►
From YouTube: 2021-01-07 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).
C
Oh
yeah,
first
time
joining
yeah,
just
seeing
how
things
go
and
potentially
trying
to
get
involved
and
interested
to
bring
it
to
work
as
well
or
introducing
it
now.
D
A
A
A
lot
it
does,
not
surprisingly
nikita
is
still
out
he'll,
be
back.
I
think,
on
monday.
E
E
E
D
E
So
the
yeah,
the
skinny,
is
basically
that
we
are
currently
creating
spans
every
time
a
a
queue
is
pulled
like
a
jms
queue
is
pulled,
and
even
if
there
are
no
messages,
we
still
are
creating
spans
and
I
think
that's
almost
never
desirable.
E
A
Yeah,
I
will
admit
in
my
exporter,
I
actually
failed
to
suppress
those
so
yeah
cause
they
are,
they
are
noisy,
they
didn't
match
what
we
used
to
capture
in
our
prior.
B
A
A
Oh
okay,
but
it
doesn't,
but
it
doesn't
conditionally
capture.
F
E
A
Yeah,
do
you
know
why
we
why
we
don't
start
it
in
onstart
there.
D
A
Yeah
and
in
the
future,
if
somebody
wants
a
span,
even
in
any
case
that
I
think
that
would
make
sense
as
a
opt-in
kind
of
a
feature.
I
I
think
this
makes
sense
for
the
as
default
at
least
cool
and
no
need
to
create
a
last
setting
at
this
point,
we'll
wait
and
see.
If
anybody
even
wants
that.
A
Sense
there
was
one,
let's
see
I
don't
know
if
anybody
has
thoughts
on.
Oh
we've
got
several
thumbs
up.
I
have
not
seen
comments
from
last
night.
A
Cool
okay,
I
was
I
wanted
just
to
get
some
eyeballs
on
this
issue,
since
it's
reverting
a
decision
that
we
had
made
in
a
prior
issue.
Yeah.
A
So
if
anybody
has
thoughts
or
concerns
about
us
going
back,
I
think
it
makes
sense
and,
and
certainly
prioritizing
supportability
is
always
always
gets
a
thumbs
up.
E
E
C
I'm
super
neutral,
so
I
don't
know
a
full
background
on
that
and
the
one
thing
that
passed
through
my
mind
was
it.
It
mentioned
different
versions
of
instrumentation
libraries.
Did
it
so
would
it
be
to
have.
C
A
Yeah,
so
have
you
seen
the
spec
repo
open,
telemetry
spec
repo,
yet
no
yeah?
Okay,
so
this
is
a
good
place.
So
a
lot
of
what
we
do
is
sort
of
defined
at
the
spec
level.
Although
more
more.
This
group
is
more
focused
on
the
instrumentation,
the
java
agent
and
library
instrumentation
instrumenting
java
libraries
jms
kafka.
That
kind
of
thing,
whereas
there's
also
the
java
repo,
which
is
all
the
languages,
are
implementing
the
spec
for
the
open,
telemetry
api
and
the
reason
I
bring
that
up
is.
D
D
A
So
basically,
I
can't
find
it
it's
somewhere
and
so
that's
kind
of
the
problem
with
the
spec
repo
also
is
it's
very
big
there's
a
lot
of
things
specked,
but
basically
we
have
two
fields.
We
do.
We
have
a
instrumentation
library
name
and
an
instrumentation
library
version
to
specify
who
we
are
we're
like
we're,
instrumenting
jms,
but
our
we're
open,
telemetry
gms
and
our
the
version,
though
we've
decided
to
populate
with
the
version
of
the
agent.
A
So
that
leaves
us
because
we
think
that's
probably
the
more
sort
of
what
it's
like
we're
supposed.
The
name
is
as
the
version
of
our
instrumentation,
but
that
does
leave
us
with
the
tricky
part
where
we
do
have
sometimes
like.
A
Yeah,
so
what
we've
done
in
general
same
with
the
maven.
A
A
So
I
think
that
so
that's
what
we've
decided
to
kind
of
do
or
that's
what
this
issue
was.
We
weren't
putting
the
version
encoding
it
into
the
artifact
name,
because
we
sort
of
initially
thought
people
will
know
which
version
they're
running
of
those.
A
But
it's
clearly
clear
that
that's
not
we're
not
clear,
but
the
the
issue
is
raised
that
from
a
support
like
a
vendor
support
perspective,
it's
really
helpful
to
have
that
info
on
the
back
end
to
see
which
version
of
the
instrumentation
library
was
applied,
because
customers
often
don't
know
without
digging
what
version
of
things
that
they're.
A
A
I
can
give
a
short
kind
of
update
on
one
zero
sort
of
some
discussion
that
we
had
before
the
holidays
about
one
zero.
I
don't
remember,
did
we
discuss
this
last?
I
don't
think
we
did.
A
I
don't
remember,
but
the
so.
The
sdk
is
tracking
pretty
well
towards
one
zero
things
are
getting
pretty
stabilized,
so
potentially
that
will
be.
I
think
the
hope
you
know
not
nothing
firm,
but
maybe
mid-february.
I
don't
know,
I'm
don't
want
to
speak,
oh
john
watson's
here.
G
Yeah
sorry,
I'm
late,
I
got
a
little
distracted
by
the
political
situation
in
our
country.
What
was
the
what
was
the
question?
What's
the
statement
that
I'm
supposed
to
tell
you
you're
wrong.
G
Yeah
I
mean
I'm
still
pushing
to
see
if
we
can
get
it
by
the
end
of
january,
but
I
mean
mid
february
is
probably
more
likely,
maybe
release
candidate
by
the
end
of
the
by
the
end
of
the
month,
cool,
that's
yeah!
That's
the
plan.
A
So
what
we're
hoping
to
do
on
the
instrumentation
side
is
to
initially
we
were
going
to
try
to
release
the
java
agent
and
the
library
instrumentation
1-0
at
the
same
time,
so
that
would
include
like
the
instrumentation
api,
the
java
agent
api.
A
All
of
those
things
go
one
zero
together,
but
we've
just
got
so
much
work
to
do
to
stabilize
the
apis
so
that
we
don't
just
immediately
break
them,
and
so,
but
we
do
want
to
get
the
the
java
agent.
We
do
feel
pretty
comfortable
with
the
java
agent,
or
at
least
it's
getting
close,
and
so
the
thought
is
to
try
to
is
to
split
those
off
and
have
the
instrumentation
library,
the
library
instrumentation
and
those
various
apis
that
we
publish
all
be
tagged
with
alpha.
A
So
it
would
be
one
zero
alpha
kind
of
the
way
the
sdk
is
tagging
alpha
some
artifact
alpha
still
like
the
some
of
the
extensions
metrics,
and
then
that
would
allow
us
to
release
the
agent
as
one
zero,
hopefully
shortly
after
I
think
our
our
rough
target
is
to
try
to
do
it.
Our
desire
is
to
do
it
shortly
after
the
sdk
goes.
A
One
zero,
and
so
what
that
means
and
you'll
see
the
issues
I'll
be
kind
of
focusing
this
month
on
taking
out
things
that
are
not
sort
of
to
this
per
spec.
A
So,
where
there's
like
attribute
span
attributes
we're
capturing
that
aren't
semantic
attributes
per
the
spec,
hiding
those
behind
experimental
flags,
potentially
config
options
that
we're
not
sure
about
basically
trimming
down
the
surface
area
of
our
public.
A
H
Yet
trust,
could
you
go
to
the
github
release
display?
This
is
top
yeah
here,
yeah
releases
yeah.
I
see
that
there
are
some
this
minor
release
and
before
also
there
was
but
not
attached,
jar
of
the
agent
no
release
notes,
and
nothing
is
I
just
questioned:
should
it
be
used
or
not
used
by
like
vendor
specific
distributions.
A
That's
a
good
question.
I
don't
remember
29
now
how
do
we
get
to
here
and,
let's
see
what's
in
there,.
A
I
have
a
feeling
that
maybe
there
was
something
in
here
that
maybe
nikita.
A
Needed,
oh
yes,
spring
batch.
I
think
nikita
and
splunk
wanted
a
release
that
included
the
new
spring
batch
instrumentation
for
their
vendor
repo.
H
A
Yeah,
I
will
bring
that
up
with
honorable
nikita
just
that
if
we
are
going
to
make
a
release,
if
somebody
wants
to
make
a
release,
we
do
need
to
have
the
release
notes
at
minimum.
The
release
notes,
I
mean
the
attachment
you
can
is
the
same
as
what's
in
maven
central.
A
That
should
be
here,
this
artifact.
H
A
A
F
A
Is
it
my
understanding
is
the
way
I
haven't
made
a
patch
release
in
our
repo,
so
I'm
not
really
sure
how
it
works.
I
from
my
recollection
that
there
you
had
to
like
put
in
certain
commits-
and
it
would
auto
build
that
patch,
as
opposed
to
having
a
patch
branch.
G
It
depends
a
little
bit
on
whether
you
can
cherry
pick
or
not.
If
cherry
picking
works
from
where
the
previous
release
was,
then
you
can
do
that,
otherwise,
you
do
have
to
create
a
patch
branch,
but
either
way
I
mean
either
way.
G
F
A
Yeah,
I
think
what
we
need
to
do
is,
and
maybe
you
could
help
by
proposing
sort
of
a
workflow
and
pushing
you
know,
push
up
pr
to
under
docks.
We
have
a
release.
D
G
Does
this
does
seem
like
something
that
is
not
just
it's
not
just
a
java
thing
right
I
mean
this
applies
across
the
entire
project
to
anything,
so
it
seems
like
this
is
something
that
would
be
really
great
to
ask
you,
and
I
probably
should
do
this-
bring
it
up
to
the
maintainers
meeting
on
monday,
because
it
would
be
good
to
have
a
a
uniform
policy
across
all
the
languages
and
all
of
the
the
projects.
A
A
Yeah,
it's
gonna
be
hard
to
have
a
policy
like
because
it
is
still
like
a
call
that
you
know
maintainers
have
to
make
of.
Is
this
bug.
I
mean
we're,
probably
not
gonna,
you
know
build
a
really
like.
If
there's
you
know,
people
out
request
a
new
patch
release
every
day
for
something
you
know,
that's
gonna
be
too
much.
A
Unless
we
can
make
it,
you
know
more
self-serviced,
which
is
sort
of
what
I
was
gonna
ask
pawel
like
if
you
can
look
at
that
patch
release
mechanism
that
we
have
in
place
and
sort
of
propose,
like
you
know,
like
the
release,
notes
like
things
that
you
know
take
time,
that
if
somebody
could
produce
sort
of
a
template
of
here's
do
most
of
that
work
and
then,
if
the
maintainers
only
had
to
click
a
button
that
would
be
more
likely
to
you
know
to
be,
then
that
would
be
easy.
F
Yeah
so
simplify
release
process.
So
it's
almost
automated.
You
just
need
a
green
from
maintainer.
I
mean
it
makes
sense
in
thinking,
jaeger
and
open
tracing.
We
used
to
just
push
a
tag
and
it
would
trigger
the
release
and
do
everything
almost
everything
there
wasn't
release
notes.
A
Yeah,
so
most
of
that
is
automated.
The
the
trick
is
a
little
bit
with
those
patch
releases
right
like
what
do
you
want
in
the
patch
release?
Do
you
have
to
create
a
patch
release
branch
first.
F
I
mean
this
can
be
delegated
to
the
one
who
wants
the
patch
release
right.
The
the
maintainer
just
has
to
create
the
the
branch
from
specific
and
then
the
the
contributor
can
contribute
the
comments
to
the
create
the
pull
request
with
everything
that
is
needed
to
arrange.
A
Right
right,
yeah,
so
take
a
look
at
how
the
current
mechanism
works,
because
there's
like
john
and
I
were
talking
there,
there's
also
a
mechanism
where,
if
it
can
be
cherry,
picked,
there's
some
automation
there.
Where
you
can,
when
you're
building
the
release,
you
don't
need
a
release
branch,
you
don't
need
a
patch
branch,
but
maybe
that's
not.
You
know.
A
A
Yeah
the
way
I
understand
it,
let's
see
if
we
can
find
the
docs
here.
I
just
haven't
done
it,
so
I
am
just
kind
of
second-hand
info
from
what
I
have
heard
onorag
and
nikita
mention.
A
Check
out
the
release?
Okay,
so
this
one
press
the
button,
then
to
the
version
number
cherry
pick
so
there's
a
workflow
that
will
automatically
you
can
give
it
commits
that
you
want
to
cherry
pick
and
it
will
automatically
build
like
a
temp
branch.
You
know
and
apply
that
cherry
picked
commit
and
build
a
new
release,
so
it
should
be
pretty
easy
and
then
this
is
if
it
can't.
If
the
cherry
pick
can't
be
cleanly
applied
to
the
release
branch,
then
this
is
the
fallback
plan.
F
E
G
E
E
A
E
E
C
G
Yeah,
that's
it's
almost
certainly
documented
in
the
normal
release
process,
but
we
should
then
either
link
over
to
it
or
copy
paste
or
something.
G
A
A
E
A
A
Cool,
so
let's,
let's
we'll
circle
back
next
week
and
update
continue
this
discussion.
F
A
F
There
is
one
pr
I
think
it's
there
for
two
weeks:
it's
it's
just
refactoring,
basically,
because
when
you
maybe
if
you
can
go
to
your
agent's
tooling,
maybe
it's
better
to
see
it
on
the.
D
F
F
F
If
you
go
so
one
of
these
has
has
let
me
open
the.
A
F
A
Right,
yeah,
that
makes
sense
to
me
to
unify
the
ones
that
are
used
by
instrumentations.
A
F
F
F
E
F
E
A
D
D
F
F
F
I
mean
today
it
is
this
one,
and
maybe
tomorrow
that
we
might
have
a
different
use
case
that
is
instrumenting,
something
else
that
is
excluded.
Then.
A
Performance,
this
gives
us
all
of
these
exclusions
because
it
is
quite
a
random
assortment.
That's
just
you
know.
It's
not
really
a
whole
lot
of
logic
to
what
we
put
in
here
like
why
why
you
know
spring
framework,
but
not
you
know,
kotlin,
let's
see
if
kotlin's
in
here
we
do.
We
do
have
kotlin
in
here,
but
do
we
have
you
know
scala.
A
A
Okay,
cool
yeah,
sorry
I
had
meant
to.
I
was
getting
through
all
the
backlog
on
tuesday
and
then
wednesday
happened
and
nothing
happened,
and
so
hopefully
we'll
get
back
to
finishing
out
the
review
backlog
but
yeah,
I
think
I
mean
the
the
general
at
least
I
don't
I
mean
I
would
support.
You
know
some
kind
of
a
it's
a
problem
that
makes
sense
to
solve
for
vendors
yeah.
A
I,
like
the
I
mean
the
the
spi
idea,
just
because
it's
yeah
it's
and
then
you
know
like
what
I
could
imagine
is
just
to
be
able
to
then
delegate
to
this
to
the
default.
Somehow,
so
you
can
make
your
own
decision
or
if
you
don't
care,
you
can
delegate
to
the
default.
A
So
maybe
that
is
what
the
spi
could
do.
Is
it
the
sbi
could
return
either.
You
know
a
thumbs
up
thumbs
down
or
neutral,
and
if
new,
if
it
returns
neutral,
then
it
would.
The
java
agent
would
call
the
the
global
ignores
matcher.
F
A
Yeah
yeah,
somehow
I'm
not
quite
sure
how
that
would
work,
but
yeah,
because
that
way
you
can
override
without
having
to
know
what
all
of
this
without
having
to
copy
paste
and
and
without
having
to
subclass,
because
I
think
it
would
be
better.
You
could
subclass
this,
but
the
trying
to
keep
the
public
api
limited.
E
A
A
Cool,
hey,
our
our
our
agenda
keeps
growing
every
time
I
come
back
love
it.
G
Yeah
I've
been,
this
is
an
idea
that
I've
been
kicking
around
in
my
head
for
like
two
months,
and
I
just
have
continually
forgotten
to
propose
it,
which
is
basically,
I
think
it
would
be
really
cool,
and
maybe
it's
only
for
me,
but
you
know
it's
all
about
me.
You
have
to
be
able
to
start
the
agent
with
no
instrumentation
enabled
but
supporting
with
span
and
context
propagation,
exactly
yeah.
A
G
A
F
G
A
G
Context-
this
is
a
great
question.
I
don't
I
don't
know
and
that's
why
I
think
this
is
a
little
bit
more
complicated
than
my
really
simple
description.
Explain
like
describes
it
as
because
I
know
the
the
over
the
wire
context,
propagation
out-of-band
context.
Propagation
is
intertwined
with
the
actual
span
instrumentation.
In
almost
all
cases,.
A
A
So
if
you
just
want
this,
I
actually
think
that
there's
a
way
to
do
this
today.
G
G
D
D
Yeah,
it
would
be
in
config.java.
D
D
A
A
A
All
right,
we
hit
the
end
of
our
agenda
with
eight
minutes
to
spare
or
three
minutes
to
spare
until
the
still
try
to
tell
myself
this
that
all
meetings
end
at
five
till
the
hour.
A
Any
any
last
thoughts,
questions.
A
All
right,
well,
hopefully,
we'll
have
back
nikita
and
mateusz
next
week
and.
G
A
G
Calendar
yeah
and
he's
of
course,
going
to
be
coming
from
u.s
time
zone,
so
maybe
it'll
be
perfect.