►
From YouTube: 2023-08-24 - Node.js Release Working Group Meeting
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
Okay,
so
we
are
live.
Welcome
everybody
to
the
node.js
release
working
group
meeting
before
I
start.
Does
anybody
have
any
announcement.
A
A
Yeah,
okay,
I
will
move
to
the
to
the
issue
880
session
at
the
collaberto
summit.
I
think
we
have
discussed
it
already
and
I
have
mentioned
that
I
won't
be
able
to
I,
don't
think
I
will
be
able
to
go
to
the
collaborative
Summit
because
it's
in
the
same
day
as
the
open,
ssf.
B
I,
don't
think
Roy
was
going
to
make
it
so
and
I
don't
I'm,
not
aware
that
any
of
us
are
not
going
either
so
I
I
think
we
did
talk
about
it
in
one
of
the
earlier
meetings
and
it
was
kind
of
if,
if
no
one's
actually
going
to
be
there,
then
there
isn't
really
much
that
you
know
that
we
can
run
because
I
think
also,
even
if
you're
able
to
go
remotely
I
think
I
think
Royce.
The
remote
is
only
one
way.
So
it's
only
like
from
the
conference
out.
B
Not
you
know,
you
can't
speak
if
you're
reviewing
remotely.
So
you
can't
even
run
the
session
remote,
so
yeah
I
kind
of
feel
that
one's
probably
probably
too
late
now
and
not
going
to
happen.
A
Okay,
that
makes
sense.
I
I.
Remember
that,
looking
to
the
thread
that
we
have
discussed
it
and
we
might,
we
once
sent
a
session
so
I
have
removed
the
label
and
I.
Think
I
can
close
this
issue.
A
We
have
the
20.6
proposal
and
this
is
stuck
because
we
have
some
machine
failures,
plus
a
flaky
test
on
windows.
So
I
sent
a
message
on
the
release
channel
on
in
the
openjs,
telling
all
the
all
the
issues
so
I,
don't
know.
Juan
sent
a
message
to
me
to
me
saying
that
he
can
help.
Oh,
this
is
to
sign
and
promote
this
release
because
I
can't,
but
we
need
to
get
to
the
green
CI
today,
which
I
think
it's
possible,
but
that
would
be
difficult.
I
guess.
C
It
will
be
a
little
bit
challenging
yeah
I
will
I
will
talk
with
Khan,
because
I
assume
that
we
need
to
thanks
the
the
things.
Look
right,
I
need
to
replace
your
name
for
the
one
one
then
so
I
will
need
to
I'm
in
the
comments.
So
I
will
trigger
everything
from
zero,
so
crossing
fingers.
Maybe
that
will
go
in
a
green
light.
C
A
Okay,
so
yeah,
if
you,
if
you
can't
do
it
today
or
the
machine,
is
not
ready,
I
suggest
to
Target
next
Monday,
because
Friday
is
quite
complicated.
Yep.
A
All
right
so
I
have
you
updated
the
the
schedule
so
I
will
I'll
leave
my
name
there
until
you
release
with
fujuan
and
then
you
feel
free
to
play
there.
So
I
replaced
all
the
all
the
list
again.
So
the
next
release
will
be.
A
I
think
we'll
be
September
1st
right,
because
I
think
we
have
a
bunch
of
things
on
on
the
Kiwi,
even
for
20
released.
C
C
A
C
A
A
I
can't
add
my
name
to
20.7
I.
Think
I
will
only
be
able
to
do
it
on
October
Maybe.
A
So
if
someone
wants
to
volunteer
to
20
release
update
the
list
all
right,
so
let's
go
to
the
node.js
18.
Let
me
see
I
think
Hui
was
doing
the
release
there,
but
he's
all
also
stuck.
Let
me
take
a
look.
B
He
did
post
this
like
that.
He
was
thinking
of
delay
into
next
week.
Okay,
I
think
there
is
a
proposal
open,
yeah.
A
Actually,
there's
one
thing
that
he
he
was
talking
to
me:
it's
because
when
we
release
a
security
release,
we
land
things
on
on
Main
as
well,
and
we
usually
learn
the
same
thing
that
on
20
to
the
main,
but
we
don't
add
the
labels,
don't
land
on
or
on
2016
or
18.,
and
so
the
chair
pick.
So
the
the
comment
that
we
use
to
chair
pick:
will
you
get
those
pull
requests?
A
So
we
need
to
remember
that
when
whenever
we
create
a
secret
release,
we
add
the
label,
don't
land
on
16,
18
or
either
20,
so
those
conflicts
will
never
happen.
I
guess.
A
Okay,
so
let
me
take
a
look
on
the
schedule
for
18.
We
have
Hui
doing
it
for
this
month,
I
guess
and
next
month.
We'll
need
another
volunteer.
A
Well,
that's
it!
Actually,
we
need
more
releases,
I
guess.
B
A
Don't
worry,
yeah
well
one
thing
we,
we
should
Target.
Also
it's
good
to
talk
to
you
all.
This
is
that
there
is
one
initiative
that
we
are
trying
to
move
forward.
A
That
is
the
automation
of
all
the
all
the
release,
so
I've
been
working
with
this,
so,
for
instance,
git
node
works
to
create
a
proposal
nowadays,
but
there
are
still
some
educates
that
we
need
to
handle
and
we
also
need
to
find
a
way
to
promote
and
sign
and
release
automatically
without
the
need
of
my
machine
or
your
machine.
A
You
know
we
need
to
discuss
this
actually,
but
the
idea
is
to
have
an
automation
for
all
of
it,
so
we
don't
need
to
have
to
to
get
people
time
to
do
what
to
do
it.
Yep
this.
A
A
C
Yeah
I
will
I
would
love
to
get
involved
here
because
definitely
well,
it's
not
in
this
agenda,
but
maybe
we
should
include
it
there
is.
There
is
the
need
also
to
change
the
notarization
for
Mac
OS
I
think
we
have
a
hard
deadline
somewhere
in
November,
so
yeah
I
need
to
check
around
that.
C
If
the,
if
the
bus
script,
that
I
did
and
didn't,
try
it
at
all
and
is
there
in
a
pull
request,
it's
going
to
work
so
yeah
I,
potentially
will
have
some
time
in
September
to
check
around
in
October,
going
to
be
like
a
little
bit
more
on
holiday,
so
potentially
going
to
be
like
more
or
less
one
week
available
in
the
full
month,
but
in
September
I
kind
of
okay,
so
yeah
yeah
I'm,
happy
I'm
happy
to
engage
with
the
automation,
so
everything
is
new
for
me.
C
I
think
it's
going
to
be
easier
to
get
involved
in
the
automation,
also
learning
the
process
so.
B
On
another,
what
I've
just
remembered
Stuart
Stuart
Addison?
He
is
on
the
list
of
potential
releases
a
long
time
ago,
and
he
did
talk
to
me
this
week
and
say
that
he's
looking
to
maybe
do
what
you
you're
doing,
Ulysses
and
yeah,
but
it
is
going
to
be
away
for
half
of
September,
so
I
probably
wouldn't
be.
B
You
know,
probably
won't
be
the
releases
we're
talking
about
right
now,
but
that
that
might
be
something
to
you
know,
maybe
or
a
October
release
or
something
I'll
I'll
have
further
discussions
with
him
and
see
when
he's
actually
going
to
be
available.
But
that
might
be
you
know
something
we
can
slotted
into
into
what
other
20
or
an
18
release
later
in.
A
B
B
C
B
A
All
right
so
I
think
he
can
make
it
now
yeah.
So
now,
let's
go
to
the
issue.
Six
five,
eight
16
maintenance,
I
think
Richard
yeah
that
one.
B
Thing
so
I've
posted
a
comment
right
at
the
bottom
of
that
issue,
which
is
listing
the
current.
It's
listing
the
states
of
the
pull
requests
and
things
labeled
LTS
watch
16.,
so
it's
basically
all
the
stuff
remaining.
That's
in
the
the
call
repo
that's
tagged
with
node.j
16.
B
B
B
A
B
Exactly
there's
a
core
pack
update,
PR
compact
is
experimental,
so
the
last
version
of
cool
pack
in
node
16
is
0
17
and
the
pr
has
now
been
updated
to
zero
19.
So
there's
been
a
couple
of
releases
since,
but
it's
all
self-contained
in
that
it's
a
complete
npm
thing
like
a
so.
B
B
So
the
LTS
watch
ones.
There
are
a
couple,
one
is
labeled,
but
you
can
for
LTS
and
I
really
don't
know
if
I
can
make
that
decision
as
to
whether
it's
boat
for
long
enough
or
whether
it's
still
considered
risky
and
then
there's
a
few
that.
B
I
I
I
would
call
it
trivial,
but
not
necessarily,
you
know
to
go
out
and
release.
There
is
one
that
will
someone
asks
for
consideration,
the
one
about
validating
arguments
which
are
process.
Someone
said
this
will
prevent
security
issues
without
sort
of
being
specific
as
to
what
I
guess
it's,
because
it's
the
validation
of
users,
Supply
parameters
or
well
I,
don't
even
know
if
it's
useless
quiet,
it's
the
application,
Supply
parameters,
but
yeah
I
mean
I.
Don't
again,
my
main
thing
here
is
the
only
things
that
I
really
saw
in
the
list.
B
That
I
would
consider
notable
for
a
16
release
was
updating
the
root
certificates.
So
that's
for,
like
the
TLs
connections
from
node,
the
last
time
we
updated
the
root
certificates
was
around
the
turn
of
the
Year
sort
of
just
at
the
end
of
last
year.
So
there
have
been
two
updates
since
so
those
two
commits
Cherry
Picked
cleaning
I've,
actually
landed
them
on
the
staging
Branch.
They
removed
three
root
certificates
and
add
two
new
ones:
well,
I'll
send
you
one
they've
had
to
to
and
remove
three.
B
So
it's
just
a
sort
of
General
tidy
up
thing.
So
again,
I
guess
it
might
give
a
little
bit
more.
Longevity
of
people
are
stuck
in
16,
but
again
it's
possibly
not
necessarily
absolutely
necessary.
One
of
the
things
I
suggested
is:
maybe
we
can
just
land
things
on
the
staging
branch
and
not
do
a
release.
So
people
to
want
to
you
know,
build
their
own
16.
They
can
and
the
things
will
be
there.
B
Otherwise,
my
window
of
opportunity
for
doing
the
releases
next
week,
and
that's
it
so
the
week
after
that,
I'm
going
to
be
busy
I'm
actually
going
to
be
sort
of
traveling
into
like
I,
don't
live
very
far
from
London,
but
I'm
going
to
be
traveling
into
there
for
sort
of
company
meetings.
So
I'm
not
going
to
be
able
to
do
anything
that
week
in
terms
of
releases
and
then
the
week
after
that
16
is
end
of
life,
so
basically
I
think
for
16
we
either
don't
release
which
I
think
is
my
preferred
option.
B
If
I
do
a
quick
release
for
next
week
and
hope
we
don't
have
to
re-spin
or
yeah,
probably
just
let
things
on
the
staging
and
say
right-
we're
not
going
to
do
a
release.
But
if
you
want
to
build
it
yourself,
you
know:
we've
updated
the
source.
There
are
a
few
things
that
were
on
staging
to
do
with
like
test
fixes
and
making
the
CI
pass.
B
So
again,
a
few
things
that
are
already
on
staging
but
I,
don't
think
they're
necessary
in
terms
of
they
would
affect
like
end
users
but
other
than
the
roots
typical
update,
so
that
I
landed.
You
know.
A
Okay,
there
is
one
just
one
thing
that
I
didn't
consider:
well
those
pull
requests
with
LTS
watch.
Yes,
we
normally
add
in
when
we
want
to
sum
that
we
believe
that
something
needs
to
be
landed
in
a
prior
release.
But
if
we
don't
add
that
label
that
should
be
automatically
shared
peaked
right.
B
Pending
for
active,
probably
but
for
maintenance-
probably
not
okay,
yeah
just
for
a
minute-
that's
a
professor
and
I
guess:
it's
kind
of
a
it's
more.
A
kind
of
the
things
that
are
labeled
are
things
that
people
think
really
ought
to
be
considered,
whereas
the
other
things
might
automatically
be
considered,
but
then,
when
you
might
make
the
decision
that
okay,
it's
not
yeah,
it's
not
that
important.
If
that
doesn't
land,
whereas
I
guess
the
watch
ones
are
more
kind
of.
B
We
think
this
is
a
demonstratable
issue,
or
this
is
something
not
not
always
I
mean
it's
up
to
collaborators
type
labels,
but
I
guess
it
does
mean
that,
like,
for
example,
for
me
to
today,
going
through
those
later
pull
requests,
I
could
just
look
at
the
things
of
the
labels.
Instead
of
you
know,
I'm
not
running,
not.
B
B
In
you
know
anyone
here
or
listening
to
the
call
please
chime
up,
because
the
opportunity
to
do
another
16
releases,
very
small
and
I
guess
the
preference
of
our
team
given
availability
is
you
know
we
wouldn't
do
one
unless
we
really
needed
to
and
yeah
that's
correct.
It
doesn't
sound
to
me
like
we
do
so
I'm
inclined
to
you
know
just
let
16
go
end
of
life
in
a
couple
of
weeks
time,
not
very
long
now,
so
hopefully
there
will
be
nothing
else.
New
coming
up.
B
That
would
also
hand
to
to
do
a
release.
Yeah.
A
And
also
considering
the
last
security
release,
I
think
that
should
be
fine
to
not
release
a
final
release
of
16..
Well,
okay!
So
if,
if
you
are
watching
this,
this
this
record
or
live
and
want
something
to
land,
please
chime
in
in
the
issue
and
tell
us
the
reason
for.
B
That
and
it's
got
to
be
something
that's
either
really
serious
in
terms
of
yeah
or
you
know,
like
a
yeah
something
low
risk
in
terms
of
co-oplexity.
A
Okay,
so
just
to
finalize
or
or
agenda,
we
have
an
issue
on
decim
to
clean
up
modules
that
are
failing.
Often
I
have
created
this
issue,
but
I
don't
have
any
update
so
far,
I
didn't
have
time
to
take
a
look
on
this,
but
my
my
initial
thought
was
okay:
let's
remove
all
the
modules
that
are
failing,
constantly
and
open
an
issue
paying
all
the
Alters.
If
they
reply
to
it,
we
add
them
back.
A
B
Always
been
a
challenge
trying
to
get
that
to
try
and
keep
it
in
some
reasonable
State,
whatever
that
reasonable
stays,
it's
never
completely
passed
anyway.
As
far
as
I
know
and
I
think
the
biggest
problem
we've
got
is
flakes
where
things
are
failing
in
modules,
but
not
either
not
consistently
or
it
fails,
but
on
different
platforms
different
you
know.
Sometimes
it
fails
on
this
platform
and
then
sometimes
it
fails
on
a
different
platform
and
the
other
problem
is
they're.
All
you
know
all
those
modules
are
running
different
test
Runners.
B
So
if
you
even
try
and
delve
into,
why
is
this
module?
Failed?
It's
not
always
easy
to
spot
the
failure
because
they
use
like
different
test
Runners,
that
spit
out
results
in
different
formats
and
it's
kind
of
I.
Think
in
some
cases
I've
seen
sit
Jim
truncate
the
standard
output
or
something
which
is
really
weird
but
yeah,
I,
I,
guess
yeah
people
have
time
to
yeah.
Well,.
C
B
Is
a
there
is
a
list
somewhere
in
situation
which
says
if
you
are
out
in
a
module,
it
has
a
whole
list
of
criteria
that
a
module
should
satisfy
in
order
to
be
added.
I
think
the
Test
passing
was
one
of
them,
it's
kind
of
implicit
but
yeah.
If
you,
if
you
look
in
the
city
and
repo,
there
is
a
file
somewhere
that
can
be
referenced
to
or
people
wanting
to
add
new
modules.
B
I
guess
the
only
question
is
if
we
start
disabling
well,
I
guess
if
the
module's
always
playing
it's
one
thing
I
suppose
the
other
thing
is:
if
we
do
blanket
disable
a
whole
bunch
of
modules,
I
guess
more,
if
they're
flaky,
rather
than
you
know,
always
failing
what
is
that
actually
doing
in
terms
of
coverage
I
mean?
Would
that
mean
that
some
parts
of
node
don't
have
any
exposure
in
sit
Jim
to
yeah
in
terms
of
ecosystem
testing,
Yeah.
A
B
Sometimes
sometimes
sit,
Jim
has
found
things
and
then
a
lot
of
the
other
time
we're
just
trying
to
cross
through
and
decipher
whether
it
has
found
anything
or
not,
but
I
think
usually
when
it
finds
something
it's
fairly
across
the
board.
Where,
like
you,
suddenly
see
a
new
module,
that's
failing
In
the
comparison
on
multiple
platforms
rather
than
a
single
one,
but
yeah
it's
one
of
those
things.
That's
going
to
be
really
hard
to
to
automate
unless
we're
able
to
get
to
a
state
where
it's
passing
and
then
it
can.
B
You
know,
find
the
regressions
which
are
this.
This
has
now
failed
when
it
wasn't
phone
before
we're
doing
the
current
comparisons.
It's
going
to
be
really
hard
to
automate
because
of
the
way
you
almost
have
to
sort
of
intuitively
look
at
it
and
it
sort
of
make
us
judgment
callers
to
yeah.
But
that's
a
that's
a
knowing
failure.
I've
seen
that
fell
before
or.
B
Maybe
okay,
okay,
yeah.