
We
need
to
figure
out
a
mechanism
for
doing
that,
supporting
like
middlewares
other
thing
is:
go
into
160
zero
scope,
because
I
think
it's
actually
like
features
like.
That
is
the
reason
I
want
to
switch
to
unstable
muscle,
to
be
able
to
add
something,
get
the
feedback
from
people
and
iterate
on
it.
A
C
A
In
in
some
ways
like
it's
totally
a
reasonable
feature
request,
it's
even
reasonable
reasonable
for
a
reference
implementation
because
other
one,
which
is
also
implemented.
It's
not
like
something
javascript
specific
or
something.
So,
as
you
probably
know,
key
issue
here
is
like
graphql:
just
was
never
built
as
well.
That
was
never
built
as
production,
radio
library
for
like
execution,
because
facebook
used
php
backend
for
executing
graphql
requires
that's.
Why
graphql.js.
A
Consists
mostly
it's
like
a
code
that
was
used
in
graphical
or
some
internal
to
link.
That's
why,
like
we
missed
some
critical,
I
would
say
critical
features
for
production
to
be
production,
ready
kind
of
people
work
around
it
by
wrapping
resources
or
doing
other
things,
but
I
think
we
need
to
solve
it
on
the
core
but
like
to
be
clear.
It's
only
related
to
providing
middleware
inside
execution.