►
From YouTube: Node.js Diagnostics WG Meeting - 2018-05-16
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).
B
B
For
the
as
being
present-
and
you
get
credit
for
attendance,
I
still
have
a
recurring
nightmare
that
like
I
I'm
in
college
and
I'm,
taking
a
math
class
and
I
didn't
don't
show
up
for
class
like
all
semester
and
and
I
show
up
for
the
final
I
have.
A
A
B
B
People
can
look
at
those
that
PR
for
the
for
the
minutes
from
last
time,
and
then
you
know
chime
in
a
few
have
comments
if
I
missed
anything,
so
the
strategic
initiatives,
185
I,
think
one
of
the
actions
here
was
to
have
a
new
label
which
I
opened,
which
is
called
tracking
issue.
So
we
can
start
tagging
things
as
tracking
issues
and
then
they,
you
know
if
they
relate
to
sort
of
another
thing,
a
number
of
other
issues,
and
so
we
can
help
categorize.
B
D
D
E
D
B
A
Yeah
I
I'd
add
that
I
also
think
that
in
each
of
these
meetings
it
would
be
good
to
get
sort
of
an
update
and
by
having
somebody
who
kind
of
owns
it
not
owns
it
but
like
whatever
you
want
to
call
it.
You
know
you
know
who
should
give
the
update
and
then
it'll
be
clear
if
we're
making
progress
not
making
progress,
that
kind
of
stuff
I.
F
A
F
A
Our
I
would
almost
you
know
if
it's
two
people,
volunteering
to
help
move
it
forward.
You
could
have
to
you
know
you
could
have
co-champions
or
I.
Think
in
the
TSC.
One
I
think
one
of
the
suggestions
that
was
being
moved
to
was
like
stakeholders,
you
know.
So
there
are
other
people
who
are
at
Hoover.
You
know
volunteered
to
be
active
in
are
participating
I'm.
B
But
it's
like
yeah
go
talk
to
Bob
right
on
and
you
know,
and
and
then
they
can
kind
of
delegate
or
deflect
or
whatever
as
appropriate,
and
if
you
know
they're,
dropping
the
ball
or
running
out
of
time
or
have
other
priorities,
and
we
can
adjust
that
right
and
I
like
the
idea
of
other
people
sort
of
being
named
as
like,
hey,
you
know,
you're
you're,
involved
you're,
watching
this
and
I
think
that's
a
fair
suggestion.
I.
I
B
All
right,
so
then,
the
next
step
here
is
to
go
through
and
assign
the
champions
these
things
that
if
people
just
want
to
kind
of
chime
in
on
you
know
as
stakeholders
they
can
just
you
know,
update
update
the
sort
of
these,
the
tracking
issue
to
say
that
they
are
a
stakeholder
in
these
things.
So
how
we
want
to
do
it
I
think
I'm
tracking
issue
to
have
a
list
of
people
it's
like
or.
A
B
B
You
know,
I,
don't
want
to
I,
don't
want
things
to
go
to
process
heavy,
but
if
you
sure
useful,
then
that's
great
I
mean
one
of
the
nice
things
about
issues
is
they're
really
dynamic,
and
anybody
who
wants
to
sort
of
listen
in
can
just
you
know,
add
notification
and
on
an
issue
and
they're
notified
that
doesn't
tend
to
reflect
people
that
are
actively
involved
but
like
if
there's
sort
of
like
a
template
for
like
hey.
This
is
a
tracking
issue.
B
You
know
and
part
of
the
template
is
like
you
know
the
list
of
stakeholders
and
you
can
just
keep
that
dynamic.
Whoever
wants
to
be
involved-
and
you
know
something
could
say
like
hey
I-
want
to
be
involved
and
then
somebody
issue
I,
don't
think
anybody's
gonna
sit
there
and
say
like
no.
You
can't
be
a
stakeholder
unless
highly
antagonist
I.
A
A
It's
something
we
can
so
some
people
hate
wikis.
Some
people
love
them,
but
it's
the
I,
don't
think
it's
it's
easy
to
turn
on.
If
we
don't
have,
it
turned
on
I
believe
I
think
we
just
want
to
link
from
the
first
readme
page.
Then
that
would
be
like
here's,
the
list
of
our
initiatives
that
are
going
on
right
and
then
in
the
wiki.
It
could
have
the
list
that
says:
here's
the
champions
and
here's
the
here's,
the
stakeholders
or
whatever.
A
B
B
If
I
understand
correctly
and
tell
me
like,
let
me
paraphrase
it-
we're
gonna
set
up
a
readme,
that's
going
to
be
in
the
root
of
the
Diagnostics
repo,
it's
gonna
say
like
hey:
these
are
the
issues
we're
currently
working
on.
It'll
say,
like
you
know,
CPU
profiling,
it'll
say
this
is
the
champion:
here's
a
list
of
stakeholders
and
here's
the
tracking
issue
and
the
tracking
issue
referenced
all
of
the
other
sub
issues.
How
and
then,
as
I
bounce
to
this,
the
wiki
page
it'll
just
pick
up
the
readme
well.
I
J
A
B
A
B
B
H
A
Mean
I
think
I
think
we
should
have
like
on
our
regular
meeting.
One
of
the
agenda
items
is
just
a
review
of
these
things,
as
if
we've
said
that
we're
volunteering
to
be
the
champion
or
whatever
we
think
we've
so
at
least,
is
doing
a
quick
update.
Even
if
that
update
is
sorry
didn't
have
any
time
is
worthwhile.
B
Yeah
I
agree
with
that
and
I
think
it
would
naturally
just
well
I,
don't
know,
I
mean
I,
think
for
the
agenda.
Stuff
gets
driven
by
the
issues,
and
so
as
long
as
there's
tracking
issues
open
that
are
tagged
on
the
agenda,
then
we'll
pulse
on
them
in
meetings,
and
so
we
should
cover
that
stuff
and
I.
Think,
as
part
of
that
is
like
you
know,
do
we
have
the
right
set
of
antagonists
in
place
and
whatnot
and
I'm
afraid
that
name
might
stick
now.
G
B
F
I've
made
that
prototype
like
several
weeks
ago,
I
I,
don't
know
if
anyone
never
looked
at
it
or
not.
I've
looked
at
it
very
briefly,.
A
I
missed
the
last
couple
of
meetings.
Sorry,
if
I
ask
a
little
bit
more
context
on
that,
is
it
like
a
fresh
start
or
is
it
leverage
some
of
the
stuff
that
the
earlier
work
on
the
diagnostic
channel
had
done
or.
F
It's
like
totally
fresh
starts,
and
it's
not
really
as
complete
as
the
other
diagnostic
channel
thing
is.
It
doesn't
have
like
actual
proper
like
monkey
patches
for
things
or
anything.
It
was
just
kind
of
like
a
theoretical
model
for
it
and
it
was
kind
of
mostly
started
as
just
a
experimentation
ground
for
a
couple.
Different
ideas.
F
I
wanted
to
try
the
the
main
one
was
just
trying
to
produce
like
it
just
generic
like
messaging
bus
around
Diagnostics
data,
so
it
treats
like
the
events
of
like
a
tres
span,
the
same
way
as
like
metrics
data,
or
things
like
that.
B
B
B
D
B
D
E
D
B
J
So
I
had
a
to
do
from
I
think
the
meeting
about
defining
the
criteria
so
I
open
an
issue
for
that.
It's
more
of
a
discussion.
This
particular
issue,
I,
think,
is
a
is
good
for
tracking
concrete
tasks
that
need
to
happen,
but
there's
I
think
some
more
sort
of
discussion
and
other
questions
about
what
does
it
even
mean?
Is
it
experimental
what
is
non
experimental
in
the
context
of
something
it's
foundational
size?
J
So
I
open
an
issue
for
that
earlier
today
and
please
take
a
look
at
that
one
on
on
the
on,
so
other
things
related
racing
hooks,
so
I
think
the
main.
So
the
biggest
thing
right
now
is
there's
a
regression
on
note
10,
where
it's
in
cooks
is
currently
broken
and
I
think
the
VA
team
is
working
on.
So
there's
optimizations
that
the
VA
to
improve
a
single
performance
in
the
last
couple
of
months.
It
turns
out
that
since
there's
no
integration,
testing
of
icing
hooks
that
happens
bite.
J
So
this
the
semantics
are
not
as
well
specified
and
that
the
fact
that
we,
it
doesn't
know
what,
if
he
mended
test
for
and
know,
doesn't
have
tests
that
those
break
that
functionality
was
broken.
So
I
think
the
B
team
is
trying
to
figure
out
what
all
do
they
need
to
back
back
out.
In
order
for
this
to
get
fixed
beyond
that,
there
was
discussion.
There
was
an
issue
open
issue
number
188
with
the
proposal
on
what
we
need
to
change
in
promised
hooks
in
or
in
order
to
address
some
of
the
performance
concerns.
J
I
think
that
discussion
has
died
down
without
preacher,
who
you
know,
I
need.
Take
a
second
look
at
that
one:
to
try
to
figure
out
how
to
move
forward
on
that.
One
as
andreas
had
some
concerns
with
the
proposals
and
I
think
the
outstanding
thing.
They
too
would
be
try
to
figure
out
what
precise
things
are
objectionable
and
whether
a
subset
of
those
are
still
doable
or
somehow
move
that
forward.
B
Okay,
so
moving
on
async
context,
formalization
diagnostic
support,
so
you
know
it
don't
have
really
anything
anything
new
here
other
than
what
I
anticipate
we'll
do
next
week
as
well.
Try
and
take
the
stuff
that's
been
in
this
sort
of
private
repro
that
we've
been
kind
of
iterating
on
and
try
and
get
it
in
some
consistent
form.
We've
had
some
changes
around
some
of
the
names
of
stuff
and
open
a
PR
into
this
working
group.
B
Yeah
but
I
kind
of
expect
something
along
those
lines.
You
know
along
the
PR
next
week
and
then
mark
is
gonna,
be
in
and
he's
not
on
the
call
today,
unless
he
enjoined
I,
don't
think
he
is
but
he's
going
to
be
in
at
the
collaborator
summit
in
Berlin.
Is
it
on
and
and
so
he
can
discuss
all
this
stuff
there
with
people
so
encourage
folks.
You
know
if
you
know,
if,
if
stuff
in
there
that
you're
like
I,
don't
understand
this
I
disagree
with
it.
B
B
J
I
can
do
an
update,
so
I
think
this
tracking
issues
needs
a
champion
to
actually
go
clean.
It
up,
I
think
lots
of
things
that
are
in
there
are
probably
already
done
or
way
soon
to
be
done
or
palookas
have
been
opened
to
address
those
I
think
James
volunteered,
but
as
a
brief
update
we
so
so
we
have
a
JavaScript
API,
the
teams
added
a
little
while
ago.
Other
things
that
have
happened
is
perf
perfect.
Those
automatically
generate
trace
events,
so
so
that
means
sort
of
it
will
get
hooked
up.
J
So
if
you
have
a
machine
setting
on
the
side
or
a
processing
is
sitting
on
the
side,
that's
controlling
tracing,
there's,
there's
a
bunch
of
other
professors
that
are
open,
more
features
being
added.
Tl
DR.
Is
that
there's
quite
a
bit
of
functionality
that
has
been
added
recently-
and
this
is
shaping
up
very
nicely-
needs
more
testing,
though
so
please
try
it
out.
J
We
have
tests
in
like
normal
JavaScript
test
the
basically
ensure
that
resonance
works,
I,
don't
know
their
comprehensive
enough,
there's
a
few,
a
there's
at
least
ten,
so
I,
don't
know
if
they're
comprehensive
enough,
but
as
people
build
dueling
on
top
of
this
I
think
that
would
be
the
best
way
of
finding
the
holes
and
finding
the
flaws.
Testing
can
only
go
so
far
with
this
one
yeah.
A
I
was
just
wondering
if
it
like
if
people
are
looking
to
help,
is
this
something
where
we
could
list
like
here's
a
list
of
X
tests
that
would
be
useful
to
write
if
that's
feasible.
It's
worth
doing.
You
know,
and
maybe
you
know
that's
a
concrete
thing
that
new
people
could
help
out
with
if
it's
more
like.
Well,
it
takes
too
much
of
a
ramp
up
to
get
to
that
point,
then
maybe
it
doesn't
make
sense.
Yeah.
J
J
There
are
edge
cases
where,
which
I
think
will
only
get
exposed
when
something
is
something
is
operating
at
the
control
plane
right.
So
let's
say:
if
you
have
a
agent
sitting
externally,
that's
going
to
repeatedly
enable/disable
or
it's
going
to
enable
disable
very
frequently,
or
it's
going
to
do
some
post-processing
of
the
events
to
do
some
inferences
and
you
realize.
Well,
you
see
some
of
the
timestamps,
don't
add
up
or
there's
a
hiccup
in
the
middle.
J
B
J
Think
Bradley
might
be
the
best
person
to
answer
that
one
I
think
he
had
some
some
concerns
about
the
ecosystem.
The
broader
JavaScript
ecosystem
leaders
needing
this
anyway,
and
that
and
I
was
out
like
not
just
in
the
diagnostic
context,
but
more
of
being
able
to
write
being
able
to
have
program
ins,
it's
understandability,
improper,
other
handling,
semantics
that
JavaScript
developers
can
use,
but
but
he's
the
right
person
to
you
answer
one,
and
he
wanted
something
like
this
to
be
taking
two
pcs:
everything,
okay,.
D
A
So
I
mean
I
think
was
this
tagged
with
diagnostic
agenda?
Just
in
terms
of
you
know
what
it
needs
is
somebody's
interested
in
investing
time
and
pushing
it
forward,
and
the
thought
was
there
as
a
synergy
with
diag,
not
the
diagnostics
working
group,
so
I
think
that's
why
it
was
tagged
as
a
place.
I
mean.
B
B
It
would
be
great
if
you
know
if
Bradley
or
whoever
else
is
championing
this
could
really
talk
to
them
at
some
point,
I
understand
where
they're
coming
from
I
think
my
point
of
view
is
that
zones
is
a
higher
level
API
than
what
we
really
have
currently
defined
semantics
to
reason
about
right
and
so
I
feel
like
in
some
sense.
It's
a
little
bit
premature
said
this
in
a
comment:
I
only
got
one
thumbs
up,
I
got
no
thumbs
down.
B
So
I
guess
that's
mildly
positive,
but
that's
my
take
right.
My
take
is
that
there
is
a
foundational
level
thing
and
that
you
can
then
sit
there
and
describe,
and
you
should
be
able
to
describe
very
crisply
zone
semantics
over
top
of
this
foundational
level
thing,
and
you
should
be
able
to
then
reason
about
its
correctness
right
without
this
foundational
level.
Thing
I
kind
of
feel
like
zones
is
a
little
it's.
B
B
If
this
continues
to
come
up
and
the
TSC
thing
I,
you
know
it's
like
and
I
think
there's
overlap
with
this
with
both
the
you
know,
the
the
async
content
or
async
context,
formalization
stuff,
as
well
as
the
tracking
issue,
that
elitist,
open,
right
and
I
think
they're
all
sort
of
hinting
around
the
same
thing
right
and
that's
my
tape.
So.
B
It's
like,
we
should
be
able
to
describe
zones
and
the
semantics
of
zone
storage,
error,
handling
across
asynchronous
boundaries
or
error
propagation
across
asynchronous
boundaries.
Right.
We
should
be
able
to
define
that
in
some.
You
know,
explain
it
in
some
well-defined
terms
right,
that
is,
we
should
have
some
fundamental.
You
know,
concepts
to
find
and
some
fundamental
data
structures
in
place
that
we
can
say
like
when
you're
doing
continuation,
local
storage.
B
F
When,
when
the
zones
proposal
started,
it
was
actually
somewhat
more
low-level
than
it
is
now,
and
it
kind
of
became
like
more
high-level,
because
it
was
just
deemed
that's.
Tc39
would
not
really
want
to
deal
with
something
that
low
level
because,
like
it
used
to
get
into
like
how
the
scheduler
works
and
all
that
kind
of
stuff
and
like
actually
exposing
the
scheduler.
B
F
B
Yeah
it'd
be
interesting
if
there's
old
revisions
of
that
somewhere.
That
would
be
interesting
to
see
right.
You
know,
I,
think
that
what
you
know
just
talking
with
our
tc39
right
here,
what
you
know,
what
we
were
told
was
you
know
it's
like
is
to
the
extent
possible
that
you
can
phrase
this
stuff
in
the
terminology
of
the
existing
spec.
That's
a
good
thing
and
you
know
I,
don't
claim
to
sort
of
be
a.
You
know
an
expert
on
the
the
ACMA
scripts
back,
but
you
know
the
I.
B
You
know,
I
think
that
it's
you
know
people
are
talking
about.
You
know
defining
what
this
API
is
I
think
at
some
level
it
needs
to
be
defined.
On
top
of
some
of
these
foundational
concepts
that
are
defined
in
already
defined
in
the
back
right,
at
least
that's
what
I
would
expect
so
and
I
don't
know
if
scheduling
stuff
is
in
the
spec.
B
J
My
point
we
wait
I
think
we
should
remove
that
agnostic
agenda
label
from
this
issue.
So
from
what
I
can
see,
there's
nobody
who's
as
far
as
I
know
nobody's
trying
to
champion
this
at
the
proposal
for
tc39.
So
without
that
I,
don't
I,
don't
think
it's
going
to
go
anywhere.
So,
okay,
if
you
have
another
proposal
that
you
want
to
take
to
tc39
I
I,
think
that's
probably
more
concrete
thing
that
we
should
talk
about
unless
somebody
wants
to
really
step
up
and
yeah.
J
B
I
mean
I'll,
let
you
know
Lee
you
or
Mike
or
somebody
I,
don't
have
I
can't
updated
since
it's
in
the
TLC
thing
so
I'll,
let
you
guys
there
close
it
or
remove
the
label
do
whatever
it
is.
You
need
to
do.
You
know,
I,
think,
with
respect
to
the
zones.
Api
thing
I
think
I'm
as
an
API
I.
Think
it's
fine
and
I
think
it
just
comes
down
this
question
of.
How
do
you
describe
what
the
API
supposed
to
be
doing
right
and
you
know,
is
that
it's
sort
of
this
you
know
kid.
B
I
J
J
The
second
question
was
how
to
join
the
conversation
so
and
I
guess
the
answer
to
that.
One
is
that
in
that
had
no
js'
work
on
github,
there's
a
diagnostic
repository
and
that
one
actually
has
a
link
to
join.
The
meeting
if
are
interested
in
joining
a
thing,
would
be
a
good
to
have
more
participants
and
the
agnostic
effort.