►
From YouTube: 2022-06-23 meeting
Description
cncf-opentelemetry meeting-2's Personal Meeting Room
A
B
B
A
A
A
A
A
A
B
B
To
get
started
in
not
necessarily
important
beats,
but
it
took
me
time
to
realize
you
you,
you
also
talk
into
the
elixirs
lag
and
even
though
I
have
an
account
was
quite
long
since
I
logged,
so
I
didn't
saw
your
messages
there.
I
would
process
to
the
sig
process
to
b2b
part
to
join
that.
You
asked
you
pointed
out
earlier
from
joining
yeah
that
you
and
brian
can
sponsor,
so
it
can
be
group
and
okay,
great
and
on
the
legs.
B
C
A
They
gave
ownership
to
me
or
we
have
combined
ownership
over
the
absent
hex
package.
So
that's
moving
along
slowly,
but
surely
the
figuring
out
how
to
merge
these
two
or
replace
that
one?
Do
you
think.
A
A
Said
the
last
time
that
some
of
the
maybe
telemetry
events
were
not
yet
in
the
release
of
absent
that
you
rely
on.
B
Yeah
yeah,
one
of
those
is
merged.
The
other
ones
still
waiting
for
review.
The
the
car
maintainers
of
absent,
say
nothing
about
not
good,
not
bad,
just
didn't
give
any
review
so,
but
even
the
one
that
got
merged,
it's
still
not
released
in
any
any
backup.
So
we
are
kind
but
of
I
I
think
in
in
regard
of
this
instrumentation
library
yeah,
like
the
the
the
plan
that
it
suggests.
A
The
their
library
into
contrib
and
then
mergers
into
it.
B
C
B
I'm
I'm
good.
I'm
good!
Okay,
because,
like
I
was
so
being
practical
here,
also
have
no
idea
when,
when,
when
absent,
will
and
if
will
merge
and
at
what
extent
of
those
changes
would
be
because
like
does
any
feedback
from
from
maintainers,
so
okay,
they
make
sense,
but
maybe
maybe
they
not,
maybe
so
you're
kind
of
weird.
B
B
Like
if
there
is,
if
there
don't
have
the
bandwidth
to
do
this
work,
maybe
I
can
do
like
I
can
open
another
pr,
just
like
copy
and
paste
our
code.
I
I
know
they
do
some
different
beats
in
terms
of
the
ci
parts.
Ci
cd.
B
Yeah
because
like
if
the
problem
is
bandwidth
on
their
side
to
do
this,
this
move,
I'm
not
sure
like.
If
legally,
I
could
do
that
because
there's
a
other
the
bits
about,
like
I'm,
seeding
the
code
to
open
source.
So
I'm
not
sure
if
I
can
do
this,
but
in
terms
like,
if,
if
needed
help
here,
I
could
do
it
like
copy
the
code,
adapt
the
ci
to
the
one
that
we
are
using
and.
A
A
A
Yet
using
any
future
flags
brings
a
pull
request
from
ice
jam
who
to
clean
it
up
we'll
do
it,
and
I
don't
want
to
keep
bringing
this
up
everywhere
to
get
as
many
eyes
since,
especially
since
I'm
not
really
familiar
with
phoenix
to
you
know,
anyways,
to
say
this
should
be
better
than
this
could
become
that
this
could
be
removed.
Stuff,
like
that,
so
yeah
just.
B
A
Yeah,
yes,
and
it
looks
like
he's-
got
a
lot
in
his
pr.
So
that's
good.
There
might
not
be
much
that
needs
to
be
done
or
anything
anymore,
but
the
phoenix
generators
it
creates
a
whole
bunch
of
stuff
and
it
needs
to
be-
and
I
don't
want
to
just
start
pulling
stuff
out
because
I
might
break
it,
remove
something.
That's
is
necessary
that
I
couldn't
tell-
and
you
know
things
are
attached
weird
so
yeah.
B
B
B
So
I
saw
your
original
message.
I
consciously
avoid
it.
It's
sort
of
like
that
feeling
I'm
doing
a
lot
of
instrumentation
things.
Many
people
want
to
contribute
to
the
project.
So
this
look
look
like
a
good
thing.
Like
hey
it's
more
like
example,
app.
You
don't
need
to
get
too
fancy
into
internals
yeah,
but
I.
A
A
Yeah,
I'm
gonna
review
it
sometime
today.
Hopefully
I'm
sure
I'll
get
to
it
sometime
today,
just
on
the
wind
but
yeah.
A
A
B
No,
maybe
not
maybe
not
much
updates.
I
have
more
of
a
question.
I
know
there's
certain
I
remember
there's
pr
opens
around
like
metrics
and
metrics
mean
a
hot
topic
internally
here.
So
it's
more
like
a
question.
Curiosity
like
where
we
are
in
this
regard
like
there's
something
that
community
can
can
help
to
do
that.
We
are
waiting
upstream,
something.
A
I
think
I
started
getting
somewhere
with
it
last
night
and
I'm
hoping
that
keeps
moving
along
today
and
I
might
have
maybe
even
like
a
draft
pr
for
the
text,
decay
that
I
can
open
that
can
it
will
be
more
of
a
trying
to
get
the
overall
structure
of
the
sdk
right,
how
it
because
there's
a
number
of
moving
parts,
because
in
order
to
support
exporters
that
are
both
like
prometheus
and
like
statsd,
is
a
pain
not
just
because,
like
one
person
wants
porno,
but
because
you
have
to
like
one
of
them
restarts
counters
after
every
collection,
every
push
every
other
one
doesn't,
and
that
kind
of
stuff
needs
to
be
able
to
sing
both
at
the
same
time.
A
So
yeah
it's
a
bit
of
a
pain,
but
hopefully
I'll
have
something
that
can
really
feel
the
view
in
the
sense
of
you
know.
Does
this
make
sense
as
it's
going
to
be
horribly
inefficient?
Is
this
you
know?
Does
it
supervise
your
treat?
Look
right.
You
know
and
stuff
like
that.
The
api
pull
request
is
up
already
yeah.
A
If
you,
if
you
or
anybody
else
that
you
work
with
one,
must
take
a
look,
that'd
be
great,
that
fred's
reviewed
it,
and
I
can't
remember
if
there's
anything
left
from
his
review,
that
I
need
to
do
but
yeah
that
would
get
that's
the
state
yeah.
B
That's
great
that's
great
today.
Thank
you.
It's
kind
of
like
I
was
really.
I
was
fearful
that,
like
what
would
be
holding
would
be
like
specification
things
or
like
yeah.
I
need
to
wait.
No!
No!
It's
just
like
the
implementation
is
quite
challenging.
Like
okay,
okay,
it's
simple,
that's
not
cool,
but
it
is
what
it
is,
and
this
is
painful.
B
Awesome,
I
don't
think
I
have
much
further
like
if
this
was
does
report
like
more
I'm
still
like
slowly
tweaking
some
of
of
the
instrumentation
things,
I'm
looking
towards
the
broad,
the
kafka
client
thing
and
looking
for
a
means
to
do
it,
but
feels
like
some
folks.
I
think
you're
included
it's
talking
about
inserting
telemetry
into
the
main
library.
So
probably
this
one
would
be
useful
to
wait
to
have
extra
telemetry
based.
A
A
Yes
again,
yeah
if
I
had
instrumented
with
open
sensors,
so
it
would
be
very
similar
to
do
the
the
c
based
one.
I
can't
remember
the
name
of
that
the
earling
app
for
that
broad.
I
don't
remember
looking
at,
but
I'm
sure
I
did
add
telemetry.
The
only
way
you
would
add
telemetry
is
it
telemetry
doesn't
provide
a
way
to
add
metadata
to
propagate
trace
content.
So
that's.
A
B
Yeah
kind
of
like
this
propagation,
like
messaging,
is
something
I've
been
playing
a
bit
here
like
a
few
different
approaches,
trying
to
avoid
requiring
to
wrap
everything
as
much
as
I
can
find
places
which,
by
the
way,
I
may
have
like
an
an
alternate
implementation
for
obama
that
requires
less
wrapping
all
right,
yeah.
I
was
talking
with
with
open
maintainers,
but
then
I
need
to
talk.
I
forgot
name
it's
something
with
I:
the
open,
telemetry,
open,
maintainer,
oh.
B
A
A
C
B
To
espresso,
I
think
I'm
good
thanks
for
the
update
with
matrix.
This
is
quite
useful.
A
All
right,
yeah,
thanks
I'll
talk
to
you
on
slack
and
next
week.