►
From YouTube: 2022-02-10 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
I
think
we
could
probably
just
wait
for
a
minute
or
two
more
looks
like
brian
enjoys
joining
the
meeting
a
few
times,
but
yeah
maybe
wait
for
a
few
more
minutes
for
more
people.
If
you're
on
the
call,
though,
please
make
sure
you
add
yourself
to
the
attendees
list
and
if
you
have
any
agenda
items
you
wanted
to
talk
about.
Please
add
them
to
the
agenda.
D
They
are,
finally
they
they
went
back
for
most
of
last
week
and
then
we
had
like
a
cold
snap
and
that
held
everybody
back
and
then
this
week
has
been
has
been
back,
we're
back
into
it
and
it
was
freezing
last
thursday
and
this
thursday,
it's
70.,
oh
wow,
okay,
yeah,
that's
a
big
difference,
big
difference
like
it's
a
gorgeous
day
out
there
right
now.
B
C
B
B
Appreciate
the
time
for
the
people
who
did
make
it
on
time,
I
think
this
is
the
right
one.
Yes,
okay,
so
jumping
in,
like
I
said,
if
you
haven't
already,
please
add
your
names
to
the
attendees
list
and
agenda
items
that
you
want
to
talk
about.
Please
add
them.
There
first
item
I
wanted
to
talk
about.
Was
this
release?
We
were
planning
on
doing
it
earlier
this
week
created
a
milestone
to
track
some
issues
I
wanted
to
include.
This
is
the
only
one
that's
outstanding.
B
I
wanted
to
get
some
opinions
on
this,
I'm
thinking
of
just
removing
it
and
losing
without
it,
because
there
doesn't
seem
to
be
a
pretty
quick
response
and
turn
around
on
this,
but
it
looks
like
there's
definitely
some
things
that
still
need
to
be
addressed.
B
C
Okay,
yeah:
let's:
let's
do
that
milestone,
that's
not
how
you
do
it.
B
Cool
I've
got
a
few
other
things
I'm
working
on,
but
hopefully
I'll
try
to
get
a
release.
Pr
started
this
afternoon
and
hopefully
get
it
merged
in
and
get
a
release
for.
The
main
repo
at
least
today
or
early
tomorrow
is
the
gold
cool.
Next
thing
I
wanted
to
double
check
on
was
the
metrics
api
stability
metrics
pr
right
here.
B
I
have
I'm
still
reading
through
it.
I've
taken
a
look
at
it
a
few
times.
I
haven't
officially
made
any
response,
but
I
I'm
wondering
aaron
if
you
had
any
status
updates
from
some
of
the
other
comments
that
have
come
in.
D
The
only
one
that
I
think
might
have
actionable
is
kind
of
the
last
one
that
j
mcd
posted.
Basically,
this
little
conversation
on
whether
or
not
we
need
to
have
record
or
what
is
it
register.
Callback
right
by
default
have
a.
D
By
default
have
a
token,
and
I'm
actually
going
to
argue
that
we
probably
shouldn't
just
because
that's
that
would
be-
I
would
assume
the
default
use
cases
to
not
worry
about
unregistering
and
we
could
provide
a
an
extra
interface
or
an
extra
access
method
that
to
s
to
satisfy
that
requirement.
So
if
you
look
down
there,
the
last
comment
that
you
have
there
that's
kind
of
like
what
I
think
we
could
get
to
once.
That's
finally
accepted.
I
could
add
it
into
this.
D
I
don't
think
it
would
be
too
hard
to
actually
implement
aside
from
removing
callbacks.
I
don't
know
if
there's
an
easy
way
in
our
current
system
to
actually
remove
callbacks
I'd
have
to
dive
into
that.
B
D
Sort
of
we
kind
of
do
we
have,
I
was
looking
at
it.
We
could
kind
of
use
the
map
key
for
the
instrument
like
I.
I
think
there
might
be
a
way
of
doing
that.
Okay.
D
That
being
said,
like
I
don't
know,
I
I'm
not
100
sure
whether
we
want
this
like
a.
If
this
is
the
interface
we
want
to,
we
want
to
accept
there
b
if
we
want
to
put
it
in
on
this
pass
or
if
this
is
something
that
we
can
agree
to
add
on
to
later.
B
Yeah,
I
think
that
makes
sense.
It
makes.
D
Sense
to
me
yeah,
so
if
that's
the
case
then
there's
it's
really
just
me.
Soliciting
feedback
there's
been
some
by
anywhere
some
anthony,
but
and
then
this
comment
from
bogdan,
but
nothing
else
really
substantial.
E
D
E
Okay,
so
I
would
say,
if
we're
going
to
target
that,
we
should
probably
just
have
a
single
interface,
haven't
returned
the
token
and
allow
users
to
ignore
it
if
they
don't
think
that
they
are
going
to
cancel
the
callback.
It
seems
kind
of
weird
to
me
to
have
separate
interfaces
for
that
unless
the
spec
says
they
need
to
be,
there
needs
to
be
separate
interfaces.
D
There's
only
a
fairly
vaguely
defined
use
case
of
like
having
it
actually
countable,
that's
in
the
spec,
but
fine
that
I
think,
is
decent
enough
reason
like
I,
I
feel
like
that's
a
good
enough
reason
right
there
to
have
it
where
the
main
pass-through
is
just
returns,
an
error
if
it
didn't
work
versus
giving
the
people
the
the
pass-through
where
you
want
to
cancel
a
an
error
and
a
token
I
mean
I
could
be
convinced
otherwise,
but.
D
E
E
I
guess
I
I
don't
know
whether
that
should
be
pushed
down
into
the
sdk
like
that
or
if
it
should
just
be
exposed
to
the
user
as
callbacks
are
cancelable.
If
you
don't
care
about
canceling,
this
callback
ignore
the
return
value
here.
D
E
Yeah,
so
I
don't
know,
I
I
think
I'm
fine
with
the
api
as
you've
got
it
proposed,
but
it
it
does
seem
like
having
two
interfaces.
There
may
not
be
ideal.
Maybe
this
is
just
something
we
should
create
an
issue
for
and
come
back
and
revisit
before
we
stabilize.
B
Yeah,
I
think
that's
a
good
idea.
Let's
do
that
and
then
make
sure
tag
it
with
required
for
ga,
or
I
don't
know
we
can
update
that
tag
as
well.
B
Okay
last
thing
on
the
agenda:
doc
is
brian.
You
are
asking
for
reviews
of
the
crosslink
build
tool.
F
Hi,
yes,
since
last
week
it
went
through
a
few
runs
of
reviews
through
anthony.
I
presented
it
in
the
collector
sig
asking
for
reviews
also,
but
I
just
want
to
call
that
out
again
that
we're
looking
for
some
more
viewers,
especially
since
the
only
owners
on
this
go
build
tools,
are
go
maintainers
and
collector
maintainers.
B
Yeah
cool
yeah,
I
think
that's
a
worthy
call
out.
B
Okay,
cool
thanks.
I
think
that's
it
actually
for
our
listed
agenda.
If
anybody
else
on
the
call
had
something
they
wanted
to
talk
about
anthony.
Just
to
recap,
we
were
planning
to
do
the
release
after
the
call
or
get
a
pr
out
and
then
start
this
release
process.
I
took
out
the
last
remaining
issue
that
was
still
in
the
milestone
for
supporting
the
sampler
environment
variables,
but
that's
I
think
the
recap.
Okay.
Did
we
make
a
decision
about
what
we're
going
to
do
about
the
metrics
exporters.
B
I
think
this
is
it
we're
going
to
split
the
experimental,
metrics
and
contribute
to
instrumentation
and
then
the
explorers
and
then
eventually
just
duplicate.
The
exporters,
I
think,
is
the
plan
yeah.
I
think
that
sounds
right.
So
that'll.
B
Yes,
that
sounds
like
a
plan
to
me:
yes,
correct,
cool,
okay,
yeah,
the
more
I
see
people
asking
questions
about
our
current
metrics
api
and
our
current
metrics
sdk.
I
realize
how
we
should
have
been
a
lot
more
brutal
in
just
ripping
things
out
that
weren't
there,
because
we
are
actively
supporting
something
that
is
actively
being
changed.
So
I
think
doing
this
strategy
is
probably
appropriate
at
this
point.
It
was
probably
appropriate
a
while
ago.
B
Okay,
yeah
now
I
will
actually
pause
and
see
if
anybody
else
is
anyway,
I
think
they
wanted
to
bring
up.
D
I
I
would
like
to
restate
that
if
you
have
time
the
api,
I
really
would
appreciate
any
kind
of
review
any
kind
of
feedback.
I
tried
to
make
it
as
simple
as
I
could,
with
this
complicated
pr
start
with
one
and
then
work
your
way
forward.
E
For
the
second
and
third,
just
look
at
the
last
commit
in
the
history.
All
the
rest
is
coming
from
the
first
one.
So
that'll
save
you
some
time.
I
think.
D
You
can
either
look
at
the
last
commit,
or
I
have
a
pr
in
my
repository.
That
is
just
the
last
commit
against
the
other
one.
But
I
can't
do
that
in
the
open
telemetry
product,
because
that's
how
github
works.
E
B
Yeah,
like
I
said,
I've
been
looking
through
it.
Unfortunately,
I've
been
distracted
but
yeah,
that's
a
pretty
high
up
on
my
priority
list
right
there,
and
that
is
not
just
a
conversation
to
me
by
the
way.
Anyone
else
on
the
call.
Please
take
a
look.
E
Yeah
one
thing
I'll
call
that
is
it:
it
is
kind
of
nice
how
the
sdk
api
made
wrapping
the
sdk
with
the
new
api,
pretty
simple
and
straightforward.
It
seemed
so,
even
though
we're
going
to
be
getting
rid
of
this
sdk
at
least
now,
we
have
something
that
can
implement
the
new
api
in
the
interim
and
it
doesn't
look
like
it
was
a
whole
lot
of
code
or
all.
D
That
ugly,
so
to
that
effect,
I
don't
think
that
was
particular
to
the
sdk
api.
We
would
have
done
a
similar.
I
like
I
would
have
proposed
a
similar
thing,
just
wrapping
the
concrete
types
that
implemented
that
api,
but
it
would
have
been
a
little
bit
more
work,
but
not
that
terribly
much.
D
But
I
I
was
very
happy
that
it
could
just
be
wrapped
like
it
works
for
now
it
it
does
terrible.
You
know
I
forgot
to
put
in
a
benchmark,
but
it
does
terrible
on
the
one
thing
that
we
expect
it
to
do,
which
is
batch,
recording
synchronous
instruments.
B
Exactly
okay,
I
think
with
that.
Maybe
we
could
ask
if
there's
any
cool
community
projects
using
the
codebase
or
some
cool
user
customer
stories
or
something
like
that.
E
E
Tweet
the
other
day,
something
about
how
they
they
had
integrated
open
elementary
into
their
go
program,
found
it
surprisingly
easy
compared
to
having
done
the
same
thing
with
java
that
took
them
several
days.
So
hopefully
that
is
good
for
us
and
not
just
bad
for
java.
I'm
not
sure
how
to
take
that,
but
I'm
going
to
take
it
as
a
win
for
us.
B
I
think
it's
good
news
yeah,
I'm
kind
of
surprised
on
that.
One
myself,
if
you
don't
mind,
if
you
still
have
it
link
it.
E
B
More
cool
david-
I
don't
know
if
you're
able
to
speak,
but
I
know
we
talked
a
little
about
kubernetes,
the
kubelet
integrating
with
opencentury
last
week.
Any
progress
on
that
was
that
actually
released.
A
B
Exactly
the
slow
wheel
of
software
development
right
there,
cool
thanks
awesome.
If
there's
nothing
else,
anybody
else
wants
to
share.
We
can
probably
end
it
here.
E
Oh
one,
one
more
thing:
let
me
see
if
I
can
find
the
issue
there
was
an
issue
created.
I
think
today,
in
the
collector
repo
regarding
go
version,
support
there
and
it
references
the
issues
that
we
had
and
our
decision
there
so
I'll
link
this
in.
If
any
anybody
has
anything,
they
feel
like
contributing
to
that
discussion
might
be
good
to
bring
our
experience
to
the
collector.
E
Okay,
but
the
the
proposal
is
fairly
similar
to
ours,
basically
use
our
policy
for
treating
the
collectors,
a
library
and
then
build
with
as
recent
a
version
of
go
as
possible,
preferably
one
that
doesn't
have
security
issues.
B
Yeah,
I'm
all
a
fan
of
that
definitely
okay,
short
and
sweet.
Today,
thanks
everyone
for
joining.
If
you
have
any
more
issues
or
want
to
talk
about
something,
please
post
it
in
the
slack
channel.
Otherwise,
we'll
see
you
asynchronously
via
issues
in
pr's
or
next
week,
bye.