►
From YouTube: 2020-04-27-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
to
the
node.js
technical
steering
committee
meeting
for
april
27
2022.
before
we
get
started,
does
anybody
have
any
announcements
that
they'd
like
to
share.
A
A
A
So
I
think
mateo
added
it
to
the
agenda.
The
comment
was
like:
did,
we
add?
Did
we
expose
fetch
too
early.
A
B
I
don't
know
the
answer
to
the
question,
but
the
context
I
hate
I
got
from
that
was
the
ques.
I
think
the
question
was
referring
to
whether
we
unflagged
it
too
early,
so
it
is
experimental,
but
it's
it's
visible
by
default
and
I
think
that
has
broken
some
packages
and
I
guess
the
question
is
whether
that
is.
A
Because
the
global
over
lands
on
top
of
other
things,.
C
Yes,
it's
smaller
other
polyfills
would
replace
fetch
if
it
didn't
exist
and
those
polyfills
provide
non-standard
features
like
they
support
proxies,
for
example,
since
they
are
based
on
those
http
implementation,
but
now
that
there
is
a
global
fetch,
the
polyfills
don't
fill
in
anymore,
so
we
are
using
nodesfetch
now
and
the
built-in
fetch
which
doesn't
have
the
non-standard
features,
because
our
goal
was
to
have
a
standard
standard,
compliant
implementation.
So
now
the
polyfills
aren't
filling
in
with
a
non-standard
compliant
implementation,
and
now
people
are
requesting
non-standard
features
from
nodes.
C
C
A
A
A
A
A
Okay
thanks
the
next
one
is
rename
default
branch
for
master
domain.
I
don't
think
there's
any
update
on
that
this
week.
Unless
anybody
else
has
one.
A
B
I
guess
with
this
one:
it's
watch
this
space
next
time,
yep
yep,.
A
Okay,
the
next
one
is
util
at
parsearg's
module.
This
is
four
two,
six
seven
five,
so
I
think
darshan
you'd
mention
this
one.
Do
you
wanna
give
some
context.
E
Sure
so
the
parsers
module
from
pkgs
was
added
to
node
in
a
pr
like
it's
still
in
a
pr
stage,
and
we
were
having
some
conversation
on
whether
the
process
object
or
the
util
module
is
the
best
home
for
this
api.
E
So
like
we
like,
currently
the
pr
tries
to
expose
this
from
the
util
module,
but
then
jordan
harban
he.
He
mentioned
that
most
of
the
things
that
we
have
in
this
in
the
util
module
are
shimmable
in
browsers,
so
something
like
parsearts,
which
is
completely
related
to
the
process.
It
doesn't
really
make
sense
in
the
browser,
so
maybe
exposing
this
from
the
process
object
makes
more
sense
so
mateo.
E
E
So
then
we
had
more
discussion
and
the
thing
is
that
it
kind
of
is
the
same
thing
from
util.
Also
because
if
you
expose
an
object
from
util,
you
can
manipulate
that
in
some
module
and
then
those
side
effects
will
be
shown
in
other
places
where
you
require
it.
A
F
I
think
we
were
in
a
similar
situation
when
the
report
was
added
to
the
core
in
terms
of
it
should
go
to
the
process
or
util,
and
eventually
we
settled
on
the
process,
but
from
the
look
of
it
looks
like
this.
Should
this
is
more
fit
for
the
util
module,
because
essentially
we
are
talking
about
parsing
stuff.
E
Modules,
but
if
you
place
it
in
the
util
module,
then
it
doesn't
really
fit
in.
If
you
look
at
the
other
functions
like
most
of
those
are
shimmable
on
browsers,
so
you
would
normally
not
use
parsers
on
the
browser,
because
there
are
no
command
line
arguments
there.
F
Yeah
but
but
node.js
does
not
run
in
the
I
mean
in
the
browser,
it's
it's
for
the
back
end,
so
it's
bound
to
have
arguments
and
util
model
is
essentially
for
the
apis
helper
apis
for
similar
executing
similar
things
right.
A
I
see
this
richard
has
his
hands
up.
B
So
so
there
is
a
difference
between
hanging,
something
off
util
and
hanging,
something
off
process
different
to
the
sort
of
semantic
high
level.
Where
does
it
conceptually
fit?
There
is
a
sort
of
operating
difference
in
terms
of
whether
you
would
have
to
import
something
to
use
it
or
use
it
directly.
E
A
I
think
what
I'll
suggest
is
you
know,
since
we
we
we
just
added
this
to
the
agenda
at
the
last
minute.
People
might
not
have
had
time
to
think
about
it.
So
let's
leave
it
on
the
agenda
and
I
think
the
ask
is
that
more
tsc
members
think
about
it
and
get
involved
in
that
issue.
Is
that
reasonable
darshan.
B
In
the
chat
saying
that,
if
matteo's
feedback
is
a
hard
block
or
asking
if
matteo's
feedback
block
is
is
hard
because
if
it
is
a
hard
block,
we
probably
need
to
you
know
hear
from
him
as
to
to
let
him
articulate
his
reasons
or
expand
on
them.
It
certainly
seems
like
a
hard
block
to
me.
A
Okay,
so
I
think
you
know
doubly
so:
let's
ask
more
tse
members
to
take
a
look
we'll
leave
it
on
the
agenda
until
we've
got
matteo.
Who
can
talk,
talk
to
it
as
well
and
we
can
go
from
there
sound
good.
A
A
So
the
plan
will
be
to
open
call
for
nominations
on
may
9th.
I
think
jarish
you'd
suggested
a
vice
chair,
and
so
maybe
do
you
want
to
just
introduce
that
we
can
have
a
discussion
on
that
that
particular
topic
so.
G
We
should
check
to
make
sure
that
it's
that
there's
nothing
in
the
charter.
That
would
somehow
require
changing
for
something
like
this,
but
I
can't
imagine
there
is,
but
it's
it's
worth
checking,
but
other
than
that,
I'm
plus
one
yeah.
G
A
G
F
G
A
You
froze
out
there,
at
least
for
me,
hope.
Maybe
it's
me
I
froze
or
richard
froze.
No.
I
think
it
might
be
me
freezing
okay,.
B
No,
I
was
gonna
say
if
we
did
want
the
the
chair
and
vice
chair
to
be
in
different
time
zones
that
may
so
just
mechanics
wise.
It
may
be,
then,
that
we
would
have
to
have
like
different
rounds
of
voting
or
something
and
that
it
would
probably
then
we'd
have
to
have
the
chair
elected
to
discover
what
time
zone
they
would
be
in
before.
G
F
G
A
Good,
okay,
because
yeah,
I
think
it
it
would
be
workable
to
say
you
know
the
two
elections
and
stuff,
but
if
we
just
want
to
start
with
like,
let's
just
elect
the
vice
chair
and
then
we
can
go
for
there
that
I
think
that's
a
great
thing
to
do
this
year.
We
can
see
how
it
works
out,
and
then
we
can
figure
out
how
to
iterate
iterate
on
that
to
improve
how
it
works
right.
G
A
A
I
will
I
think,
just
you
know.
It
sounds
like
from
the
discussion
here
everybody's
in
favor
of
vice
chair.
A
G
And
bethany
suggests
that
you
know,
maybe
just
whoever
comes
in
second
for
chair
could
be
asked
to
be
vice
chair,
which
I
think
makes
a
lot
of
sense.
It
might
even
make
sense
that
anybody
who
runs
for
chair
you
know
anybody.
Anybody
who
put
their
name
in
the
hat
to
run
for
chair
can
be,
can
be
a
vice
chair
or
something
you
know,
and
then,
as
long
as
nobody's
on
objections,
you
know
we
can
have
like
four
vice
chairs
or
something,
but
anyway.
G
A
A
Okay,
that's
good
thanks,
jerish
for
bringing
that
up
and
reminding
us
that
we
sort
of
plan
to
do
that,
and
this
is
a
good
time
to
do
that
on
to
the
next
issue.
The
node.js
admin
so
createpat
for
blocklist
import,
export.
A
You
know,
basically,
I
think,
there's
a
lot
there's
there's
enough
approvals
in
there,
because
I
think
we
only
require
a
couple
plus
ones.
A
I
think
maybe
the
reason
it's
still
here
is
like
somebody
wait,
a
sec
blockless
token
right.
Okay,
so
maybe
we
can
just
take
this
off,
because
I
can
see
that
some
people
have
been
added
and
have
made
it
move
forward,
because
I
think
that's
the
the
last
I'd
heard.
It
was
just
waiting
on
somebody
to
be
able
to
help
out
anybody
feel
otherwise,
otherwise
I'll
just
remove
it
for
the
end.
Yet
from
the
agenda
issue,.
A
Okay,
great
the
next
one,
which
is
the
node.js
build
tool
chain
initiative,
needs
a
champion
number
13.,
so
rich.
I
think
you
put
that
on
the
agenda
right.
G
Possibly
probably,
but
just
you
know,
mary
has
you
know
on
on
on
an
indefinite
hiatus
and
is
not
does
not
have
the
time
or
or
or
inclination
or
whatever
it
may
be,
to
to
continue
to
lead
that
initiative.
So
we
need
somebody
else
to
lead
it
or
we
should
remove
it
from
the
list
until
we
have
somebody
to
lead
it.
A
G
And
I'm
sorry,
but
I
I
didn't
realize
this,
but
I
have
to
go
so
thanks
for
having
me.
H
A
Do
we
think
like
do
we
need
somebody
like
somebody
going
to
be
able
to
come
from
who's,
not
part
of
the
the
the
tse
or,
I
guess
the
collaborator
base,
and
it's
a
good
question
whether
collaborator
base
would
see
that
see
this
here
and
be
successful
like
should
we
be
just
asking
like
who
would
we
ask
to
step
up?
Is
it
anybody
or
is
it
maybe
one
of
the
collaborators
or
what
do
you?
What
do
you
think
on
that
front?.
A
A
A
A
So
so
from
that
perspective,
I
I
think
we
can
work
to
change
it
so
that
it
could
be
anybody
on
the
the.
What
I
just
wonder
is
in
terms
of
the
the
success
path
to
changing
our
build
thing.
How
how
how
close
or
non-close
do
you
need
to
be
to
the
original
prod
to
the
existing
product?
Like
would
somebody
just
who's
not
been
contributing
to
the
project
at
all,
be
able
to
come
in
and
do
that.
B
In
that
there's,
no
reason
that
they
have
to
be
linked,
and
in
that
that
sort
of
respect
I
could
see
you
wouldn't
need
to
be
so
invested
into
sort
of
how
node
core
is
built
in
order
to
look
at
how
add-ons
are
being
built
and
sort
of
you
know
make.
Maybe
the
advantages
to
not
be
embedded
in
the
node
ecosystem
to
to
have
a
fresh
look
at
how
add-ons
are
built
for
node,
but
definitely
for
building
node.
B
B
H
I
thought
the
two
were
linked
in
that
the
add-ons
use
the
common,
the
guipi
to
yes,.
B
That's
because
that's
the
way
it's
currently
done
so
that
could
be
divorced
from
from
the
core
build
system.
There's
there's
a
long
standing
open
issue
in
no,
I
mean
there
was
another
issue
about
maintainers
with
no,
but
there
is
an
issue
in
no
grp
about
moving
away
from
using
common
gypi,
and
that
would
break
the
dependency
on
the
sort
of
node
build
output,
at
least
at
least
as
far
as
the
tooling
is
concerned,.
H
I
said
that
would
be
some
kind
of
prerequisite
to
moving
no
core
from
jeep.
I
guess.
B
Potentially
it
doesn't
have
to
be,
it
would
make
it
harder
if
we
had
to
maintain
common
gypi
when
we
were
not
using
it
to
build
core,
but
I
mean
that
is
an
option
as
a
sort
of
compatibility,
so
I
mean
the
problem
with
the
the
main
issue
with
the
ecosystem
is
there's
an
awful
lot
of
stuff
out
there
that
doesn't
get
updated,
often,
and
but
it's
still
popular
so
trying
to
do
a
migration
of
the
ecosystem
from
to
something
else
would
be
very
challenging.
B
A
I
think
the
past
discussions
have
always
landed
on
will
have
to
be
supported
for
add-ons
for
some
probably
substantial
period
of
time.
No
matter
what
you
do,
I
think
they
can
be
decoupled,
but
I'd
I'd
be
hesitant
to
say,
let's
choose
a
new
add-on
thing
until
we
know
until
we
figure
out
what
we
want
to
do
for
core,
because
I
think
one
might
inform
the
other,
even
though
they
can
be
separate.
G
A
Yeah
and
definitely
the
note
ad
on
epi
work,
there's
been
work
related
to
supporting
you,
know,
cmake,
and
I
know
you
know
reasonable
number
of
people
use
that
instead,
but
you
know,
I
guess
my
my
opinion
here
is
just
as
like
the
first
part
of
figuring
out
what
we
would
use
for.
Node
core
is
kind
of
the
one.
You'd
want
to
do
first,
but
just
my
opinion,
yeah,
the.
C
Existing
work
in
that
repository
that
the
issue
is
in
is
all
about
not
core,
so
there
are
three
pr's
where
mary
did
some
work
on,
arguing
for
and
against
certain
two
chains,
and
those
are
all
just
concerning
note
core,
not
not
any
npm
packages
and
there's
also
a
separate
issue
in
the
same
repository
about
forming
a
group.
C
So
there
were
a
few
people
interested
in
joining
that
group
and
that
might
be
and
the
long-term
yeah,
regardless
of
whether
we
have
a
volunteer
for
championing
that
group,
we
might
still
have
people
interested
in
working
on
it.
So
finding
a
champion
might
just
be
a
formality.
A
C
A
A
H
I
guess
we
can
maybe
use
the
twitter
account
to
promote
this
this
this
work
a
little
bit
and
see
if
there's
any
more
volunteer
to
join
before
we
move
it
like
out
of
the
the
static
strategic
initiatives
yeah.
H
Suggestion
yeah,
if
there
are
more
people
like
joining
this
effort,
maybe
we
can
postpone
that
a
bit
until
we
found
and
they
will
find
someone
who's
willing
to
be
a
champion
even
if
they're
not
on
the
tsc.
H
But
if
there's
no
one,
no
one
who's
going
to
actually
actively
working
on
it,
then
I
guess
we'll
have
to
just
remove
it.
A
A
A
Okay,
that
sounds
good.
We
have
a
next
step
and
then
we'll
see
how
that
goes
and
can
talk
about
it.
Next
time
I
see
under
strategic
initiatives
joey
that
you
filled
in
some
info,
which
is
the
next
thing
on
the
agenda.
So
why
don't
we
start
with
that?
While
I
bring
up
the
others.
H
Okay,
so
I
it's
been
a
while,
since
I
update
about
the
startup
performance
initiative
in
the
tsa
meeting,
so
I've
been
adding
embedded
snapshot,
support
to
workers
and
hopefully
the
vm
contacts
as
well
later-
and
I
opened
an
issue
to
request
feedback
about
the
useless
javascript
api,
which
allows
users
to
add
user,
define,
deserialize
and
serialize
callback
for
their
own
objects,
as
well
as
setting
a
main
function
in
the
snapshot,
so
that
when
you
deserialize
the
snapshot,
you
don't
have
to
like
specify
another
main
script
from
the
command
line.
H
You
can
just
like
the
the
resulted
binary
can
just
enter
a
main
function
and
do
whatever
the
user
wanted
to
do.
So.
I
guess
that
would
effectively
make
the
binary
some
sort
of
single
executable
file,
but
it's
still
a
little
bit
different
from
the
the
packaging
stuff,
which
also
has,
for
example,
virtual
file
system
support.
So
I
also
opened
an
issue
about
snapshot
and
packaging
interop.
A
Okay,
thanks
antoine
anything
on
the
our
promise.
Api
side
of
things
actually
should
double
check.
We
didn't
sorry
antoine,
didn't
make
it
this
week.
So
let's
get
that
next.
10
is
the
other
one
that
we
have
somebody
here
from.
So
we
have
submitted
sessions
for
opengs
world
collaborator
summit,
we're
planning
like
the
mini
summit,
deep
dives
for
esm
and
observability
as
well
as
if
there
is
like
you
know,
a
common
core
track,
like
we've
done
in
the
past,
submitted
a
session
for
that
as
well.
A
There
is
also
you
know,
you
know
we
just
had
the
meeting
today.
So
one
of
the
other
updates
is
one
of
the
outcomes
from
the
earlier
mini
summit
on
types
was
to
try
and
add
in
some
extra
requirements
into
our
our
docs
so
that
they
could
be
structured
so
that
we
can
more
easily
parse
and
generate
the
type
information
that,
like
you,
know,
typescript
needs
and
anybody
else
would
need
so.
Tyranny
has
a
pr
to
do
that.
A
I
think
it's
now
in
a
form,
that's
ready
to
land
and
the
goal
is
to
get
that
landed
and
then
tierney's
actually
going
to
submit
a
a
pr
to
add
I'm
updating
the
docs
as
a
strategic
initiative.
I
see
that
is
kind
of
like
it's
a
longer
term
thing
where
we're
docked
by
doc.
We
can
look
to
update
them
to
fit
into
that.
A
That
structure-
and
I
think,
like
I,
you
know
earlier
we'd-
looked
at
the
parser
that
will
parse
them
and
we
can
make
it
so
that
it
can
parse
just
the
ones
that
have
been
transitioned
and
but
I
think
that
that'll
be
an
interesting
strategic
initiative
to
sort
of
go
through
and
do
those
updates,
as
joey
mentioned,
there's
still
some
discussion
back
and
forth
in
terms
of
the
outcome
from
the
single
executable
binary,
I
put
together
a
brief
prototype.
A
I
haven't
had
any
time
to
spend
any
more
time,
but
there
hasn't
been
some
really
good
discussion
in
that
that
issue,
and
I
think
the
other
one
that
we
planned,
I
guess
related
to
the
opengs
world
is
mateo,
was
also
submitted,
a
http
follow-up
session
to
sort
of
say.
Okay,
we,
you
know
exposing
fetch
was
one
of
the
first
steps
for
modern
http
in
terms
of
the
plan
we
talked
about.
But
let's
talk
about
other
steps
and
review
how
it's
going
so
far.
A
A
If
not
thanks
for
everybody's
time,
please
stick,
please
do
stick
around
for
a
short
private
session
and
many
thanks
to
everybody
who's
watching
the
video
we'll
see
everybody
next
time
and
just
before
I
stop
the
stream
I'm
going
to
grab
the.