►
From YouTube: Node.js Diagnostics WG Meeting - 2018-03-21
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
There
is
an
action
around
the
deep
dive
for
the
profiling
thing,
which
it
happened
last
week,
I
wasn't
able
to
stay
on
that.
There
was
some
action
of
gang
about
the
proposal
for
promise
books,
changes
that
we
can
talk
about,
and
that
looks
like
it
as
far
as
the
actions
out
of
the
last
time,
so
I
just
go
on
to
the
agenda.
So
number
168
is
deep
dive
topics.
It
sounds
like
the
first
one
happened,
I
wasn't
and
how
did
that?
Go?
I.
A
I'm,
just
gonna
leave
this
open
for
people
to
kind
of
chime
in
with
some
other
stuff,
and
you
know
some
other
topics
here
so
that
we
can
try
and
get
the
next
one,
and
maybe
we
can
go
through
the
ACA
context.
I
know
that
we've
been
talking
about
that
on
our
end,
serving
more
deaths
that
and
some
different
applications
of
it,
and
so
you
know
maybe
we've
missed
one
thing.
We
can
do
deep
dives
on
that.
Some
of
those
things
crystal.
B
D
B
D
For
example,
so
we
have
this
five
action
items
to
finish:
much
booty
quest,
v8
I
think
it's.
Basically
we
can
say
it's
Mattias
and
the
v8
team
who
is
working
on
the
and
after
that
we
have
to
come
up
with
the
CI
config
for
test,
come
up
with
basic
testings
canarios
integration,
testing
scenarios
and
performance
benchmark
story,
but
we
don't
have
to
discuss
them
now.
Just
we
shouldn't
forget
them.
Yep.
D
A
A
I
mean
I
think
last
question
here:
I
was
like
okay,
where
do
we
start
and
I?
Think
at
the
summit
there
was
some
questions
around.
Like
you
know,
do
you
use
that
diagnostic
channel,
the
existing
diagnostic
thing
or
you
start
fresh
with
something
or
start
with
a
clean
slate
and
try
and
you
know,
build
some
stuff
in
is
build
it
up.
One.
You
know
from
scratch
us
yeah.
B
E
C
C
C
C
That's
a
good
question:
I,
don't
know
so.
The
point
of
putting
a
document
together
is
to
to
gather
feedback
but,
like
I
said,
unfortunately,
I
didn't
have
time
to
do
that
yet
yeah.
Another
thing
we
discussed
and
I'm
not
entirely
sure
whether
it
makes
sense,
though,
is
in
the
spec.
There
are
quite
a
few
temporary
premises
being
created,
but
a
lot
of
those
are
not
actually
available,
then
regular
JavaScript.
They
are,
however,
through
promised
hooks.
The
question
is:
can
we
skip
those
or
promise
as.
F
So
more
or
less,
if
it
comes
out
to
say,
comes
out
the
correct
one
after
the
chain
and
we
skip
Samantha
Vigo
absolutely
like
this
is
my
those
are
not.
Those
does
not
do
not
do
not
surface
at
all
in
JavaScript,
so
or
even
in
in
user
control
C++.
So
why
would
like
it's
just
something
internal
to
to
to
JavaScript
itself,
to
to
be
eight,
so
I
mean
like
as
long
as
out
of
it
a
lot
as
long
as
out
of
it.
The
current
context.
We
have
reigned.
E
C
G
A
C
So
most
performance
improvements
that
we
have
implemented
for
promises
do
not
affect
the
case
when
promise
works
are
enabled,
because
you
know
there's
just
no
way
to
cut
corners
yeah
for
asking
crooks-
or
at
least
in
many
many
cases,
there's
no
way
to
do
that,
which
means
that
the
the
gap
between
when
icing
cooks
are
enabled
or
disabled
is
just
growing
us
further.
Okay,
so
I
think
he
was
throwing
some
numbers
like
50
percent
around.
C
A
A
No
okay,
so
the
async
context,
formalization
so
I
don't
have
I've
got
some
code.
That's
trying
to
sort
of
fill
the
dynamic
monkey-patching
of
all
of
the
node
libraries
that
will
generate
the
various
events
that
we
talked
about
and
it's
it's
close
I
haven't
had
a
lot
of
time
to
work
on
it
I'm,
hopefully,
next
week,
I
can
actually
have
something
to
share
just
as
sort
of
a
contrasting
point
of
view
of
a
model
you
know
versus
what
async
hooks
provides
and
it
other
than
that
I.
Don't
think
I
have
anything
to
share
what.
A
If
you
look
at
that,
there's
a
branch
on
that
async
context,
definition
thing,
I
can
say
it's
like
whatever
github
mike
coffman
definition,
and
there
should
be
a
public
branch
on
there
called
add
monkey,
bacitracin
or
tracking,
and
so
you
look
at
that
and
there's
kind
of
like
this.
Some
of
the
codes
really
like
I,
don't
know
some.
We
had
an
intern
a
few
years
ago
and
and
he
did
a
really
good
job
in
identifying
all
of
these
different
transition
points.
And
you
know
some
of
the
code
is
a
little.
A
You
know
a
little
head
scratchy.
Maybe,
but
the
key
thing
is:
we've
got
this
date
file
that
we
can
then
just
sort
of
you
know
monkey
patch,
all
these
different
places
inside
of
node
one
time
or
node
libraries,
JavaScript
libraries
and
and
generate
the
right
events
and
so
I'm
sort
of
working
through
that
a
little
bit
I
think
it's
pretty
close.
A
A
G
So
so
you're
playing
on
talking
about
this
in
more
depth
in
two
weeks,
right,
I,
hope.
A
That,
like
I,
can
say
like
hey.
If
you
want
to
try.
You
know
this
automatic
generation
of
the
event
stream
on
code
grab
this
thing
and
and
use
it,
and
then
we
can
actually
go
through
the
code
and
we
can
talk
about
some
different
abstractions
that
like
sort
of
introduce
this
notion
here
of
a
in
that
code
of
a
continuation
rate
which
is
sort
of
a
function-
that's
gonna.
A
You
know
it
gets
created
and
linked
in
one
context
and
then
it
gets
executed
in
another
asynchronous
context,
right,
which
is
kind
of
the
thing
that
we're
all
talking
about,
and
so
there's
sort
of
some
constructs
that
can
quote
if
desired,
hoist
it
up
into
user
code.
So
there's
more
visibility
to
the
user
who's.
Reading
the
code
on
and
you
can
trans
translate
a
function
will
convert
a
function
into
one
of
these
continuations.
G
A
Hopefully,
we
can
use
that
kind
of
use
this
code
to
sort
of
yeah
as
a
sort
of
an
example
and
use
it
to
discuss
some
of
these
some
of
these
concepts
in
more
detail
and
use
it
to
say,
like
hey,
here's
a
real-world
program.
What
is
the
asynchronous?
It's
probably
not
going
to
work
with
native
promises
and
it's
not
gonna
work
with
async/await,
so
you
probably
have
to
run
it
on
some.
G
Yeah
I
take
a
look
at
a
before
Chanelle
I,
so
I
was
I've,
been
spending
a
little
bit
time
like
coming
with
up
with
a
sort
of
busy
zation
for
well
for
async
planter
tracking
and
which
is
trying
to
be
informed
of
both
like
concepts
in
the
a
syncline
maximization
paper
ends
in
async
hoax,
and
this
is
something
that
I
think
I'd
like
to
talk
about
in
two
weeks
as
well.
Yeah.
A
Certainly
I
think
when
I
don't
know
like
I
would
say
you
know
just
try
man-eater
on
the
call
whenever,
because
this
is
still
gonna
be
sitting
on
the
agenda
right,
so
I,
just
chiming
the
call
or
you
know,
chime
in
whenever
the
meeting
gets
supposed
to
being
like
you
want
to
discuss
what
suppose
you,
okay
but
yeah,
I'd
love
to
see
and
feel
free
to
reach
out.
You
know
early.
If
you
want,
you
know
preliminary
feedback
on
any
points
right
sure.
B
B
A
I,
don't
know
where
they're
at
the
last
I
recall,
seeing
is
that
there
was
a
James
was
doing
some
some
playing
around
and
there
was
some
Perth
pain
points.
I
think
what
some
of
the
sort
of
I
think
he
was
trying
to
test,
not
mistaken
kind
of
testicle
category
was
enabled
or
something
and
then
and
that
was
slowed
from
JavaScript
in
to
trace
events
and
I,
don't
know,
there's
any
other
update.
If
anybody
knows
more
detail
on
this
yeah.
A
B
A
B
One
thing
I'd
throat
is
I,
I
haven't,
had
I've
been
pretty
busy
so
having
that
chance
to
sort
of
chase
down
people's
actions
from
the
diagnostic
summit,
but
I
do
hope
to
get
to
that's
not
next
week
because
I'm
on
vacation
but
the
week
after.
So
if
you've
got
an
action
and
you
can
spend
some
time
on
it,
that
would
be
great.
A
B
So
I,
you
know
I'll,
try
and
spend
some
time
doing
a
little
bit
of
rounding
up
to
see
you
know
which
ones
we've
completed,
which
ones
we
haven't
and
what
we
might
start
to
be
pushing
on,
like,
for
example,
I'm
trying
to
find
somebody
internal
the
to
work
on
pulling
node
report
into
core,
but
on
the
other
ones.
You
know
I'll,
try
and
see.
Whoever
had
the
action
to
like
do
the
write-up
and
and
make
sure
we
actually
make
some
progress
on
those.
F
F
So
it's
I
will
ask
David
to
sit
out
to
put
an
issue
on
on
it.
I
wanted
to
have
some
feedback
from
this
from
from
yang,
but
it
probably
faded
away,
and
anyway,
that's
that's
about
it.
Okay
check
it
out
because
it
just
changes
quite
a
bit
from
the
previous
release.
So
it's
actually
worse
was
checking.
It
does
some
some
inter
interesting
thing,
especially
with
thin
line
that
inlinable
functions
and
and
stuff
like
that.
F
F
F
F
It
is
lightly
better
than
what
it
was
before.
So
I
am
a
PMA
already
happy
by
the
way.
It's
also
because
it
works
on
Windows,
so
that
covers
some
use
case
that
it
wasn't
working
before
because
it
doesn't
support
it.
Dw
into
double
support
is
out
of
a
question,
but
it's
yeah
whatever
it
took
me
two
weeks
to
get
something
out
of
that
and
it's
a
mess,
but
it's
still
working
very
well.
It
also
some
interesting
stuff.