►
From YouTube: 2023-02-08 meeting
Description
cncf-opentelemetry@cncf.io's Personal Meeting Room
A
A
A
B
E
All
right
can
everybody
see
it.
E
Well,
I
guess:
I'm
going
to
start
with
the
first
three
items,
all
together
since
they're
all
related
the
other
maintainers
and
I,
not
sure
if
any
of
the
others
are
on
the
call
right
now,
they're,
actually
not
the
other
maintainers
and
I
were
talking
about
our
our
the
state
of
our
current
approvers
and
maintainers
and
decided
to
make
some
changes.
E
E
E
I
was
waiting
for
Martin
and
had
us
to
comment
on
their
own
PRS
before
merging
them,
but
looks
like
that's
already
done
so.
I'll
merge
those
after
this
meeting
and
we
can
move
on.
E
For
those
that
have
not
yet
seen
it,
the
contrib
release
went
out
yesterday,
which
contained
the
newest
bug.
Fix
releases
of
the
SDK
packages
seem
to
have
gone,
just
fine
haven't
had
any
complaints
yet,
but
we
will
see
the
current
contrib
release
is
actually
being
held.
Anthony
reached
out
to
me
in
Slack,
there
is
potentially
an
issue
with
a
PR
that
was
merged
a
few
hours
ago.
It
has
not
yet
been
released,
but
he's
looking
into
it.
A
E
Released
yesterday,
it
shouldn't
be
too
big
of
a
problem.
E
Moving
on
to
the
exponential
histograms
PR's
I
know
we
talked
about
this
every
week,
but
please
review
this
part
to
PR
I
reviewed
it
yesterday,
I'm,
not
sure
Matt.
If
you've
seen
my
comments
yet
or
made
any
changes,
but
everybody
please
review
these.
A
F
That
this
afternoon,
thanks
for
the
review,
yep.
E
Sounds
good
another
topic
that
we've
been
talking
about
every
week
so
far
the
the
splitting
of
the
logs
and
events
API
is
merged
as
of
I.
Believe
yesterday,
maybe
today
I
don't
remember
exactly
yeah.
E
Probably
not
much
to
say
there
I
assume.
The
next
step
is
to
work
on
the
logs
or
event
SDK.
D
Yeah
I
think
the
next
next
step
is
going
to
be
the
log
SDK.
There
is
an
open
PR
already
from
another
contributor,
so
we'll
I'll
be
helping
them
finish.
That
PR
oud,
okay,.
E
A
E
All
right,
the
first
topic
from
you
know
other
than
the
normal
ones,
is
the
the
sync
resource
PR.
This
allows
both
synchronous
and
asynchronous
resources.
E
A
E
I
figured
I
would
give
one
last
one
last
chance
for
people
to
speak
up
if
they
have
concerns.
E
Seems
like
no,
the
next
item
on
the
agenda:
Mark
deprecating,
the
Yeager
exporter,.
C
Yeah
so
just
say
a
few
words
about
this
one,
so
the
Jaeger
export
has
been
deprecated
in
the
spec
and
I
opened
the
pr
that
just
adds
deprecation
documentation
and
a
date
or
like
hit
General
time
frame
to
when
that
is
expected
to
this
PR,
and
there
is
an
ongoing
discussion.
That's
also
linked
in
the
issue
about
how
we're
going
to
deal
with
deprecation
of
G8
packages
in
the
future.
C
So
yeah
I
put
the
date
now
to
July
2023,
as
that
is
the
same
time
that
it
will
be
deprecated
in
the
spec.
But
in
the
issue
there
there's
already
a
discussion
going
on
to
possibly
prolong
that
for
a
year
after
the
deprecation
notice
has
been
given,
so
that
we
keep
releasing
it
and
then
maybe
archive
it
that
in
some
way
shape
or
form.
E
Okay,
until
the
related
conversation
is
finished,
I
think
we
probably
shouldn't
put
a
date
yeah.
Maybe
if,
if
that's
from
the
spec
is
deprecating,
it
maybe
say
that
the
specification
will
be
deprecated
to
July.
A
A
C
Yeah
the
number
changed
it
up
and
just
refer
to
the
spec,
and
let
everybody
know
that
it
is
deprecated
for
now,
but
not
how
long
we
are
going
to
wait
until
it's
actually
and
until
it's
actually
stopped
releasing
okay
yeah.
Thank
you.
E
The
next
item
is
related.
It's
actually
linked
from
the
Jaeger
deprecation.
This
is
the
discussion
we
were
just
talking
about
and
it
is
essentially
to
come
up
with
a
a
more
formal
deprecation
policy
for
when
we
deprecate
GAA
components.
E
In
the
past,
we've
only
ever
deprecated
experimental
components
which
we've
done
with
you.
A
E
But
that's
not
fair
to
do
on
GA
components.
The
specification
doesn't
have
any
real
guidance
on
this.
It
does
have
guidance
for
when
you
go
to
the
next
major
version,
so
it
the
the
spec
says
when
you
rev
your
major
version.
You
should
still
support
the
previous
version
for
at
least
a
year
and
I
think
it's
probably
fair
to
carry
that
policy
over
to
deprecating
packages
as
well,
but
there
may
be.
E
Maybe
that
should
also
be
added
to
the
spec
or
at
least
get
the
the
approval
of
the
the
spec
maintainers
that
that's
a
a
reasonable
interpretation
of
it
before
we
move
on
there's
not
much
discussion
in
this
PR,
it's
all
or
in
this
issue,
it's
all
relatively
straightforward.
But
if
you
have
thoughts
about
this
or
if
you're
interested
in
in
what
the
policy
will
be,
this
is
a
good
place
to
go
to
look
for
that.
E
Sounds
like
no
again
somewhat
related
last
week,
I
created
a
milestone
for
tracking
changes
that
we
will
want
to
make
breaking
changes
that
we
want
to
make
when
we
go
to
2.0
I,
just
added
a
couple
that
were
on
my
mind
at
the
time,
but
anyone
can
please
feel
free
to
add
issues
to
this
Milestone.
E
Just
the
existence
of
the
issue
in
the
Milestone
does
not
mean
that
we
will
100
make
that
breaking
change.
It's
just
changes
that
we
want
to
consider
when
we
go
to
2.0
because
they're
breaking
so
we
may
consider
them
and
decide
not
to
do
them
or
we
may
consider
to
do
them.
So,
if
you're,
if
you're
unsure
of
something
belongs
in
this
Milestone,
go
ahead
and
add
it,
it
can
always
be
removed.
E
If
we
decide
not
to
do
it
or
or
something
along
those
lines,
not
sure
if
anybody
has
any
immediate
ideas
for
stuff
to
go
in
here,
but
I
just
want
to
make
sure
that
we
don't
lose
track
of
those
things.
E
B
Yeah,
so
this
is
just
a
an
update,
so
while
I
originally
planned
to
have
the
sandbox
all
up
and
running
right
now,
it's
stopped
good
in
the
way
as
usual,
but
we
did
have
a
a
major
step
forward
where
Martin
and
I
went
through
the
script
and
we
merged
that
yesterday
and
it's
now
effectively
emerging
from
the
staging
repo
into
Maine.
So
this
is
the
an
example.
B
This
is
the
pr
that
it
created,
but
you
can
see
it's
got
everyone's
history,
so
this
is
only
emerging
well,
this
is
merging
the
entire
history
of
API
and
contrib
up
until
from
I
think
November
last
year
into
into
Main
in
the
sandbox,
which
is
why
it's
like
2000
odd
files.
B
It
is
moving,
so
the
auto
merge
holders
that
you
see
listed
there
is
effectively
there's
the
JS
is
the
entire
JS
repo
contributes
the
entire
country
repo.
This
PR
then
moves
the
individual
components
that
we're
concerned
about
for
web
and
puts
them
into
pkgs.
So,
in
this
case,
it
creates
pkgs
API
activates,
a
bunch
of
tests
for
both
browser
and
web
workers,
I
plan
to
enable
CI
testing
for
this
by
the
end
of
the
week,
so
that,
as
these
PRS
get
created,
they'll
also
be
tested.
B
There
will
be
some.
There
is
already
some
known
issues
once
I
bring
them
in
some
of
the
other
packages
from
local
testing,
where
there
are
some
tests
that
are
designed
to
work
in
the
browser
that
don't
work
in
a
worker
that
will
fail
so
I'll
have
to
contribute
those
back
into
the
JS
and
control
repos
so
that
they
can
be
merge
back
into
here.
B
For
example,
one
of
them
is
there's
a
a
test
for
I.
Think
a
document
load
that
has
visibility,
change,
visibility
change
is
not
defined
as
far
as
typescript
is
concerned,
when
you
compile
it
for
a
web
worker.
So
therefore
that
test
fails
so
fails
to
compile
just
stuff
like
that.
B
There's
there's
three
minor
changes
that
that
I'll
have
to
get
before
we
get
every
package
in
here,
I
haven't
yet
tested
every
single
package
for
the
web
and
I
haven't
yet
brought
over
the
examples,
but
the
plan
will
be
to
have
those
coming
as
well.
B
You
don't
see
it
well.
If
you
actually
look
at
the
test.
Actually,
there's
no
test
run
if
you
could
run
the
test
locally.
You'd
see
this
also
generates
bundles
for
everything,
for
every
packager
that
creates
in
both
es5
and
the
es6
or
ES.
Next,
so
yeah,
it's
getting
very,
very
close
down.
The
plan
will
be
we'll
have
main
main
will
just
be
this
Auto
merge
system,
and
then
we
can
go
and
create
experimental
branches
for
anyone
to
go
and
play
and
do
what
they
want.
B
One
of
the
ones
I'll
be
running
with
will
be
a
minification
branch.
We
talked
in
the
romsig,
but
we
could
have
used
a
branch
War.
If
we
had
it
set
up
in
time
was,
we
could
have
done
the
logs
and
events
apis
as
a
experimental
branch
that
muck
around,
because
this
is
like
all
of
the
code
and
then
once
we're
happy
with
it.
We
could
then
contribute
it
back
to
JS
and
controvery
points.
So
it's
things
like
that.
B
E
Okay,
as
far
as
this
PR
goes
obviously
reviewing
a
2300
file.
Pr
is
not.
B
An
option
yeah,
so
if
you
look
at
the
commits
anything
tagged
with
the
auto
merge
in
square
brackets
is
really
what
the
script
does.
Everything
else
is
everyone
else's
stuff.
B
The
one
issue
that
I
found,
which
is
why
I
haven't
pushed
this
in
yet
who
is,
is
because
it's
emerging
from
JS
contrib
into
this
one
it.
Actually,
this
merge
deletes
the
readme
I
think
this
is
a
one-off,
so
I'm
I'm
planning
to
I
think
push
this
in
I'll
follow
up
with
a
manual
one
to
push
put
the
readme
back
into
the
the
sandbox
and
then
subsequent
mergers
should
then
not
go
and
delete
the
readme
again
now.
B
If
they
do
I'll,
let
him
fix
it
up
and
figure
out
a
way
to
not
cause
I
get
merged
to
go
and
crash
files
that
shouldn't.
E
Okay,
why
is
the
review
deleted?
I
guess
I,
don't
understand,
but.
B
That's
historical,
so
it
was
part
of
the
repo
staging
branch
when
it
merges
in
the
Json
control
repos
into
that
Branch.
To
avoid
merge
conflicts
I
deleted
the
readme
out
of
the
out
of
that
one.
Okay,.
E
Okay,
all
right
and
is
there,
are
you
looking
for
reviews
here
or
is
this
just
a
FYI?
This
is
happening
type
of.
B
This
is
an
FYI.
This
is
created
by
the
open,
Telemetry
bot.
So
when
I'm
happy
I
I
was
just
gonna
like
approve
it
and
then
push
it
in
and
then
I'll
follow
up
with
activating
the
CI
and
putting
that
readme
back
in
and
then
I'll
start
when
I
get
time.
I've
got
like
back-to-back
meetings
today,
I
think
I
have
a
half
hour
window
of
no
meetings
today,
I'm
doing
well.
B
So,
by
the
end
of
the
week,
I
should
have
the
CI
and
that
running
here,
which
is
why
I've
said
the
new
plan
is
now
the
end
of
Feb
because
of
all
this
other
stuff
that
got
in
the
way,
but
I
hope
to
have
all
the
other
packages.
I've
been
testing
this
locally.
B
If
you're
interested.
If
you
look
in
the
tools,
scripts,
folder,
I'll,
I'll
put
it
in
the
chat
window.
Actually
I'll
put
it
there
in
the
meeting
notes,
you
can
see
the
config
that
is
required
to
you
know,
get
the
appropriate
package
merged.
It's
pretty
straightforward.
B
A
E
B
Okay
yeah
as
part
of
that
merge
script,
it
actually
renames
every
package
that
to
include
sandbox
Dash.
So
any
code
that
says
import
from
open,
Telemetry
API
becomes
open,
Telemetry
sandbox,
API
Etc,
but
it
does
go
and
touch
every
import.
It
renames
the
packages
in
in
every
package
that
Json
updates
all
the
dependent
fees
as
well.
It
also
fix
fixes
the
version
so
that
Russia
effectively
just
creates
some
links
to
everything.
So
there
is
no
backward
compatibility
checks
that
will
be
occurring
in
the
samples.
E
E
Let's
see
this
was
an
old
one
with
information
requested
but
looks
like
it
has
updates.
In
the
last
week.
E
We
will
be
able
to
test
this
in
the
next
two
weeks.
Okay,
so
we're
just
waiting
on
feedback.
There
still
can't
update
to
SDK
node
34
using
yes
build.
E
E
E
So
I'm
not
sure
that
this
is
necessarily
a
bug.
It
looks
like
he's
already
got
a
PR
created
here
so.
C
I
think
this
is
the
the
pr.
Oh
you
just
mentioned
it
that
that
he
yeah
that
he
originally
found
out
about
the
issue.
C
E
C
E
I
got
it
okay,
so
I
guess
this
is
a
bug.
I,
guess
it
results
in
incorrect
instrument.
Incorrect
instrumentation
output.
If
you're
expecting
it
to
be
there
I
guess
it
could
be
either
P2
or
P4.
It
doesn't
affect
end
user
applications.
So
it's
not
one
or
three
I'll
put
P4
for
now
and
I
will
add
a
unwanted
label
up
for
grabs.
E
All
right
user
action,
instrumentation
too
many
traces
per
user
interaction.
C
I
would
have
to
read
into
the
stuff
that
they're
using
there,
but
I
may
be
able
to
tackle
this
one.
They
have
provided
some
information
on
how
to
how
to
reproduce
it,
but
I'm
not
familiar
with
the
instrumentation.
C
Yeah,
it
seems
they
they
are
also
talking
about
that.
They
thought
that
might
be
related
to
strict
mode
and
reactions,
ethical
events,
but
that
doesn't
seem
to
be
the
issue,
so
definitely
more
investigation
necessary.
So
it
definitely
shouldn't
be
happening
this
in
in
this
way,.
E
Okay,
will
you
have
time
to
look
into
this?
Can
I
send
this
to.
C
You
yeah,
you
can
assign
it
to
me.
I
will
look
into
it
tomorrow.
A
E
I,
don't
remember
who
is
exactly
in
charge
of
the
Azure
SDK
instrumentations
these
days.
Mev
is.
A
E
E
It
looks
like
Florida
map
thinks
this
is
a
bug
on
Azure
as
well.
If
it
is
a
bug
on
the
Azure
side,
I
would
close
it
here,
although
maybe
there's
a
reasonable
workaround
that
we
can
make.
E
Browno
assigned
himself
to
this
he's
on
vacation
until
the
end
of
this
week,
so
we
can
talk
about
this
one
again
next
week
when
he's
back
and
then
this
one
also
has
a
PR
open
already,
so
you
should
be
and
she's
still
there's
a
PR
open.
I,
don't
want
the
issue
to
be
stale
and
there's
the
last.
E
E
All
right
well,
thank
you,
everybody
for
your
time
and
I
will
talk
to
you
in
two
weeks.
I'll
be
gone
next
week,
but
Mark
will
be
here.