►
From YouTube: 2022-5-10 Node.js Technical Steering Committee 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
So
welcome
everybody.
I
am
joined
here
with
the
breast
good
chunk,
a
good
bunch
of
the
tsc
crowd
and
yeah.
Let
me
actually
do
a
tweet
about
it.
So
join
us
about
join
the
tsc
meeting,
see
meeting
on
youtube
here
we
go
yay,
perfect,
okay.
A
So,
let's
start
with
a
little
bit
of
the
announcements.
So
first
announcement
is
that
the
note
conf
you
call
for
proposals
is
now
open
on
session
eyes
and
also
early
bird
tickets
for
the
conference
are
on
sale.
If
you
are
willing
to
come
to
the
to
the
conference
and
not
speak,
of
course,
speakers
have
travel
and
accommodation
covered
so
yay,
please,
no,
no
collaborators
send
your
proposals.
You
almost
always
have
the
best
speakers.
A
I
also
want
to
congratulate
to
livia
menderos,
as
she
is
now
one
of
the
new
collaborators.
Congratulations,
okay,
and
for
the
new
to
be
to
be
part
of
us,
so
welcome
now,
let's
start
with
our
usual.
Does
anybody
have
any
other
announcements
to
make.
A
Cool
amazing,
let's
start
with
the
issues
attend.
So
the
first
issue
that
we're
going
to
cover
is
issue
number
43014.
A
They
revert
the
change
of
network
interface
family
from
string
to
integer
these.
It's.
A
Long,
this
change
is
actually
important
so
that
we
take
a
look
because
essentially
it's
a
breaking
change.
Okay,
because
we
do
that
we
december
major
change
on
node
18
that
is
being
caused
to
be
to
break
it
through
to
breaking,
and
now
we
are
wondering
if
it's
better,
if
we,
if
we
should
revert
there,
has
been
values
number
of
issues
opened
on
that
one.
I
actually
feared
when
those
some
major
changes
were
proposed,
that.
A
That
would
cause
the
problem.
Okay,
massive
breakage,
but
I
don't
know
canada
in
the
gold
mine
turned
out
fine.
So
I
don't
know
we
probably
need
to
add
a
few
more
modules
there
to
cover
it.
Okay,
so
I
don't
know:
is
there
anybody
that
have
a
good
opinion
on
that
topic?
Maybe
antoine
you
put
the
changes
in
so
you
might
want
to
talk
about
them.
The
revert
you
also
open
the
revert
so.
B
Yes,
the
original
change
was
about
to
harden
the
dns
function,
because
previously
it
would
accept
anything
for
the
family
and
we
wanted
to
throw
if
the
family
was
not
a
number,
but
that
caused
some
incompatibilities
with
the
net
modules,
which
turns
out
has
unforeseen
consequences
and
the
revert
is
also
breaking
so
it
has
landed
in
18.00
and
18.0.
B
So
we
can
guess
that
not
many
forks
have
a
depend
on
the
new
behavior,
but
still
it's
it's
a
breaking
change
and
yeah.
The
community
feedback
so
far
seems
to
to
be
on
the
side
of
we
should
revert.
B
C
I
have
a
question
yeah:
is
it
the
whole
pr
or
it
so
most
of
the
complaints
are
about
the
the
family
attribute
on
the
network.
Return
network
interface
object
and
I
wasn't
it
wasn't
clear
to
me
from
the
pr
whether
that
was
just
a
small
part
of
the
pr
or
whether
it's
all
so,
intrinsically
linked
that
we
would
have
to
revert
the
entire
change.
C
So
I
understand
the
deprecation
was
for
setting
that
value
as
if
you
were
passing
it
into
like
the
the
options
object,
but
it
seems
to
have
had
the
whole
pr
seems
to
have
also
affected
that
return,
that
the
type
of
that
field
in
that
object-
and
that
seems
to
be
the
one
that's
caused
the
break
in
the
in
the
ecosystem
and
I'm
guessing.
The
question
is:
is
there?
C
Is
it
too
complicated,
for
example,
to
keep
most
of
the
most
of
the
change
but
to
to
to
change
to
to
restore
that
particular
bit?
You
know
that
that
family
attribute
on
the
network
interface
is
object
back
to
us
to
the
string
that
it
was
before,
or
is
it
so,
inter
intrinsically
linked
with
the
intel?
You
know
the
entirety
of
the
pr
that
it
is
and
either
or
revert
everything
or
you
know,
keep
it
as
it
is
in
18.
At
the
moment.
B
So
yeah,
I
have
a
pr
open
on
that.
I'm
not
sure
if
you
referring
to
that.
So
what
this
pr
does.
It
reverts
the
change
of
like
from
from
an
integer
back
to
a
string,
and
it
changes
also
the
dns
that
look
up
to
accept
the
string,
ipv4
or
ipv6.
D
B
Folks
are
trying
to
just
make
an
eq
quality
comparison
to
see
if
it's
ipv4,
v6
or
they're,
using
dot
slice
or
string
yeah
string
related
methods.
B
So
yeah
I've
tried
to
keep
the
the
change
as
minimal
as
it
can
be.
Okay,.
A
I
have
not
captured
all
of
that
in
the
minutes,
so
if
you
folks
can
richard
and
write
down
your
your
paragraph
and
stuff,
that
would
be
great
because
I
I
have
not
okay,
so
I
have
totally
not
okay,
cool
okay.
I
don't
think
there
is
any
problem
with
this.
Okay,
so
is
there
any
decisions
that
we
need
to
make
apart
from
that,
should
we
so
should
we
bring
it
to
github
and
just
land
it
normally
or
what
exact
question
and
one.
A
I
don't
think
we
need
a
vote,
we
just
need,
but
we
don't
have
quorum
here
today.
So
yeah
I'll
keep
it,
but
we
need
to
move
fast.
To
be
honest,
so
I'm
going
just
to
post
it
at
today's
meetings.
A
A
A
Yeah
cool
okay.
So
let's
move
to
the
next
topic,
which
is
auto
close
old
stale
pr.
A
Opened
by
michael
I'm
not
familiar
with
this
issue,
so.
A
That's
the
question
and
sorry
older
than
one
year
no
comments
in
the
last
six
months,
close
and
then
pr
older
than
when
you're
not
commenting
last
five
months,
yeah
comment
and
then
close
yeah
is
there
anybody
that
has.
A
There
is
some
of
one
objection
from
jordan
about
it,
so
I
don't
know
if,
if
you
can
just
want
to
chime
in
on
github
and
if
you
just
want
to
say
something
about
it,
otherwise
we
can
move
to
the
next
one.
I
don't
see
any.
A
Okay,
so
the
next
one
is
issue
428104,
which
is
they
make
it
possible
to
fetch
with
proxy
and
other
agents.
Okay,.
A
This
is
an
open
question.
To
be
honest,
we
came
to
a
resolution
to
this,
so
I
think
I
can
just
remove
the
tsc
agenda
tag
the
fixes
I've
landed
for
this
issue
as
landed
in
undigi.
A
They
will
probably
require
some
documentation
work,
but
the
the
feature
is:
is
there
so
people
could
essentially
use
custom
dispatchers,
custom-managed
dispatchers
with
fetch,
so
basically
they
can
install
undigi
separately
and
that
will
bring
it
up.
A
So
yeah
I
will
open
up,
I
will
end
it.
I
will
do
a
mundus
release
and
then
maybe
michael
can
you
do
they
not
pr.
A
It
amazing
yeah,
sorry
I
didn't.
I
didn't
know
that
we
had
so
fixes
landed
on.
A
A
Before
graduating
it
took
to
support
it.
A
Michael
so
today
to
know
js.
A
A
Okay,
next
one
antoine
you,
it's
the
vote
on
par
sharks,
so.
B
A
B
Results,
so
not
everyone
has
voted
yet
some
people
on
the
call
have
not
just
a
vote.
So
if
you
want
to
to
do
that,
that'd
be
great.
I
think
that
just
like
half
of
the
tfc
that
has
voted,
which
means
we
can
close
it
but
yeah.
If
you
want
to
vote
please
please.
B
Yeah
pasting
the
link
to
the
pr.
B
I
want
to
take
care
of
that.
It's
it's
a
really
quick
watch.
You
just
have
to
to
decide
whether
we
want
this
api
to
name
the
kill
or
what
is
process.
D
Cool
thank.
A
D
B
Yeah,
okay,
like
that,
if
you
click
on
the
the
first
commit
you
you
can,
you
can
read
them,
but
yeah,
because
it
would
be
easier
in
the
erp.
A
C
C
Yeah
so
michael
and
I've
been
knocking
out
some
of
the
build
repositories,
so
we've
done
we
did.
The
unofficial
builds
repository
this
week.
No
last
week
week
just
gone.
We
are
planning
to
do
the
build
repository
next
week
after
the
next
set
of
releases
that
are
due
out.
I
think
it
should
be
okay,
but
I'm
you
know,
I'm
almost
always
a
bit
bit
cautious
when
it
comes
to
changing
something
and
build,
but
yeah
that
that
is
a
bit
of
update
for
that
one.
C
The
unofficial
builds
node.js,
that's
where
we
shunted
some
of
the
platforms
that
we
don't
regularly
test
build
in
the
in
releases.
So
things
like
the
point
of
compression
build
is
an
unofficial,
build
and
the
moosel
based
build.
Is
there.
D
D
C
C
So
I
don't
know,
maybe
maybe
I
need
to
create
like
a
test
repo
and
try
out
there
or
something.
But
yes,
there
are
branch
protection
rules
that
you
can
now
enable,
but
it
seems
to
sort
of
suggest
that
if
you
were
an
admin,
you
could
bypass
that
rule,
and
I
don't
know
if
you
get
a
warning
or
a
question
mark
if
you
try
it
or
whether
it
just
allows
you
to
do
it.
C
I
didn't
include
anyone
in
the
rule,
it's
just
when
you
look
at
it.
It
tells
you
that
people
with
that.
C
C
E
C
May
then
may
still
be
a
small.
You
know
it
may
not
prevent
a
tsa
member,
for
example,
from
accidentally
pushing.
C
I
think
that's
probably
what
we
should
do,
but
yeah
I
I
it
is
good.
You
know
we
should
enable
that
when
we
do
renames
anyway,
but
just
a
sort
of
warning
that
it
may
it
may
or
may
not.
I
guess
one
of
us
will
test
it
afterwards.
It
may
or
may
not
sort
of
prevent
owners
from
pushing.
A
A
Amazing,
okay,
so
now
we
are
back
to
everybody's
favorite
topic,
which
is
the
open,
ssl
1.1.1
end
of
life
date.
A
And
I
know
it's
everybody's
favorite
topic,
because
you
know
it's
it's
what
it
is.
Okay,
it's
a
it's
just
something
that
nobody
wants
to
deal
with,
but
we
have
to
so
I
don't
think
there
is
much
to
I
don't
know
richard
want
to
say
anything.
I
think
we
can
just.
C
Update
I'd
like
to
sort
of
hunt
the
decision
by
another
week,
I'm
having
some
talks
inside
red
hat
too,
to
explore
a
possibility,
it's
not
a
firm
commitment
or
anything,
but
there's
a
possibility
that
perhaps
there
might
be
an
avenue
where
we
would
be
able
to
pick
up
say,
for
example,
the
openness.
So
that's
maintained
within
red
red
hat
red
hat
eight.
C
C
This
is
the
actual
sort
of
official
supported
statement
of
that.
So
it's
supported
for
rel,
but
it's
just
just
it's
just
to
see.
If
we
have
another
option
before
pulling
that
that
trigger,
we
have
had
a
bit
of
pushback
regarding
moving
the
dates
of
an
lts
release,
but
yeah.
It
is
what
it
is.
I'd
like
to
just
push
the
decision
back
by
week,
but.
A
So
and
he's
exploring
if
red
hat
would
be
willing.
C
A
A
It's
the
dates
are
that
a
call
for
nomination
is
may
9
call
for
nomination.
Currently
there
are
two
people
that
have
been
nominated,
that
are
that
nominate
themselves,
so
michael
dawson
is
running
again
as
tsc
chair,
I
put
my
name
in
in
the
hat
for
the
vice
chair
position,
so
we
still
have
four
more
days
for
more
dominations.
Please
apply
so
somebody
else
can
do
the
vice
chair
role,
please
so
at
least
we
have.
We
can
vote
on
something.
If
not
it's
it's.
A
It's
also
fine,
but
I
don't
know
it's
just
weird
to
call
for
votes
if
you're
running
so
anyway,
I'm
flagging
the
fact
that
we
have
a
vote
coming-
and
I
don't
know-
I
don't
think,
there's
anything
else
to
say
here
so.
A
No
okay,
great,
let's
move
to
v8
currency!
So
michael,
would
you
like
to
say
something
about.
D
A
D
D
That
gives
us
time
to
to
include
it,
and
another
thing
about
this
version
is
that,
according
to
the
chromium
schedule,
which
I
I
pasted
in
in
the
notes,
it
will
be
the
version
of
v8
present
in
an
lts
release
of
chrome
os,
and
I
don't
really
know
about
how
chrome
os
is
managed,
and
especially
it's
lts
support.
C
I
I
I
think
it
sounds
like
a
good
idea
in
the
absence
of
more
information
about
what
lts
support
for
chromos
means.
D
Yeah,
I
don't
really
know
what
it
means,
but
as
a
as
one
of
the
people
who
have
access
to
some
security
reports
for
v8,
I
see
sometimes
even
in
the
last
week,
I
saw
messages
about
about
version
9.6,
which
is
the
previous
lts
of
chrome
os.
So
they
definitely
maintain
it.
C
Yeah,
so
it
sounds
sounds
like
a
promising
from
some
idea
to
pursue.
Then
I
mean
you
know
with
a
release
hat
on
less
less
v8
versions.
We
update
to
in
node
18,
probably
the
easier
it
will
be
to
to
maintain.
C
I
know
you
do
a
really
really
good
job
of
keeping
v8
updated
in
all
the
release
lines,
but
yeah.
You
know
it
would
avoid
things
like
having
to
work
out
abi
compatibility
for
new
v8
versions,.
C
Yeah,
so
I
you
know,
I'm
yeah,
I
I'm
in
favor
of
you
know
of
keeping
the
version
v8
to
that
to
that
version
when
it's
available
in
in
node
18.
C
It
would
be
good,
not
not
not
not
blocking,
but
it
it
might
be
nice
to
try
and
find
more
information
about
chrome,
os's,
long-term
strategy,
whether
we
can
maybe
align
future
releases
on
particular
v8
versions
or
something.
C
A
D
A
A
D
Know
when
it
becomes
lts
in
october,
I
think
yeah
yeah,
it's
more
than
two
versions.
It's
five
versions.
A
D
But
then
to
be
clear,
the
next
version
that
will
be
in
an
lts
chrome
os
is
is
going
to
be
released
in
end
of
november
and
yeah.
It
might
be
very
difficult
to
get
to
that
version.
In
note,
18.
D
We
could
stop
at
any
point
like
we
did
before.
E
D
F
I
think
the
the
argument
for
blocking
locking
the
node
version
together
with
a
va
lt
s
version
is
that
if
there
is
another
bug,
a
security
bug
introduced
in
the
updated
va
version
that
is
not
lts,
then
they
want
back
port
security
fixes
to
that
branch,
and
then
we'll
have
to
do
that
by
ourselves,
and
it
will
be,
you
know,
more
work,
so
it's
probably
better
to
just
keep
it
in
sync
with
one
via
lts,
and
they
will
do
the
security
fixes
back
ports
and
we'll
just
need
to
keep
inside.
D
F
F
Yeah,
so
he
basically
has
like
three
months
as
a
long-term
support
candidate
and
then
another
six
months
as
I
long
to
support
the
versions
of
10
months.
I
think.
A
So
10
months,
okay
yeah-
I
like
it,
please
still
leaves
us
significantly
without
patches,
as
I
don't
think,
this
is
improved
too
much
compared
to
the
current
situation.
To
be
honest,
like
we
have
a
few
months
more
support,
but
not
it
does
not
cover
for
the
full
lts
line.
A
D
A
A
A
F
Okay,
so
we're
currently
moving
more
into
the
embedded
snapshot
for
workers
and
probably
will
do
some
work
for
the
vm
contest
as
well,
but
we'll
need
to
investigate
investigate
whether
that
will
break
the
global
proxies
for
vm
contexts,
and
I
did
some
investigation
into
the
memory.
Corruption
bug
in
newsline
snapshots
and
it
looks
like
it's
unsafe
to
make
the
code
cache
generator
separated
from
the
snapshot
generator.
F
So
the
first
step
is
to
merge
it
to
so
nk
cache
is
going
away
and
there
is
a
workaround
for
the
current
memory.
Corruption,
reproductions,
that's
going
to
be
a
way
patch
and
also
we'll
need
to
work
on
a
more
complete
fix
to
guarantee
consistent
rate
only
space
layout
in
va.
So
basically
before
that,
it's
possible
that
adding
more
stuff
to
the
snapshot
could
cause
a
memory,
corruption
and
right
now,
there's
no
other
way
to
know
that
other
than
doing
some
checks
into
the
snapshot
contents
by
ourselves.
A
A
A
No,
we
think
you're
good.
Thank
you.
Thank
you
very
much,
joey.
Okay,
so
I
think
that's
all!
For
today
I
have
a
light
tiny
bit
of
stuff
to
talk
about
in
the
security
about
to
talk
about
in
the
private
section
and
one
there's
a
question.
B
Yeah,
can
we
can
we
maybe
go
back
to
the
the
first
item
of
the
list
now
that
I
think
we
have
quorum.
A
B
B
Yeah
revert
the
network
interfaces;
okay.
No,
we
have.
A
B
Didn't
so
do
do,
do
you
guys
maybe
need
context
or
sorry
for
putting
you
on
the
spot
like
this,
so
we
have
a
breaking
change
that
landed
on
not
the
18
and
a
few
folks
have
complained
that
it
was
too
breaking
now
we're
looking
if
we
want
to
revert
it
in
the
sembler
minor
and
we
we
need
the
tsc
to
to
decide
on
that.
B
To
run
this
in
the
meeting,
I
think
it's
because
there's
a
release
schedule
for
today,
maybe
yeah.
A
The
there
is
a
little
bit
of
urgency
anatoly
as
because
we
have
the.
A
A
If
can.
B
And
beth:
do
you
know
if
it
will
be
like
when
does
it
need
to
land
for
to
be
on
time
for
the
release.
H
I
I'm
working
with
raphael
on
this
release
and
we
can
push
it
like
next
tuesday
and
say
that
I
think
we
also
need
the
open
ssl
to
land
so
anyway.
So.
B
A
Okay,
beth
I'm
going
to
tag
to
put
the
label
on
this
pr
as
sambar
miner
and
with
don't
do
not
back
part.
Otherwise,
if
we
keep
somewhere
major,
I
think
it's
going
to
break
a
lot
of
our
automation.
H
Yep,
okay,
we
should
also
add
notable
change,
because
we
probably
want
to
explain
what
we've
done,
though,.
A
Sorry
nikita
says,
is
I
see
ten
plus
one?
Is
there
anybody
that
has
not
approved
this.
A
Perfect
great,
thank
you
everybody.
I
just
want
to
remember
that
you
can
follow
subscribe
to
the
node.js
project
calendar
so
that
will
include
you
can
get
notified
about
these
meetings
and
perfect
and
that's
it
guys,
and
you
know,
bye,
bye
and
thanks
for
tuning
in
please
everybody
stays
on
for
a
little
bit
of
our
private
section.