►
From YouTube: Node.js Diagnostics WG Meeting - 2018-10-31
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
Yeah,
so
we
had
was
that
I
guess
we
talked
about
it.
I
think
there's
a
few
things
in
the
comments
here.
Michael
you
said
you
were
gonna.
Have
some
reach
to
drive
I,
don't
know
like
I
think
somebody
just
needs
to
go
through
and
wrangle
this
and
be
like
here's,
where
it's
at
right.
B
B
B
A
We
have
quorum
ends
is
like
worth
it.
If
you
know
we've
got
like
a,
you
know,
I
mean
I,
don't
know.
Maybe
the
majority
of
people
here
are
on
West
Coast
time,
because
this
is
a
good
time
for
people
on
the
west
coast
right
and
we've
had
it
at
5
a.m.
we
get
a
different
audience.
I,
don't
know,
I
mean
from
your
experience.
Cfcs
are
like
the
right
level
of
the
12
and
5s.
B
A
B
A
You
know
I
think
somebody
needs
to
just
sort
of
go
through
and
you
know
figure
out
like
how
we're
going
to
deal
with
conflicts
with
other
stuff
on
the
node
calendar
great
and
then
figure
out
what
the
best
time
you
mean.
If
people
want
to
do
a
five
and
it's
like
I
get
up
early
I'm.
Okay
with
that,
but
that's
rare
and.
A
A
A
B
C
A
A
A
D
A
C
A
I,
don't
I
don't
want
to
commit
to
going
through
the
scheduling
stuff
because
I've
got
a
number
of
other
things.
I
need
to
just
close
out
now.
Somebody's
like
hey
I
went
through
all
the
work
on
and
all
I
need
to
do
is
kind
of
show
up
and
walk
through
that
deck
I'm
happy
to
do
that,
and
that
would
be
good
but
yeah,
some
of
the
other
stuff
I,
don't
I,
don't
know.
Maybe
it
wouldn't
take
me
a
long
time.
I
just
feel
like
I've
got
a
pile
of
post-it
notes
with
little
to
do.
D
A
A
Yeah,
but
it's
if
people
don't
think
it's
necessary
or
we
think
that
there's
not
an
audience
that
that's
interested
in.
Seeing
that
then
I
don't
know
what
else
there
is
to
deep
dive
on
at
this
time.
At
some
point,
I
would
expect
that,
like
this
has
moved
along
in
some
fashion
and
we've
got
a
different
set
of
things
to
talk
about,
but
we're
not
there
today,
I.
A
E
A
E
A
B
E
E
A
A
A
A
F
So
there
was
a
call
yesterday,
I
think
with
a
few
pants
were
there
as
well
and
there's
some
concrete
next
steps
that
are
to
be
taken,
I
think
there's
some
words.
I
think
this
is
going
to
do
to
resurrect
materials
here
and
then
there's
welcome
to
be
insight
to
help
improve
the
performance
of
promises.
A
A
D
A
F
That
was
quite
a
bit
of
the
Preakness
has
been
solved.
I
think
there
is
a
one
issues
to
loop
and
that
was
opened
by,
but
someone
who's
running
it
inside
I.
There
are
a
q,
mu
or
a
qu
environment,
and
it
only
fails
one
and
on
500
times
there
so
trying
to
figure
out
whatever
we
can
from
that
one
we
didn't
see
but
I
think
wouldn't
use.
Given
the
scenarios
are
getting
work
lot
more
complex,
I
think
the
flakiness
is
getting
moved
to.
You
know
we're
moving
in
the
right
direction.
Okay,.
A
D
Sorry,
you
probably
can't
hear
me
I
I
I
just
copy
the
contents
of
the
email,
a
discussion
to
a
github
issue,
so
we
can
open
that
up.
Yeah.
A
D
D
What
like
in
general,
would
it
be
good
to
just
open
new
issues
for
for,
like
separate
discussions,
I
feel
like
there's
there's
a
few
different
things
to
talk
about
there.
There's
like
brazing
context,
there's
like
well
I
guess
like
GC
is
one
them
and
then
how
promises
work
with
it
and
I
guess
the
fields
are
there
are
the
best
fields
to
represent
the
data.
A
Certainly
I
mean
I
think
we
can
and
I
think
some
of
this
stuff
is
easier
with
a
whiteboard,
but
but
we
can
certainly
we
can
certainly
try
and
address
it.
It
may
be
the
case
that
you
know
it's
like
given
that
that
some
of
this
some
of
the
context
is
in
email,
not
everybody
has
contacts
but
I.
Think
if
you
people
are
have
no
idea.
What
we're
talking
about
and
just
shout
and
I'll
try
and
clarify
I
guess
I
could
draw
on
my
whiteboard
and
point
the
camera
at
it.
A
D
So
I
just
the
just
to
confirm
about
continued.
If
I
like
I
know
now,
they
wrote
a
JavaScript
Bert
only
version
of
it
and
I
I've
done
the
same.
I
was
to
Mike,
but
I
also
share
on
the
issue
when
the
tracking
issue,
when
Mike
opens
it,
but
so
is
that
yeah
that
continue,
if
I
supposed
to
be
completely
implanted
by
the
VM
in
the
future.
Well.
A
I
think
that
the
ideal
scenario
is
that
we've
got
some
lumpy
patchable
thing
that
has
the
same
semantics
right:
that
people
can
go
in
and
and
do
for
older
versions
of
things
or
potentially
older
versions
of
a
browser,
or
something
like
that
right
so
that
now
the
question
is,
can
you
set
up
this
ready
context,
stuff
quickly
and
I?
Think
you
know
pretty
madeley's
comment?
It's
like
you
can
get
most
of
it.
Maybe
maybe
you
know
I,
don't
know
what
80%
you're
not
being
too.
A
A
Just
said
it
like
it's
just
kind
of
hand
wavy,
you
know
yeah
most
the
time
you
know
it
works
every
time,
sort
of
thing
right.
What
is
the
80
percent
of
the
time?
It
works
every
time
right
later
it
just
kind
of
made
it
up,
but
at
some
point
there's
some
gaps:
I,
don't
I,
don't
have
a
clear
picture
of
where
there
may
be.
Natalie
can
speak
to
specifically
what
the
gaps
are
prevalent.
Those
are
for
I
think
for.
A
F
A
There's
a
question
I
think
you
know,
which
is
that
like
and
I
think
it's
there's
I
think
there's
valid
reasons
why
you
want
to
push
these
concepts
so
that
they
are
available
to
the
VM
right.
I
mean,
but
I
think
that
the
question
is
is
if
you
can
prove
the
value
of
this
stuff
outside
of
the
VM,
and
you
can
do
that
efficiently.
Then
I
think
that's
a
win
and
then
you
can
start
saying
like
oh
look,
the
same
concepts
or
you
know
interesting
to
the
garbage,
collector
or
interesting
to
the
allocators
or
interesting
you.
A
You
know
whatever
these
other
VM
level
components
are
motivation
to
push
it
down
right
and
it's
a
much
it's
much
more
work
to
get
it
into
the
VM
right.
That
is,
you
know.
You're
gonna
write
all
this
stuff,
a
native
code
or,
if
you're
doing
a
v8,
you
got
to
write
all
this
stuff
in
you
know
torque
or
whatever,
right
and
and
so
it
is
just
extra
work
there
right.
Not
so.
B
D
So
now
the
would
you
but
you're
okay
with,
like
us
like
say,
commenting
on
on
the
code
in
your
repository
like
alright,
the
one
wit.
You
know,
fine,
okay,
so
I
think
what
that's
a
good
place
to
like
iterate
on
thoughts
on
it.
Oh.
A
A
Other
aspect
too
nattily
stuff
is
you've
got
some
code
that
knows
how
to
go
in
and
hook
all
the
places
in
the
node
api's,
so
that
all
these
things
are
being
continued
if
I'd
at
the
right
spot
right
and
that's
code
that
you
leverage
from
the
New
Relic
agent.
So
there's
some
there's
there's
some
confidence
in
that
code.
Right
yeah.
D
So
I'm
not
sure
what
Natalie's
view
is
on
how
promises
work
with
the
current
unify
implementation,
but
I
guess
one
thinking
is
that
like
the
well
do
you
expect
that
the
promise
class
itself
would
have
cause
to
continue
flying
within
it.
E
D
E
D
D
Okay,
I
guess
I
I
think
I
agree
that
those
are
the
points
to
be
continued,
find
I,
actually
think,
there's
a
way
to
do
it
that,
where
you
don't
need
any
like
extra
external
logic
to
make
it
like.
Okay,
sir
promises
like
I,
think
if
you
continue
find
the
right
places
than
you
never
need
to
like,
like
ready,
if
ready
contacts
and
always
be
so
automatically,
there's
nothing
where
I'm
trying
to
come
up
with
a
with
a
real.
The
code
example
like
right
now
so
I'll
share
on
the
issue.
Probably
a
little
I.
E
Think
it
promises
our
difference
in
normal
callback
sequences
that
they
do
have
sort
of
tuning,
and
so
what
they're
going
to
require
extra
bits
like
they
have
your
one
link.
If
you
do
call
then
or
wage
call
new
promise,
your
other
link
is
well.
When
did
my
stuff
actually
become
ready
to
start
executing?
A
A
What
would
happen
is
at
some
point
and
I
think,
and
maybe
this
is
because
I
read
the
spec,
and
that
was
the
wrong
thing
to
do,
because
now
I'm
confused,
but
the
there's
there's
places
in
the
spec
when
it
actually
says
like.
Oh,
you
know
going
and
queue
this
thing
in
this
promise
reaction.
Job
in
you
know
the
micro
tasks
you
know,
and
those
are
the
places
where
we
actually
want
to
say.
Oh,
what's
the
context
where
this
thing
is
getting
in
queue,
that's
the
ready
context.
A
Uh-Huh
now,
there's
maybe
some
argument
like
if
you
say
if
you've
got
an
already
resolved
promise
and
you
change
something
on
to
it,
then
it's
ready
context,
as
implied
as
the
current
context
based
on
that.
Maybe
that
thing
correct
right,
there's
the
potential
that
different
people
can
have
a
different
view
of
what
the
right
semantics
should
be
there,
but
I
do
think
it's
the
kind
of
thing
that
if
you
can
phrase
it
at
that
sort
of
level
of
a
whenever
something
is
in
cued.
A
Whenever
a
promise
reaction
isn't
cued,
then
you
will
have
very
clear
semantics
about
what
is
the
ready
context
right?
No
different
people
may
agree
with
whether
or
not
that's
the
right
thing,
but
at
least
your
your
explicit
about
correctness
right
I,
don't
know
how
you
get
that
from
monkey
packing
and
maybe
that's
what
you're
talking
about
Natalie
is
like
there's
some
gaps
and
what
you
can
get
from
monkey
patching
yeah.
E
F
E
A
And
if
you
think
that
there's
another
there's
another
bit
of
information
there,
that
people
want
I
think
it
would
be
good
to
sort
of
like
identify
that
like
or
come
up
with
a
code
example
and
say
like
no.
No,
in
this
case,
what
I'm
really
interested
in
is
being
able
to
walk
back
to
this
context,
for
whatever
reason,
I
think
as
we've
been
discussing
about
it,
we've
claimed
that
there's
you
know
this
idea
of
their
reading
context
as
sort
of
we've
defined
it
where
the
thing
is
in
Hue
and
but
I.
E
A
That
resolved
then
right,
then
it
would,
since
the
proper
anytime,
you
already
have
the
promise
are
resolved
and
it
would
just
become
this.
The
current
context
right
and
then
in
context
right,
which
may
or
may
not
be
what
you
want
and
I
think
there
was
some
different
like
when
the
first
iterations
of
this
we
were
kind
of
talking
about
it.
People
would
raise
questions
about
that
and
I
think
it
sort
of
came
out
of
like
that
original
paper.
That
Mark
was
working
on,
which
was
that
you
know
they
just
always
picked
it
as
being.
A
Promise
wraps
right
and
what's
going
on
with
promise,
growls
and
a
bunch
of
different
questions
that
you
can
answer
about
promises
that
may
be
interesting
from
a
diagnostics,
perspective
and
I.
Think
one
of
the
papers
like
in
those
slides
they
talk
about
like
doing
a
bunch
of.
A
Analysis
on
on
promises
and
being
able
to
say
things
like:
oh
here's,
a
promise
and
it
was
resolved
more
than
once
with
different
values
or
a
promise
was
rejected
after
it
was
resolved
or
like
all
of
these
things-
and
arguably
these
are
you-
know,
sort
of
like
at
a
minimum
like
you
can
imagine
like
some
dynamic
promise.
Linter,
that's
gonna
sit
there
and
tell
you
like
hey.
What
do
you
do
it
like?
Is
this
really
what
you
want
to
be.
B
A
A
You
can't
we
don't
have
the
fullness
of
that
level
of
information.
I
suspect
that
there's
potentially
like
as
somebody
started,
looking
would
start
looking
into
this
you'd
find
specific
things
that
you
want
to
do,
and
you
won't
start
augmenting
this
graph
for
more
information
about
the
promises.
I
think
it's
an
open
question,
whether
or
not
that
I
that's
useful
or
that
makes
sense,
but
I
think
it's
worth
conversation.
A
D
A
All
right
awesome:
do
we
have
anything
else
or
are
we
done
I.