►
From YouTube: 2020-10-08 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).
A
A
Just
wait:
wait
a
little
bit
for
alex
to
join.
A
C
A
If
y'all
could
fill
out
the
attendees
list,
that'd
be
great,
so
we
can
find
you
guys
and
blame
you
for
being
present
in
the
future.
A
Okay,
well,
I
guess
we
could
just
start
cool.
So
a
couple
of
topics
to
talk
about
today,
so
ga
is
quickly
approaching.
A
I
believe
the
actual
hard
goal
that
you
people
are
kind
of
just
hitting
would
be
the
kubecon
which
was
mid-november,
so
we
expect
to
like
have
all
of
our
code
complete,
like
ga
stuff
ready
by
around
like
the
first
week
of
november,
so
that
gives
us
like
about
a
month
to
get
all
of
our
stuff
in
there's
still
quite
a
bit
of
open
issues,
especially
because,
like
the
these
specs
up
until
today
are
still
changing.
A
So
it's
quite
quite
difficult
right
now,
but
the
like
alex-
and
I
discussed
what
we're
gonna
be
trying
to
do
is
like
close
all
of
the
current
issues
that
we
have
so
far
like
try
to
push
those
and
then
do
like
one
more
sweep
of
all
the
spec
issues,
and
hopefully
we'll
have
a
pretty
good
velocity
and
like
have
an
active
month
this
for
the
next
few
weeks.
A
With
that
being
said,
we're
really
trying
to
focus
only
on
ga
tasks.
Now
I
don't
believe
python
especially
has
a
real
lack
of
resources
and
like
we
only
literally
only
have
like
four
or
five
active
approvers,
so
really
try
to
like
if
you're
gonna
be
picking
up,
work
like
it
would
be
appreciated.
If
you
like,
were
to
focus
on
the
require
for
ga
tasks
and
if
not
just
like,
don't
be
like
too
surprised.
E
Oh
yeah,
that
was
just
wanted
to
call
out
that
I
one
of
the
processors
that
we
wanted
to
try
it
at
this
meeting
was
to
ensure
that
all
of
the
open
pr's
that
are
marked
for
required4j
have
assignees
to
kind
of
carry
them
across
the
finish
line.
A
Sure
yeah,
so
we're
gonna
have
to
try
to
like
for
the
open
prs,
as
alex
said,
try
to
like
at
the
end
of
by
the
end
of
the
meeting.
We
should
have
like
actionable
things
to
do
to
move
prs
forward,
because
some
of
them
have
been
just
floating
around
for
a
while,
and
you
know
we
really
need
to
get
those
in
some
of
them
only
need
like
one
more
approver
to
do
it
or
it's
like
just
an
open
comment,
so
we'll
be
going
through
those
as
we're
going
through
the
pr's
today.
A
Another
thing
I
want
to
mention
is
like,
if
you
want
to,
if
there's
an
open
issue
that
you
want
to
discuss
or
like
a
problem
with
one
of
the
pr's
that
you
don't
really
know
how
to
deal
with,
I
really
emphasize
like
you,
you
have
to
join
the
sig,
or
else
most
of
the
time
like
we
won't
be
able
to
be
exposed
to
that
problem
like
alex,
and
I
and
the
other
approvers
are
actively
reviewing
all
the
pr's.
A
But
if
you
really
want
to
accelerate
you
know
the
merging
end
of
your
pr
and
to
have
discussion
for
it.
It's
imperative
that
you
join
the
join
the
sig
and
also
like
call
out
your
pr
issue
like
like
it,
there's
no
point
of
just
sitting
there
and
like
having
the
pr
sit
there
and
not
being
exposed
so
really
try
to
yeah
like
at
least
put
it
in
the
the
the
meeting
notes,
so
we
can
see
it
and
get
to
it.
A
Cool
second
thing
I
want
to
talk
about
is
a
certain
aws
responsibilities,
so
I
know
we
have
a
lot
of
like
interns,
this
semester
and
you
guys
have
been
doing
really
really
awesome
work.
You
know
like
putting
in
a
lot
of
like
the
the
required
for
ga
stuff.
I
kind
of
do
want
to
sync
up
on
like
the
kind
of
expectations
from
at
least
the
maintainers
group.
A
It
was
discussed
that,
like
we
really
need,
you
know
you
guys
to
have
like
your
main
mentors
or
managers
more
actively
involved
in
at
least
your
prs
or,
like
you
know,
the
general.
You
know
what
it
would
take
to
make
you
successful
this
semester
and
already
with
like
the
rest
of
the
the
maintainers,
the
approver
is
kind
of
swamped
with
actual
ga
tasks.
A
It's
difficult
for
us
to
provide
you
a
lot
of
help
without
like
assistance
and
directions
from
your
your
your
mentors,
and
this
was
like
explicitly
said
by
the
maintainers
meetings.
It's
like
the
mentors
it's
their
responsibility
to
be
like
the
actual
drivers
behind
you
guys,
not
us
so
kind
of
just
wanted
to
like
talk
to
the
interns
here
like
I'm,
not
sure
it
wasn't
expected
for
them
to,
like
you,
know,
join
the
cigs
or
like
be
more
active,
but
I
haven't
received
any
word
from
you
know.
A
Anyone
yet
so,
especially
like
I
I
see
like
nathaniel
put
his
next
topic.
The
contra
reviews,
repo
stuff,
like
that
really
needs
to
be
talked
about
in
like
a
collaborative
design
plan
needs
to
be,
you
know
established,
and
it's
difficult
without
any
communication
with
the
other
party.
So
I
can
see
like
amos
is
in
the
call
and
nathaniel's
on
the
call.
So
I
guess
do
you
guys
have
any
updates
on
that
or
anything
to
say.
B
Hey
on
on
my
side,
like
it's
the
same
update
that
I
I
mentioned
in
the
gita
channel
that
you
know
I
we
we're
just
trying
to
get
these
things
for
our
release
soon
as
well,
and
when
I
mentioned
that,
oh
you
know
like
the
maintainers,
don't
want
it's
in
the
main
repo
they
wanted
the
contribution,
but
the
contributor
needs
some
some
work
to
get
it
through
like
they
were
all
for
that
right.
B
They
said,
okay,
sure
like
let's
like,
like
I
mentioned
they
assigned
one
week
of
my
time
to
to
help
out
how
I
can,
and
you
guys
said
that
you
guys
needed
a
bit
of
time
to
work
on
it
and
get
it
ready,
and
they
said
they
were
fine
with
that
too.
So,
like
we're
here
to
help
too
how
we
can
and
if,
if
it
means
like
going
going
another
way,
then
that's
totally
cool
too
so
right,
I
don't.
A
Oh,
like
specifically
like
in
in
the
maintainers
meetings,
I'm
not
even
specifically
talking
about
your
contributor
test,
like
it
was
expected
for
like
mentors
to
be
the
first
approver
like
specifically
that
and
like
we've
had
a
lack
of
that
which
is
like
fine
for
the
time
being
because,
like
you
know,
we're
trying
to
rush
things
for
ga,
but
we
don't
really
want
to
set
the
precedence
that,
like
we're,
able
to
do
this
constantly.
A
It's
really
expected
for
the
mentors
to
be
actively
reviewing
and
looking
at
everything
that
you
you
ship
and
be
involved
in
a
lot
of
the
discount
design
discussions
like
specifically
for
this
contra
repo
thing
like
like
alex,
and
I
can
drop
a
design
document
and
everything.
But
it
could
be
totally
not
what
you
guys
want,
but
we
wouldn't
even
know
that,
because
there's
no
one
for
us
to
talk
to
on
your
side.
So
does
that
make
sense.
Like
do
you
see
the
kind
of
the
issue
here.
A
B
Yeah,
I
think
that's
something
that
that
I
brought
up
and
they
said
yeah
like
they
it's
true.
They
haven't
had
a
chance
to
to
get
involved
on
the
python
side.
They've
been
they've,
been
on
the
java
and
the
net
on
the
javascript
side,
but,
like
I
think,
I'm
the
only
one
working
on
the
python
side
right
now
and
they
don't
have
like,
like
the
rest
of
the
people,
aren't
aren't
able
to
join
right
now.
B
They,
I
think,
there's
plans
for
them
to
join
in
like
a
month
or
so,
but
like
it,
it's
not
fair
to
you
guys
either,
because
you
guys
have
a
deadline
for
november
anyways
to
ga
right
so
right,
I
think
their
their
plan
isn't
to
come
in
and,
like
start
saying,
oh
like
we
have
to
do
it
this
way.
Now
we
have
to
change
things.
It's
it's
kind
of
see
like
how
how
can
like
our
lines,
arm
goals
align
and
if
they
do
like
we'll,
contribute
and
we'll
do
it
like
stuff,
that's
easy!
B
A
Okay,
yeah,
I
think
for
your
specific
case,
like
I
could
talk
to
you
offline
and
at
least
get
you
started
and
like
in
the
right
direction.
But
I
think
in
general,
like
like
to
to
emails
too,
you
know
like
this
is
kind
of
the.
A
E
E
E
The
approvers
get
overwhelmed
by
you,
know
new
contribution,
and
it
just
makes
it
really
difficult
for
for
both
people
contributing
because
it
feels
like
they're
not
getting
any
of
the
approver's
time,
but
then
the
approvers
also
don't
have
any
like
bandwidth
to
to
provide
that
feedback.
So
I
think
you
know
this
is
something
that
layton
brought
up
at
the
maintainers
meeting,
and
I
think
I
think
the
the
end
goal
is
to
really
have
more
involvement
from
the
different
organizations
in
these
sigs.
So
yeah
anyway.
Just
wanted
to
call
out
that
this.
E
B
Totally
and
on
my
side,
I
can
I'll
ping
my
manager
again
in
my
mentors
and
ask
them
to
like
you
know,
review
the
the
youtube
video
that
this
gets
uploaded
to
him,
just
see
how
how
maybe
they
can
afford
to
bring
some
people
more
over
here
and
help
out.
I
think
they'd
be
willing
to
do
that,
like
especially
in
the
next
coming
of
months,
because,
like
they
want
to
see
this
grow
into
like
being
something
awesome
too.
A
Sure
that
sounds
good
sure,
yeah,
hey.
G
Yeah
I
just
wanted
to
clarify,
like
I'm,
not
on
the
same
team
as
nathaniel
on
my
seat
side.
I
guess
my
team
doesn't
really
have
any
specific
goals
for
open,
telemetry
python,
I'm
more
of
here
just
on
my
own
volition.
So
I
guess,
like
I,
don't
really
have
any
like
company
goals
to
bring
to
to
the
open,
telemetry
python
repo
from
my
side.
So
I'm
more
just
like
a
contributor.
I
guess
I
see.
A
Okay
makes
sense
cool
all
right
thanks
if
any.
If
I
have
any
more
questions
I'll
I'll
talk
to
you
about
offline,
but
let's,
let's
just
keep
going
so.
Does
anybody
have
any
more
other
topics
that
they
want
to
talk
about?
You
know
not
specifically
related
to
pr's
or
issues
in
general.
F
Yeah,
I
have
one
one
thing,
so
I
was
thinking
if
we
should
speed
up
releases
for
now
until
ga
we
have
a
few
distributions
out
there
and
some
companies
might
be
using
a
hotel
right
now.
So
if
we
speed
up
like
release
every
week,
it
might
help
detect
bugs
early
enough
and
not
have
them
creep
up
into
energy.
So,
for
example,
since
since
last
release
we
landed
a
ton
of
prs
and
none
of
the
distributions
have
those
features.
F
So
if
you
wait
one
month,
that'll
be
only
after
one
month,
we'll
get
feedback
from
from
people
who
are
using
it
so
until
ga,
maybe
it
might
be
helpful
to
release
every
week
or
so,
and
I
can
help
with
that.
E
Yeah,
I
mean
I
think
we
had
set
up
for
like
for
a
while.
We
were.
I
was
trying
to
do
every
every
two
weeks
and
I
think
currently
with
the
the
process
that
we
have
for
releasing.
That
was
that
was
probably
as
fast
as
we
could.
We
could
do
it.
I
don't
know
if
yeah,
if
you
have
any
thoughts
like
you
know,
what's
involved
now,
I
guess
it
takes
like
a
a
couple
hours
or
whatever.
A
Yeah
so
like
the
the
bi-weekly
thing
was
was
okay,
you
know
like
it's
in
terms
of
like
time
wise
and
like
we
kind
of
know
what
to
do
now.
However,
away
does
bring
up
a
good
point
about,
like
you
know,
being
able
to
catch
like
issues
and
bugs,
and
we
can.
We
can
try
it
and
then
I
guess
provide
feedback
on
it,
but
again
away
like
right
now,
the
only
people
who
could
do
the
releases
are
me
and
alex.
A
So
it's
just
it's
just
really
like
how
much
time
we
have
like.
We
know
that
we
know
the
positives
and
benefits
of
this.
It's
just
like
we
physically
don't
have
the
time
to
do
it.
So
yeah.
A
Yeah,
really,
the
only
thing
that
we
need
from
other
people
is
like
just
as
soon
as
we
have
release
candidates
out
or,
like
you
know,
merging
master
in
like
we
just
need
approvers
for
those.
So
that's
like
the
best
thing
you
can
do
to
streamline
that
help.
Us
streamline
cool.
H
Yeah,
I
was
just
gonna
agree,
and
I
was
gonna
also
say
so
like
since
google
has
the
our
exporters
and
stuff
in
a
separate
repo,
we're
also
kind
of
like
blocked
on
releases
if
we
implement
features,
but
I
was
wondering
kind
of
like
what's
what's
like
the
manual
work
part
of
it
like.
Is
it
possible
to
automate
it.
E
Yeah,
so
I
I
did
try
to
automate
it
a
long
time
ago
when
we
were
using
the
old
cla
to
deal
with
cla
check
and
it
wasn't
possible
to
fully
automate
it
because
I
was
using
github
actions
to
indicate
have
actions
bought
to
create
prs
that
I
wanted
to
to
get
merged
and
because
the
bot
wasn't
like
cla
approved,
we
couldn't
do
anything
with
it.
I
haven't
tried
it
since
we
upgraded
to
this
easy
cla,
which
I
think
allows
you
to
register
bots
more
easily,
but
yeah.
E
H
Yeah
that
makes
sense
is
isn't
there
kind
of
like
some
issue
with
some
of
the
scripts
too
or
like?
Sometimes
there
can
be
corner
cases
or
or
is
that
all
handled
too?
Because
I
mean
that's
something
we
could
probably
help
out
on.
This
is
just
code
right.
E
Yeah,
I
think
we're
yeah.
I
I'd
have
to
review
it.
There's
an
open
issue
that
I
think
I
have
assigned
to
me
around
around
figuring
out
the
automation
for
for
releases
that
I
can
probably
pick
up,
hopefully
in
the
next
few
days
here,
to
figure
out
if
we
can,
if
we
can
address
it
now
that
we
have
the
different
cla,
bye,
cool.
A
Yeah
some
of
the
issues
with
the
scripts
like
they
were
only
caught
in
like
after
the
fact
like
in
hindsight,
you
know
so
honestly,
it's
it's
kind
of
difficult
to
see
whether
or
not
they
will
catch.
Everything
for
the
for
the
bot
thing
is
that,
like
is
that
something
that
someone
else
can
like
investigate
alex
or
is
that,
like
you
need
kind
of
permissions
or
something
to
try
that
out.
E
A
Yeah,
I'm
just
thinking
it's
like
it
just
seems
like
it's
just
more
work
for
you.
You
know
it's
not
really
helping
in
terms
of
like
you
know,
like
delegating
work
to
other
people.
E
E
A
Yeah,
okay,
yeah!
I
can
do
it
sometimes
this
week,
it's
a
friday
tomorrow,
don't
really
want
to
release
on
fridays,
even
though
we're
end,
I
think
I'll.
Do
it
monday,
probably.
A
E
A
Okay,
sorry.
H
Sorry,
I
just
want
to
say
one
other
thing:
real
quick.
I
think
I
remember
it
was
oa,
but
somebody
or
somebody
else
brought
up
like
the
release.
Numbers
are
kind
of
weird
like
it's
like
zero
dot,
whatever
be
zero
for
all
them,
with
no
space,
there's
not
really
like
semantic
versioning.
I
know
it's
not
directly
related
to
this,
but
is
that
something
that
we
should
also
change
for
this
release?
Or
was
there
a
reason
for
that?
A
No
sorry,
I
wasn't
aware
of
the
the
release
numbering
issue.
Can
you
explain
it
a
bit
more.
H
A
Mean
you
mean
like
after
the
13
yeah
yeah,
I
think
it
was.
I
think
it
was
originally
like
that,
like.
A
A
Yeah
like
if
you
could
just
bring
it
up
in
like
an
issue
or
a
pr
like
we
have
a
discussion
there,
yeah
nice
cool,
all
right,
cool
nice,
nice.
So
I
guess
we
kind
of
went
through
the
contra
repo
updates
nathaniel
like
is
there
anything
else
you
want
to
talk
about
specifically
for
that.
E
I
didn't
I
didn't
want
to
talk
about
the
the
doc
that
I
think
you
and
I
laden
kind
of
started
working
on
a
little
while
ago,
just
to
kind
of
cover,
maybe
some
of
the
general
goals
and
benefits
of
what
we're
trying
to
do.
Maybe
that'll
help
nathaniel,
if
you
want
to.
If
you
wanted
to
get
started,
I'd
link
the
doc
inside
the
sig
notes
as
well
yeah.
This
looks.
E
Yeah,
so
I
think
I
think
the
main
the
main
concern
around
the
moving
things
into
contrib
is
really
the
ci
pipeline,
because
right
now
we're
really
benefiting
from
having
ci
run
against,
like
basically
every
pr
that
we
have,
and
so
you
know
I
think
today
we
have.
We
have
this
w3c
trace
context
check
that
runs
on
every
pr,
and
I
I
was
kind
of
imagining.
E
The
same
thing
would
be
true
of
the
like
the
contrib,
the
contrib
repo,
so
that
might
be
kind
of
a
nice
to
have
so
to
ensure
that
we're
not
breaking
functionality
between
prs,
because
it
definitely
caught
a
lot
of
issues
for
us
and
like
even
this
week
when
I
was
making
a
change,
I
was
breaking
some
instrumentations
that
if,
if
I
didn't
have
that
check,
I
would
have
like
potentially
committed
code
that
broke
the
other
repo
and
then
the
other
people
would
have
to
catch
up
and
all
that
kind
of
stuff.
E
So
I
I've
seen
this
be
a
pain
in
some
of
the
other
sigs
as
well.
I
know
the
the
go
repo
had
some
some
challenges
there
as
well.
So
if
we
could
come
up,
if
you
could
come
up
with
some
kind
of
plan
on
how
to
address
that,
that
would
be
really.
E
Yeah,
so
it's
all
driven
through
it's
all,
driven
through
github
action,
so
you
can
see
the
workflows
inside
the
the
repo
inside
the
github
directory.
It's
all
in
in
here.
A
Yeah,
there's
no
docks
or
anything
but
yeah
everything
is
encapsulated
in
here
and
you
can
take.
A
E
Yeah
for
sure
yeah,
I
spent
some
amount
of
time
on
the
ci
stuff.
I
know
danielle,
who
isn't
here
right
now,
also
spent
a
bunch
of
time
on
this,
so
you
can
probably
ping
us
and
we
can
help
you
out.
A
Nice
yeah
thanks
alex
for
putting
that
together,
okay
cool.
If
there's
any
other,
no,
not
any
other
topics,
we
can
jump
right
into
the
prs.
A
What
we're
going
to
be
trying
to
do
is,
like
you
know,
for
the
ones
that,
like
are
active
and
like
we
need
for
ga.
We
try
to
like
at
least
walk
out
of
here
with
an
assignee
or
someone
that
can
like
take
some
actions,
action
items
so
yeah.
These
are
the
ones
that,
like
don't,
have
any
assignee
yet.
E
Yeah,
I
guess
some
of
these
are
still
drafts
right,
like
two
of
them
are
drafts
and
another
one
is
a
work
in
progress.
So
there's
really
only
three
here
that
we
need
someone
to
kind
of
take
on.
A
A
Also
like
on
a
side
note,
how
do
people
usually
treat
like
drafts
or
work
in
progresses
like?
Why
do
people
put
those
up?
Because,
usually
I
don't
look
at
them
until
they're,
actually
open.
B
A
Okay,
cool
cool,
so
it's
expected
for
us
to
not
to
review
it.
Yet
it's
working
progress:
okay,
cool
cool,
nice,
okay,
cool!
Let's
take
a
look
at
this
pr
convert
text
map
propagator,
get
it
okay,
keys,
method,
okay,
so
look
like
away-
and
I
have
already
made
comments
on
this.
A
I
think
this
issue
was
opened
by
alex
for
one
of
the
ga
tasks
this
I
I
don't
know
if
he
is
in
the
call,
but
like
just
an
example
of
this
like
if,
if
he
wanted
to
get
this
pr
in
like
quickly
like
he
should
join
the
trying
to
stick
meetings
so
anyways.
It
looks
like
away-
and
I
have
already
been
commenting
on
this-
so
yeah
thanks
away.
A
So
I
guess,
like
you-
can
just
assign
me
to
like
actively
look
at
this,
because
that's
what
I
was
going
to
do,
anyways
so.
A
F
F
So
we
keep
repeating
here
that
people
should
join
the
sig
to
sort
of
drive
the
apr's,
but
new
contributors
don't
know
that
until
they
join
the
second
hear
the
message,
so
maybe
we
should
put
a
message
on
the
pr
in
the
readme
or
somewhere
in
the
contributing
guidelines.
If
it
isn't
there
already.
I
don't
know
if
it
is.
A
Yeah,
that's
a
good
idea,
yeah
most
of
the
time
like
we
have
people
coming
in
and
out,
and
I
repeat
it
every
single
time
so
like
I,
I
was
hoping
that
I
could
like
kind
of
just
drive
by
and
like
catch
people
as
they
come,
but
yeah
some
people
don't
ever
join
so.
A
Yeah,
okay
makes
sense.
Okay,
this.
A
E
This
one's
already
got
one
approver,
it
looks
like
we
just
need
someone
else
to
do
a
second
review
and
an
approval.
If
someone
wants
to
pick
this,
one
up
looks
like
it's
pretty
pretty.
F
H
All
right
cool:
do
we
ever
do
you
ever
discuss
having
like
something
automatically
assigned
just
a
random
one
of
us.
E
You
should
just
clean
up
the
approvers
list,
yeah
yeah
I'll,
take
that
as
a
takeaway.
A
H
E
Erin,
do
you
any
chance?
You
want
to
take
a
look
at
how
to
automate
the
the
signing
process.
Awesome.
A
A
E
Cool,
oh
man,
all
right,
so
this
is
the
last
unassigned.
It
looks
like
aaron,
bladen
and
myself
all
have
requests
for
changes
in
here.
Yes,.
A
A
I
think,
like
there's
just
an
ongoing
conversation
with
with
my
issues
with
it
just
yeah.
I'm
not
sure
if
he's
addressed
your
comments
yet
so.
E
A
I'm
like
actively
waiting
and
I'll
take
a
look
at
his
most
recent
comments,
so.
E
A
Okay,
I
guess
aaron
still
has
a
request
for
changes
like
we'll
still
be
blocked
on
that
all
right.
Thanks,
aaron.
E
F
D
I
was
talking
about
the
your
return
value
of
exit,
which,
if
you
click
on
that
link,
that
I.
C
D
E
A
E
E
E
Yeah,
this
is
nathaniel's.
B
B
B
For
example,
I
can't
create
opentelemetry.sck.extensions
because
it
conflicts
with
the
other
one
being
a
regular
package,
but
if
we
were
to
convert
the
init.py
file
to
be
an
init.pyi
file,
then
that
allows
mypi
to
understand
that
this
is
a
namespace
package
and
allows
any
other
developer
to
create
a
package
and
even
publish
it
on
pip.
To
say
I
want
to
have
something
that
lives
underneath
the
opentelemetry.sck,
dot,
namespace
and
so
like
I
can
mention
like
erin
always
were
commenting
and
they
voiced
some
valid
concerns.
B
I
think
the
most
recent
one
was
always
saying
like
if
concerned
that,
if
we
ever
want
to
expose
something
at
the
top
level
of
open
telemetry.sdk,
which
would
mean
to
like
get
something-
that's
nested
in
open
telemetry.sdk
like
something
that's
in
the
trace
or
metrics
folder
and
just
expose
it
at
the
top.
That's
something
that
would
have
to
be
done
on
the
init.py
file.
But
I
haven't
had
a
chance
to
respond.
B
But
I
guess
I
would
say
that,
like
that's
a
valid
concern,
but
on
the
api
side,
we've
already
committed
to
not
doing
that,
because
we
removed
the
init.py
file
on
aaron's,
pull
request
that
I
mentioned,
and
I
don't
think
we
would
want
to
do
that.
Anyways,
because
I
think
that
the
whole
point
of
the
package
is
that
it
provides
these
sub
things
like
trace
and
metrics
in
util,
where
you
would
get
these
things
and
then
now
by
making
it
a
namespace
package.
A
Nice,
that's
good
explanation,
yeah.
I
was
wondering
why
why
why
did
we
include
the
init
dot
pi
in
the
sdk
package?
Again
I
remember
there
was
an
issue
surrounding
this
or
something.
B
I
can't
speak
to
any
issue,
but
it
looked
from
what
I
saw
that
the
file
wasn't
doing
anything.
So
it
was
just
exposing
the
folders
that
were
right
on
the
next
level
of
the
sdk
package.
But.
B
Yeah,
but
that
that
shouldn't
make
a
difference,
so
you
can
already
do
that
right,
like
they
basically
said.
Oh
now,
you
can
import
the
trace
folder
from
open
telemetry.sdk,
but
since
it
was
already
on
the
next
level,
you
could
already
do
that
anyways.
You
could
already
do
opentelogen.sdk.trace
folder,
so
like
I
deleted
it,
and
I've
got
all
the
checks
to
pass
and
like
it
functioned
exactly
the
way
I
wanted
it
to.
I
think
the
only
thing
left
on
this
pr
is
to
address
os's
concern
like
will.
A
H
Yeah,
I
just
wanted
to
add
that
I
think
there
is
like
a
breaking
change
there
with
removing
it,
maybe
in
our
repo,
it's
okay
like
we're
not
using
import,
open,
telemetry,
dot,
sdk
and
then
calling
like
opensource.sdk.whatever
whatever,
like
there's
a
difference
between
whether
or
not
you
import
it
as
just
like
a
top
level
thing,
and
then
the
package
has,
or
rather
yeah.
The
package
has,
like
those
sub
directories,
the
sub
packages
as
fields
on
it
or
whether
you
do
like
the
python
full
import
statement.
H
So
I
I
mean
I
don't
know
if
anybody
else
is
depending
on
it
and
and
not
like
we're
not
like
we're
shy
about
making
breaking
changes
at
this
point.
So
I
don't
know.
A
That
is
a
good
point,
but
yeah,
I
think,
if
we're
making
like
smart
design
dishes
decisions
like
now
would
be
the
time
but
yeah.
That
is
something
good
to
call
up
on.
So
I
guess
in
like
in
case
of
like
other
approvers,
I
guess
aaron
away
like.
Would
you
be
okay
with
just
seeing
this
finished
as
well?.
H
I
have
kind
of
mixed
feelings
because
I
think
two
things
one
is
you
could
like.
If
you
have
an
amazon
extension,
you
could
just
do
like
your
own
package
right,
like
there's,
no
real
hard
requirement
to
have
that
it
just
changes
like
the
like
the
things
you
type
right
unless
there's
some
behavior
that
that
actually
changes.
H
When
you
do
this,
in
which
case
I
could
I
could
be
wrong,
but
I
that
said,
I
know
we're
doing
this
for,
like
the
exporters
and
the
instrumentations
like
there's
like
an
open
telemetry.exporter
with
telemetry
instrumentation
package,
so
we
are
already
doing
it
for
those
but
yeah
like
like
you
notice,
it's
really
easy
to
break
it
so
like,
for
instance,
with
google
we
have
for
python
packages.
B
B
I
see,
I
guess
I
guess
yeah
like
that's.
That's
the
kind
of
the
point
that
I
wanted
to
emphasize
the
most
was
that,
like
we
already
do
this
with
instrumentation
and
exporters,
so
it's
just
the
way
I
was
seeing
it
right
like
if
we
do
add
vendor
specific
stuff,
and
it
goes
into
the
contrib
repo
that
like
it,
would
be
nice
to
have
them
all
and
they're
like.
Oh,
if
I
want
an
extension,
I
follow
the
same
pattern
that
I
have
already
for
instrumentation
and
exporters,
but
yeah,
that's.
A
Hey
so
aaron,
can
we
put
you
as
an
signee
as
well
like
yeah.
H
E
Oh
man,
you're
assigned
to
this
lady
yeah,
my
bad
actually
I'll,
add
myself
to
this
one
too.
E
This
one,
diego,
do
you
want
to
talk
about
this?
One.
D
Yeah
so
right
now
we
are
not
adding
the
timestamps
to
the
objects.
The
otop
metrics
folder
exports.
So
this
is
adding
them.
It
also
includes
a
question
refactoring
for
the
aggregators.
I
think
later
already
commented
this
pr.
I
replied.
There
is
only
one
comment
I
would
like
to
discuss
here.
Maybe
didn't
you
mention
that
something
should
be
max
instead
of
min.
D
D
Yeah,
so
when
we
merged
these
two
objects,
I
was
under
an
impression
that
we
wanted
to
use
the
minimum
checkpoint
stamp
timestamp
so
that
the
merged
timestamp
will
be
us.
A
Well,
it
depends
because
this
last
checkpoint
timestamp
is
a
new
field
right.
It
depends
on
what
you
want
that
to
represent
on
your
side.
I
thought
it
was
literally
just
like.
When
was
the
last
time
that
it
got
updated
and
like
the
convention
is
like
the
later
updated.
One
would
be
the
real
real
one
unless,
if
I'm
mistaken,
about
the
behavior
of
what
you
want,
the
last
checkpoint
timestamp
to
represent.
D
A
A
I
think
we
we
had
this
merge
thing
it's
just
for
like
concurrency
stuff
like
if,
if
the
two
things
get
checkpointed
at
the
same
time
and
we
merge
them-
oh
sorry,
not
two
things,
two
things
get
updated
at
the
same
time
or
call
checkpoint
at
the
same
time.
It's
like
we
merge
them
and
then
like
we
want
the
later
one
to
be
at
least
for
the
update.
We
want
the
later
one
to
be
like
the
real
one
of
what
that
is.
A
Like
just
just
it
looks
just
kind
of
confusing
like
when
I
look
at
what
it
means
like.
I
know
what
argument
you're
trying
to
say
like
that.
I
know
what
you're
trying
to
do,
but
it
seems
like
it's
like
implicit,
like
I
wouldn't
really
know
by
just
looking
at
the
word
last
checkpoint
timestamp.
You
know.
D
Yeah,
well,
I
guess
we
can
discuss
this
offline.
If
somebody
else
has
something
to
say,
please.
A
Yeah
but
anyways
everything
else
looks
good.
I
was
just
kind
of
confused
on
this
part.
A
All
right
thanks
for
the
review
yeah
does
anybody
else
want
to
take
a
stab
at
this.
H
A
E
Unless
we
yeah
we've
already
talked
about
this
one
sweet.
Does
anybody
else
have
any
other
pr's
that
they
want
to
talk
about
that
we
haven't
talked
about
already.
F
Yeah,
I
have
a
long-standing
one,
the
one
two
that
allows
configure
exporters
automatically.
So
I
simplify
that
further
this
week
and
removed
extensions
is
added
to
the
bootstrap
command.
So
now
it
only
strictly
sets
up
an
exporter
and
tracing
pipeline,
not
even
metrics
right
now.
F
E
That,
okay,
thanks,
I
guess
we'll
need
a
second
person
to
take
you
guys.
You
sure
laden
feels
like
you've,
you've
already
taken
quite
a
few
of
them.
E
E
Was
actually
I
was
hoping
to
to
see
if
diego
would
be
interested
in
also
reviewing
us
since
he's
all
about
the
automatic
instrumentation
yeah.
H
E
All
right
cool
any
other,
any
other
pr's
want
to
talk
about.
I
feel
like
this
one
mario
brings
up
every
so
often
and
we
just
it's
just
been
sitting
here.
It's
not
late.
A
A
E
E
It's
in
the
instrumentary
category.
E
Yeah,
okay,
cool!
Do
we
want
to
kind
of
take
a
stab
at
getting
some
of
the
issues
getting
through
some
of
the
issues?
Maybe
getting
some
of
those
assigned.
E
So
for
those
who
who
haven't
looked
at
this
little
project
tab
here,
this
is
kind
of
the
board
view
of
all
of
the
issues
that
are
currently
marked
as
required
for
ga,
and
you
can
kind
of
see
at
a
glance
if
things
are
assigned
or
not
assigned
or
whatever.
E
D
A
I
A
Yeah
we
just
came
across
this.
As
you
know,
we
were
kind
of
briefly
scanning
the
the
tracing
specs,
as
we
said,
like
there's
cost,
the
specs
are
constantly
changing.
So
definitely
like,
there's
a
lot
of
stuff
that
we
missed
and
we'll
probably
have
to
do
another
pass-through
again.
Once
we
closed
a
bunch
of
the
pr's
that
we
have
open
and
some
of
these
issues,
but
this
is
what
just
one
of
them
that
we
kind
of
just
stood
out.
A
So
it
says
that,
like
the
the
propagators
should
exist
in
a
separate
package,
it's
it's
pretty
much
just
moving
some
of
the
stuff
out
from
the
propagators
folder
in
the
api
and
in
the
sdk,
I
believe,
into
their
own
separate
folders
and
have
them
as
like,
installables
yeah.
That's
pretty
much!
It.
A
A
E
A
Yeah
not
much
to
talk
about
for
these
ones.
It's
literally
just
like
our
our
implementations,
don't
comply
to
the
specs,
so
yeah,
so
if
any
of
them
like
catch
your
eye
or
something
or
like
you
have
questions
about
it,
just
yeah
yeah.
This
one
was
like
pretty
recent
too
we're
not
using
grpc
status
codes
anymore.
A
They
just
map
to
like
success,
error
and
some
some
other
thing.
So
all
of
them
there
are
like
linked
with
specs,
pretty
straightforward
too
so.
A
Oh
just
the
thing
that
call
out-
I
guess
something
I
I
recognize
is
like
when
some
people
assign
them
or
like
take
interest
in
an
issue
or
something
it's
very
frequent
in
which
they
have
a
lot
of
questions
afterwards
and
yeah.
So
I
guess
like
that,
is
you
know
going
back
to
the
you
know?
The
sig
is
a
perfect
place
to.
I
guess:
ask
these
questions,
so
I
guess
back
to
what
owen
was
saying.
A
We
really
need
to
have
somewhere
to
mention
that
you
know
that
this
place
is
the
best
place,
and
speaking
of
that
I
feel
like
modifying
the
con
contributing.md
or
like
some
readme
is,
is
not
explicit
enough
because
me
personally,
like
I
don't
ever
look
at
that
that
file,
I
think,
what's
more
explicit,
is,
if
like
it
would,
it
would
like
be
tacked
on
to
issues
that
we
would
create
or
something
like
that
like
like
in
the
template.
E
E
A
E
F
F
Yeah,
I
think
yeah
yeah.
I
think
I
remember
it
yeah,
okay,
so
so
so
here
I
think
we
take
a
dependency
indirect
dependency
on
on
trace
provider
from
tracer
as
the
source
attribute.
F
So
that's
mostly
the
issue
that
the
tracer
is
accessing,
trace
provider
instance,
but
accessing
undocumented
attributes
that
are
not
part
of
the
api.
So
the
solution
I
was
proposing
was
to
maybe
so
it
accesses,
I
think,
sampler
and
some
other
things.
So
if
tracer
really
needs
those,
it
could
take
those
as
explicit
dependencies
and
accept
the
api
spec
for
those
things,
and
then
any
third-party
implementation
could
just
pass
custom
implementations
for
those
things,
and
it
will
work
is
that
did
I
did
it
make
any
sense.
G
E
I
I
think
it
makes
sense.
I'm
I'm
curious,
though,
have
you
looked
at
the
spec
to
determine
whether
or
not
this
was
a
like
a
thing
that
we
oh-
and
I
just
realized,
we're
over
time
now.
Maybe
we'll
continue
this
discussion
and
take
it
asynchronously
so
that
we
don't
block
this
zoom
meeting
for
the
folks
that
are
just
joining
yeah
all
right
thanks.
Everyone,
okay,.
E
C
C
J
K
I'm
heading
up
community
over
at
traceable,
so
we
just
are
in
the
middle
of
getting
hypertrace
out
the
door.
J
J
Yeah
and
I
I'm
chris-
I
haven't
actually
joined
these
office
hours
before
about
one.
I
wanted
to
kind
of
see
how
how
they
generally
went
and
then
I
wanted
to
also
check
on
some
of
the
things
that
have
been
kind
of
informally
communicated
around
scope
of
ga
timing,
announcements,
and
I
realized,
I
think,
commute
the
communications.
J
C
That
so.
J
C
Well,
yeah
I
mean
so.
The
format
here
is
totally
unstructured
and
we
just
by
chance
a
bunch
of
folks
in
the
governance
committee
who
are
supposed
to
we're
supposed
to
attend
this
meeting
have,
I
think,
multiple
people
are
sick.
Most
people
have
conflicts
and
one
person's
late.
So
I
think
there's
five
people
who
were
not
expecting
to
be
here,
so
I
don't
think
we
need
to
wait.
That's
a
fairly
impressive.
This
is
being
recorded
and
you
know
there
is
like
an
agenda
talk
that
we
aren't
very
good
about
that.
C
I
can
link
to
if
you
want
to
add
stuff,
but
you
just
want
to
go
with
your
questions.
I
think
that's
all
good.
J
Yeah
sure
yeah,
first
of
all,
it
looked
like
in
getter
that
that
there
was
some
clarification
of
what
was
in
scope
for
the
november
ga
and
that
that
metrics
had
enough
going
on
that
that
we
were
planning
on
now
having
that
as
a
separate
ga,
a
follow-on
ga.
So
I
wanted
to
just
clarify
whether
that
was
that
was
a
decision
and
and.
I
There's
been
no
official
decision
about
splitting
the
ga
like
doing
a
tracing,
ga
and
then
a
metrics.
Ga,
though
I
mean
I
suppose
we
could,
but
but
that
hasn't
been
communicated
or
decided
upon,
but
for
release
candidates
and
that's
when
we
effectively
lock
the
specification
and
and
and
the
implementations
will
go
built
from
that
that
then
yes
you're,
correct.
J
Okay
yeah,
maybe
I
was
misinterpreting
something
that
I
saw
from
joshua
in
gitter,
but
okay
and
so
then
so
release
candidates
staggered,
but
then
both
planned
to
ga
on
november
9th.
J
The
date
I
mean,
I'm
guessing
that's
the
date
of
kubecon-
is
that
correct.
Kubecon
is
november
17th.
I'm
basing
this
on
something
that
andrew
hsu
put
together
in
in
a
meeting
recently.
I
think
it
was.
I
I
think
the
ninth
is
a
date
that
we're
trying
to
hit
and
we're
trying
to
use
to
hold
people
accountable
to
actually
get
things
done
by
but
like
I
wouldn't
go
like
back
to
like
new
relic
or
something
like
and
tell
your
co-workers
it's
going
to
be
on
november
9th
right,
because
this
is
more
of
just,
I
think,
andrew's
put
it
in
it's
just
like
an
aspirational
goal.
We
could
be
earlier.
We
could
be
later
we're
just
using
that
to
drive
development.
J
I
I
think
andrew
and
I
had
chosen
the
ninth
just
because
it
gives
us
a
bit
of
time
before
kubecon,
and
so
it's
an
easy
thing
to
rally.
People
around,
like
you
know,
certainly
there'll
be
vendors
involved,
who
want
to
make
announcements
that
are
tied
to
kubecon,
but
I
think
a
lot
of
it's
just
it's
it's
an
easy.
It's
an
easy,
forcing
function
for
the
contributors
to
to
push
them
towards
and
say
no,
no.
If
this
isn't
going
to
make
the
ninth,
then
we're
gonna
have
to
reduce
the
scope.
J
J
Yeah
and
then
the
other,
if
nobody
else,
I
don't
want
to
dominate
the
the
office
hours,
but
if
nobody
else
has
any
other
questions,
the
other
thing
I
wanted
to
check
on
was
looking
at
the
kubecon
schedule.
There's
the
in
the
in
the
main
stage,
sort
of
presentations
and
breakouts
there's
actually
not
much
on
open.
Telemetry.
Surprisingly,
you
know,
there's
there's
no
like
main
talks
on
it,
so
we
got
community
day
and-
and
obviously
you
know
we'll
all
be
jumping
into
that.
But
are
there
other
things
that
are
coming?
J
I
Yeah,
so
certainly
any
announcements
that
we
make
as
a
community
would
be
generally
not
separate
from
kubecon,
but
like
we
can't
guarantee
ourselves
like
announcement
talks
or
things
at
cubecon.
One
of
the
challenges
I
think
with
kubecon
this
year,
is
that
we're
still
technically
an
incubation
project
bennett.
You
probably
know
the
terminology
better
than
I
do
and
we're
in.
I
C
C
Most
of
the
kubecon
stuff
is
in
incubation
and
some
are
graduated,
but
we're
kind
of
this
weird
project
in
that
from
an
activity
standpoint,
we're
like
up
there
with
kubernetes,
but
we're
still
standing
on
this
project.
So
the
just
the
kind
of
importance
did
you
get
are
different
depending
on
your
maturity
level.
Within
that
matrix
we
are
going
up.
I
mean
we're
basically
blocked
on
just
like
process
in
terms
of
becoming
incubated
projects.
C
F
C
Think
is
probably
the
the
main
forum
for
actual
like
cfp
style
talks
of
my
doorbell
right.
Csp
style
talks
about
open
telemetry.
J
I
C
At
if
you
look
at
the
cncf
guidance,
they
there's
a
long
version
of
this,
but
the
short
version
is
that,
even
if
you
do
everything
right
it
takes,
I
think
it
was
like
four
to
six
months
or
something
to
actually
get
through
the
whole
process,
so
that
clock
had
started.
I
think
maybe,
but
if
even
but
if
it
did
start,
it
was
very
recently.
So
I
would
say,
like
the
actual
incubation
status.
Change
would
probably
happen
in
like
april,
or
something
like
that.
C
I
mean
don't
hold
into
that,
but
that's
kind
of
the
guess
unless
they
get
through
their
backlog
faster.
I
think
the
toc
for
cncf,
which
makes
these
decisions
it's
just
literally
backed
up
and
that's
kind
of
the
yes,
that's
the
bottleneck,
but
for
kubernetes
there's
no
way
that
will
be
an
incubation
status.
For
that
reason
and
and
obviously
the
schedule
is
already
set.
I
believe
that
constance
is
on
the
governance
committee,
but
is
sick.
C
Today
she
is
doing
a
keynote
at
kubecon
she's,
one
of
the
program
committee,
one
of
two
program
committee
chairs
for
kukan
this
year,
and
I
I
wouldn't
be
surprised
if
there
was
some
mention
of
hotel
on
that.
But
I
I
can't
speak
to
that
specifically.
J
Okay,
yeah,
I
mean,
I
guess
I
mean.
Probably
you
all
share
the
same
view.
I
mean
we
got
this.
It's
we've
made
this
huge
effort
as
a
community
over
the
last.
You
know
year
and
a
half
and-
and
this
is
kind
of
the
big
splash-
oh
yeah
yeah
it's
getting
to
it's.
Finally,
a
ga.
It
just
seems
like
a
shame
to
not
use
that
forum
to
really
amplify
and
celebrate
that
that
message
and
get
the
word
out.
C
I
I
think
you're
you're
speaking
increasingly
the
choir,
but
there's
there
are
rules
or
whatever
I
mean
the
yeah,
it's
a
little
bit
awesome
for
not
like
I
mean
it
is
actually
genuinely
a
consortium.
E
C
C
Budget
even
much
less
a
plan
for
open
telemetry
per
se
beyond.
C
Of
nice,
in
a
way,
I
think
it
makes
the
it
sort
of
speaks
to
the
bona
fides
of
the
project,
but
yeah,
but
but
chris
I
I
think,
from
a
from
a
coupe
kind
of
location
standpoint.
I
totally
agree.
J
Yeah
and
of
course
you
know,
individual
vendors,
we'll
all
be
we'll
all
be
amplifying
the
message,
and
you
know
in
our
own
way,
and
so
that'll
that'll
create
some
some
amplification
and
momentum,
but
yeah
it's
it's
nice
to
be
able
to
point
to
the
community
announcements
as
well
and
say:
hey
this,
isn't
just
you
know
one
vendor
talking
about
the
stuff:
it's
it's
really
a
new
community
and
that
that's
that's
really.
The
biggest
news
of
this
is
the
way
that
everyone's
gotten
together.
I
So
I'm
putting
something
together
for
the
release
candidates,
I
mean
we'll
get
to
beta
when
it
happens,
but
I
am
putting
something
together
later
like
starting
today
for
the
release
candidates
that
we're
going
to
go
over
at
the
maintainers
meeting
on
monday
for
the
rc
announcement,
at
least
for
tracing
just
again,
spelling
out
our
plan.
That
tracing
is
rc,
the
spec
is
rc.
Rather,
the
sdks
are
building
up
their
rc
capabilities
and
we're
going
to
go
through
the
same
process
for
metrics
stay
tuned
for
ga
and
similar.
C
Thanks
I
mean
chris.
I
do
think
that
there's
I
mean
the
the
question
behind
the
question
for
kubecon
is
really
like.
What
are
we
doing
to
amplify?
The
transition
to
ga
right
is
that
packing.
J
Yeah
yeah
yeah
that
you
know
that
there's
still
a
lot
of
a
lot
of
talk
about
open
tracing
and
open
census,
and
so
you
know
the
more
we
can
amplify
that
you
know
open
telemetry
is
the.
C
I
mean
I
I'm
just
sort
of
shooting
from
the
hip,
but
I
wonder
if
it
would
be
worth
putting
something.
I
guess
it
would
be
in
the
community
repo
or
something
just
have
an
issue
tracking.
The
kind
of
I
guess,
the
lack
of
a
better
word,
the
marketing
plan
for
the
ga,
because
I
I
it's
fine
for
it
to
mostly
be
discussed
or
acted
upon
a
maintainers
meeting
or
something.
C
C
Yeah
yeah,
I
mean
arguably
the
most
important
like
state
change
for
the
project.
I
think
it's
more
much
more
important
than
incubation.
What
it's
worth
like.
C
Going
to
have
in
a
year
in
the
next
year,
so
it
might
be
worth
just
like
if
nothing
else
making
sure
that
everyone
who's
talking
about
it
is
using
the
same
language
and
and
that
we
know
when
it's
happening
and
stuff
like
that,
because
I
think
it
can
have
issues
it's
easier
to
track
than
like,
knowing
which
of
the
many
meetings
someone
has
to
show
up
to,
and
so
I
would
I'd
recommend
that
we
do
that.
If
no
one,
I
think
that's
a
bad
idea
for
somebody.
C
J
Yeah
that
sounds
good
and
then
and
then
I
haven't
attended
the
communications
sig,
but
I
looked
at
the
notes.
It
looks
like
that's
more
on
you
know,
actual
documentation
rather
than
kind
of
marketing
amplification
right.
I
It
probably
will
become
more
focused
on
the
marketing.
I
I
mean
we
were.
I
wasn't
planning
on
using
it
for
the
conversation
for
like
the
rc
announcement,
just
because
hey
that's
kind
of
low-key
and
b,
the
maintainer
sig
has
probably
the
better
audience
for
right
now.
Typically,
the
communication
group
is
like
me
and
austin
and
emilia
it's
it's.
It's
not
doesn't
have
huge
attendance
and
it's
been
historically
more
focused
on
the
website.
I
I
think
we're
trying
to
change
that,
but
I
wasn't
gonna
use
that
as
the
main
event
for
reviewing
the
rc
blog
post,
at
least
just
because
it
would
literally
be
the
three
of
us
looking
at
it.
If
we
did
it
that
way,.
I
J
F
I
J
I
It's
not
so
much
time
and
attention,
it's
just.
I
think
the
channel
just
being
at
least
for
the
rc.
It's
just
up
until
last
week.
We
we
had
ambitions
for
when
that
would
be,
but
not
a
whole
lot
of
confidence
in
when
that
would
be,
and
now
for
the
rc
things
are
looking
great
for
for
doing
an
announcement.
The
next
two
weeks.
I
K
Not
really
we're
just
starting
to
get
involved
really
just
made
a
couple
hires,
who
can
actually
have
some
cycle
to
do
some
things
so
just
trying
to
figure
out
where
to
start
really,
we've
started
playing
around
with
the
the
betas
and
we've
been
using
a
lot
of
the
earlier
versions,
but
now
we're
having
cycles
to
make
some
contributions,
though
just
let's
try.
C
K
Yeah
we've
got
two
products:
we've
got
a
a
sas
product
called
traceable,
which
is
more
of
a
security
product
based
on
understanding,
what's
happening
with
all
your
apis
and
what
is
a
security
product,
and
then
we
open
source
the
the
framework
that
collects
the
data.
So
it's
it's
sort
of
similar
to
like
a
zipkin
or
jaeger.
That
kind
of
thing
so,
and
that
is
still
not
completely
ga,
but
it's
available.
J
D
K
Yeah,
so
I
you
know
if,
if
it
sounds
like,
you
guys
have
a
lot
of
volunteers
and
people
doing
things,
but
if
there
were
something
I
know
that
we're
looking
at
the
the
java
side,
our
platform's
written
in
java,
so
we're
looking
at
the
the
java
libraries
and
probably
will
contribute
there
somewhere
or
help
out,
and
I
know
we're
doing
some
work
around
go
but
again,
I
know:
there's
lots
of
people
working
on
things.
I
Yeah
awesome,
hey
any
more
assistance
with
the
sdks
is
always
very
welcome.
So
thank
you.
J
Yeah
awesome,
yeah
and
my
information
is
a
little
bit
out
of
date,
but,
like
last
time
I
talked
to
john
watson
who's,
one
of
the
maintainers
on
the
java.
Sig
yeah
definitely
help
wanted
there
and
contributions
welcome
and
then
also
auto
instrumentation
for
java
could
use
a
lot
of
contribution.
K
I
A
I
Yeah,
no,
I
apologize.
We
didn't
have
more
members
of
the
committee
here.
Usually
it's
the
the
full
house
I
was
checking
through.
It
looks
like
a
lot
of
people
at
sick
today
are
speaking
at
conferences
right
now,
so
it's
a
smaller
crowd
than
usual,
but
still
a
very
welcome
meeting.