►
From YouTube: Node.js Diagnostics WG Meeting - 2018-06-13
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
For
the
summit
you
know,
maybe
maybe
it
would
be
helpful
to
like
get
some
numbers
to
like
quantitatively
say
like
oh,
like
only
50%
of
people
want
to
come,
can
or
maybe
only
80,
but
whatever
that
is
I'm
super
empathetic
of
people.
Obviously,
who
can't
make
it
obviously
I
think
you
want
to
be
as
possible,
but
we
have
some
numbers
around
that
date.
The
dates
right,
I
think.
B
People
have
been
chiming
in
on
the
issue.
Is
that
sufficient
to
get
a
head
count
there?
Is
it
what
I
mean?
One
approach
is
to
go
on
the
issue
and
say,
like
hey,
you
know,
make
a
comment
that
says:
July,
26
or
whatever,
and
get
a
tally,
a
thumbs-up
thumbs-down,
and
that
might
be
the
most
efficient
way
to
do
it
and
throw
like
a
handful
of
dates
out
there.
You
know
I'm.
C
D
Other
thing
that
sort
of
rolling
around
in
my
head
and
I
don't
know
alley.
You
can
comment
on
whether
this
makes
any
sense
is
like
we
could
still
try
and
push
the
you
know
a
full
summit
out
a
bit
to
give
more
planning
time
and
then
possibly
have
us.
You
know
the
subset
of
people
who
are
going
to
be
around
just
meet
to
talk
about
things
in
person
as
opposed
to
making
it
an
official
summit.
Is
that
it's
not
any
different,
or
is
that
still
not
excluding
people?
Just
because
they're,
not
there
I.
C
Think
like,
if
anybody's
free
to
meet
whenever
they
want
with
the
people,
they
want
to
meet
right.
So,
but
if
it's,
if
it's
the
no
diagnostic
working
group
that
wants
to
meet
and
and
get
stuff
done,
then
I
think
we
should
do
some
more
effort
to
make
sure
people
have
time
to
make
plans
and
again
I
mean.
A
Would
it
be
helpful
to
the
like?
Perhaps
we
can
just
identify
I,
think
specific
issues
that
we
want
to
work
through?
That,
perhaps
are
can
be
more
inclusive
like
I
know,
there
are
specific
things
that
maybe
you
know
Matias
and
us
have
been
working
on
or
other
things
where
you
know
it
could
be
like
a
subset
of
the
group
anyway,
who
be
interest
in
that
discussion,
so
I,
like
Michael
kind
of
like,
if
there's
this
opportunity,
where
a
bunch
of
folks
are
going
to
be
around,
and
so
you
know,
yeah.
B
Ally's
point
is
that
somebody
who
may
want
to
be
involved
in
that
conversation
doesn't
have
the
opportunity
to
participate
because
short
planning
right
so
I
think
you
know.
The
reality
is,
is
that
if
you
know
there's
two
people
in
a
room
who
are
interested
in
a
topic,
they're
gonna
have
a
conversation
on
that
topic.
Right
and
I.
Think
that
that's
going
to
happen
it.
What
what
is
it
no.
A
Think
then,
like
I,
don't
I
don't
know
if
folks,
who
can
attend
because
they're
physically
just
not
able
to
attend
or
they
have
conflict
or
if
they
have
conflicts,
and
obviously
we
should
find
the
time
a
better
time,
but
if
it's
because
they're
somewhere
else
or
whatever
and
they
can
dial
in,
would
that
be?
Would
that
make
it
I?
A
C
B
I
mean
I'm
sympathetic,
both
ways,
I
think.
If
somebody
has
a
strong
opinion,
like
you
know,
I
I
think
you
should
voice
it.
I
think
the
strongest
argument,
I've
heard
so
far,
is
from
Ally
right,
which
is
that
we
should
give
fair
warning,
and
you
know
we
should
have
people
be
able
to
plan
to
attend
these
things
and
I
do
think
that
sort
of
in
alignment
with
you
know
this
overall.
D
Go
ahead,
the
thing
all
that
is
I,
don't
think
we're
gonna
get
as
broad
a
group
as
we
did
the
last
time,
given
the
shorter
notice.
So
the
other
thing
I
think
we
want
to
avoid
is
having
one
which
then
says
of
what
we've
already
had
one:
let's
not
have
another
one
for
six
months
right
right
exactly
because
then
then
we
you
know,
then
those
people
don't
haven't
had
a
haven't
had
the
chance
to
participate.
D
D
A
Like
if
we
give
you
like
September,
then
that's
maybe
it
maybe
it's
like
okay,
so
we
could
have
some
informal
meetings
that
note
summit.
Then
it
sounds
like
you
know
we
want
to.
We
want
to
make
sure
everyone
can
attend,
and
should
we
talk
about
kind
of
where
we
want
to
have
when
we
want
to
have
the
next
next
summit.
B
B
You
know-
and
you
know
say
with
those
two
proposals
on
and
putting
those
up
on
this
issue
and
getting
a
thumbs-up
thumbs-down
count
might
be
the
best
approach
right,
I,
think
folks,
coming
in
from
Europe
or
traveling
far,
you
know
coming
in
to
North
America
and
then
two
or
three
weeks
later
coming
back
might
might
be
a
pain,
and
that
might
might
be
something
to
avoid
on
so,
if
it's
possible
to
align
this
stuff
with
Kaos,
interactive
and
I,
think
that
would
be
ideal.
So
there's.
A
Already
going
to
be
like
a
collaborator
somewhere
right,
so
I
is
that
do
we
feel
like
there's
enough
time
to
do
that
and
a
two-day
diagnostic
summit
I
just
want
also
like
in
terms
of
I,
just
think
like
folks,
are
gonna,
be
giving
talks
and
gonna
be
attending
the
conference.
There's
already
a
collaborative
summit
and
like
are
we
are
we
at
risk
of
like
people
boarding
out
because
I
don't
personally,
like
after
a
summit
and
at
a
conference
I
basically
need
to
check
out
for
quick?
A
A
So
one
possibility
is
I,
know
yang
I,
said
know:
if
he's
on
the
call
yeah
yes,
so
he
I
think
yang
had
proposed,
like
hey
they're,
having
a
hosts
one
in
Munich
next
February
or
March,
and
I
told
that
there's
there's
some
appetite
to
host
the
next
one
in
Europe.
So
could
we
do
that
in
September
instead,
I
know
September
in
Germany
is
probably
better
weather
than
say
February
March,
but
mom.
This
video,
hello
September,
might
be
possible
yeah,
okay,
so
it's
does
that.
A
B
E
A
Happy
to
host
whatever
I
was
just
more
trying
to
be
more
conscientious.
I
know:
Michael
Dawson
had
talked
about
like
hey
folks.
Are
some
interesting
not
always
holy,
seeing
it
in
North
America,
but
obviously
we're
happy
to
host
a
venue
whenever
really
weather
here
is
always
good.
So
you
know
maybe
I
think,
let's
start
an
issue.
We
can
propose
some
dates
and
locations
and
then,
let's
just
figure
out
from
there
like
what
works
for
people
yeah
does.
B
B
B
Like,
let's
pick
the
dates
for
the
next
one,
and
you
know
and
let's
let's
like
figure
out
a
process
of
like
you,
know
how
tally
votes
for
what
people
are
in
agreement
with
or
not
in
agreement,
you.
A
B
E
A
A
C
D
B
D
G
B
G
B
The
there
may
be
an
informal
thing,
note
summit,
I,
think
the
the
opposition
is,
there
wasn't
sufficient
time
for
people
to
sort
of
plan
and
schedule
a
different
for
an
actual
diagnostic
summit.
So
there's
gonna
be
some
proposal
of
some
dates.
This
will
start
in
September
timeframe
and
there
will
be
an
up
down
those
on
various
dates.
Locations
for
the
summit
on.
D
G
D
B
B
B
D
B
B
B
E
D
E
G
G
G
G
Profile
of
course,
but
if
I
need
to
find
where
I
weigh
my
thing
is
at
whatever
it's,
where
my
application
is
act,
that's
very
it
works
more
or
less
so
I
would
not
I
would
not
concern
I'm
present.
That's
the
fact
that
we
get
did
the
interpreted
ones
back,
that's
fantastic,
so
I,
wouldn't
I,
wouldn't
consider
I.
Don't
consider
that
a
bug!
That's
what
okay.
E
C
C
So
at
the
climate
summit
there
was
discussion
of
whether
we
are
ready
to
make
it
stable
and
I
think
there's
a
issue
discussing
some
things
that
must
happen
before
I
think
it
should
become
stable
there.
There
is
this
question
of
the
of
a
formally
defined
semantics
that
I
think
you're
working
on
Mike
and
Mark,
but
they
were
also
concerned
raised
that
I
create
it's
probably
moved
anyway,
because
people
are
using,
you
think,
exciting
yourself.
It
doesn't
matter
there.
D
G
G
D
Way
the
way
I'd
characterize
it
was
there's
a
general
feeling,
maybe
not
from
everybody,
that
you
know.
We
need
the
the
semantics
and
terminology
to
be
able
to
reason
about
it
before
we
want
to
make
it
non
experimental,
but
at
the
same
time,
what's
already
there
in
existing
releases,
is
you
know
depending
it
on?
So
we
don't
necessarily
want
to
change
the
api's
and
back
port
that
so
I
could
see.
You
know.
D
Maybe
the
way
forward
is
that
once
we
have
the
semantics
once
we,
you
know
we're
comfortable
that
we
can
actually
say
that
you
know
what
we
have
or
something
different
is
the
right
thing.
We're
we're.
Leaving
is
experimental.
So
we
have
the
flexibility
to
do
that,
but
there
would
be
opposition
to
back
porting
those
changes
to
previous
releases,
because
you
know
people
actually
are
using
it
in
production.
D
So
it's
it's
like
Mateo
Mateo
was
saying
it
we're
kind
of
in
a
in
a
middle
ground
where
you
know
we
need
to
acknowledge
to
not
break
people,
but
at
the
same
time
we
need
to
keep
the
flexibility
of
actually
changing
the
API.
If
we,
if
we
discover
what
we
have,
isn't
the
right
thing
based
on
the
terminology
and
and
the
the
stuff
that
you
guys
are
working
on
to
help
us
reason
about
it,.
D
D
Similarly,
with
monkey-patching
right,
you
know
internal
things,
so
there's
an
appreciation
there
that
they
will
need
to
make
updates.
It's
just
that
if
we,
if
we
force
those
updates
on
non
major
release
boundaries,
you
know
that
won't
be
won't,
be
a
good
thing
and
I
know.
I
know
that
the
thing
would
be
the
December.
You
know
the
the
major
ones
are
cember
major,
so
we
can
already
break
things,
but
we
do
work
not
to
break
people
too
much,
even
though
cember
major
is
allowed
right.
G
B
F
C
I
know
I
mean
II,
think
hooks.
It's
a
p.m.
vendors
that
we
that
are
using
it
not
end
users
directly
and
a
p.m.
Enders
have
to
adjust
anyway.
If
they
are
going
to
break
it,
then,
on
on
the
latest
release
on
the
current,
so
they
need
to
adapt
anyway.
They
might
as
well
do
it
for
all
really
hammock
special
cases.
It's.
G
C
But
continually
just
anyway
right
so
continuation,
local
storage,
yeah,
yeah,
so
I
don't
think
it's
so
for
further
I
think
we
have
to
change
the
API
in
the
past
like
the
way
promise
hooks
were
were
done,
but
the
the
promit
in
it
callback
what
the
information
which
they
receive.
That
has
changed,
and
we
back
ported
that.
C
D
G
G
B
Think,
to
summarize
this:
it's
that
it
is
still
in
experimental
there's.
You
know
we
don't
there's,
there's
not
unlimited
freedom
to
make
changes,
given
the
fact
that
this
is
being
used
relatively
widely
for
a
experimental
feature
and
there's
pros
and
cons
to
anytime.
Somebody
wants
to
back
port
changes
and
if,
and
those
things
shouldn't
be
considered
case-by-case
basis,
is
that
a
fair
summary.
G
B
G
B
G
In
there
I
talked
about
it
for
a
while
I.
Finally,
a
time
I
finished,
my
traveling
after
the
nod,
10.4
came
out.
I
had
to
actually
wait
the
new
v8
to
actually
verify
this
was
actually
working
because
I
needed
in
order
to
file
the
PR,
at
least
because
otherwise
it
would
have
been
a
little
bit
moot
because
of
how
different
it
was
before
and
after
the
work
that
we
did
for
the
throw
away
I
promise
in
in
a
sink
await.
G
So
I
put
the
PR
in
gives
us
some
some
performance
benefit
and
add
a
completely
new
API
to
us
and
cooks,
which
is
the
current
corner
source
blast.
Note
that
this
is
I,
don't
know,
I,
don't
think
this
is
big
portable
as
it
is.
If
you
want
to
back
port
it
I
have
the
other
code
that
can
run
in
in
no
date
or
something.
G
E
G
G
It
has
some
benefits
and
some
discussion,
some
interesting
things
and
some
discussion
points
so
the
first
thing
that
it
stores
the
current
a
sync
resource
into
the
the
context
into
the
v8
context,
which
is
a
different
away
from
how
the
other
things
has
been
done
now
announcing
in,
is
pointing
out
that
there
might
be
kissing
that
this
does
not
really
work,
but
I
don't
know
so.
I
would
really
want
to
understand
what
those
are
you.
G
D
G
It
is
the
internal
stuff
they
in
turn
out
before
signature,
and
so
the
way
it
works
is
when
you
create
an
ass
when
we
create
our
a
synchro
source,
the
object
new
objects
get
flagged
as
persistent
anyway,
and
so
that
gets
retained
so
when
before
gets
emitted,
I
can
take
that
object,
that
I
have
and
stick
into
the
complexed,
and
that's
it
so
and
then,
when
there
is
after
I,
just
remove
it
and
I
just
rent
it
down
and
I
just
wind
it
down.
So
that's
how
I
say,
which
works
is
simple,
probably
simplistic.
G
Now
the
naming
and
so
on.
It
can
actually
be
whatever
okay,
so
I'm.
Not
it
is
not
a
good
name,
so
I
just
I
just
wanted
to
get
it
out.
So
somebody
else
can
tell
me
what
name
to
to
put
it
in,
but
it's
that's,
that's
more
or
less
the
crux
of
it.
It
has
some
benefits
because
it
removes
a
lot
of
suppose
class
to
JavaScript
back
and
forth.
So
it's
actually
pretty
good,
pretty
good
and,
and
so
on.
G
G
G
G
G
Yeah
I
know
yeah
yeah
I,
know
I
know.
So
what
I'm
just
saying
is
the
reason
why
it's
implemented
in
that
way.
It's
because
it
matches
100%
what
you
do
within
it
within
it,
and
destroy
okay,
okay,
that
same
logic,
okay,
because
you
can't
do
anything
else
at
this
point
in
time.
If
you
don't
do
something
else,
what
you
want
really
it's
that
when
you
create
a
transaction
or
something
like
that,
you
want
that
context
to
be
the
the
same.
G
B
I
pointed
out,
nonetheless,
is
that
you're
you're
just
copying
the
object
versus
creating
a
new
instance
for
that
property,
bag
right
and
you'll,
just
overwrite
stuff.
It's
not
because
it's
in
the
in
at
the
beginning,
it
every
one
right,
so
it
creates
one
and
then
every
subsequent
thing
it
yeah
right
yeah.
B
So
at
some
point,
if
you
have
some
timer
calls
in
there
something
that's
spun
off
some
additional,
you
know
async
Forks,
those
all
end
up
with
one
copy
for
a
price.
Yes,
like
I,
want
to
set
a
value
on
here.
It'll
write
it
in
the
same
property
back
yeah.
So
exactly,
but
that's
how
most
most
of
the
PMS
works
today
anyway,.
G
Okay,
so
that's
how
people
that,
if
you
look
at
the
init
a
bill
to
destroy
example,
that's
exact
same
thing
and
that's
code,
that
I,
more
or
less
was
very
much
inspired
from
elastic
EMV.
So.
B
B
B
G
G
B
B
Okay,
so
so
moving
on
so
async
context,
formalization,
there's
a
peeler
posted
there's
been
a
number
of
comments
there
that
some
I
still
need
to
get
back
to.
You
I
think
there's
a
few
themes
that
have
popped
up
and
my
plan
is
to
update
the
document
at
some
point
here
with
to
sort
of
be
explicit
about
what
these
themes
are.
So
one
theme
is:
how
do
these
definitions,
work
in
generators?
I?
B
Think
that's
fairly
straightforward,
I'll
go
into
the
dock
at
some
point
here
and
add
a
example
that
explains
that,
and
in
this
case
it's
like
the
the
the
context
is
defined
by
a
frame
on
the
stack
and
I
think
that
works
fairly
nicely
for
generators
for
async/await
I,
believe
it
works
nicely
as
well.
B
So
that's
on
me:
there
I,
don't
know
if
anybody
else
has
read
it.
I
know
Michael.
You
had
some
comments
the
other
day
yesterday,
maybe
yep
or
I,
need
to
go
through
those
and
respond
to
those.
If
you've
read
it,
I'd
appreciate
feedback.
If
you're
like
I,
don't
understand
it
I'd
like
to
know
that,
because
I
think
the
goal
here
is
that
we
end
up
with
a
set
of
terminology
and
concepts
that
people
are
like
yeah
I
understand
what
this
means
right.
B
People
like
Joe
JavaScript
can
reason
about
asynchrony
and
use
the
same
set
of
terminology
that
you
know.
Everybody
else
is
using
right.
It's
not
a
mystery
right,
and
so,
if
somebody's
reading
it
and
you're
looking
at
it
and
you're
like
I,
don't
get
it
please
reach
out
I'm
happy
to
get
on
a
call
with
anybody.
You
know
privately
or
you
want
to
get
on
with
a
small
group
and
talk
through
this
stuff,
I'm
happy
to
do
that.
D
B
So
if
you
specific
things
where
you're
like
this
tripped
me
up
or
displaced
over
here,
it's
like
like,
let
me
know,
if
there's
specific
things
about
the
terminology,
you
know
that,
like
hey,
this,
this
term
means
something
else
in
my
head.
This
is
something
that
needs
to
be
done
better
like
please.
Let
me
know
I
think,
what's
one
of
the
harder
things
about,
this
is
finding
the
right
set
of
names
that
resonating
people,
the
other
thing,
I,
think
that
came
up
and
in
some
conversations
with
Ali
and
Kelvin,
was
this
idea
of
like
well.
B
B
What's
the
right
word,
I'm
I
feel
like
what
we're
coming
up
with
can
sort
of
fit
nicely
in
that
model
of
the
spec,
at
least
in
the
spec
language
on
and
so
I'm
sort
of
I
feel
you
know
humbled
from
reading
this
pack
and
I'll
just
also
feel
confident
about
I
think
the
ability
to
get
some
of
these
concepts
sort
of
smoothed
out
and
then
incorporate
it
into
spec
language,
which
I
think
is
enough.
So
anybody
have
any
comments
on
that
at
all.
D
G
B
G
B
Works
whatever
works
for
you
and
if
there's
specific
things
where
you're
like
I
lost
you
here,
then
please
reach
out,
and
it's
true
for
anybody.
It's
like!
Please,
don't
think
that
like
you're
reading
it
and
if
you
don't
get
it
that
you
know,
somebody
else
must
get
it
so
you're
just
going
to
be
quiet
like
I,
really
want
to
hear
people
are
getting
lost
in
it.
Right
like
it's,
not
the
intent
that.
B
B
C
Yes,
it's
an
teams
volunteer
to
be
the
champion
and
I've
been
doing
some
work
on
it
as
well.
So
I
can
give
a
small
update
so
on
trace
events.
They
are
some
changes
landed
on
master
that
add
some
some
new
features
to
it.
So
specifically,
there
is
a
node
inspector
for
the
hall,
so
it's
available
through
the
inspector
protocol
that
was
brand
new
I
think
that's
available
in
10.
Now
as
well.
If
I'm
not
mistaken,
there
are
other
changes
happening,
such
as
all
the
all.
C
The
threads
now
get
names,
and
there
are
also
some
bugs
that
I've
been
fixing
in
and
James
has
been.
Fixing
James
have
also
been
working
on
intrinsic
in
v8
or
that
he's
prototyping,
and
the
idea
is
that
intrinsic
would
make
it
easy
for
JavaScript
programs
to
to
use
trace
amounts
in
a
low
overhead
way.
That's
a
work
in
progress
so
stuff
happening,
but
but
trial
trace
events,
especially
through
inspector,
and
if
there's
a
reportable.
B
C
Don't
know
if
it's
perfectly
wide
together
at
this
point,
but
so
dev
tools,
it
has
a
performance
tab
and
then
our
stab
is
supposed
to
like.
It
has
been
heavily
focused
on
the
browser
at
the
book.
At
the
moment,
yeah
comments,
tab
gives
you
a
profile
and
traces
in
one
view,
so
I
think
what
they're
working
on
is
making
that
support
node
as
well,
but
I,
don't
know
if
it's
wild
up
perfectly
at
this
point.
Okay,.
B
C
That
would
be
the
place
where
this
would
be
connected
up.
The
alternative
is
like
you,
we
can
generate
a
file
and
that
file
can
be
open
in
chrome,
creasing
or
in
the
performance
tab
so
that
that
works.
But
the
idea
is
that
debuggers
can.
Like
automate
all
of
this.
You
open
your
you,
our
debugger
UI.
It
connects
to
the
node
process,
it
does
tracing
for
you.
It
does
the
same
way
it
does
profiling.
C
Okay,
okay,
so
these
are
so
new
features
and
I
think
they're
a
bit
buggy,
so
people
are
and
I
know
the
regular
protocol
right
now
it
is
buggy
that
I'm
working
on
a
fix
that
it
unless
you
started
right
at
startup,
some
of
the
categories
get
permanently
disabled,
so
very
brand
new,
but
please
play
with
it
and
open
bug
reports.
Okay,.