►
From YouTube: Node.js Tooling Group Meeting 2021-08-06
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
All
right,
I
guess
I'll-
do
the
agenda
today.
All
right
all
right!
Welcome
everyone.
This
is
the
node.js,
tooling
working
group
meeting
it's
august
6th
today.
A
So,
first
up
on
the
agenda,
we
have
some
way
to
alert
end
users
to
features
that
were
added
to
node
core
when
they're
using
a
package
or
a
shim
or
something
so
this
is.
I
was
supposed
to
be
working
on
this
one.
I
have
not
made
any
additional
progress
on
this.
I've
been
giving
it
more
thought
over
the
last
little.
While
I
do
think
that
we
kind
of
talked
last
time
like
should
this
be.
A
I
think
it
should
be
a
standalone
package,
but
should
it
kind
of
implement
all
this
logic
on
its
own
or
should
be
built
on
top
of
something
like
eslint,
and
I
do
think
it
probably
makes
sense
to
build
on
top
of
eslint,
because
the
more
I
keep
thinking
about
it,
the
more
I
realize
you
would
just
be
rebuilding
a
ton
of
that
functionality.
A
Basically,
to
make
it
something
standalone,
we
can
still
run
it
as
a
standalone
package,
but
I
think
I
think
it
was
ben
that
suggested
that
in
our
last
meeting
and
I
yeah
I've
been
giving
it
some
thought
since
then
as
well,
and
I
think
that
is
a
good
way
to
go.
But
I
have
not
had
a
chance
to
start
on
that.
Yet.
A
A
Yeah
no
point
in
reinventing
the
wheel
right,
even
even
as
a
starting
point.
I
think
it
would
just
be
easier
to
do
that
way.
So,
okay,
so
that's
it
for
that
up
next
we
have
recursive
copy,
which
I
know
you
have
some
updates
on
that
ben.
C
A
So
well,
while
you're
doing
that,
I
noticed
the
name
got
changed
from
copy
to
cp.
I
missed
the
discussion
around
that,
though,.
C
Targos
had
asked
for
that
as
well,
and,
oh,
my
god,
is
there
another
comment
on
this.
I
don't
know
it
seems
like
the
right
way
to
go.
C
C
A
Extra
like
like
ads
on
top
of
the
existing
fs
module,
like
it
patches
it
or
something.
C
Yes,
that's
one
of
the
most
abusive
plus
I
I
kind
of
agreed
that
it
was
like
this
thread
is
a
I
kind
of
agreed
with
targos
and
he
he
was
supportive
of
the
pr.
If
we
made
that
change.
A
C
Yeah,
a
few
folks
had
suggested
it,
including
you,
and
I
thought
it
was
reasonable
and
I
kind
of
like
the
the
other
thing
I
do
kind
of
like
about
the
name
change
is
there
already
is
a
copy
file
method?
Oh
the
links
are
in
here.
There
already
is
a
copy
file
method
and
cp
can
copy
a
file
or
a
folder,
so
it
felt
like
it
felt
a
little
weird
to
have
both
of
them.
C
But
if
we're
claiming,
if
it's
more,
that
it's
behavior
similar
to
copy
like
rm
was
behavior
similar
to
rm,
I
felt
like
it
would
just
felt
consistent.
It
felt
consistent
and
like
a
good
good
extension
to
the
api
and
basically
cp
and
rm
end
up
being
kind
of
like
slightly
more
complex
kind
of
a
layer
on
top
of
the
simpler
fs
operations.
So
it
seems.
A
C
A
C
A
Yeah
anything
outstanding
on
the
pr
or
anything.
C
B
C
Oh
yeah
yeah.
I
guess
we
learned
from
from
our
first
pass
right
all
these
drive.
My
code
reviews
are
driving
me
nuts,
so
like
this
is
like
I've
fixed
like
100
things
that
are
like.
Maybe
move
this
up
one
space
or
change
this
thing
to
this,
and
I'm
like.
Oh
my
I've
run.
It
takes
the
node
test,
suite
like
three
hours
to
run
each
one
of
these
drive-by
little
myths
about
my
spacing
on
something
means.
It's
gonna.
C
B
That's
and
that's
a
good
thing
and
that's
you
know
I
think,
a
bit
of
a
perhaps
a
blocker
for
some
folks.
You
know
I'm
reminded
of
chris
hiller,
of
course,
not
that
it's
the
same,
but
you
know
one
of
the
things
that
we
were
talking
about
in
the
next
10
stuff
is
like
how
to
make
it
easier
to
be
a
contributor
and
what
are
the
hurdles
and
the
blockers,
and
so
maybe
that's
something
we
should
be
taking
to
that
conversation
as
well.
I
can
try
to
surface
it
next
time.
C
Yeah,
I
don't
know
the
right
answer
like
like,
I
think
like
it,
is
good
to
have
a
high
level
of
quality
so
like
to
to
one
degree
it
can
be
like
we
don't
want
to
land
stuff
before
it's
really
polished,
getting
a
code
base
used
by
so
many
people.
But
it's
really
frustrating
with
something
like
this,
where
it's
been
open
for
30
days
and
like
you'll,
be
a
day
away
from
merging
and
then
once
one
person's
like.
Oh
what
about
this
one
additional
change
or
this
one
additional?
C
C
B
I
was
gonna
say
I
think
he
was
really
invested
like
he
did
a
lot
of
work
and
up
front
work
and
planning
and
and
and
documentation,
and
then
some
people
just
really
kind
of
I
don't
want
to
say
poo
poo'd
it
but
like
wanted
him
to
do
something
different
than
what
he
had
like
invested
all
this
energy
into.
So.
C
B
C
C
C
B
Right
I.
C
C
So
it
defaults
to
defaults
to
only
doing
files
until
you
say
recursive
true,
which
is
how
copy
works
on
the
command
line,
but
it
is
overrating
by
default
because
on
account
of
the
fact
that,
like
copies
overrating
by
default,
so
if
you
do
like
copy
foo
to
bar,
it
will
copy
foo
over
bar.
If
there's
already
a
bar.
C
Yeah,
the
only
thing
that's
a
little
weird,
but
I
called
it
out
of
documentation
is
copy,
has
behavior
that's
different.
If
the
folder
already
exists,
then,
if
the
folder
didn't
exist,
it
will
copy
the
folder
to
a
subfolder
of
the
folder.
If
it's
there,
I
believe,
and
it
will
copy
it
to
the
folder
itself.
C
If
there's
not
a
folder
there
already
that
behavior
people
are
probably
used
to
it
from
using
copy,
but
it
complicates
the
algorithm
a
bit
and
also
feels
a
little
weird
for
an
api
like
I
feel
like
the
api
should
just
copy
to
where
you
set
it
you're
copying
to
like,
if
you
say,
copy,
folder
a
to
folder
b,
regardless.
If
there's
already
a
folder
b,
just
copy
folder
a
to
folder
b,
don't
copy
it
to
a
sub
folder
of
b.
A
C
Yeah
you
well,
you
end
up
needing
to
basically
implement
globs
right
cause
like
because,
like
you,
you
can
copy
f
a
folder
without
the
slash,
a
folder,
with
a
slash
which
has
different
behavior
than
a
folder.
Without
a
slash,
and
then
you
could
say
like
copy
foo,
slash
star.js
into
another
location,
which
is,
it
means
you
need
to
start
doing,
glob
handling.
C
C
So
I
think
it's
reasonable.
That's
the
behavior
that
fs
extras
copy
went
with
okay.
I
think
we
could
revisit
it
if
we
ever
added
glob
in
and
then
we
could
and
then
we
could
add
glob
to.
C
Glob
to
rim
wrap
as
well
maybe.
C
Yeah,
I
still
find
that
behavior
a
little
weird
weird
for
copy
like
it's.
It's
like
I'd
rather
be
explicit,
like
if
you
want
it
in
a
subfolder
of
b,
then
indicate
that
you
want
to
copy
to
a
subfolder
of
b
right.
I
feel,
like
I
always
forget.
I
always
accidentally
do
that
occasionally
and
didn't
mean
to
when
I'm
using
copy.
So
I
didn't
know
that
I
wanted
to
reproduce
that
behavior,
but
maybe
seven
people
will
we're
calling
it.
Oh.
C
The
other
thing
we're
saying
is
we're
calling
it
experimental
for
now
we're
calling
out
the
feature
as
experimental.
So
if,
like
20
people
comment
after
the
fact
and.
C
Or
like
why
the
heck
does
it
behave
like
this?
We
can
revisit
that.
B
C
B
A
Well,
yeah,
I
actually
thought
about
that
earlier,
like
we
were
talking
about
like
what
could
we?
What
could
we
do
about
this
this
process
at
some
point
in
the
future,
like
maybe
that
is
kind
of
an
option
where
it's
like
last
call
yeah
like
last
call
for
for
fixes
and
then
then
we're
gonna
leave
this
up
for
some
period
of
time,
just
to
make
sure
everyone's
had
a
chance
to
comment
on
it
or
something
I
don't
know,
yeah.
A
B
C
Okay,
yeah,
maybe
it's
just
something
you
have
to.
I
mean
at
the
end
of
the
day,
25
days.
Isn't
that
long
to
build
a
big
feature,
yep?
What
I
will
say
is
it's
slower.
I
will
say
feedback
for
like
the
next
10
years.
It's
definitely
a
little
slower
than
like
some
other
big
code
bases.
I've
worked
in,
like
I
think
you
can
move
a
bit
faster
in
chromium
or
like
v8,
but
it's
much
harder
to
contribute
to
chromium
or
v8.
You
have
to
write
c
plus
plus
so,
like
I
don't
know
like.
C
C
C
B
C
We
pulled
an
fs
extra
and
then
one
of
the
refactors
was
to
rewrite
it
to
these
promises,
which
was,
I
didn't
actually
do
that
that
was
a
submitted,
pull
request
and
then
node.js
wants
you
to
document
all
your
errors
in
errors.js.
If
you
have
your
own
errors
that
you're
throwing
so
kind
of,
we
took
all
the
errors
and
made
them
kind
of
the
way.
C
Node
does
errors,
and
then
you
know
node
has
you
know
ways
that
you
add,
like
there's
markdown
files,
you
have
to
update
with
all
the
new
api
methods
for
any
file
operation.
You
have
to
update
three
places:
the
synchronous,
asynchronous
and
promise
version
of
the
api,
okay
and
then
there's
validators
as
well.
A
Another
one
that
I
remember
from
rimraff
is
the
way
that
node
does
argument
validation
as
well
in
functions
like
that
was
something
that
they
didn't
have.
So
I
had
to
add
that
in
in
addition
to
all
the
things
you
were
kind
of
mentioning
yeah,
so
there's
a
few,
a
few
things
for
sure
that
need
to
be
kind
of
yeah
made
into
the
node
style.
B
C
Yeah
yeah
and
I
think
because
because
working
with
the
fs
extra
people
like
I
rewrote
the,
I
also
rewrote
the
test
suite
because
I
wanted
to
completely
understand
all
the
functionality.
So
I
kind
of
went
from
first
principles.
B
C
And
actually
found
a
few
bugs
in
the
process,
so
I
should
make
sure
I
upstream
those
depth
as
extra
yeah
one
or
two
little
bugs
that
were
fixed
cool
cool
got
still
97
code
coverage.
I
don't
know,
I'm
stoked
yeah,
that's
great,
really
great
nice.
Maybe
I
just
won't.
C
A
That
that
was
just
before
I
reviewed
it
earlier
today,.
C
C
A
Yeah,
nice,
okay:
let's,
let's
move
to
the
next
item
on
the
agenda,
which
is
source
map
v3
support
any.
I
assume
no
updates
on
that.
C
C
C
A
Is
that
the
one
in
the
in
our
github
or
no.
C
B
C
C
That's
the
source
map
update,
though
it's
it's,
I
think
it's
doing:
okay,
okay,
cool,
there's,
there's
two
or
three
issues,
though,
if
anyone
does
want
to
contribute
that
would
be
good.
First
issues.
B
Yay,
I
don't
have
any
update,
I
I
don't
know
if
anyone
else
does.
A
A
B
Well,
we
know
some
people
over
there,
so
if
you
want
to
see
if
we
can
get
it
reviewed,
you
know
considered
certainly
an
option.
A
Yeah,
I
thought
about
that.
I
I
tweeted
about
it,
but
I
don't
know,
I'm
not
sure
I
don't
know.
What
do
you
guys
think
is
it?
Is
it
worth
it
or
sure?
A
B
Yeah
I
have
a
meeting
with
bernadette
from
near
form,
who
I
think
does
events
and
stuff
for
them.
I
mean
she's
doing
I'm
working
with
her
on
a
different
event,
so
I
could
ask
her
about
it
before
meeting
on
monday
or
tuesday
sure
it's
worth
a
try,
yeah.
A
B
Yeah
I'm
meeting
with
her
on
wednesday,
but
I'll
just
send
her
an
email.
Do
you
want
to
send
me
something
that
I
can
send
over
to
her?
I
mean
I
know,
there's
that
blurb
in
the
slack
that
that.
A
B
Okay,
I'll
shoot
her
an
email
when
we're
done
with
this
meeting
and
mention
it
to
her.
A
A
C
That
sounds
good.
I
think
we
can
find
some
other
forcing
punch
function
independently
of
that.
I
think
yeah.
If
we
push
comes
to
shove,.
B
C
Well,
if
I
get
this
fs
copy
thing
done,
maybe
I
can
pitch
in
too.
Maybe
that's
because
I
think
that
gets
us
through
the
that
will
be
our
next
big
ticket
item.
A
Yeah,
I
agree
one
one
thing
I
think
I
raised
this
last
time
that
would
at
least
make
it
easier
to
to
start
trying
out
is
actually
publishing
the
npm
package,
because
we
have
that
we
have
the
repo,
and
I
think
we
have
some
working
code
in
there,
but
it's
not
published
npm
at
all.
A
B
B
A
C
Yeah,
oh
sorry,
go
ahead.
I.
A
So
I'm
currently
a
use
case
that
I
think
would
be
interesting
for
this.
Actually
is
that
I
we
have
a
slack
bot.
I
think
I've
mentioned
this
before
we
have
a
slack
bot
and
has
like
a
slash
command
that
you
can
use
to
run
various
commands
and
some
of
the
commands
have
a
bunch
of
arguments
and
we
basically
just
made
like
a
you,
know
very
crude,
like
parser
of
our
own,
for
those
arguments,
and
actually
someone
on
my
team
right
now
is
currently
trying
to
to
jam
yards
into
there
to
parse.
A
B
C
No,
I
think,
that's
I
opened
an
issue
for
that
feature
that
hadn't
been
implemented
yet
for
source
maps
so
that
we
don't
lose
track
of
it
just
now.
Oh
this
is
the
rebel
one
yeah.
C
A
B
B
B
Yeah,
I
think
I
might
I
might
end
up
managing,
like
you
know,
cloud
native
computing
foundation,
work
folks
who
are
working
on
that
side
of
things,
but
but
yeah
anyway.
One
other
thing
that
comes
to
mind
and
actually
is
sort
of
a
little
related
to
that
side.
Comment
is,
I
wonder
about,
and
this
is
a
little
bit
from
the
next
10
conversations,
but
it's
also
from
website
redesign
work.
B
I
wonder
about,
and
I
know
there
used
to
be
a
documentation
team,
but
I
wonder
about
kind
of
reviving
that
and
improving
nodes
docs.
I
know
that
they're
decent
for
like
reference
stuff,
but
not
necessarily
for
people
who
are
new
to
node
and
building
stuff.
I
know
we've.
B
Some
of
us
in
a
variety
of
places
have
heard
comments
that
no
docs
don't
aren't
really
helpful
and
I
think
they
mean
like
when
they're
trying
to
build
stuff.
You
know
yeah.
A
B
A
A
C
A
Yeah,
yes,.
B
B
B
B
We
want
to
bring
the
docs
over
to
the
new
website
and
we
need
to
figure
out
how
we
do
that,
because
there's
some
parsing
that's
going
on
in
there
that
we
would
need
to
make
happen,
but
also
like
adding
some
front
matter
and
being
able
to
kind
of
manage
some
of
the
integration
into
the
the
new
platform.
B
So
docs
is
one
area.
Localization
is
another
that
we
need
to
flesh
out,
which
which
probably
isn't
a
big
thing,
but
I
and
the
internationalization
team
is
aware
of
it
and
has
it
on
their
radar
shoot.
What
was
the
third
one?
I
have
a.
I
have
a
dock
for
it
as
well.
If
anybody's
interested
is
there
a?
B
Is
there
a
repo
for
that
project
in
the
org
yeah,
it's
nodejs.dev,
which
is
not
a
great
name,
because
that's
also
the
name
of
the
development
work
in
the
openjs
org,
but
here
I'll
give
I'll
give
you
guys
this
link.
If
you're
curious,
it's
a
hack,
md
file,
but
it
does
have
links
to
the
repositories
sort
of
some
of
the
current
status
stuff.
So
docs
localization,
oh
a
news
and
blog
feature.
B
You
know
we
have
that
for
security
releases
and
some
announcements,
but
we're
using
gatsby.
So
we
could
have
the
full-fledged
blog
happening
there,
which
I
think
would
be
better
than
using
medium,
or
at
least
we
could
syndicate
to
medium
and
dev
too.
But
I
think
having
the
canonical
blog
on
the
website
is
probably
a
good
thing
and
then
sort
of
as
an
aside
just
making
sure
that
we
have
no
broken
links
when
we
move
from
one
to
the
other.
B
You
know
some
things
have
moved,
but
I
just
want
to
make
sure
nothing
gets
abandoned
or
overlooked
or
whatnot.
So.
B
Yeah
we
can
certain
there's.
There
are
a
few
people
who
are
active
and
that's
great
the
people
who
are
sort
of
the
leaders.
There
are
not
great
at
like
really
managing
and
being
sort
of
leaders
in
a
way
they're
they're
they're
good
at
helping
to
get
the
work
kind
of
keep
going,
but
sometimes
they
overlook
the
big
picture
and
sort
of
jump
the
gun
and
do
some
other
things
that
I
I
try
to
help
them
with.
B
But
it's
it's
getting
to
the
point
where
I
think
we
could
start
to
work
more
with
the
nodejs.org
people
and-
and
you
know
that's
why
I
was
thinking
about
a
documentation
team
if
we
needed
to
kind
of
flush
anything
out
or
think
beyond
that.
So
that's
that's!
That's
where
it's
at.
C
A
C
I
think
that's
it,
I'm
gonna
go.
I
think
I'm
gonna
go
with
my
plan
of
waiting
a
couple
more
days
to
see
if
a
few
more
things
come
in
on
that
pr
and
then
I'll
merge
people's
suggestions
on
monday
and
get
it
out
so
in
two
weeks,
that'll
be
exciting
yup.
One
of
the
one
other
thing
I'm
excited
about
this
week
is
mozilla,
said
they're
good
to
prototype
the
random
uid
work
I
was
doing
so.
We
have
support
from
mozilla
and
from
chromium
which
cool
I'm
excited
really
excited
about
it.
C
C
Cool
when
I
saw
actually
adjacent
work,
I
guess
I
we
already
have
random
uid
and
node.js,
but
I
guess
it
wasn't
exposed
on
the
crypto
namespace
because
it
wasn't
seen
as
being
like.
You
know,
fully
standardized
yet
so
I
I
thought,
targos
had
just
added
it
to
the
exposing
it
on
the
web.
Crypto
object,
which
I
thought
was
neat
wait
is
that
is
that
the
crypto
object,
or
is
it
a
different
name?
I.
A
C
I
think
it's
it's
like
splitting
hairs,
you
could
import
it,
but
I
don't
think
it
was
like
actually
on
the
object
itself,
if
you
brought
it
in
as
a
like,
if
you
as
he
brought
it
in
the
entire
name
space.
Basically.
B
Well,
you're,
looking
obviously
too,
I
don't
know
if
you
all
saw
tierney's
issue
about
pre-built
binaries.
B
Oh
no
yeah,
another
kind
of
conversation
we're
having
in
the
around
like
contributors,
work
having
like
pre-built
node.js
and
node
images,
or
a
cache
of
sorts.