►
From YouTube: 2020-09-02 JavaScript SIG
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
B
B
E
Are
you
new
yeah
I
work
at
new
relic.
I
think
you
and.
E
E
E
B
Okay,
yeah.
We
can
always
use
help,
especially
on
the
browser
side,
because
I
think
right
now,
bart
from
lightstep
one
of
the
maintainers
is
our
only
person
that
is
really
browser
focused
at
the
moment.
So.
B
All
right
about
five
minutes
after
start,
I
think
we're
probably
good
to
start
real
quick.
I
wanted
to
just
mention
we
did
release
version
11
this
week.
So
thank
you
to
everybody
involved
in
that,
and
I
wanted
to
draw
everybody's
attention
to
the
spec
compliance
matrix.
B
I
have
a
pr
open
for
the
spec
compliance
matrix.
I've
been
trying
to
keep
it
up
to
date.
I
know
that
bart
has
also
made
a
couple
of
updates,
but
just
so
that
everybody
knows
that
this
exists.
It's
sort
of
a
list
of
of
spec
features
that
we're
supposed
to
implement.
B
B
So
I
just
want
to
make
everyone
aware
that
that's
there
and
please
look
at
it
and
I've
been
trying
to
to
keep
it
up
to
date.
But
if
I
missed
anything,
please
let
me
know
I
wanted
to
also
mention
the
synchronous
sdk
update.
B
I
opened
a
new
pr
for
this.
I
know
there's
been
a
couple
that
have
opened
and
closed
in
the
past,
but
the
idea
here
is
that
the
sdk
package
currently
requires
you
to
await
the
startup
before
you
can
start
your
application,
and
this
change
makes
it
so
that
the
resource
input
to
the
tracer
and
meter
providers
can
be
a
promise
to
a
resource
as
well,
which
means
that
the
startup
can
be
completely
synchronous.
B
So
you
can
start
the
sdk
and
immediately
start
tracing,
and
it
will
await
the
resources
behind
the
scenes
appropriately.
So
I
realize
it's
a
big
change,
but
honestly
most
of
it
is
tests,
so
I
would
appreciate
it
if
people
would
review
this.
I
know
it
is
kind
of
a
lot.
B
B
B
So
if
anybody
has
anything
that
they'd
like
to
bring
up
before
we
do
that,
I
think
it
will
take
quite
a
while.
So
normally
I
ask
at
the
end,
but
for
now
today
I'm
gonna
ask
now:
does
anyone
have
anything
that
they
want
to
talk
about
today,
or
should
we
go
directly
into
the
triage.
B
So
we
have
18
open
issues
here.
Some
of
them
are
assigned,
most
of
them
are
unassigned
and
I'm
just
going
to
go
through
them
one
by
one
and
we
can
decide
which
ones
to
assign
which
ones
to
close
things
like
that.
B
This
is
assigned
to
someone,
that's
not
on
the
call
today,
but
I
think
that
this
is
actually
superseded
by
the
parent-based
sampler.
I
know
that
matt
you
have
some
familiarity
with
the
sampling
spec.
Am
I
correct
in
saying
that.
F
F
Possibly
like
does
this
predate
yeah?
I
think
this
predates
the
most
recent
set
of
changes
I
think
we
would
have
to.
F
I
would
have
to
review
what
what
just
happened,
but
I
think
this
is
probably
it's
probably
out
of
date.
B
Yeah,
I
think
the
probability
sampler
was
renamed
to
trace
id
ratio
based
sampler
and
is
a
thick
probability
based
on
the
trace
id
now
and
then
all
of
the
parent
stuff
is
handled
by
a
parent
based
sampler
with
a
delegate
which
you
would
then
be
the
trace
id
ratio
based
sampler.
I
guess
so
between
these
two
updates.
I
think
this
issue
is
superseded.
So
let
me
there's
that
one.
B
B
And
I
think
I
have
duplicates
here
so
1487
and
1457.
so
I'll,
just
close,
whichever
one
I
don't
think
there
was
any
interaction
with
either
of
them,
so
it
should
be
okay
to
just
close
them.
E
E
B
The
hmp2
semantic
convention
for
content
size-
this
is
just
a
semantic
convention
that
I
noticed
we
aren't
following
because
it
was
updated.
Does
anyone
want
to
volunteer
for
this
one,
so
I
can
assign
it.
C
Okay,
I
can
do
that.
Okay,.
B
Okay,
high
cardinal
high
cardinality
hdb
span
names.
I
think
this
one
is
already
fixed
for
the
xhr
plug-in
and
it
looks
like
there's
an
open
pr
for
http
plugin
and
the
https
plugin
inherits
from
this.
So
I
think
it
should
be
all
done.
I'm
going
to
assign
this
to
morig's
because
he
opened
that.
B
B
Rename
plugin
to
instrumentation.
I
think
this
is
low
priority.
I'm
just
going
to
move
this
to
the
bottom.
For
now,.
B
All
right
matt,
this
issue,
the
resource,
auto
detection.
Otec.
Can
this
issue
be
closed?
Now?
Are
we?
Are
we
compliant
here.
F
So
I
kind
of
went
through
and
wrote
up
the
differences
between
our
implementation
and
the
otep.
I
think
the
spec
I
think
spec
changes
have
actually
come
through,
which
I
don't
know.
I
could
do
like
a
final
pass,
but
I
think,
like
I
think,
we're
compliant,
but
I
should
probably
make
sure.
F
Yeah
I'll
close
it
or
just
mention
whatever
else.
D
B
F
B
H
H
I
H
G
F
B
Plugin
grpc
does
not
filter
outgoing
requests.
I
think
this
may
be
fixed
now,
but
the
pr
was
not
linked
to
it.
H
B
B
B
The
attribute
implementations,
so
this
is
just
to
make
sure
that
we
are
accepting
the
various
data
types
that
we're
supposed
to
the
string
number
null
and
I
think
list
there's
a
handful
of
data
types
we're
supposed
to
allow.
B
B
Resource
auto
detection
and
sdk
initialization.
I
think
this
is
the
issue
that
the
pr
for
the
synchronous
sdk
is
actually
supposed
to
fix.
It's
just
not
linked.
So
let
me
link
that
and
I
believe
this
is
actually
not
required
for
ga,
so
I'm
going
to
remove
it
from
the
project.
B
B
B
Okay,
so
now
we're
into
like
the
end
of
the
unassigned
ones
here,
so,
according
to
the
spec,
it
should
be
possible
to
propagate
span
context
using
only
the
api
without
an
sdk,
which
means,
if
you
create
a
span
with
a
parent
from
a
propagator.
B
B
Is
there
anybody
that
wants
to
volunteer
to
take
this
or,
if
I'm,
not
explaining
it
properly?
Please
let
me
know.
B
Okay,
so
implement
text
map
propagator
fields.
I
noticed
this
when
I
was
updating
our
spec
compliance
matrix.
I
actually
don't
know
what
this
specification
is,
but
I
think
that
we
are
not
following
it.
So
the
predefined
propagation
fields
of
carriers
were
used.
You
should
delete
the
fields
before
calling
inject.
B
So
I'm
not
entirely
sure
whether
we're
following
this
or
not.
I
guess
I'll
assign
this
to
myself
before
telling
somebody
to
fix
this,
and
if
it
turns
out
not
to
be
an
issue
so
I'll
take
this
one
all
right.
So
the
getter.keys
call
right
now
our
propagators
take
a
getter
and
a
setter
argument,
which
is
just
a
function
that
takes
the
the
carrier
and
the
key
and
the
value
and
sets
it
or
takes
carrier
and
a
key
and
gets
the
value.
B
But
apparently
we
need
more
operations
than
just
that.
So
I
think
we
should
change
those
to
be
an
interface.
B
Then
rename
probability
sampler
to
the
trace
id
ratio
based
sampler.
We
already
talked
about
this
one.
Does
anyone
want
to
take
this,
or
should
we
wait
until
we
hear
back
from
legendicus
about
the
other
one?
B
B
B
Okay
and
then
rename
plug-in
to
instrumentation,
so
this
one
is
going
to
be
a
big
pr.
It's
not
really
any
functionality,
but
it
requires
package
renaming
and
things
like
that.
You
guys
think
that
we
should
just
bite
the
bullet
and
handle
this.
It's
we're.
Gonna
have
to
do
it
before
ga,
so
I
don't
really
see
a
reason
to
delay
it
any
longer.
B
Yeah
just
to
make
fewer
conflicts,
I.
A
I
mean
I've
already
like
a
bigger
pr
that
and
now
we're
naming
anything
like
this.
It's
always
causing
like
lots
of
headaches.
So
maybe
just
choose
the
the
moment
when
to
do
this.
B
Yeah
so
in
terms
of
open
prs,
there's
a
couple
big
ones.
I
think
I
know
the
sdk
one
that
I
made
is
pretty
big.
Are
there
any
other
like
big
prs
that
are
making.
A
B
C
We
might
want
to
finish
landing
that
migration
of
node
plugins
all.
C
C
D
A
G
I
mean
we
should
use
11,
that's
fine.
When
we
I
mean
you
won't
be
able
to.
A
I
mean
because
the
plugin
all
version
0.11
is
already
released
yeah,
so
you
won't
be
able
to
release
it
from
the
country
under
the
same
version.
So
you
need
to
release
it
as
at
least
11.1
okay.
So
this
is
what
we
don't
want,
probably
to
be
over.
A
You
know
to
have
greater
version
number
than
than
than
incorrect
people.
I
was
just
wondering
I
know.
Maybe
we
should
rename
this
plugins
all
to
something
different
name
or
come
out
with
another
solution.
If
you
have.
A
G
A
A
Because
either
we
do
the
release
of
the
car
with
next
number.
By
already
removing
the
note
all
and
then
it
will
give
us
possibility
of
releasing
under
the
same
version
with
the
next
version
of
from
the
country.
So
we
can
cut
the
next
release
11.1
without
the
node
all
and
then
to
the
country
release
as
11.1
with
the.
A
A
No
no
yeah
but
just
add
a
contrib,
so
put
the
same
name,
but
just
a
dash
country,
for
example.
I
don't
know.
C
A
C
Yeah,
I
don't,
I
don't
quite
follow
what
we're
doing
here
and
that's
probably
on
me,
apologies.
So
the
the
problem
is
the
plugins
node
all
which
I'm
in
the
process
of
moving
into
the
contrib
library
is
a
couple
versions
ahead
of
everything
else
in
contrib.
B
B
C
That
that
works
for
me
and
are
we
so
originally
when
we
talked
about
this,
this
migration
work,
I
was
told
that
I
needed
to
move
it
to
contrib,
do
a
release
and
then
remove
it
from
open,
telemetry.js.
C
It
kind
of
sounds
like.
Maybe
I
should
be
removing
it
from
open
telemetry.js
sooner
than
later,
because
we're
we're
having
these
ongoing
version
issues.
Is
that
accurate?
I.
A
Mean
we
want
heavy
if
we
put
this
under
different
name,
so
I
would
like
to
avoid
basically
the
situation
like:
let's
someone
discover
it
back
now
and
we
want
to
make
a
release
and
we
are
blocked
because
we
just
moving
this
around
and
so
on.
If
we
put
this
under
a
different
name,
we
can
do
the
next
releases
on
the
curve
branch,
and
once
once
this
plugin
is
approved
to
release,
then
we
can
duplicate
the
main
package
and
remove
it.
C
C
Okay,
that's
that's
where
I
thought
we
wanted
to
go,
but
I
wasn't
quite
following
so:
okay,
so
yeah
I'll
I'll.
Do
that?
Should
I
get
that
pr
landed
and
then
do
the
rename
or
do
you
want
me
to
do
the
rename
as
a
part
of
that
pr?
That's
that's
pending.
F
Okay,
we
did
this
in
ruby
not
too
long
ago,
and
one
of
the
things
that
did
come
up
was
instrumentation
versus
instrumentations
and
I
guess
instrumentation
is
the
is
also
plural.
Without
an
s
and
the
main
spec
seems
to
use
that
terminology.
So
we
we
went
with
that.
That
instrumentation
is
the
plural
of
instrumentation,
so,
like
our
our
all
package
is
called
instrumentation
all
without.
B
On
us
yeah,
I
think
I
like
it
better
without
the
s.
I
understand
why
some
people
would
want
the
s
there,
but
I
think
I
prefer
it
without.
F
B
Okay,
so
it
looks
like
we're
through
our
required
for
ga
if
she
is,
except
for
this
content
size
one
did
I
miss
that
one
or
does
anyone
want
to
volunteer
for
this,
or
should
I
just
take
it.
A
B
D
I
E
A
I
I
Oh
yeah
sure
it's
yeah,
it's
it's
very
long.
Actually
it
has
yeah
I'll
either
deflect
yeah.
Just
add,
though
I
know
it's
long
but
just
add.
B
The
full
one
there
because
it's
hard
to
right
right,
hard
to
guess
what
this
would
be,
but
I
don't
think
the
base
plugin
changed.
A
I
It's
a
actually,
you
know
only
for
this.
Plugin
has
issue,
like
others
like
a
new
xml
http
request,
plugin
that
one
has
no
issue
so
like
only
this
one
has
some.
D
I
Me
yeah,
oh
from
the
previous
version,
it's
a
just
small
version
bomb
yeah.
I
don't.
I
think
I
can't
remember
the.
Let
me
check
the
log.
Actually,
the
old
version
is.
I
Yeah
yeah,
I
got
yeah
god.
I
think
it's
a
type
here.
Actually
you
know
I
remember
yeah.
I
will
try
exactly
and
comment
back
there.
Okay.
B
B
B
B
B
H
Yeah
sure
on
this,
not
not
really
on
the
on
this
issue,
especially,
but
I
think
we
should
add
something
on
the
on
the
default.
When
someone
is
roboting
a
bug,
because
it
can
happen
that
someone
isn't
using
jsdk
properly
and
almost
all
of
the
time
I
just
directly
asked
the
the
person
that
opened
the
issue
like
how.
D
H
Oed
initially
sdk
did
use
correctly.
The
propagator
correctly
declared
the
enable
js
in
cook's
contact
manager
because
it
still
needs
to
be
initialized
and
stuff
like
this.
So
I'm
not
sure
what
you
think
about
it,
but
I
think
it
will
be
quite
helpful
to
under
all
the
bugs
issues.
H
Yeah,
I'm
not
I'm
not
saying
that
it's
all
of
the
time,
but
most
of
the
time,
especially
with
the
context.
It's
some
people
that
just
doesn't
initialize
the
contextual
manager.
H
So
I
I
I'm
not
sure,
but
I
think
it
will
be
helpful
to
just
ask
the
people
if
they
can
just
copy
past
the
code
they
are
using
to
initialize
the
sdk.
H
H
B
Okay,
I
will
add
that
to
the
issue
template
today,
then.
H
B
Error
loading
plug-in
when
it's
already
being
patched
by
replayer
in
the
middle-
oh
this
one.
So
this
happened
because
one
of
the
plugins
required
the
http
module
that
the
htp
plugin
required
the
http
module
which
made
it
a
circular
require.
I
have
a
pr
for
this.
I
just
haven't
opened
the
pr
yet
so
I'll
do
that
today
we
already
talked
about.
I
don't
think
this
is
a
bug
anymore,
but
I
think
we
do
still
want
to
move.
Plugins
know
it
all
to
contrib.
So
I'll
just
remove
the
bug.
D
D
B
Socket
destructuring,
error
and
plug-in
http,
so
I
tried
to
look
into
this,
but
I
didn't
get
very
far
on
this.
Has
anyone
else
looked
at
this
bug.
B
I
think
oliver's
not
on
the
call,
but
he
wrote
this
plugin
initially.
Let
me
mention.
G
B
B
Yeah,
so
I
think
the
types
are
wrong
so
that
the
issue
is
that
he
tried
to
it
tries
to
get
the
local
address
from
a
socket,
but
the
socket's
not
and
then
the
types
right.
So
this
is
the
line.
B
So
we
need
to
do
some
null
checking
here,
but
I
was
not
able
to
reproduce
it
because
I
couldn't
I
couldn't
reproduce
the
situation
where
the
socket
was
actually
null.
B
Right
zipkin.
A
B
The
the
deep
merge
issue
who
wrote
the
original
deep,
the
the
original,
merge
logic
I
forgot-
who
it
was,
I
think
it
was
nastine.
A
But
from
what
I
remember,
we
already
mentioned
this
somewhere
about
this
deep
match.
B
Yeah,
it
was
in
a
pr
somewhere
yeah,
so
I
think
I'll
just
mention
him
on
this
issue
and
hopefully
I
can
assign
it
to
you
and
then
you
can
fix
this.
If
not
I'll,
probably
take.
B
Oh
yeah,
so
the
waving
integration,
since
this
view
is
creating
and
boarding
the
three
headers
which
open
telemetry
is
not
expecting
so
I'll
assign
this
to
myself.
I
think
it.
This
is
a,
I
think,
he's
just
using
the
sdk
incorrectly,
so
valentine
again.
That
comes
back
to
what
you
said.
I
think
we
should
add
that
to
that
template.
B
G
A
A
B
B
An
eps
exporter,
observer
aggregator,
looks
like
valentin
you're
assigned
to
this.
Are
you
still
working
on
this,
or
is
this
done.
H
I
think
it
was
no
oh
no
right
well,
it
was
a
really
really
old
and
I
think,
since
the
prometus
exporter
was
rewards,
I'm
not,
I
think
it
was
implemented,
but
I'm
not
sure
because
I've
met
the
pr
that
I've
opened
the
time
with
was
with
the
from
the
prom
client
from
its
clients
and
since
it
was
removed
by
one
another
pr.
I
just
guessed
that
it
was
down.
B
Okay
mark
are
you
on
the
call.
B
It
looks
like
yes,
hey
yeah.
Is
this
something
that
you
want
me
to
assign
to
you,
or
are
you
just
trying
to
bring
this
to
people's
attention?
Let.
J
Me
think,
okay,
so
this
is
related
to
the.
I
guess,
the
other
issue
that
we
talked
about
earlier
with
the
filtering
methods.
I
think
the
thing
I
was
just
flagging
here
was:
there
was
a
central
config
for
all
plugins
to
ignore
urls,
but
I
don't
think
any
of
the
any
of
our
plugins
are
actually
using
this.
D
J
J
So
yeah,
like
the
the
I
guess,
the
confusing
part
here
was
yeah.
There's
the
the
thing
the
I
guess,
the
difference
between
incoming
and
outgoing
and
whether
or
not
we
should
have
a
filter
for
those
also
or
we
should
like
separate
those
out
and
which,
which
ignore
your
health.
Doesn't
let
you
do
so.
I
think
right
for
bar
for,
like
the
xml
http,
there
is
no
incoming
so
ignore
your
roles.
You
should
say
I'm
not
good
or
else
right
anyway.
Yes,.
J
A
Just
checking
their
to
just
checking
the
code,
because
I
don't
remember
exactly,
but
so
the
ignore
url
is
basically
is,
is
taking
into
consideration
for
your
xml
http
request,
I'm
not
sure
about
other
plugins.
B
A
A
C
A
But
we
don't
have
it
here
on,
we
don't
have
yet
to
ignore
outgoing
and
ignore
incoming.
You
don't
have
it
yet.
On
the
plugins.
B
D
B
I
would
prefer
not
to
have
it
on
the
global
config
honestly,
because
we're
going
to
run
into
situations
like
this,
where,
like
some
plug-ins,
will
implement
it
and
some
might
not
or
they
implement
it
differently.
B
Yeah
enabled
obviously
has
to
stay,
but
I
mean
like
behavior
options.
I
mean
only
only
keeping
global
options
that
actually
apply
to
every
single
plugin.
B
And
it's
past
one
o'clock
so
I'll,
let
you
guys
go.
We
can
do
the
rest
of
these
next
week,
they're
all
we're
getting
into
the
pretty
old
ones
now
anyway.
So
if
anyone
has
anything
that
they
particularly
want
to
bring
up
before
we
go.
B
B
B
Okay,
thank
you,
everybody
for
your
time.
I
realize
it
was
a
sort
of
a
draining
meeting
this
week
won't
be
like
this
every
week.
I
promise,
and
I
will
talk
to
you
all
next
week.