►
Description
A
C
E
Yeah
I
was
just
being
myself:
yeah
I
continue
to
work
on
the
e
testing
in
the
node
tree.
It's
working
on
several
Linux
platforms,
but
have
to
figure
out
why
it's
not
working
on
arm
still
working
on
the
release
machine
for
for
big
endian
BBC,
some
libya
v
clean
up
for
AIX
and
some
preparation
for
the
upcoming
VMA,
a
p.I
face-to-face
and
next
is
Trevor.
Noah
Norris.
F
H
A
Seeing
a
patentee
lots
of
reviewing
going
on
so
I've
been
spending
in
a
northerly.
The
woman
amount
of
time
on
the
OSX
build
slaves
they're
having
some
drama
over
there.
Hopefully
that's
resolved
for
the
time
being,
but
we
are
in
the
process
of
trying
to
find
a
new
supporter
of
the
OSX
build
slaves,
so
somebody
that
can
put
some
money
towards
either
Colo
situation
or
a
large
corporate
that
can
actually
host
them
somewhere.
So
that's
yeah,
that's
ongoing
stuff.
A
C
C
D
So
I've
been
working
on
some
tools
to
try
to
understand
or
wide
what
the
activity
is.
So
there's
a
few
in
the
tsc
repo
about
that.
That's
like
a
30
day
report
across
the
org,
but
I
think
it's
going
to
be
good,
we'll
be
able
to
get
some
good
kind
of
metrics
and
graphs
and
maybe
even
some
kind
of
reporter
alerts
on
a
regular
basis
about
activity
Oregon
project
wide,
because
this
is
getting.
D
M
N
P
A
Right
that
des
landes
landed
in
all
the
LTS
branches-
yeah,
that's
right!
Yep
now
josh
Josh
Quebec
from
Microsoft
we,
this
is
normally
a
ssl
stand
up
section
as
you've
seen.
This
is
probably
good
time
for
you
to
introduce
yourself
and
tell
us
what
you
do
and
how
you
might
be
able
to
contribute
here
in
future
and
welcome
by
the
way.
C
C
A
Well,
just
will
let
you
fill
with
that
end
and
if
we
get
your
audio
back
at
some
point,
will
the
little
different
introduction.
But
let's
move
on
from
here,
so
a
quick
review
of
last
main
last
meeting.
We've
got
the
docs
test,
pester
the
testing
working
group,
ratification
that
was
done
by
the
CTC
and
then
yesterday
it
was
done
by
the
build
working
group.
So
that's
done
quick
update
on
illegals,
nothing
substantive.
A
First
of
all,
the
es
module
stuff
now
I
didn't
Bradley
I
didn't
tell
give
you
a
chance
to
do
any
stand
up
or
anything
but
I'm,
hoping
that
most
people
know
who
you
are.
So
what
Bradley?
Why
don't
you
tell
us,
ladies,
didn't
tell
us
what
the
status
of
that
EP
is
and
and
then,
where
you
think
we
go
from
here.
Q
Q
File
resolution
has
undergone
tweaks
because
of
browser
requirements,
which
is
a
little
weird
to
say,
but
it
does
give
us
some
advantages
and
now
we're
at
two
final
decisions.
Really,
if
one
is
how
we're
going
to
detect
the
mode
of
any
given
source
code
and
the
other
is
if
we
want
to
allow
inner
assets
to
be
available
from
packages-
and
that's
talking
about
things
like
readable
stream,
/a
through
or
pass
through,
I
think
it
is
thought
jeaious
where
you
can
require
a
file
inside
of
a
different
package.
Q
Those
are
the
only
two
things
really
in
discussion.
Now.
The
rest
is
ironed
out.
Circular
dependencies
are
a
little
bit
of
a
gotcha,
but
they're
documented
in
can
throw
errors
with
nice
error
messages.
So
that's
the
current
status
of
things
and
I
guess
I
should
discuss
what
we're
talking
about
when
we're
detecting
file
mode
and
why.
A
Q
Q
Q
Q
So
both
have
some
Universal
disadvantages
just
because
we
have
to
default
to
common
jayus,
but
the
ones
unique
to
the
file
type
are
going
to
be
be
the
fact
that
anything
that
uses
like
a
HT
Access
file,
a
that
points
to
star
j
s
will
have
to
change
and
that's
where
some
people
are
very
afraid.
That
servers
won't
update
that
and
we
don't
have
a
registered
file
extension
against
IA
na.
But
Jordan
Harvard
has
promised
to
fast-track
us
through
that.
If
we
choose
that
so
so,.
D
D
Suffered
one
of
these
file
changes
successfully
in
j/s.
So
if
you
look
at
like
other
plugins
that
have
come
about,
that
did
say,
action,
script
and
other
stuff,
they've,
basically
failed
and
part
of
that
was
because
of
this
new
files
extension
or
at
least
that's
what
they
would
say.
It's
it's
much
more
than
a
c-axis,
literally
anything
that
does
file
detection,
DN
extension.
So
every
code
editor
every
IDE,
every
compliance
tool,
checker
everything
that
ever
looks
through
code
and
looks
at
that
extension
to
figure
out
what
type
that
is.
That
would
all
need
to
change.
Q
J
Q
D
C
D
C
G
L
Q
Okay,
it's
we're
really
talking
about
scripts
and
things
that
don't
parse
it
don't
parse
the
JavaScript
itself
right
now
are
going
to
be
the
ones
that
are
heavily
affected.
That
would
not
be
affected
by
package.json,
so
rewrite
rules,
HT
access
files,
bash
scripts,
the
rails,
asset
pipelines,
a
big
one.
Django
has
a
similar
thing.
G
G
What
I'm
wondering
is
like
did
you
consider
something
like
you
strict,
like
you
know,
a
basically
a
string
literal
in
the
top
of
the
package
that
indicates
the
type
of
it
or
like
in
the
individual
javascript
file
that
we
want
the
question
one
and
question
two
is
like
what
are
these
edge
cases
that
require
us
or
to
like
lift
every
single
javascript
file
in
the
package.json?
That
exposes
a
module
interface?
F
G
D
D
Says,
there's
really
only
two
options
like
really
believe
that
there
only
to
rob's,
it's
like
everything
possible,
is
considered
and
thrown
out
like
on
the
way
to
also
just
just
to
make
it
a
little
bit
clearer
on
the
on
the
detection
part,
if
it's
the
entry
point
for
the
module
right,
so
most
modules
are
really
only
exported
via
like
one
file,
and
so
it
would
be
module
and
that
file.
If
you
want
people
to
be
able
to
do
like
blah
/
another
file,
then
those
would
need
to
be
in
there
as
well
right.
G
Q
So
it's
not
just
like
readable
stream
/
pass
through
jas
that
we
need
a
whitelist
it
any
file.
That
is
the
module
goal,
because
there
is
a
grant.
There's
a
grammar
ambiguity
between
the
two
and
so
I
think
it
was
Chalker
tried
to
run
npm
test
because
you
auto
upgrade
to
strict
mode
on
the
top
100
mpm
modules,
and
it
was
a
horrible
disaster
through
silent
failures,
things
would
parse
properly,
but
they
acted
incorrectly.
Q
Otherwise,
we
can
have
silent
failures,
which
is
why
we're
talking
about
glob
patterns
and
the
goal
the
way
to
mitigate
this
eventually
would
be
npm
in.
It
sets
a
star
star,
j/s
or
something
where
every
file
defaults
to
the
module
goal,
though,
but
neither
solution
is
a
silver
bullet,
which
is
why
we're
here
right.
C
Q
A
Ok,
so
for
moving
forward,
the
TTC
needs
to
need
to
be
comfortable
with
this
proposal
and
the
image
of
photonic,
so
the
people,
the
city,
see
members
here.
Do
you
feel
like
you,
have
access
to
enough
information
to
make
that
call,
or
is
there
some
way
you'd
like
to
be
able
to
dive
deeper
into
these
things?
I.
G
R
A
We
just
this
is
just
the
start
of
a
process,
so
just
want
to
make
sure
that
we
are
yeah.
It's
a
complex
issue
that
the
EP
document
is
pretty
is
pretty
complete,
but
we
want
to
make
sure
that
there's
enough
comfort
here
to
be
able
to
make
a
decision
at
some
point
whether
to
move
forward
with
modules
for
start
and
then
exactly
what
shape
it
looks
like
so
whether
we
accept
the
EP
as
it
is.
So
if,
if
anyone
has
any
concerns
about
the
way
this
is
being
conducted,
so
I'm
laying
you
don't
feel
like.
A
C
Q
D
A
Q
That's
the
thing
preliminary
is
going
to
be
necessary,
but
we
gave
a
timeline
really
that
next
week,
so
you
have
a
week
to
think
about
it
and
read
the
EP.
But
roll
up
and
babble
want
to
hear
a
decision
pretty
soon.
This
doesn't
affect
v8.
I
was
talking
to
them
two
weeks
ago.
So
they're,
not
a
big
concern
and,
like
I,
said
what
WG
and
tc39
stepped
out
of
the
issue.
So.
C
Q
So
that
relates
to
a
difference
where
tc39
built
the
syntax
and
semantics
of
the
module
system.
They
did
not
build
the
loader.
So
when
we're
talking
about
inspection,
we
can
inspect
modules
themselves
pretty
well,
but
inspecting
the
registry
and
things
like
that
you're
going
to
want
to
talk
to
the
what
WG
loader
people
in
particular
carroty,
but
no
they're
in
spectacle,
I,
don't
know
what
else
to
say
about
that.
You've
got
a
metadata
pointer.
You
can
throw
anything
in
there.
So
yeah.
D
J
C
Q
When
you
go
and
you
read
through
the
discussions
on
the
CP,
there
are
a
few
cases
where
package.json
does
not
support
all
the
workflows
that
currently
exist
in
node.
So
that's
something
to
keep
in
mind.
The
package.json
approach
is
really
saying
that
we're
not
going
to
change
the
file
type
and
that's
the
big
advantage
to
it.
Q
A
Well,
it's
an
assumption
from
some
people
that
when
they
all
JavaScript
what
tools
you
know
it
will
be
this
new
module
type
and
therefore
all
of
our
files
will
either
have
to
have
an
extension
or
we
can
remove
that
thing
from
the
package.
Jason
I,
don't
know
how
realistic
that
is,
but
I
guess
that's
as
business
option
that
some
people
are
bringing
to
the
discussion.
G
All
right,
I
think
I
have
an
opinion
too,
but,
like
whatever
we'll
talk
about
it
next
week,
I
think
everybody
should
just
read
up
on
a
discussion.
Yes,
we
are
to
have
a
make.
A
decision.
I
would
like
to
ask
one
more.
It's
kind
of
unrelated
to
the
other
things,
but
one
of
the
things
that
note
does
is
like
view
past
modules,
as
you
look
them
and
I
have
always
been
conflicted
as
Fergus.
You
know
whether
this
was
a
good
or
a
bad
thing
kind
of
flows.
G
Q
G
Q
A
A
H
Early
would
up
service
50pc
in
terms
of
process.
You
know
Michael
us
BMC,
TC
TC,
but
you
know
when
we
I've
been
acquired
strong
loop,
we
pull
down
in
the
spec
and
Michael
Dawson
stuff
and
now
that
we
expand
it
again.
The
limit
is
little
little
broader
and
we
can
look
at
that
again
back
on.
So
that's
the
idea
and
I'll
leave
it
up
to
the
rest
of
the
group
to
decide
on
buzzer.
C
A
H
Basically
here
recently
Mason
some
feedback
from
Deb
Wilson.
I
expanded
that
to
include
a
little
bit
broader
handling
of
deprecation
events
and
its
really
intended
to
align
with
some
of
the
things
user
modules
are
doing
either.
My
mother
was
doing
specifically
the
deputy
module
that
is
used
by
Express
based
on
those
changes.
Your
mind
provide
some
feedback.
I
implemented
the
changes
that
Jeremiah
suggested
this
morning,
so
he's
happy
with
us.
Then
I
think
we
could
probably
save
everyone
some
time,
I
think
the
other
you
move
to
Jeremiah's.
You
give
any
remaining
issues.
C
C
Come
so
the
before?
What
before
my
feedback,
there
was
two
different
events.
There
still
is
two
different
events,
so
there's
warning
and
there's
deprecation.
Those
are
two
events
and
the
default
is
to
just
logged
out
to
console
and
that
depending
to
which
event
would
be
emitted,
depended
on
the
type
of
like
the
instance
type
of
the
warning
you
admitted
or
the
name
of
it.
So
if
it
was
a
deprecation
warning,
it
would
not
go
to
warning
at
all,
but
it
would
go
to
deprecation
now.
C
Let's
changed,
though
everything
pipes
to
warning,
but
we
still
have
this.
We
still
deprecation
event
net
right
now,
because
I
guess
some
user
modules
use
it
or
you
have
used
the
similar
thing
and
that
would
hook
that
in
might
help
and
then
we're
keep.
I
guess
the
current
plan
is
to
keep
the
current
deprecation
flogs
and
also
add
warning
flag.
So
if
you
ignore
all
warnings,
you
won't
receive
deprecation
'he's.
Currently,
however,
it
used
to
be
that
if
you
ignored
warnings,
you
would
not
necessarily
ignored
effort
occasions.
C
H
Change
that
I
made
this
morning,
I
suggested
with
Jeremiah,
was
to
make
it
so
that
if
you
use
process
starting
that
warning
to
issue
a
deprecation
warning,
the
warning
event
will
be
triggered.
If
there
are
no
other
handlers
registered
for
the
warning
event
other
than
the
default
one,
then
the
deprecation
warning
will
be
forwarded
to
a
deprecation
event.
H
G
C
H
G
Was
I
was
in
favor
of
the
warning
event,
but
I
do
not
understand
why
there
is
a
need
for
all
this
complication.
Like
you
know,
if
there's
no
warning
handler
will
limit
deprecation
or
like
deprecation
are
emitted
separately,
but
also
as
a
warning.
I
am
I
missing
something
here.
Should
I
read
up
on
that
discussion
as
well.
I
like.
H
That's
actually,
where
I
started
and
that's
what
I
would
put
both
is
just
having
a
single
morning
event
a
if
deprecation
warning
was
added
based
on
the
feedback
from
from
Doug
Wilson
who
dip
the
depth.
D
module
currently
already
emits
process
on
deprecation
events,
so
it
hooks
into
the
fact
that
that
process
is
an
event
emitter
and
uses
that,
instead
of
creating
its
own,
you
got
emitter
for
this
thing.
So
the
idea
here
was
to
get
some
alignment
between
what's
happening
in
node
versus,
what's
happening
in
New
Zealand.
H
R
C
R
A
C
A
H
A
Next
week,
if
you're
going
to
be
expressing
an
opinion
on
this,
then
you
probably
should
have
read
up
on
what
the
status
of
is
so
maybe
James
you
can
provide
a
status
in
the
original
post
of
what
the
cut.
What
the
current
thing
is
in
couple
dot
points.
So
if
you,
if
you
think
you're
gonna
care
about
this
next
week,
we
have
a
vote.
Please
go
when
engaging
the
issue
now,
so
that
we
don't
have
to
go
in
circles
next
week,
even
if
that
means
you
just
don't
want
this
added
like.
A
S
Here
sure
so,
I
am
a
technical
program
manager
in
dev,
dev,
developer
division
here
at
Microsoft
responsible,
I'm
right
now,
coordinating
across
several
of
our
product
teams,
stuff
related
to
node
and
I'm
working
hard
to
get
them
contributing
more
and
more
involved
directly
in
the
and
the
core
project.
Of
course,
I
work
pretty
closely
with
Alexis
for
a
note
on
Windows
and
note
on
windows
server,
and
I
also
I'm
actually
part
of
the
Visual
Studio
code
team
at
this
point.
S
So
I
do
work
with
there's
a
studio
videos
to
your
code
and
some
of
our
browser
tools,
team
and
our
chakra
team,
just
pretty
close
to
me,
I'm
attending
these
meetings
have
to
start
with
to
be
able
to
better
info
people
here
and
to
in
to
coordinate
and
find
ways
where
our
teams
can
contribute
more
I
definitely
am
happy
to
help
right
from
the
start
here
with
admin,
stuff
and
logistics
and
whatever
and
I'm
hoping
to
make
gradually
make
more
direct
contributions
myself.
So
that's
it
thanks.
A
D
Yeah
so
based
on
feedback
last
week
and
really
good
conversation
with
James.
Actually
there
was
a
relatively
small
textual
it
to
the
proposal,
but
a
pretty
substantial
one.
Essentially,
we
have
a
bunch
of
working
groups
that
are
really
just
teams
and
we
have
a
lot
of
unnecessary
process
for
creating
those
teams
because
we're
calling
them
all
working
groups.
So
the
idea
is
basically
just
say
like:
let's
get
rid
of
this
whole
chartering
thing
and
all
this
extra
process
around
organizing
his
team's
and
just
make
it
a
lot
easier
for
people
to
create
a
team.
D
Critter
repo
go
off
and
do
their
own
thing
not
really
have
to
be
chartered,
so
that
dramatically
reduces
the
number
of
sort
of
like
top-line
projects
and
working
groups
and
allows
that
the
teams
to
be
kind
of
more
freeform.
So
that's
the
main
change.
So
now
the
only
things
that
would
you
know
kind
of
bump
up
to
the
top
level
would
be
you
know,
nan
and
I
think
the
build
working
group.
A
D
A
A
Okay,
let's
move
to
a
public
QA
session,
a
public
discussion
we
have
a
number
of
forms:
we've
got
is
eternal.
No,
no
dev
I
think
we're
using
for
this.
So
if
you're
in
my
RC
on
freenode-
and
you
can
hit
us
up
with
questions
or
comments,
there
I
think
also
Jeremiah,
you
might
be
monitoring
the
YouTube
channels
elements.
C
A
Think
Bradley
has
opinions
on
this,
but
no
I'll,
just
quick,
throw
in
mind,
which
is
that
experimenting
with
different
ways
to
attack
these
large
change
proposals.
I
think
it's
healthy
for
us
and
and
and
with
Bradley,
pursuing
a
different
method
for
this
than
the
promise
of
stuff.
I
think,
is
good
for
us
to
learn
which
is
most
effective
in
in
finding
proper
closure
Bradley
at
least.
Q
This
is
an
interrupt
thing.
It's
more
of
a
one-off
deal
once
we're
done
with
the
interop.
There
shouldn't
really
be
any
reason
for
there
to
be
a
module
working
group
after
that
point,
so
we
would
also
have
to
think
about
disbanding
and
what
that
means
for
working
groups
and
that's
just
a
little
complex.
M
For
me
and
I,
dad
I
guess
that
my
impression
has
been
there
working
that
working
groups
form
bottom
up.
The
CTC
does
not
does
not
decide
that
you
know
there
ought
to
be
a
modules
working
group
and
then
forms
one.
People
band
together
form
a
modules,
working
group
and
then
come
to
the
CTC
to
get
chartered.