►
From YouTube: Node.js Tooling 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
So
we
have
a
couple
things
to
update.
Okay,
oh
look
at
that
streaming
on
youtube.
It
worked
great.
We
should
probably
say
with
what
we're
here
to
do.
You
wanna
go
ahead:
yeah,
yeah,
sure,
hello,
everyone,
hello,
everyone
joining
on
youtube.
This
is
a
meeting
of
the
node
tooling
working
group,
fairly,
sparsely
attended
today.
I
think
folks
are
starting
to
take
summer
vacations.
So,
yes,
let
us
go
through
the
agenda.
A
A
A
A
I
think
we
should
find
the
preview.
I
posted
a
comment
on
there
check,
though
yeah
you
did.
I
got
it
here.
We
discussed
in
the
cooling
meeting
today
the
idea
of
creating
a
small
standalone
package.
You
could
run
in
a
project
that
would
scan
your
code
and
dependencies
and
generate
a
list
of
recommendations.
I
like
that
idea.
A
A
Yeah,
I
mean
like
starting
with
its
own
tool
if
it
makes
sense
to
combine
it
in
with
something
like
if
it
eventually
made
sense
as
a
linting
rule
or
a
free
check-in
node
or
something
that
was
one
thing
that
I
so
I
have
put
some
thought
into
like
how
you
would
build
this,
and
I
started
to
realize
that
basically,
I'm
just
building
eslint
but
yeah
so,
like
you
said,
maybe
maybe
eventually
it
it
does
make
sense
to
you
know,
incorporate
it
as
like
a
plug-in
or
something
like
that
in
eslint,
but
I
I
do
still
think
it's
actually
reasonable
to
start
with.
A
Just
like
I
mean
it's
easy
enough
right
to
build
a
little
cli
tool
that
just
scans
like
the
contents
of
your
project
for
certain
things,
so
yeah
I'll.
I
will
still
start
with
that
plan
and
then
yeah,
maybe
it
maybe
it
goes
in
a
different
direction
in
the
future.
But
I
do
still
think
that's
a
reasonable
starting
point.
A
Little
standalone,
cli
yeah,
I
mean
that's,
that's
not
a
bad
idea.
I've
I've
built
other
tools
as
well
that,
like
scan
projects
like
I
built
one
that
scans
all
your
source
code
for
credit
card
numbers
for
compliance.
A
I
kind
of
got
already
like
a
basic
application.
That
does
that.
So
I
might
I
might
start
with
that,
but
yeah,
that's
not
a
bad
idea
as
well.
If
it
gets
tricky
to
like
plug
the
rules
in
or
something
like
that,
I
might.
I
could
try
that
approach
yeah.
I
just
think
you
might
like
you
might
want
to
use
babel
or
a
supreme
or
something
just
so
you
can
actually
detect
the
like.
Has
someone
passed
in
the
parameter
recursive?
Has
you
know
what
I
mean
right,
yeah?
A
A
Okay
source
code,
anyways
yeah
I've
got
a
few
ideas
there,
so
yeah.
I
might.
I
might
hack
on
that
a
little
this
weekend,
since
I've
got
some
unexpected
free
time.
A
Yeah
I
saw
this
looking
up.
It
has
pretty
good
feedback.
I
I
feel
like
there's,
not
nothing.
I
mean
still
in
drafts,
so
they
might
get.
You
know
a
lot
more
people
pile
on
it
has
people
it
has.
It's
actually
gotten
a
lot
of
review
already
and
it's
been
mostly
addressed.
People
seem
happy,
but
once
I
take
it
out
of
draft,
maybe
we
get
more
feedback,
but
I'm
a
little
blocked
right
now
on
failing
windows
tests,
because
I
don't
have
a
windows
environment.
A
So
I've
been
spending
the
last
week
kind
of
getting
security,
exemptions
and
and
stuff
to
be
able
to
actually
put
the
build
chain.
I
need
on
it.
So
a
little
blocked
right
now,
just
waiting
on
some
additional
exemptions
to
go
through
and
then
I
will
have
a
windows
machine
up
and
that
I
can
run
from
my
macbook.
A
I
actually
had
similar
issues,
so
I
actually
bought
like
a
windows
machine
a
few
months
ago,
which
I'm
still
in
the
process
of
setting
up
for
like
dev
work,
but
this
is
part
of
the
reason
why
I
got
it
so
maybe
maybe
I
could
help
out
there
as
well
yeah.
I
was
actually
gonna
ask
if
you
had
windows.
A
If,
if
I
basically
start
next
week-
and
I
still
don't-
have
my
windows
machine
runnable
like
maybe
we
could
pair
together
on
a
session
or
something
sure
the
frustrating
thing
is,
I
think
it's
going
to
take
five
minutes
to
fix.
Once
I
have
a
windows
machine
up
and
running,
I
think
it's
probably
like
a
hopeful.
It's
a
bug
in
the
path.
A
Munching
logic,
that's
just
like
treating
a
slash
the
wrong
way
or
something.
So
that's
my
punch
or
maybe
there
is
an
actual
behavioral
difference
on
windows,
but
at
least
we
can
see
it
and
skip
the
test
or
something
so
does
the
other.
Does
the
package
that
you
brought
the
recursive
copy
from
have
windows
tests?
A
A
It's
probably
the
silliest
thing
from
like
some
of
our
little
bit
of
refactoring
or
something
right,
but
it's
been
really
really
smooth.
So
far,
I
actually
think
I
found
a
bug
in
the
source
material
we
were
using.
So
if
so,
my
like
clean
room,
re-rated
tests
helped
me
find
a
bug,
an
existing
bug,
which
I
think
was
good.
A
Like
you
know,
most
of
the
comments
there
there
didn't
seem
like
there
were
any
like
major
like
blocking
issues.
So
that's
definitely
positive
yeah.
I
think
we'll
flag
in
more
people
once
it
goes
out
of
draft
and
but
I
think
in
general,
it's
to
me
it's
like
it's
convenient
right
like
it's
funny.
I
was
writing
the
test
for
it
and
you
know
again
using
mr
recursive
made
writing
the
test
for
this
way.
Easier
and
and
rimra
for
cursive
is
being
used
to
clean
up
the
testing
environment.
A
My
gut
is
copy
it's
like,
like,
I
think,
because,
because
we're
not
trying
to
completely
copy.
Oh
sorry,
I'm
overloading
the
word
copy,
we're
not
trying
to
have
cp's
behavior
like
like
our
discussion
around,
should
it
clobber
by
default.
Like
I
wasn't,
I
don't
think
we
need
to
reference
how
how
copy
does
it
out
of
the
gate.
A
Open
discussion
we
could
we
could
discuss
that
in
a
future
in
a
thread,
because
if
we
did
copy
cp,
then
we
probably
would
want
like
a
force
and
a
recursive
option
right
aren't
don't
those
options
exist
in
the
current
implementation
there
was
a
an
override
flag
or
something
right.
F,
extra
hat
fs
extra
called
overrate
yeah.
I
called
it
force,
but
there's
no
recursive
flag,
because
it's
always
recursive
because
you
use
copy
file.
A
If
you
okay,
so
here's
an
argument
for
copy
actually
is
we
already
have
copy
file,
so
it
feels
like
it
feels
like
kind
of
natural
to
have
copy
and
then
copy
file
copy
file
for
a
single
file
copy
for
one
to
n
files.
I
don't
know
yeah,
that's
fair.
Like
I
said
I
don't
I
I
don't
feel
strongly
about
one
or
the
other.
It's
just
about
aiming
for
consistency
in
some
way:
yeah,
maybe
just
rename
rm
to
remove
yeah
yeah.
A
We
do
our
alias
it
to
remove
or
something
but
we've
renamed
that
thing
seven
times
at
this
point.
So
right
like
I
don't
want
to
do
that
either.
So
I
know,
but
I
don't
want
to
claim
that
it's
a
port
of
the
copy
tool,
whereas
I
think
when
we
wrote
remove,
we
were
actually
really
trying
to
closely
match
rm's
behavior,
so
right
yeah.
So
I
don't
know.
Maybe
it's
a
slightly
different
problem.
A
Yeah
yeah,
I
noticed
it
didn't
come
up
in
the
discussion,
so
I
didn't
want
to.
I
didn't
want
to
mention
it
and
to
kind
of
derail
the
whole
thing.
So
I
I
might
I'm
kind
of
leaning
towards
copy
right
now,
because
if
you're
looking
up
the
documentation
for
copy
file,
you
can
see.
Oh
here's,
the
analogous
thing
to
copy
many
files
right.
It's
kind
of
you
could
kind
of
line
up.
Well,
so
plus
it's
what
it's
called
an
fs
extra
already
calls
it
copy.
A
So
it's
like
bringing
over
the
naming
convention
of
the
library
that
we
used,
whereas
the
library
we
used
for
recursive
remove,
was
was
rim
raft,
so
it
kind
of
already
had
rm
baked
into
the
into
its
name
right.
So
yeah
I
mean
that
seems
like
a
good
enough
logic
to
me
yeah,
but
yeah.
That's!
I
was
happy
to
see
that
I
was
actually
surprised
that
that
ended
up
being
as
straightforward
as
it
was
yeah.
It
was,
I
don't
know
it
took
like
that,
was
in
the
details.
A
It
definitely
took
like
eight
hours
of
coding
just
to
get
it
ported
over,
but
I'm
happy
with
how
it
looks
because
the
nice
thing
is
we
could
use
the
work
that
was
done
on
rimroaf
already
was
a
good
starting
point,
because
it's
like
set
the
path
for
how
we
would
pull
in
a
library
how
we
would
vendor
a
library
for
this
sort
of
thing
right.
So
so
I
used
a
lot
of
the
same
approach.
A
We'd
used
for
rimroaf,
even
like
the
validation
approach
we
used
when
you
and
I
sat
down
and
pair
coded
on
some
of
that,
so
right
yeah.
I
was
just
surprised.
I
know
we
had
lots
of
issues
with
like
rimraf,
where,
like
it
had,
you
know
initially
like
a
dependency
on
glob,
just
pleasantly
surprised
that
it
didn't
seem
like
those
issues
existed
in
the
copy.
A
No,
it
was
very
nice
and
self-contained.
So
that
was
great.
Do
you
want
to
so
so
kind
of
put
a
pin
in
that?
Is
it
okay?
If
I
flag
you
in
our
chat
room
later
this
week,
if
I,
if
we
need
to
pair
on
it
yeah
for
sure
I
need
to
that's
another
one
of
my
weekend,
projects
is.
I
need
to
finish
setting
up
my
I
kind
of
got
the
windows
machine
set
up
just
working
on
like
the
wsl
and
like
dev
environment
kind
of
set
up.
A
Well,
I
guess
you
need
not
that
if
you
want
to
test
like
windows
proper,
I
don't
know
what
I
need.
I
haven't
used
windows
in
years.
I
only
use
windows
when
I
either
yeah,
so
I'm
happy
to
help
with
the
caveat.
I
have
literally
no
idea
what
I'm
doing
yeah,
that's
good,
we'll
figure
it
out
together.
All
right.
A
Let
me
see
here
I
might
just
take
a
look
if
there's
any
open
issues
on
node
for
source
map.
I
think
I
did
see
a
couple.
A
A
It's
basically
someone
asking
for
source
maps
to
be
enabled
by
default,
because
it
would
be
useful
for
a
programming
environment
vitejs
which
looks
it's
cool
cool,
another
javascript
framework,
what
front-end
framework,
but
just
like
this
here,
that's
the
only
outstanding
source
map
issue.
Right
now
I
see
oh
one,
cool
source
map
related
thing
that
happened,
though
it
can
now
be
turned
on
programmatically
as
of
recently
okay.
A
A
A
In
my
opinion,
it
kind
of
comes
up
in
that
thread
that
was
shared,
which
is
it
certainly
can
impact
performance
because
you're
loading,
this
big
additional
file
and
also
can
make
errors,
take
longer
to
pump
out,
although
you're
already
you
know,
you're
about
to
shut
down
already,
when
you're
printing
out
an
error
a
lot
of
the
time,
but
also
we've
still
been
finding
some
bugs
occasionally,
where
we
just
screw
up
the
source
map,
handling
or
load
them
incorrectly.
A
So
I'd
rather
have
you
have
explicitly
said:
you're
in
a
testing
environment
or
an
environment
where
you
want
source
maps
where
you've
opted
into
them
before
we
just
break
people
who
aren't
even
trying
to
use
the
feature
right
now.
I
think
so
today
does
the
source
map
only
get
referenced
when
you
print
a
error
to
the
console?
A
A
A
So
it's
not
like
happening
all
the
time,
but
it
does
mean
like
if
you
have
a
large
typescript
application,
your
the
amount
of
space
you're
going
to
take,
but
take
up
in
memory
when
you,
when
your
application
starts,
is
maybe
doubled
because
you
have
each
of
those
files
has
a
relatively
large
source
map
that
sits
next
to
it
right.
So
I'm
a
little
concerned
about
that
like
I'd.
A
Rather,
it
seems
to
me
like
a
feature
you
more
want
when
you're,
when
you're
testing,
so
that
your
testing
environment
actually
maps
back
to
the
original
line
of
code
or
when
you're
debugging,
something
in
production.
Maybe
you
turn
it
on
temporarily,
because
you,
you
know
you're
having
trouble,
reading
the
stack
traces
right
right
or
if
you
know
your
application's,
not
super
performance
sensitive.
You
ought
to
turn
it
on.
A
A
I
think
it
would
mostly
hit
cold
start,
which,
which
version
of
node
is
that
available
in
it's
been
available
for
quite
some
time.
So
experimental
though
right,
but
I
guess
not
well
not.
A
lot
has
really
changed
since
we
dropped.
Not
much
has
changed.
Yeah,
okay,
yeah
but
you'll
have
you'll
still
have
the
difference.
You'll
still
have
the
like.
A
I
don't
think
you
get
the
stack
trace,
that's
consistent
until
node
16,
maybe
okay,
like
the
one
that
doesn't
deviate
from
a
normal
stack,
trace
right,
because
we,
you
tried
back
porting
that
right
and
it
was
tricky
yeah.
I
could
not
get
the
test
asan
to
pass
and
gave
up
eventually
so,
okay,
so
whatever
in
a
few
years,
it
will
be
this
the
same
everywhere.
A
A
It's
not
a
clean
back
port
yeah
the
final
agenda
item,
but
you
should
yeah
you
should
play
with
it.
Anyways
like
if
you're
like,
if
your
environment
doesn't
really
I'd,
be
curious
to
know
your
benchmarks
anyways
it's
in
like
a
real
production
environment.
A
So
it's
really
useful
for
us
to
be
able
to
just
turn
on
the
source
map
flag
in
mocha
so
that
we
map
back
to
the
rate
line
of
code
in
our
unit
test
failures
right.
So
that's
a
really
good
use
case.
Everything
about
turning
on
in
our
production
environment,
because,
like
my
team's
production
environment,.
A
Is
low
traffic
because
it's
just
we
just
do
like
tooling,
for
tons
of
teams
at
google
so
like
we
we're
like
a
large
pro
bot
deployment
that
gets
maybe
a
few
million
events
a
day,
but
not
like
you
know
we're
not
talking
about
thousands
of
events
a
second
or
anything
like
it's
like
millions
a
day,
so
we
can
take
probably
take
the
performance
hit.
Fine
and
like
just
have
nicer
stack
traces
need
to
be
good,
I
think,
to
like
benchmark
the
performance
hit
in
in
different
applications.
A
A
I'd
still
be
a
little
concerned
like
because
I
feel
that
it
has
a
lot
of
potential
to
impact
like
just
memory.
Footprint
of
typescript
apps
would
significantly
increase
right,
I
think
so
I
don't
know,
maybe
I'm
being
too
conservative,
I'm
still
not
necessarily
advocating
like
enabling
it
by
default.
I
just
mean
it'd
be
interesting
to
know
like
yeah,
that
makes
sense,
yeah
memory,
overhead.
A
A
A
I
don't
think
it's
been
published
to
npm,
though
maybe
prod
darcy
on
that
like
and
see
if,
because
he
probably
has
the
permission
to
where
we
wanna
like
we'll
need
to
make
sure
we
have
access
to
the
package
organization
right
yeah,
I
don't
publish
it
access
right
now,
so
yeah
I
can.
I
can
always
message
him
and
see
if
we
can
get
like
the
initial
version
published.
A
A
A
Awkward,
hey
there.
You
are
hey
sorry
about
that.
I
don't
know
what
happened.
It's
just
me
just
here
by
myself
for
a
couple
minutes.
That's
nice
and
awkward
awesome!
Well,
I
think
that's
everything
on
the
agenda,
so
yeah
I'll
be
trying
to
get
the
copy
recursive
over
the
finish
line.
A
I've
definitely
been
inspired
to
like
keep
looking
down
the
just
list
of
top
npm
modules
and
seeing
if
there
are
ones
that
make
sense
either
in
the
javascript
standards
like
either
in
to
be
standardized
in
w3c
or
in
tc39
or
to
like
are
good
candidates
for
like
us
to
standardize
with
regards
to
tooling
in
the
node.js
code
base,
because
right
this
has
seemed
this
seems
to
have
borne
fruit
so
far,
which
I'm
excited
about.
I
agree
I
mean
I
noticed
random
uuid
is
available.
A
I
think,
like
everywhere
now
it's
available
in
chrome
and
node
and
dino,
and
I'm
really
hopeful
that
firefox
will
adopt
too,
like
I
think,
we've
at
least
been.
You
know
asked
for
their
position
on
standards
and
I've
been
talking
with
them
so
yeah.
I
think
it's
super
useful
yeah,
that's
something
like.
I
think
we
use
that
in
like
every
single
one
of
our
services
and
most
of
our
libraries
so
like
just
having
it
built
in
it
seems
great
so
yeah.
This
has
been
fruitful
for
sure
yeah.
A
I
think
we'll
have
to
maybe
take
a
step
back
after
we
get
there.
The
reason
I
brought
that
up
is
I'm
really
excited
with
the
command
line.
Argument.
Parser,
I
think
they'll,
be.
I
think,
that's
really
in
the
similar
vein,
where
it
will
be
really
be
a
quality
of
life
improvement
for
for
known
people,
but
yeah,
then
we're
gonna
be
like
man.
Then
we've
checked
some
boxes,
though,
like
we,
it's
it's
been
slow
but
like
when
you
look
back,
though
over
you
know,
it's
like.
A
Oh,
we
got
recursive
copy
rim
raft,
make
derp
command
line
argument,
parsing
source
maps.
Yes,
you
know
like
yeah,
actually
pretty
good
yeah,
not
too
shabby
right
like,
and
I
think
it's
stuff
that,
like
is
like,
I
said,
quality
of
life
improvements
for
users
like
because,
if
you
come
in
and
you're
haven't
been
tuned
into
the
community
for
years
like
having
just
a
few
more
batteries,
is
nice,
like
you,
don't
want
to
like
immediately
have
to
go,
buy
10,
double
a
batteries
to
get
your
toy
working
right
so
yeah.
A
Maybe
we
need
to
do
a
little
bit
more
of
a
road
map,
kind
of
exercise
and
figure
out
like
what's
next,
yes,
yeah
like
pull
the
community
and
try
to
see
each
other
again,
hopefully
in
person
at
some
point
and
yeah
yeah
yeah,
and
hopefully
there
will
be
some
conferences
coming
up
where
we
could
do
you
know,
like
the
the
session
we
did
at
node
interactive,
like
the
you
know,
kind
of
open
session
where
people
can
discuss
topics
and
stuff.
That
would
be
helpful.