►
Description
B
Welcome
to
the
new
foundation
CTC
meeting
for
the
seventh
in
February:
where
is
the
82
febri?
We're?
Always
that's
a
bit
of
a
mix
between
seventh
and
eighth,
the
suspect
2017.
We
have
a
fairly
small
group
here,
but
it's
a
difficult
time
for
a
lot
of
people,
I
think,
and
we
only
have
really
been
one
of
them.
Probably
we
can't.
Let
me
discuss
that.
B
C
D
E
B
A
F
H
Hey
everyone,
so
I
talked
with
as
many
people
as
they
could
about
modules.
In
the
last
week,
I
worked
on
the
ABI
smoking
for
v8
5.5,
which
has
just
landed
in
to
v7
staging,
which
is
super
awesome.
I
updated
the
v4
RC
to
include
the
custom
search
stuff.
There
was
a
little
bit
more
work
that
needed
to
get
that
ready
to
be
landed
and
I
worked
on
my
mentee
on
some
note.
Serial
related
stuff.
B
Yeah,
so
no
chousen,
so
the
main
thing
is
I
moved
the
arm
costs
most
of
it
anyway,
still
don't
have
a
permanent
internet
connection
there
dangling
at
the
end
of
a
temporary
one
that
should
be
resolved
tomorrow.
So
hopefully
this
one
tomorrow
there
will
be
all
back
up
and
running
again,
but
I
will
update
that
issue
in
nodejs.
Let's
build
with
that.
J
K
E
Yeah
she
has
a
pretty
clearly
fine
task.
We
have
a
bunch
of
the
M
related
issues,
the
vm
module
and
we
recently
changed
the
v8
api
which
landed
with
55.
So
she
can
fix
a
bunch
of
them
just
working
very
independently
hacking
away
asking
questions,
and
then
we
meet
once
a
week
but
she's
making
progress.
So
she
just
landed
us
sort
of
bigger,
refactoring
and
we'll
see
actual
fix
us
very
soon.
E
H
Could
talk
about
what
I've
been
working
with
my
mentee
Jessica
about,
so
the
project
was
started
as
kind
of
an
examination
of
how
we
can
improve
the
native
module
story
specifically
for
mid
serial.
So
there's
been
a
lot
of
coordination.
That's
been
having
to
happen
between
maintain
errs
of
note
serial
port
as
well
as
maintain
errs
from
Libya
V.
The
idea
that
this
food
is
that
this
would
go
into
something
called
Libya
the
extras
and
breaking
ground
on
something
like
this
is
not
an
easy
task.
H
So,
in
the
meantime,
the
mentee
has
been
spending
a
bunch
of
time
getting
familiar
with
the
note
serial
port
codebase,
getting
into
how
native
modules
in
the
binding
system
works,
which
has
been
a
fun
thing
to
explore
with
them.
We've
also
spent
part
of
the
time
she
spent
a
bunch
of
the
time
looking
at
the
stream
system
and
has
been
putting
together
a
pretty
long,
medium
post
on
the
history
of
streams
and
node,
the
various
iterations,
and
you
know
why
different
changes
needed
to
be
made
and
has
also
been
working
on
the
streams.
K
B
H
B
To
maybe
do
a
note
up
about
this
whole
topic
once
once
it's
finished
anyway,
let's
move
on
and
do
grant
through
the
agenda.
So
first
item
in
the
agenda
is
update.
Copyright
issue
number
174
of
the
tsc
repo,
which
links
to
some
pull
requests
on
the
nodejs,
the
node
repo
for
changing
copyright
hitters.
This
went
to
a.
B
Sorry
asura
done
the
previous
meeting
review
before
this,
so
this
went
to
a
legal
committee
meeting
that
was
held
this
week.
I
didn't
attend
it
because
it
was
I
was
asleep,
but
I
believe
James
was
invited
to
that.
Don't
know
whether
he
did
and
I
haven't
seen
any
notes
from
that.
Yet
so,
unless
anyone
on
this
call
has
information
about
how
that
went,
I
don't
believe
we
can
go
further.
I,
don't
do
anything
you
get
over
better.
B
My
request
was
for
the
to
the
legal
committee
for
it
to
be
a
discussion,
not
just
throwing
things
over
the
wall
at
each
other.
So
hopefully
we
are
at
least
beginning
to
enter
into
a
discussion
about
how
to
proceed
with
this
stand
and
not
just
be
dictating
one
way
or
the
other,
because
I
don't
believe
it
needs
to
be
just
purely
dictating,
because
it's
not
it's
not
purely
black
and
one.
B
Up
to
reviewing
previous
meeting,
and
so
there
was
the
ESM,
the
question
synchronous
async,
which
we
are
dealing
with
again
now
deep,
strict,
equal
and
enforcing
type
check
there.
That
was
hard.
What
did
we
decide
on
that?
One?
That
was
all
that
came
back?
That's
right.
They
came
back
to
the
assert
modules
is
locked,
so
the
question
was:
should
we
unlock
it
and
in
there?
So
that's
for
a
discussion
on
github
issue
number
there.
B
Is
it
changing
the
definition
of
locked,
which
is
in
oj
f,
/
node,
if
you're
number,
1,
1200
I,
believe
miles
is
leading
that
one.
So
should
we
change
locked
for
the
definition
of
that
to
include
things
that
include
es6
new
language
features,
so
it's
locked
except
for
new
language
feet
to
changes
so
see
discussion
there
and
CTC.
B
Please
participate
in
that
one
because
it's
non-trivial
the
next
one
we
talked
about
was
note,
inspect
and
that
was
adding
note
inspect
into
foundation
and
then
whether
it
should
go
in
the
repo
or
be
rendered
in
the
next
I
was
resolved,
yet
must
think
it
might
have
been
leaning
towards
being
vented
process.
Glop
clobbering
process
on
global
internally
in
nodejs
and
whether
we
should
protect
against
that
I
think
we've.
The
way
we
were
going
with.
B
B
M
M
B
M
So
I
have
the
APR
updated
for
the
first
one,
assuming
none
of
those
files
disappear
we're
doing
now,
and
then
that
pair
should
be
ready
to
land.
The
second
PR,
with
the
revised
editor
I'm,
making
a
few
additional
changes
to
it.
We
will
committee
asked
that
we
put
linting
in
place
to
ensure
that
all
files
that
are
in
source
inland
lib
have
that
header
and
also
make
sure
we
document
in
our
contribution
guidelines
that
copyright
header
should
not
be
removed
from
files.
M
H
B
M
H
M
B
M
K
M
B
Okay,
okay,
so
if
you
have
anything
outstanding
there,
please
let
james
know
cuz
he's
in
direct
contact
with
the
relevant
people
or
raise
it
in
I.
Think
your
best
to
use
that
original
tsc
issue.
That's
flagged
for
this
agenda,
yeah
to
discuss
so
onto
esm.
We
don't
have
Bradley
today,
but
the
unbelievably
outstanding
question
here
is
still
a
synchronous
sync
for
the
new
Lola
and
tc39
and
is
what's
on
that
and
it's
quite
vexed.
But
we've
had
some
discussion
that.
How
do
we
proceed
here
is
I?
Don't
have
any
clue
how
we
proceed.
M
Was
just
going
to
say
that
I
think
at
this
point
we've
only
talked
about
last
week
was
that
we
get
folks
at
another
week
to
think
about
it
and
come
up
to
a
vote
miles.
I
know
you
do
is
important
and
you'll
be
other
spread
because
of
a
bit
of
direction
with
the
choir
import
book.
I
think
we
should
come
to
a
vote
if
we
are
able
to,
but
wouldn't
a
see
where
we're
at
in
the
discussion
so
miles
me
in
your
life
I.
Don't
you
have
a
big
I'm
worried.
H
H
H
You
know,
as
it's
used
at
the
top
of
the
module
I'm,
not
necessarily
talking
about
import
as
braces,
which
would
turn
a
promise,
but
I
think
it's
important
that
we
keep
the
asynchronous
loader
to
match
the
browser
behavior.
One
thing
that
I
did
find
out
was
that
the
initial
implementation
that
chrome
is
planning
to
do
while
the
resolve
phase
is
going
to
be
asynchronous.
H
The
actual
execution
phase
of
the
graph
after
it's
resolved
is
going
to
be
deterministic
and
synchronous.
There
are
plans
to
look
into
doing
the
potential
deterministic
resolve
with
deferred
Xin
between
modules,
but
that
will
not
be
part
of
the
original
implementation
and
even
if
the
browsers
do
end
up
implemented
that
the
differences
between
different
sessions
introduce
you
know,
differences
between
each
run,
that's
not
the
same
as
just
a
straight-up
difference
in
behavior
between
platforms.
H
The
other
thing
that
I've
been
thinking
a
lot
about
and
I'm
not
going
to
do
too
much
into
it
because
is
tangential
to
this
particular
vote.
Is
that
I?
Don't
think
that
having
the
top-level
import
and
the
import
keyword
as
using
an
asynchronous
loader
precludes
us
from
maybe
looking
into
some
sort
of
synchronous.
A
A
A
A
M
A
A
M
M
A
M
H
B
H
A
It
doesn't
preclude
us
from
doing
it,
but
you
kind
of
have
the
potential
to
run
into
issues
if
that
dep
tree
has
a
common
j/s.
That
requires
es
modules
or
the
other
way
around
somewhere
down
the
line,
because
in
an
async
or
a
sink
loader
there
is
differences
in
how
those
will
function.
So
one
could,
as
one
of
those
dependencies
down
the
tree,
might
assume
it's
a
sink
and
then,
if
you
try
to
load
it
in
what
would
be
a
sink
way
that
just
won't
really
work
I
mean
we.
M
H
So
the
circular
deb
stuff
is
definitely
where,
like
my
brain,
starts
breaking
on,
unlike
thinking
about
how
this
would
be
solved,
but
you
know
III,
guess
it
really.
Just
to
me-
and
I-
and
I
think
it's
worth
saying
right
now
is.
I
think
that
this
is
a
different
conversation
from
the
conversation
that
we're
having
about
the
import
keyword.
A
Like
to
do
one
set
of
things,
you
kind
of
need,
like
the
other
set
of
things,
and
it
just
gets
weird
like
I
mean
top
level
await,
would
technically
be
like
a
different
conversation,
but
then
like
if
we
do
a
sink
loader
that
could
get
extremely
complicated
with
like
underneath
threading
in
node.
That's
like
sort
of
invisible
stuff.
K
M
M
There's
a
helpful
opinions
that
have
to
say
mean
more
towards
race
encoder
for
success
and
just
keep
it
isolated
from
the
contest.
I
mean
I.
The
easiest
thing
we
are
going
to
be
able
to
do
is
just
not
Nick
sees
these
things.
If
we
can't
import
now
we're
going
to
be
able
to
import
a
conscious,
but
if
we
can't
require
any
effort,
then
okay,
we
just
follow
that.
To
take
it,
we
would
still
require
land
college,
the
Atlanta
and
move.
B
Right
so
we're
heading
for
a
lot
of
abstains
here,
which
is
concern
for
a
significant
vote
like
this,
and
those
abstains
are
either
because
of
limitations
or
technical
as
Daniel.
Simply
people
are
saying
they're,
both
terrible
so
whatever
it
seems
to
me
that
the
sentiment
is
sleep
slightly
is
leaning
more
towards
they
think,
and
the
votes
are
as
well,
but
I'm
not
sure
how
we
should
handle
this
and
whether
we
should
throw
it
back
to
tc39
and
say
that
this
is
all
terrible.
K
M
K
That's
that
that's
actually
exactly
what
we're
doing,
if
I
understood
bradley
correctly
last
week
like
this,
is
not
like
a
final
decision.
This
is
like
telling
bradley
what
to
do
the
next
couple
weeks
to
like
you
know,
and
and
and
we
can
always
like
your
two
weeks
from
now-
be
like
oh
wow.
That
was
a
terrible
decision
we
need
to
reconsider.
Is
that
am
I?
Am
I
am
I
summarizing
that
correctly
Jeremiah
or
anybody
else
who
was
part
of
the
conversation
last
week.
A
A
N
B
C
A
A
K
N
B
H
For
a
second
sure,
so
I
Anna
and
I
brought
this
up
last
week
for
a
second
but
I.
Think
one
of
the
things
for
me
that
I
keep
having
to
remind
myself
is,
we
may
be
a
node
community
here,
as
a
group
of
people
working,
you
know
on
the
CTC
and
the
tc39
may
primarily
be
run
by
browser
vendors
or
people
working
on
you
know
the
various
and
betters,
but
but
the
actual
people
at
the
end
of
the
day,
who
are
writing
code
in
JavaScript
for
the
most
part,
are
running
things
in
both
environments.
H
But
we
lose
the
reason
why
it's
so
easy
to
jump
between
the
two
platforms
and
I
would
fear
that
we
could
lose
a
lot
of
the
users
that
we
have
because
they
use
it
for
built
tools.
They
use
it
for
support
tools
for
the
things
that
they're
doing
in
the
browser
and
if
they
can't
just
jump
between
two
environments
anymore,
why
would
they
use
node
0
for
something
else?.
H
M
M
If
we
can't
then
yeah
we
diverge
and
we
need
to
make
sure
there
will
be
very
deliberate
about
diverting
any
inside.
This
is
exactly
why
in
this
particular
case,
we
have
two
extremely
complex
and
their
tree
incompatible
model
systems
that
we're
trying
to
get
to
work
together
and
every
time
any
of
us
looks
at
this.
We
cannot
find
a
way
to
reasonably
make
it
work
together
without
jumping
through
some
hoops
are
compromising.
It
is
someone
we
decide
at
what
point
are
we
going
to
call
it
and
say
this?
M
Is
the
minimal
set
that
we're
going
to
be
able
to
reasonably
do,
and
this
is
where
books?
If
we
ultimately
come
down
and
say
well,
we
have
to
have
a
simple
odor,
because
we
just
hack
it
right
and
we
just
need
to
be
able
to
be
buried
a
little
bit
about
that
and
and
have
you
know,
rot
assignee
just
stand
up
and
say
this
way
this.
B
B
We
should
always
do
things
that
are
in
the
interests
of
users
like
they
are
foot.
There
are
trivial
primary
concern,
but
it's
also
in
our
users
interests
that
node
is
a
sensible
platform
that
is
performant.
That
makes
sense
that
is
internally
I'm
unified,
not
just
you
know,
there's
some
hack
job
because
it
has
to
conform
to
this
completely
different
environment,
which
is
the
browser
I'm
so
but
either
way.
I'm
not
like
I,
still
I'm,
still
leading
towards
the
async
side
of
this,
but
I'm
standing
can
at
convergys
a
school.
B
M
A
cute
come
on
coming
around
to
that
I
I,
don't
think
we
are
going
to
be
able
to
make
adequate
progress
until
we
actually
have
code
to
look
at
to
see
how
a
reasonable
reasonable
this
approach
is,
and
if
this
decision
right
now
is
to
say,
okay,
let's
put
the
Devourer
team,
you
get
some
code
implemented
to
see
if
it
works
and
I
would
have
to
say,
exciting
a
thing
crap,
let's
see
how
feasible
it
actually
is,
only
get
code
written
if
it
turns
out
to
be
Q
unreasonable.
You
unworkable.
B
B
M
M
B
B
B
B
L
K
Let's,
let's
not
I
mean
it
sounds
like
that.
We
seem
to
the
consensus
we
seem
to
be
reaching
here
is
that
we
should
make
a
tentative
decision
and
proceed
down
a
path
and
I,
don't
know
that
we
necessarily
need
to
go.
You
know,
go
to
vote
root
for
that.
It
does
seem,
like
you
know,
of
the
people
who
have
opinions,
Aly,
Myles,
etc.
You
know
the
you
know
with
with,
with
the
exception
of
Evan,
you
know
it's
all
on
the
async
side.
B
M
E
M
Not
going
to
know
for
sure
we're
not
going
to
be
sure
until
we
get
code
yeah
until
we
actually
start
going
it
down
a
path.
I
I
have
my
suspicions
that
this
is
going
to
just
prove
to
be
too
too
difficult.
It's
possible
I
mean
everything
possible,
throwing
up
code
edit,
but
I
think
it's
going
to
be
too
difficult,
and
my
personal
hypothesis
is
that
we're
going
to
switch
we're
going
to
go
back
to
can.
K
M
K
Really
trying
all
right
so
so
I,
don't
I,
don't
I.
Don't
expect
us
to
make
a
whole
lot
of
progress
on
this
at
this
point,
so
I
wonder
if
I
would
like
to
propose
that
what
we
do
is
I'll
go
ahead
and
they'll
take
the
the
agenda
label
off
of
the
EP,
since
it's
not
really
on
the
EP.
It's
not
really
about
the
EP
anyway
and
open
up
a
separate
issue.
That
is
just
you
know
us
monitoring
the
async
implementation
and
being
sure
to
LIKE.
K
Although
I
can't
imagine
we
would
forget
to
talk
about
this,
but
to
you
know,
discuss
it
every
week
and
you
know
especially
once
Bradley's
back
from
his
vacation
or
whatever
he's
doing
you
know
and
and
sort
of
just
try
to.
You
know
revisit
this
decision
in
miniature
every
week
and
hopefully
maybe
we
can
like
try
to
time
box
the
conversation
to
15
minutes
and
then
allowed
buzzer
sounds
and
we're
not
allowed
to
talk
about
it
anymore
or
something.
H
I
B
B
Okay,
let's
close
that
one
up
in
got
five
minutes
left,
but
is
there
anything
else
that
not
only
generative
people
would
like
to
cover
in
the
public
meeting
and
in
while,
while
I
ask
that
as
well,
anyone
listening
on
the
public
under
the
YouTube
feed?
If
you
have
something
to
ask
or
say
to
us,
then
please
leave
a
comment
there
or
in
one
of
us
in
IRC
and
we'll
give
that
equal
two
minutes.
But
is
there
anything
else
anyone
has
anything
getting
anything
to
discuss
the
public
meeting
here,
I.
M
A
M
Flanked,
oh
we're
waiting
on
that.
The
other
thing
as
I
opened
a
PR
today
kind
of
restarting
the
effort
or
continuing
the
effort
on
the
error
codes
part.
I
took
a
separate
PR
just
to
land
the
base
changes,
not
any
of
the
actual
error
changes.
Just
the
the
base
coat
for
creating
the
errors
need
to
have
that.
Take
a
look
at
and
the
plan
would
be
to
do
all
of
the
actual
error
codes
themselves
as
separate
PRS
and
small
chunks.
H
So
to
the
person
watching
on
YouTube,
who
asked
the
question
about
sensors,
no
doubt
of
the
box
did
not
do
anything
with
serial
or
with
hardware.
There
are
a
bunch
of
projects
that
you
should
check
out.
One
is
Johnny
Five
and
the
other
one
is
tehsil
and
tehsil
is
published,
a
bunch
of
modules
for
specific
sensors
that
you
can
NPM
install,
and
that
sounds
like
the
kind
of
solution
that
you're
looking
for
I'm,
so
I
highly
advise
checking
that
out.
It's
tehsil
te,
SSG,
l,
dot
io,
I
believe,
is
the
web.