►
From YouTube: 2020-08-18 Java SIG
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
A
A
Hey,
do
you
know
if
the
gradle
I've
noticed
when
I
run
gradle
assemble
the
java
doc?
It
seems
like
the
javadoc
isn't
benefiting
from
the
gradle
cache.
B
A
Oh
okay,
is
there
a
gradle?
I
guess
I
can
do
a
dash
x
to
exclude
what
it
like
what
gradle?
What
do
you
run
typically
like
to
verify
the
problem?
A
Is
I'm
doing
a
lot
of
things
that
cross
over
a
lot
of
instrumentation
and
things
so
normally,
I
would
just
like
run
the
tests
on
a
given
instrumentation
that
I'm
messing
around
with,
but
for
a
bunch
of
them,
I'm
just
kind
of
running
like
gradle
assemble,
and
if
that
works
then
using
ci
to
run
the
tests,
do
you
have
better
thoughts
on
that?
B
B
Run
it
depends
on
depends
on
which
machine
do
do.
I
run
it
like
from
eight
minutes
to
20
minutes,
okay
and
and
depends
on
what
I
changed
yeah,
if
you
change
some
tooling,
for
example,
which
requires
to
all
tests
to
run
that.
A
Yeah
I
haven't
tried
that
cool
yeah.
I
should
try
that
now
with
the
gradle
cache
in
place,
because
in
theory,
if
I'm
changing,
even
if
I'm
changing
10
different
instrumentations
as
long
as
I'm
not
changing
the
tooling
or
the
yeah.
B
A
C
B
B
A
There's
been
some
good
discussion,
what's
yeah
anrag,
what's
your
thoughts
currently
about
java,
7
and
android
or
java
7
desktop
android
any
of
that
yeah.
C
C
C
C
A
Well,
I
think
that
we,
I
don't
think,
there's
any
disagreement.
Well,
I
don't
know.
I
definitely
don't
disagree
that
we.
I
think
that
supporting
android
is
a
must
how
we
support
it.
A
I
don't
certainly
supporting
java
7
keeping
the
baseline
at
java.
7
is
the
easiest
way.
A
So
yeah
it
would
take.
I
think,
a
lot
of
effort
to
potentially
have
sort
of
what
like
ben
evans
was
commenting
in
the
in
the
java
repo
that
he
had
seen
with
other
projects
sort
of.
C
C
A
On
the
instrumentation
side,
so
I
mean
oh
yeah
did,
let
me
share.
Did
you
both
see
the
issue
kind
of
the
text
that
I
was
proposed
that
I'm
gonna
thinking
to
post
to
the
community
repo
sort
of
requesting
android
trying
to
get
some
android
expertise?
A
A
Okay,
we
were
having
a
discussion.
D
A
The
java
x
maintainer
channel
there
about
a
couple
things.
Oh.
B
A
So
yeah,
just
trying
to
from
the
android
angle.
A
Yeah,
since
I
don't
know
android,
I
know
nikita
doesn't
android,
it
seems
like
you
have
some
or
just
from
other
projects,
sort
of.
C
C
B
That's
my
problem
with
android:
do
we
have
any,
but
any
future
user
at
all
like
expressing
the
interest?
In
that
I
mean
I
mean
in
general,
it
seems
like
a
must
to
have
the
support
for
for
android,
because
we
want
to
support
everybody,
but
do
we
have
any
user?
Any
company
who
says
I
am
waiting
for
open
telemetry,
because
I
want
to
monitor
our
android
applications.
D
B
C
B
C
C
C
C
A
Yeah,
maybe
maybe
I
should
expand
this
post
to
almost
more
be
like,
for
we
want
to
form
an
android
working
group
or
android.
I
mean
to
gather
not
just
necessarily
developer
expertise,
but
like
product
manager
vendor
in
basically,
anybody
who
wants
to
have
who
wants
to
who
wants
us
to
support
android.
B
I
I'm
a
little
bit
afraid
of
that
because
well
in
my
head
there
are
two
more
or
less
separate
issues.
No
the.
If
we
want
to
have
a
good
android
support,
we
need.
B
D
B
C
A
So
I
was
looking
at
the
okay
http,
their
git.
They
talked
about
the
api,
the
android
api
levels.
A
And
on
java
8
on
a
panel
and
on
okay.
B
B
A
A
C
C
C
A
A
C
C
A
Here,
do
you
want
to
why
don't
you
share,
I
think
you
got
a
better.
C
A
A
So
we
could
do
the
same
for
because
I
I
I
don't
mind
so
much
so
from
the
testing
perspective.
A
The
thing
that
worries
me
most
is
the
java
agent
test
only
testing
the
java
agent
instrumentation
on
java
8,
but
saying
well,
we
compile
it
to
java
7,
so
it
works
on
java
7..
It
should
work
on
java
7.,
but
for
the
library
instrumentation.
A
A
B
So
we
are
now
saying
that
we
don't
aim
to
support
okay
yeah.
We
can
support
out
in
auto
instrumentation.
That's
on
the
right
java
agent,
yeah.
D
C
A
If
we
could
do
it
with
confidence
with
the
what
but
the
tests
I
mean,
if
we're
I.
D
A
Supported
ship
not
supporting
java,
seven,
okay,
except.
B
A
C
B
D
B
B
A
Yeah,
so
maybe
for
ga
I
mean
it
would
be
nice
to.
A
I
think
it
would
be
worth
trying
to
do
one
like
okay
http,
just
so
that
we
have
the
template
for
it
and
with
this
approach
it
seems
like
it
would
be
relatively
straightforward,
and
that
would
make
sure
that
we
don't
our
instrumentation
api
doesn't
grow
outside
of
the
android
level
that
we
want
to
support.
A
A
So
do
you
think
I
I
mean:
is
it
worth
starting
that
discussion
now
in
the
community
repo
and
trying
to
get
those
people,
or
would
we
rather,
since
those
people
aren't
coming
to
us,
postpone
that
you
know
until
ga.
B
I
think
what
it
it
probably
is
a
good
thing
to
start
a
discussion
with
the
specific
goal
right
now
before
jay,
to
understand
what
what
paths
we
don't
want
to
block
so
like
we
don't
want
to
migrate
everything
to
this
for
to
java,
8
or
11,
because
we
need
that
android
api
levels
support.
So
what
is
the
minimum
to
run
at
all.
A
C
C
C
A
C
A
C
So
there's
no
stream
api
or
any
of
a
completable
feature
any
of
the
apis,
but
the
language
features
work,
including
the
default
and
sonic
interface.
So
it's
a
bit
weird
like
that,
would
be
confusing
for
java
developers.
I
think,
but
as
long
as
they
have
animals
before
running,
I
guess
we
wouldn't
have
problems
so
it
it's
definitely
something
we
have
to
think
about.
I
don't
know
about
this
option
until
right
now,.
A
Yeah,
do
you
know
how
does
android
like
do
you?
Do?
You
know,
use
a
normal,
you
compile
to
normal
byte
code
and
then
the
dalvik
vm
goes
from
there
or
you
have
to.
C
A
A
C
A
Wait
thread
local
random.
I
thought
that
was
one
of
those
things
that
the
they
kicked
out
of
the
sdk
early
on
because
of
android.
A
A
C
A
C
A
D
C
A
So
does
that
seem
to
be
the
popular
way
to
to
check
the
android
level
as
the
animal
sniffer.
C
B
D
C
A
Your
thing
can
we
tell
what
okay
http
is
doing
for
this.
C
A
Okay,
so
it's
manual,
it's
not
like
automated.
B
C
B
A
A
B
So
I
think
your
your
call
to
the
community
for
give
us
actual
people
who
are
who
are
interested
in
using
that
and
help
us
with
establishing
that
baseline.
That's
very
good!
We
need
that
before
gga,
but
we
certainly
will
not
support
more
than
bare
minimum
ourselves
before
gega
and
probably
even
after
that,
we
will
need
the
android
community
to
take
stewardship.
Well,
not
stewardship,
to
take
to
take
lead
on
that.
A
So
sorry,
back
to
that
nikita's
about
I
mean
like
okay
http
right.
If
we
we
want
to
do
library
instrumentation
for
okay
http
already
so
as
long
as
we
for
that
library
instrumentation,
we
stick
in
animal
stiffer
for
ok,
http
library,
instrumentation
and
for
our
instrumentation
api
were
done
right
I
mean,
maybe
we
think,
would
we
don't
really.
D
B
C
D
A
C
C
B
B
A
All
those
little
classes
that
we
bridge
back
and
forth,
I
want
to
turn
those
into
interfaces
with
static
builder
methods.
I
think
that's
true.
I.
A
C
A
Well,
we
definitely
made,
I
feel
like
we
know,
or
I
feel
like
we
know
exactly
what
we
want
to
do
on
the
instrumentation
side.
Now
that
makes
we'll.
C
A
To
java
8
we're
going
to
pin
the
instrumentation
api
module
at
pretty
much
the
low
I
mean
we
may
as
well
just
be
conservative
to
start
with
and
pin
it
at
the
lowest
android
api
level
and
get
you
know
we'll
get
the
sugaring
stuff.
The
de-sugaring
stuff
that's
available
at
any
api
level.
A
Yeah,
so
I
think
that's
on
the
instrumentation
side,
that's
our
takeaway
and
then
ok
http.
It
would
still
be
nice
to
have,
I
think,
for
ga
to
have
the
library
instrumentation
for
that.
So
we
have
at
least
one
thing
that
we
can
say:
hey
try
this
out
start
getting
feedback,
give
the
community
something
so
they
can
get
involved
and
and
I'll
open.
So
that's
that's!
A
On
the
instrumentation
side,
I
will
make
that
call
to
the
community
repo
to
try
to
gather
whatever
we
want
to
call
it
a
loose
working
group
stakeholders,
stakeholders,
yeah,
just
gathering
the
stakeholders,
so
we
can
have
discussions
and
then
the
big
open
question
is
what
to
do
at
the
api
sdk
side,
but
I
I
I
can
take.
I
can
write.
I
can
respond
on
that
thread
just
sort
of
about
our
findings.
A
Unless
one
of
you
want
to
and
propose
that
you
know
that
it
feels
like
we
have
a
way
forward
where
we
could.
If
we
drop
java,
7
server
side
support,
we
could
get
java
most
of
java
8.
Some
of
the
the
good
thing,
some
of
the
a
lot
of
the
good
things
we
want
out
of
java
8,
at
least
plus,
maintain
full
android
support.
A
Yeah,
I
can,
I
don't
mind,
pushing
on
pushing
that.
I
think
it
makes
a
lot
of
sense
for
to
do
it.
Now
I
mean
for
a
new
a
brand.
It's
a
brand
new
project.
C
Brand
new,
like
a
like,
open
tracing
shim
like
these
sort
of
things,
if
they're
java
7,
they
won't
work
either
right,
and
so
it's
like.
I
wonder
what
our
backwards
like.
I
guess
that's
what
carlos
is
working
on.
Maybe
it's
a
point
to
add
there,
but
one
could
argue
that
our
compatibility
requirements
require
java
7.
If
these
two
projects
are
what
we're
bringing.
A
But
why,
as
long
as
somebody's
using
open
tracing
with
I
mean
they
have
to
be
using
java
8
still,
but
why
does
the
open
tracing
shim.
C
Runtime,
that's
right!
It's
not
a
compound
yeah.
I
guess
runtime
is
a
reasonable
requirement.
That's
true.
A
C
A
A
Yeah
and
I
nikita
I'm
kind
of
with
you
on
the
these
legacy
apps
that
are
still
running
you
know
old
java
7,
like
I'm,
not
following
exactly
why
somebody
would
be
touching
them
to
add
new
monitoring,
especially
the
using
the
api
which
requires
going
in
and
changing
that
code.
I
mean
if
it's
that
legacy,
don't
touch
don't
break.
I
understand
I
totally
get
the
don't
touch,
don't
break
I've
been
there,
but
part
of
don't
touch.
Don't
break
is.
B
A
So
yeah
that
was
an
auto
java
agent
user.
D
B
About
right
agent,
but
well
that
that
I
don't
understand
either
if
it's
too
hard
for
you
to
test
your
application
on
java,
8
and
migrate
to
java
a
why.
Why
do
you
think
that
adding
some
extreme
bytecode
modification
to
your
application
is
easier
and
less
risky.
A
What
happens
is
there's,
there's
certain
libraries
versions
of
libraries
that
work
on
java
7,
some
popular
ones
that
fail
on
java
8,
and
so
you
get
pinned
in
and
you
don't
upgrading
library
versions
feels
very
risky
and
sometimes
requires
you
know
that
has
api
breakage
yeah,
whereas
throwing
a
java
agent
on
you
know
you
can
test
that
sort
of
separately
or
vendors.
D
A
C
C
C
A
If
we're
on
the
agent
side,
if
we're
going
to
java
8,
then
I
don't
see
why
the
api.
B
B
B
We
will
be
using
an
open,
telemetry
based
solution
for
our
for
our
own,
auto
instrumentation.
That
is
the
exact
work
of
christian
for
from
that
issue.
B
A
Any
all
right
and
the
I
know,
you're
a
little
hesitant
honorable
any
thoughts
before
I
tried
to
push
forward
with
java
8
on
the
api.
C
C
A
That's
what
you
know
new
relic
does
when
they
bump
up
their
java
agent
version.
Is
you
know
they're
like
okay?
You
want
to
use
java
6
here
we
have
this
old
java
agent.
You
know
that,
still
that
a
lot
of
customers
are
using
on
java
6,
you
can
use
it.
We
don't
maintain
it
anymore,
but
you
can
send
us
data
and
pay
us
money
still
yeah
yeah,
there's
other
options.
We're
not
the
only
monitoring
solution.
D
A
A
All
right
was
there
anything
else
that
either
of
you
wanted.
A
Yeah,
so
the
no,
I
didn't,
try
it
so.
C
A
C
A
Me
nervous
the
whole
putting
the
putting
the
the
api
unshaded
in
the
bootstrap
class
letter
makes
me
nervous.
B
A
Do
if
we
can,
if
we
can,
if
we
convert,
trace
id
and
span
id
to
interfaces,
I
can
we
can
bridge
without
without
going
back
and
forth.
D
B
D
A
A
Okay.
Okay,
so
you
you
still
yes
yeah!
No,
it's
a
very
interesting
idea.
Sorry,
it
just
makes
me
nervous,
I'm
having
a
hard
time
getting
over
my
nerves,
about
sticking
the
pinning
stuff
into
the
bootstrap
class
loader.
But
you
know
we
would
have
good
tests.
I
if
we
can
show
that
it
works.
A
And
that
would
yeah
so
I
mean
if
we
do.
I
should
work
on
that,
because,
if
we
can
do
that,
that
also
removes
part
of
I
mean
my
my
desire
for
the
java
8
on
the
api,
at
least.
A
A
A
Yeah
keep
ping
me
if
I
don't
get
back
to
you
on
that
in
the
next
couple
days.
Nikita.
B
A
Will
oh
and
you're
you're
driving
thursday's
meeting.
C
A
On
thursday,
so
yeah,
that's
why
I
was
on
the
community
repo.
I
was
asking
about
the
the
videos,
because
I
couldn't
find
our
video
from
last
week's
meeting,
and
I
want
to
make
sure
that
this
week's
meeting
gets
recorded
because
new,
relic
jason,
plum
from
new
york,
is
going
to
give
a
code
overview
of
their
of
their
agent,
which
should
be
cool.
A
If
it's
up
by
the
time
our
our
meeting
starts
honorable,
we
can
just
watch
the
meeting
together.
A
Oh
yeah
nikita
the
the
last
sort
of
tracer
refactoring
pr
from
helen
is
ready
for
review
yeah
and
then
she's
gonna
work
on
some
internal
stuff
for
a
little
bit
after
this.
So
the
tracers
are
all
yours.
A
Oh
yeah,
so
you
wanted
to
so
as
far
as
those
using
those
the
auto
generated
semantic
spans,
we're
calling
them
semantics
bands.
B
A
Yeah,
they
were
always
calling
them
typed
spans.
Okay,
and
do
you
think
what
what
are
we
hoping
to
get
out
of
those
using
those.
B
B
What
was
the
word
discoverability?
So
if
you
just
if
you
just
autocomplete,
you
see
which
attributes
rst
are
still
like
unused,
so
we
we?
What
attributes
you?
Don't
you
don't
like
set
because
semantic
attributes
they
they
don't
give.
I
believe
such
good
overview.
Okay,
I'm
currently
writing
instrumentation
for,
like
kafka
consumer,
which
are
what
are
messaging
related
attributes
that
they
have
to
say
to
set.
A
Yeah
yeah,
so
the
semantic
attributes,
those
constants
I
mean
everything-
would
be
started
with
the
message
yeah.
No,
I
it
makes
sense.
I
I'm
just.
I
want
to
make
sure
we're
getting
enough
out
of
it
to
be
worth
the
additional
allocation
right,
because
that
the
way
I
mean
we
we're
gonna
have
to
instantiate
a
new
those
wrappers
essentially.
D
B
A
I've
just
been
really
happy
with
the
new.
The
tracer
stuff
has
been
cleaning
up
so
much
of
the
instrumentation
and
it's
getting.
You
know,
and
I
mean
certainly
there's
still
a
lot
more
cleanup
that
we
can
do.
A
A
Oh
yeah,
honorable
just
checking
the
armeria
instrumentation
context:
propagation
yeah.
Oh
nikita
reviewed
it
okay,
just
as
long
as
you're,
not
waiting
for
us.
You
take
take
as
long
as
you
want
yeah.
A
Much
faster
yeah,
so
yeah
going
through
it
also
there's.
I
was
noticing.
A
Oh
yeah
yeah,
plus
it
motivated
me
to
go
through
and
do
more
cleanup
on
those
that's
what
my
currently
failing
draft
pr
is.
A
But
now
that
it's
like
I
don't
know,
I
find
same
with
the
tracer
refactoring
stuff
like
once
it
starts
to
get
better.
It's
like
motivation
to
go
in
and
even
like,
you
can
start
to
see
where
things
are
still
weird,
so
cool
any
anything
else.
A
B
A
Yeah
after
steve,
when
so
stephen
and
I
recorded
it,
and
we
finished
and
he's
like-
oh
well,
you
know
like
we
could
go
back
and
we
could
edit.
We
could
redo
it
like
nope,
that's
good!
Whatever
happened,
it's
all
good!
I
don't
want
to
read.
I
don't.