►
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
A
Collaborators
summit-
and
we
have
minutes
from
that
and
I-
don't
know
if
it
was
recorded.
I,
don't
know
if,
like
that,
was
something
that
I
was
supposed
to
do
myself.
I,
don't
know,
but
I
didn't
do
anything.
So
if
somebody
has
a
recording,
maybe
that
would
be
cool
to
share
but
not
entirely
necessary.
A
A
Alright,
so
one
thing
that
came
up
and
was
there's
some
talk
about
if
you
look
at
issue
78,
there's
some
criticism
of
the
renderer,
recursive,
API
and
so
I
think
we
need
to
take
that
into
account
when
we're
thinking
about
implementing
these
recursive
api's
and
there's
in
that
issue,
78,
there's
I
added
some
quotes
that
I
think
we
need
to
answer.
Maybe
there
are
more
questions
that
we
need
to
answer,
and
this
may
be
a
topic
that
is
better
suited
to
like
a
deep
dive
session.
A
Yeah,
if
we
want
to
spend
time
trying
to
trying
to
hash
this
out
now,
we
can
it's
just
kind
of
top
of
mind.
I,
don't
know
what
do
you
think
it
would
be
nice
to
like
schedule
that
meeting
I
think
later
so
maybe
more
more
stakeholders
or
what-have-you
would
show
up
instead
of
just
kind
of
going
over
it
right
now,
like
maybe
the
fs
group
in
node,
and
maybe
somebody
the
you
know,
people
who
open
that
that
issue,
if
there's
a
reference
to
an
issue
in
the
node
repo
about
it.
A
C
B
C
A
Well,
it's
still
marked
experimental
so
and
I
didn't
even
realize
that
it
was
marked
experimental,
but
but
it
is
and
I
don't
think
it's
necessarily
because
it's
full
of
bugs,
though
it
may
have
bugs
it's
it's
that
you
know
the
the
API
is
not
like
it
doesn't
do
what
some
people
would
expect
that
it
should
do.
I
think
is,
or
it's
just
like
a
the
quote
was
a
the
API
is
clumsy,
which
I
mean
I.
A
Yeah
I
can
see
issues
with
it,
but
then
again
we
came
to
a
decision
to
do
what
we
did,
because
you
know
it's
going
to
be
very
difficult
to
to
like.
Maybe
all
this
stuff
should
go
in
a
new,
a
new
module,
but
that's
gonna
be
hard
to
get
done.
So,
if
not
impossible,
I
think
there's
just
like
a
yeah.
We
we
tried
but
yeah.
So
it's
like.
What
can
you
do
if
you
can't
do
that?
D
A
C
She
would
update
the
documentation.
I,
don't
know
like
I
feel,
like
I've
had
seen
a
lot
of
pushback
on
this
recently
on
Twitter
I
think
is
revenue.
Reading
is
just
this
like
there's
this.
No
one
was
built.
C
Technic
and
I
do
agree,
I
if
I
could
couch
my
comments,
but
but,
like
you
know
what
kind
of
like,
how
does
this
philosophy
of
tiny
modules,
but
there's
now
this
this,
like
dualistic
kind
of
pushback,
that's
happening
that
people
are
trying
to
write
these
pure
DSM
modules,
and
now
they
realize
they
have
a
module
tree
of
200
things
that
they
would
move
over.
We
need
to
move
over
to
ESM
before
they
can.
C
You
know,
and
they
have
to
talk
to
people
from
15
years
of
notes
or
10
years
of
nodes
history
before
they
can
update
that
entire
module
trees.
There's
like
this
push
and
pull
right
now
the
keeping
dependencies
to
a
fairly
small
number
of
dependencies,
which
isn't
necessarily
an
agreement
with
small
modules
they're
like
a
tiny
module
philosophy
with
kind
of
the
pushback
we
sometimes
get
on
like
it.
Extending
any
of
the
API
is
in
the
core
language
itself
to
be
more
functional,
so
I,
don't
know
how
to
keep
both
like
keeping
both
camps.
A
C
C
E
I
kind
of
made
this
comment
on
the
ticket,
but
I
think
it's
good
to
be
by
default.
It's
kind
of
one-to-one,
but
I,
don't
really
see
a
problem
with
having
a
you
know:
optional
flag
that
can
be
specified
that
will
make
the
function
do
more,
which
that's
the
case.
With
everything
recursive
we've
done,
you
have
to
specify
an
extra
option
to
get
more
than
what
POSIX
API
is
would
give
you.
F
C
Not
chartered
but
like
I
feel
like
I,
like
the
philosophy
of
that
like
alright
I'm
gonna
choose
my
like
I'm
gonna
choose
Express
versus
FASTA
file
versus
you
know,
rest
rest
of
hi
or
whatever.
That
philosophy
makes
a
ton
of
sense
to
me
about,
like
am
I
really
gonna
bring
I.
Have
my
own
favorite
recursive
file
remover,
like
that's,
not
don't
need
the
options
there
and
what's
happening
as
a
community
is
we're
seeing
as
we
mature
to
be
ten
years
old
or
whatever,
like
a
lot
of
these
alternative
options.
A
C
E
E
Basically,
it's
the
AAP
I
should
not
be
able
to
directly
remove
a
file.
It
should
be
able
to
remove
a
directory
that
contains
files
but
like
if,
if
I
do,
FS
dot
are
under,
you
know,
dot,
slash
package.json
I
mean
the
documentation,
says
that
that's
an
error,
adding
the
reversed,
like
I,
think
shouldn't
change.
That
I
think.
C
What's
kind
of
confused
I
agree
with
that
Corey
well,
I
would
say
like
what
happens
if
I
do
that
with
look?
We
want
to
go
to
the
UNIX
arguing
what
happens
if
I
do
RM
dash
R
and
give
it
a
file
like
you
know,
to
me
like
there
is
a
there's,
a
pretty
good
logical
argument
for
recursive
as
a
plague,
which
is
that
that's
how
RM
yeah.
E
That's
just
it,
though,
the
the
RM
is
a
seal.
I
command,
which
you
can
do.
Rm
file,
dot,
j
s
and
it'll
delete
that
so
adding
the
recursive
to
delete
file
doesn't
really
I,
think
change
anything.
But
if
you
drop
down
to
the
well
like,
if
you
use
the
RM
durka
man,
then
I'd,
imagine
that
that
should
fail,
and
if
you
use
the
like
the
actual
API
is
it'll
definitely
fail.
A
A
G
E
C
It
is
annoying
that
it
creates
an
extra
work
for
someone.
No
like
it.
Cuz
like
I,
mean
the
main
use
case
that
looked
when
I
mentioned
of
people
who
do
tooling
at
Google.
The
these
recursive
functions
trying
to
exist
I
think
that's
the
most
excited
thing.
They
look
like
it's.
The
most
excited
extension
to
the
node,
API
I
think
that
they
have
had
on
their
radar
recent
years
because
they,
this
is
the
reason
they
pull
in
a
Tim
raft
and
every
module
they
write
or
make
trip
and
every
module
they
write.
C
C
H
D
H
There's
there's
this
whole.
We
have
this
history
of
a
bunch
of
the
commands,
mapping
fairly
directly
to
lower-level
either
system
calls
or
unix
commands
right
and
a
bunch
of
them
that
don't
necessarily
now
copy
file
does
correspond
to
a
low-level
call
on
on
most
platforms,
but
not
all
platforms,
if
I'm,
if
I'm,
not
mistaken,
so
I,
think
adding
another
top
level
function
to
FS
to
perform.
Some
functionality
that
could
be
replicated
with
other
functions
is
totally
within
scope
and
totally
within
the
history
of
node
right.
H
C
C
C
D
Yeah,
this
has
been
on
my
mind
as
well,
just
because
I
think
I
volunteered
to
start
looking
into
the
recursive
file
copying
so
as
soon
as
I
started
thinking
about
it
every
day,
I
realized,
where
do
I
put
this
but
I
I'm,
definitely
in
favor
of
it
being
like
a
different
like
copy,
derp
or
something
but
I.
Remember
what
happened
with
rim
raff
and
I.
Don't
exactly
wanted
to
do
that
again,
I.
A
C
Mommy
just
come
back
with
some
numbers
and
say
like:
let's,
you
know
what
I
think
this
is
a
fair
point
if
I
would
have
deleting
a
file
like.
Maybe
we
should
think
about
the
behavior
around
this,
but
look.
We've
run
numbers
on
github
and
we
see
this.
This
we've
seen
a
recursive
flag
used
on
40,000
github
repositories
like
like.
Let's
find
that
number
I
know
how
to
I
have
I
know
how
to
run
that
number.
E
So
another
comment
just
about
like,
for
example,
creating
a
separate
function.
I
think
that
might
make
sense,
for
you
know,
copy
tree
or
copy
directory,
since
that's
kind
of
a
different
function
than
copying
an
individual
file,
whereas
our
ender,
a
recursive
you're.
Basically
just
saying
it's:
okay,
if
there's
stuff
in
this
directory,
just
delete
it
and
everything
in
it.
E
Taking
the
RM
durand
the
MK
door
and
removing
the
recursive
flag
and
creating
separate
functions
for
each
of
those
is
two
functions
in
FS
and
a
function
and
stop
promises.
So
at
a
certain
point
like,
if
we're
looking
to
do
this
for
chmod
CH
own,
how
many
functions
are
we
gonna
get
away
with
adding
to
the
FS
namespaces.
D
D
Yeah
yeah
I,
just
I
think
in
some
cases
like
that
makes
sense,
whereas
like
adding
recursive
to
copy
file
just
because
of
the
name,
it
doesn't
make
sense
right.
So
maybe
it's
just
like
case
by
case
we,
maybe
maybe
the
recursive
flag,
is
kind
of
the
default
if
it
makes
sense
and
if
not,
we
just
make
a
I.
C
It
looks
like
we
have
experimental
called
out
pretty
loudly
for
a
render.
We
don't
have
experimental
on
Victor,
which
isn't
currently
liked
I
mean
that's,
not
I.
Don't
think,
there's
currently
an
issue
litigating
that
anyway,
so
it's
more
like
how
do
you
get
Rimmer
have
to
make
sure
people
are
happy
with
it
so
flag.
A
A
C
A
E
I
just
want
just
add,
as
far
as
searching
for
users
of
the
recursive
are
under
I'm,
using
it
on
private
projects
and
I'm
sure
others
are
as
well
and
in
the
case
of
private
projects.
I'm
have
no
obligation
to
anybody
to
support
older
versions
of
notes.
So,
basically,
as
soon
as
api's
are
available
and
functional
I'll
tend
to
use
them,
but
obviously
not
the
first
concern,
but
it
exists
and
really
can't
be
counted.
C
I
will
just
say
where
we
may
be
screwed
up
here
and
we
should
scripts
around
things.
We're
learning
experience
here
is.
We
should
have
been
loudly
printing
experimental
to
the
terminal
for
RIM
rouse,
because
then
yeah
as
long
as
you're,
not
getting
ugly
warnings
and
your
terminal
people
are
gonna
use.
It.
A
A
A
All
right,
so,
let's,
let's
move
along
the
next
one.
Man
is
73,
deep
dive
meeting
proposals,
and
so
there
have
been
a
couple
comments
here
about
some
cool
ideas.
Maybe
there
haven't
been,
there's
been
one,
it
was
been
so
there's
two
one
of
them
would
be
this
recursive
FFM
der
stuff
or
maybe
extending
that
to
just
I.
A
A
A
A
C
Paul's
gonna
be
really
good
topics:
cuz
I'm
good,
yes,
everyone
might
be
good
just
because
I
feel
like
we
keep
running
in
circles
around
it
like
like
is
the
it's
the
hooks
to
catch.
It
are
the
super
mega
hooks
or
whatever
you
called
them.
Is
that
or
is
that
gonna
be
the
I
mean
we
never
need
to
do
anything
with
yes,
I'm
reloading
into
super
mega
hooks
can
do
it.
We
keep
I
feel
like
I'm
blathering.
Yes,
let's
talk
what
yes
I'm
reloading.
A
E
A
E
Isn't
that
kind
of
already
supported
by
essentially
using,
like
you
know,
question
mark
random
ID
equals
you
know
incremental
number?
Yes,.
C
That's
what
I
was
poorly
trying
to
say
hey?
Is
that
like
it
would
be
interesting
to
have
a
deep
dive
where
we
talk
about
those
hacks
and
decide
if
we
could
protect
people
from
it
being
too
horrible
like
it,
you
know
we're
basically
saying
we're
gonna,
give
you
a
memory
leak.
If
you
want
to
use
something
like
proxy
query,
you
can
do
with
the
memory
leak
rate
right.
A
E
I
E
C
A
C
A
A
C
F
C
F
A
A
A
An
attendee
I,
don't
know
how
do
you
say
your
your
name
mentioned
this
one.
He
commented
here.
I,
don't
know
the
person
who
created
this
issue,
but
it
is
71,
so
recommendation
for
config
file,
location
formats,
I
mean
I.
Don't
think
we
can.
One
thing
I
can
tell
you,
it
would
be
silly
is
to
try
to
come
come
together
of
what
what
a
config
file
the
the
config
file
format
should
be.
There's
fifty
thousand
ways
to
make
a
config
file
and
I.
A
Don't
think
it's
worth
time
spent
like
trying
to
think
about
it,
but
I
I
do
see
here
that,
like
it,
would
be
helpful
to
recommend
things
right.
So
one
of
the
things
here
is
this
xdg
base
directory
spec,
which
is
cool
if
you're
on
Linux
I
have
no
idea
what
what
that
is
on
Windows.
It's
probably
nothing
probably
has
no
meaning.
You
know.
Different
systems
want
to
put
this
type
of
thing
in
different
places,
and
so
is
this
I
don't
know.
Is
this
cool
to
talk
about?
Do
we
want
to?
A
Should
we
take
it
upon
ourselves
to
try
to
say
well
if
you're
gonna
use
a
config
file
in
your
tool?
Maybe
you
should
put
it
here
and
if
you
can't
put
it
there,
put
it
somewhere
else,
maybe
if
you
need
a
cache,
you
should
use
this
directory.
Would
it
be
helpful?
You
know,
for
the
I
mean
yes,
I
think
it
would
be
helpful
for
the
community
to
have
some
sort
of
guidance.
I,
don't
know
like
how
to
provide
that
guidance,
or
if
this
we
should
be
the
people
necessarily
doing
it.
I
don't
know.
C
I've
seen
the
module
working
groups
do
some
stuff
along
these
lines
where
they
give
kind
of
one
form
recommendation
for
which
was
when
I
was
reading.
I
read
a
really
good
recommendation
about
you
know,
always
adjacent
to
the
module
working
group
and
a
lot
of
people
are
involved,
but
I
think
wasn't
a
separate
org,
but
they
had
given
a
recommendation
on
what
node
version
uses
support
and,
like
long
term,
long
term
version
support
policies.
It
was
a
good
read
that
was
great
I,
like
this
idea.
G
But
I
spotted
this
on
this
group
and
that's
why
I
joined
and
I
think
it's
just
as
good.
If
it
happened
in
any
other
group,
what
it
is,
it
needs
people
who
are
maintaining
these
modules,
standing
behind
that
and
yeah
there's
a
suggestion
of
a
module
here.
I,
don't
know,
I,
never
used
it,
but
just
if
one
or
more
modules
started,
you
know
looking
in
the
dot
car
think
folder,
for
example,
before
looking
at
the
root
folder
or
even
after
looking
at
the
root
folder.
C
Jason,
so
that's
like,
on
my
mind
immediately
when
I
we
talked
about
this
topic,
is
the
finicky
logic
I'm
using
the
library
like
find
up
to
detect
the
nearest
config
to
ever
the
heck
you're
trying
to
do
something
and
like
mocha,
has
to
implement
that
Justin
will
have
to
implement
that
it
ca
implements
that
so
it's
this
finicky
logic
that's
implemented
fifty
times
over
by
people
and
and
like
those
types
of
these
kind
of
baked
in
assumptions
about
where
it
finds
those
config
files.
So
that's
an
interesting
topic.
A
Yeah,
it's
I,
so
yeah
I
mean
I.
Think
all
of
us
probably
have
have
been
frustrated
with
just
the
the
sheer
amount
of
crap
that
ends
up
in
the
root
of
our
working
copy.
Right,
I'm.
Sure
half
of
us
have
probably
written
their
own
tool
to
try
to
do
something
about
it.
You
know,
but
ok,
so
a
couple
things
in
no
order.
One
ultimate
mega
hooks
could
solve
this
because
you
could
fake
fight
the
file
system
right.
A
So
when
you
go
and
you
want
to
look
for
something
in
the
root-
well,
maybe
it's
not
actually
there,
but
you
could
tell
the
phony
file
system
to
look
in
this
other
directory
and
that
would
that
would
solve
that,
and
so
there's
there's
there's
also
been
two
different
things.
There's
one
there's
this
community
like
recommendation:
this
is
what
you
should
do.
A
Custom
loaders
to
load
y
am
o
files
tamil
files,
instead
of
just
j
s
and
jason
stuff
right,
and
so
that
that
goes
out
into
all
sorts
of
territories
of
various
groups
around
node.
But
you
know
we're
we're
what
what's?
What
should
we
be
focusing
on?
You
know.
The
package
maintenance
group
is
a
really
good
group
for.
A
So
yeah,
it's
I
know
it's
I
think
that
would
be
a
really
fun
one
to
look
at
further,
and
you
know,
maybe
even
this
is
something
that
we
might
want
to
spend
more
time
on.
Just
kind
of
you
know,
maybe
brainstorming
and
and
there's
certainly
tons
of
things
that
we
could
do,
which
of
them
that
we
actually
should
be
doing
and
like
what.
What
do
we
want
to
spend
our
limited
time
on.
G
G
It
doesn't
need
like
what
a
good
starting
point
be
at
least
and
I'm
sure
some
of
you
also
maintain
some
tools.
Would
a
good
starting
point
would
be
to
start
some
issues
to
start
using
that
folder
in
the
tooling
that
we
know
that
could
potentially
use
that
just
start
putting
stuff
in
the
not
putting
stuff
reading
stuff
from
that
folder
start
updating
that
and
see
if
that
picks
up,
as
opposed
to
doing
something
and
something
official
and
something
that
you
know.
Here's
the
group.
Here's
a
committee,
we
decided.
A
C
F
C
Mean
I
like
having
my
config
under
version
control,
so
the
thing
that
I'm
even
jumped
out
at
me
is
why
the
heck
don't
I,
have
like
a
dot
config
folder
in
my
repository
or
something
that
I
shove
all
those
things
in
and
then,
as
long
as
like
I
mean
yards
used
by
a
lot
of
people,
and
it
has
the
opinion
it
uses,
the
yard
says,
find
up
built
into
it.
I
doesn't
find
up
looking
at
config,
folder
I
don't
have
to
when
they
gonna
shove.
C
A
E
A
A
E
So
my
comments
on
this,
like
I,
wouldn't
necessarily
be
opposed
to
making
this
possible
in
NYC,
for
example,
but
I'd
want
to
kind
of
get
a
bunch
of
tool
maintainer
x'
together,
you
know.
Obviously
we
have
mocha
here,
but
you
know
just
roll
up,
wet
pack
and
basically
say
hey.
Are
you
willing
to
make
it
so
that
this
file
will
be
found
by
default
and
I'd
like
to
see
a
critical
mass?
You
know
before
I
essentially
make
it
so
NYC
is
stuck
with
that
forever.
A
C
Anyways
yeah
I
agree
with
Corey's
like
I.
Think
it's
something
worth
talking
about
more
though
I
think
that
would
be
nice
to
clean
up
that
folder
slightly
different
than
the
xdg
based
or
like
I
think
I'm
talking
about
like
our
C
files
in
a
repository
not
so
much
about
like
our
credentials
file,
that's
found
at
the
system
level.
This
is
like
a
different
problem:
yeah
yeah
yeah,
the.
D
H
If
we
can't
get
enough
of
a
critical
mass
behind
this,
because
I'm
worried
about
things
like
various
CI
tools
and
maybe
things
that
have
not
as
much
necessarily
a
thing
to
do
with
no
Jess
they're,
just
part
of
our
build
life
right
or
our
tooling
life,
and
you
know
trying
to
agree
as
a
node
community
to
go
through.
This
kind
of
effort
might
be
one
thing,
but
it
would
have
to
be
much
bigger
in
order
to
actually
deal
with
the
problem
in
any
sort
of
completion.
A
I
get
what
you're
saying
Brian
I.
Yes,
it's
it's
not
just
our
little
ecosystem,
our
big
ecosystem.
That
would
have
to
do
this,
but
it
doesn't
mean
hey,
I,
don't
think
it
would
mean
it's
not
worthwhile
to
do
if
JavaScript
community
is
the
only
one
who
adopts
this,
but
also
we've
led
the
way
on
on
other
things.
Before,
like
we
led
the
way,
maybe
even
I
don't
know
semantic.
Versioning
is
a
virus
now,
because
it's
all
over
the
place,
I
was.
C
Gonna
make
that
point
like
I
think,
like
I've
seen
these
these
kind
of
grassroots
things
work
really
well
like
coverage
like
not
so
many
people
know
no
community
did
coverage
until
NYC
was
out
and
just
made
it
so
easy
people
started
doing
it
like
and
I've
seen
this
happen
a
few
times
NPM
was
viral
like
MP
MP
Isaac
just
started
putting
package.json
than
a
ton
of
a
ton
of
project.
So
it's
a
really
good
idea.
I,
don't
know
I
put
up
with
so
many
configs
in
my
darn
root,
folder
I'm
now
angry
about
it.
D
C
C
B
So
I
quickly
know
that
I've
been
mulling
over
a
NPM
RC
JSON
file
as
an
RFC,
so
an
expansion
on
NPM
RC
files,
which
actually
would
potentially
allow
us
to
clean
up
config
for
NPM
in
M
PM's
world
I'm,
not
sure
if
we
could
expand
on
that
too.
You
also
have
you
identify
your
config
folder
if
that
for
no
world
plan,
but
it's
so
yeah.
This
is
I,
don't
know
if
you
can
ever
solve
this
unless
you're
gonna,
like
symlink
everything
back.
C
We
didn't
cover
that
many
topics
in
this
week's
meeting
funny
enough
I've
gotten
very
excited
about
this
one,
that's
easy
to
derail
this
week,
but
this
is
to
me
this.
This
is
a
great
topic
of
conversation.
I,
don't
know
to
me
honestly.
It
sounds
like
probably
getting
that
momentum
like
Cori
suggests
with
package.
Maintainers
would
make
sense,
but
a
bunch
of
them
are
in
this
meeting.
So
this
is
a
great
place
to
bring
that
up.
E
Yeah
so
part
of
the
point
of
me
bringing
that
up
so,
for
example,
to
me,
dot
config
make
sense,
but
maybe
the
author
of
roll-up,
you
know,
says
hey.
You
know
that
make
sense
because
of
this
and
I
like,
for
example,
for
NYC
to
support
loading
in
the
config
from
a
subdirectory
I'd
like
to
get
it
right.
The
first
time
would.
B
E
B
C
I
think
that
I
mean
there's
a
value
in
consolidating
config,
nm,
pm's
level,
but
I
think
to
exactly
the
Cory's
point
like.
If
we
pick
something
really
logical,
then
I
think
you'd
start
to
see
it
a
little
by
rail
on
github
and
so
laying
the
go
community
would
be
doing
it
and
the
Ruby
community,
with
you
doing
with
if.
D
That
I
guess
two
things
there
one.
It
might
be
good
to
see
if
there
is
any
kind
of
convention
for
this
any
other
ecosystems
and
well
I.
Personally,
really
like
that.
Config
I
guarantee
there's
some
like
framework
or
something
out
there,
I
like
that's
where
they
store
their
config
and
that's
gonna
break
for
them
or
whatever,
but
yeah.
G
D
C
G
A
C
I'd
be
curious
to
have
someone
research,
this
smart
I
like
it's
not,
it
seems
like
something
where
maybe
it
is
just
a
matter
of
people
saying
hey
shoot,
we
do
have
10
things
in
our
top
of
a
folder.
Maybe
we
should
just
have
a
different
convention
for
that
I.
Think
it's
a
great
conference.
I
know:
let's
do
some
of
the
research
sounds
like
you've
been
bitten
by
it
really
hard
you
in
and
I'm
gonna
pronounce
the
name
wrong,
but
it's
a
domine
or
dominant
that
dumb,
because
it's
the.
C
A
Darcie
it
was
your
question
answered.
The
scope
of
the
problem
is
beyond
node
packages,
not.