►
From YouTube: 2021-04-14 meeting
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
Yeah,
which
item
are
you
talking
about.
B
B
One
I
pretty
much
only
updated
the
the
last
one
about
drc.
B
B
A
Yeah,
okay,
so
let's
start
hello,
everybody
daniel
is
not
today
he's
off
till
the
end
of
the
week,
so
I
will
run
the
meeting.
Let's
maybe
hear
the
person
who
did
that
the
first
item.
D
Yeah,
so
this
is,
will
I
edit
the
first
few
this
is
related
to
last
week
we
talked
about
the
contrib.
Automated
release.
Workflow
was
having
failures,
so
I
had
just
added
this
pr
to
try
to
fix
it.
So
I
don't
know
when
the
next
contrib
release
is
scheduled,
but
it
would
be
great
to
have
this
merged
in
before
then
so
that
we
can
retry
the
workflow.
D
D
The
next
one
yeah
it
was
on
it
was
I
linked
it
on
the
on
that
pr
yeah,
I
was
yeah.
It
was
this
one
and
daniel.
I
think
commented
on
this,
but
it
was
basically
saying
that
there
might
be
like
a
more
general
problem
with
publishing
to
like
name
spaced.
D
Or
was
it
not
this
one
that
I
counted
on?
Maybe
it
was.
I
think
I
linked
it
in
the
initial
pull
request,
description.
A
A
D
Yeah
there
I
think
up
there
dan
commented
on
it
yeah,
so
it
was,
but
I
I
can
talk
with
dan
about
it
offline
like
next
week,
if
he's
out
for
the
rest
of
the
week
on
slack,
to
see
if
he
thinks
it'll
be
an
issue
again
and
worst
case.
The
next
publish
will
also
just
have
to
be
manual
if
the.
If
the
workflow
fails
due
to
this
issue
again.
A
D
Yeah
right
I
mean
we
might
need
to
add
some
more
secrets
to
the
repo
to
to
get
it
to
succeed,
since
it's
like
an
authorization
issue
of
requiring
like
npm
login,
as
opposed
to
just
providing
the
token.
B
I
see
I
see
that
daniel
commented
on
the
when
he
commented
the
original
issue.
He
said
that
if
you
provide
a
flag,
the
issues
don't
happen
and
he
made
a
pr
on
the
core
repo
to
fix
that
actually.
B
And
pretty
much
it
only
added
a
flag
to
to
I
mean
yeah,
that's
what
he
said
on
the
comments.
So
did
you
try
to
add
the
flag
and
did
it
fix
it
or?
Oh.
D
Like
different
people
had
different
results
and
specifically
with
the
namespace
like
having
the
app
open,
telemetry
namespace
seemed
to
be
causing
some
of
the
issues,
and
so
I
couldn't
really
test
that
because
I
don't
have
okay,
a
name
namespace
packages.
I
can
publish
to
test
that
so
yeah.
So
if
it's,
hopefully
with
that
flag,
it'll
it'll
work
but
yeah,
if
not
it'll,
require
probably
a
second
look.
A
Okay,
I
need
to
make
sure
so
either
daniel
or
me
will
keep
it
to
try
with
the
next
minutes
and
see
how
it
goes.
The
next
issue
is
this
is
also
added
by
you.
I
I
assume.
D
D
Yeah
yeah,
okay.
So
that's
all
it's
just
providing
context
yeah
for
the
next
two,
I'm
just
kind
of
raising
them
on
behalf
of
my
colleague,
anurag
who's
on
he's
in
japan.
So
he's
not
able
to
make
this
meeting,
but
they
are
basically
both
pr's
for
yeah.
Just
for.
A
B
You
know,
but
I
think
he
addressed
all
the
document
by
flana,
so
I
think
it's
fine
to
argue
it
now.
A
Okay,
so,
besides
that
towards
sdkfc,
the
only
thing
that
I
think
left
is
is
removing
the
plugins
from
country.
Diane
was
working
on
this
one,
so
he
just
removed
the
country
but
we'll
take
over
this
one
and
we'll
continue
updating
the
readme
and.
A
So
that
we
can
have
the
pr
this
week
here
and
besides
that.
B
Environment-
I
can
talk
about
this
since
I
try
to
fix
them,
so
those
are
pretty
much
what
we
need.
What
is
what
is
missing
from
the
spec
in
terms
of
configuration
from
the
environment,
so
we
have
two
things
that
we
need
to
be
able
to
configure
from
the
other
the
environment.
So
we
have
the
all
the
exporter
that
should
be
comfortable
from
the
environment.
B
So
we
need
to
have
both
zipskin
and
jager
in
in
the
not
tracing
sdk,
which
is
quite
heavy,
especially
because
we
need
to
bring
the
protobuf
the
world
protocol
dependency
to
be
able
to
send
them
off
with
grpc
so
yeah.
I
I
think
we
should
discuss
that
pretty
much
with
the
spec
when
they
will
be
reviewing
if
we
are
compliant
or
not.
B
A
B
Yeah,
that's
that's
my
main,
my
main
problem
with
the
with
the
crown
spec,
because
we
for
the
spec.
We
should
bring
all
the
exporter
with
the
web
and
so
they
so
they
will
be
comfortable
by
by
their
environment
and
also
by
default.
B
It
specified
that
we
should
enable
the
the
collector
exporter,
which
I
I
think,
it's
quite
weird,
because
if
I
mean
if
we
just
only
enabled
the
export
the
sdk
without
specifying
an
exporter,
we
will
try
to
export
them
to
the
to
the
collector
which
which
not
is
not
really
there.
I
mean
that's
sorry.
A
I
don't
think
should
yes,
yes,
sir,
I
mean
because
we
have
three
different
collector
exporters.
B
Yeah
because
so
the
the
they
when
you
use
the
the
exporter
collector
for
them,
you
just
specify
that
you
want
the
exporter,
the
export,
the
collector
exporter
and
then
with
another
environment
viable.
You
can
configure
if
you
want
the
json,
the
grpc
or
the
http
or
the
printer
over
http,
one
which
is
really
really
hard
to
to
to
do
for
us,
because
we
can
just
bring
them
all
as
dependency,
especially
with
the
browser
so
yeah.
I
I
think
it's.
We
can't
really
implement
it
right
right
now
as
such
as
specified.
A
So
maybe
for
the
js,
we
should
do
this
a
bit
differently,
like
you
could
still.
A
This
by
the
environment,
but
in
fact
you
should
have
something
like
I
don't
know
somewhere
for
the
just
to
be
able
to
register
the
exporter,
and
then
once
you
register
the
certain
exported,
it
will
basically
track
them
for
the
existence
of
the
of
the
variable.
If,
if,
for
example,
user
will
tell
okay,
I
want
the
zipkin,
we
can
simply
tell
him
hey.
You
didn't
register
to
see
this
and
then.
F
Yeah
right,
you
know
that
I
suggested
that
twice.
It
was
strongly
opposed
to
twice
that
was
for
propagators,
but
I
think
it's
a
same
principle.
A
Because,
honestly
for
the
js,
I
don't
really
see
any
other
way
of
doing
this
correctly,
especially
in
the
web,
unless
we
really
dump
this
for
the
environment,
and
we
talk
to
the
specification
that
doing
this,
for
the
js
doesn't
make
sense.
A
G
I
think
yeah,
I
think,
not
only
does
this
not
make
sense
for
js
like,
I
think
it
doesn't
make
sense
for
most
dynamic
languages
where
the
different
exporter
implementations
are
basically
different
packages.
So
it's
kind
of
like
the
package
that
you're,
including
in
your
app,
determines
the
export
format
and
not.
E
G
F
Can't
be
done
for
web,
like
we
look
into
you
many
ways
of
doing
it,
it
just
can't
be
done.
A
20
megabytes
only
so
yeah,
I
know
we
should
probably
raise
the
issue
to
the
spec
to
really
think
this,
and
the
only
other
way
if
they
really
want
this
to
be
done
by
other
environment,
then
probably
we
should.
We
should
think
of
some
name
registration
or
something
like
this
yeah.
B
Well,
I
guess
I
can
raise
an
issue
on
the
spec
about
this
and
explain
a
little
bit
about
the
the
reason
about
that.
C
I
guess
the
other
way
to
look
at
is
on
you
know:
in
a
browser-based
environment
you
don't
really
have
access
to
it
to
the
environment
variables,
so
we
could
effectively
introduce
a
config
which
is
the
environment
for
initialization.
And
if
we
go
down
that
path,
then
the
object
can
contain
a
reference
to
the
to
the
exporter,
but
we
need
to
either
get
the
exception
from
the
specs
or
have
that
as
a
as
an
option.
F
In
the
browser
we
take
these
from
window,
so
you
can
look
at
get
emp
implementation
for
web.
B
B
B
The
last
one
is
pretty
much
the
just
configuring
the
collector
values
like
the
endpoint
and
stuff
like
this
via
the
environment
again,
so
that
was
not
really
simple
to
do
it,
because
we
again,
we
have
like
three
packages
that,
depending
on
the
protocol,
that
we
want
to
use
to
export
the
data,
so
yeah
pretty
much
that
needs
review.
So
yeah,
that's
pretty
much.
F
A
F
So
I'll
do
it
later.
I
don't
think
it's
necessary
right
now.
F
No,
I
haven't
opened
the
pr
yet
I'm
rebasing.
B
B
Yeah
yeah
I
forgot
to
understand,
is
that
an
issue
for
that
yeah
yeah,
because
at
the
time
we
opened
only
one
issues
for
the
general
configuration
of
the
sdk.
I
think
it's
just
actually
the
same
the
same
issues
as
for
the
for
the
sdk.
This
one.
F
B
Yeah
I
tried
to
split
issues,
but
this
was
the
main
one.
I
will
add
it
into.
B
Okay
and
yeah
so
jacob
made
some
comments
about
what
was
missing.
I
responded
and
I
opened
a
different
pr
and
stuff
like
this,
so
so
yeah.
The
only
thing
that's
needed
is
the
is
the
sampler
stuff.
F
Yeah
you're
right,
hey
valentine's
day,.
G
F
Sorry,
no
very
good:
are
you
gonna
open
the
pr
about
the
limits
spam
limits,
or
should
I
do
it?
I.
A
B
The
for
the
zip
kicking
sparked
out-
that's
appear
on
yesterday,
too.
No,
no,
the
matrix,
yeah
yeah
sure
this
one.
This
one
is
not
needed
right
now.
I
was
talking
about
the
the
next
one
on
the
list.
It
was
the
zip
endpoint.
A
B
The
three
exporter
is
the
issues
that
I
spoke
about
just
before
the
span,
the
hotel
response
limits.
We
already
updated
them.
So
it's
pretty
much
only
the
the
first
item
in
the
list
on
the
issues
which
is
basically.