►
From YouTube: 2020-09-18 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).
B
Can
you
might
see
my
screen
yep
great,
I'm
because
I'm
in
between
jobs,
I'm
having
to
use
my
gaming
computer
for
all
my
meetings,
I've
never
shared
screen
on
it
before.
B
B
Oh
well,
that's
why
I
got.
I
got
let
go
immediately
as
soon
as
I
gave
them
my
notice
so,
but
I
mean
it
wasn't
difficult.
B
Well,
I
you
all
are
probably
too
young,
but
back
in
the
90s
swatch
had
this
thing
called
swatch
time
that
there
was
without
it
was
metric
time,
basically
a
thousand
ticks
in
a
day,
and
it
was
universal
set
in,
of
course
geneva
as
the
zero
at
midnight.
B
Are
there
still
countries
that
you,
I
know
that?
Well
maybe
some
of
the
I
think
like
the
the
orthodox
churches?
Might
I
don't
know
if
there's
any
kind
of.
C
B
B
B
D
Good
point
makes
sense
what
is
splunk's
vacation
policy.
D
E
We're
just
waiting
for
people
to
show
up
and
now
you're
here,
carlos
yeah?
But
it's
you
know
like
holidays
policy
is
always
an
interesting.
D
E
Okay,
I
have,
I
have
four
rumors
but,
like
you
know,
you
never
know
well
right
now,
I'm
still
unemployed,
not
until
monday.
Okay.
So
thanks
for
joining
no
really
I
don't
know,
I
don't
know
you
know
a
lot
of
developers
when
we
are.
You
know
we
are
working
so
hard
sometimes,
and
then,
when
you
have
time
between
jobs,
you
just
want
to
disappear.
B
That
I
can't
go
outside
because
because
of
the
smoke,
so
what
else
am
I
going
to
do?
I
mean
I'll
doing
this
meeting
and
then
I'll
probably
I'll,
go
through
my
to-do
list
and
then
watch
movies
for
the
rest
of
the
day.
Probably
so,.
E
B
Cool
all
right:
well,
let's
get
started,
it
looks
like
I
think
we
have
a
quorum.
We
have
two
of
the
four
maintainers
giovanni.
You
had
the
first
item
on
the
agenda.
A
Yeah
I
was
going
through
the
spec
and
I
will
notice
that
I
noticed
that
we
have
different
naming
convention
with
what
the
spec
says
and
the
wonder.
Should
we
change
or
not?
I
agree
with
your
comment,
john.
I
also
found
that
our
naming
is
more
correct
in
terms
of
java
like,
but
I
want
to
hear
also
the
opinion
of
the
others.
B
Yeah,
I
also
will
say
that
metrics
there's
still
like
the
internal
pieces
of
the
metrics
sdk
are
still
haven't
been
specified.
So
I
wouldn't
worry
too
much
about
that
stuff.
Yet
josh
and
I
are
working
on
that
part,
but
I
think
we
should
be
as
much
as
possible.
Conformant
keep
our
names
as
close
as
we
can
to
the
spec
within
the
within
the
you
know
what
normal
java
practices
are.
That's
my
opinion.
F
B
E
D
D
A
B
Or
we
could
just
add
a
note
to
the
spec
that
says:
object-oriented
languages
may
use
nouns
and
verbs
differently.
I
think
I
think,
because
I
think
this
is
the
spec
is
based
on
what's
in
go,
which
just
has
a
function
called
record
batch
so.
B
B
I
don't
think,
there's
necessarily
a
100
general
rule,
though
okay,
the
next
item
meeting
times,
I
think
that
people
in
europe
don't
love
the
fact
that
we
meet
late
on
a
friday
evening.
B
So
I
think
we
should
figure
out
some
different
meeting
times
and
I
think
it's
probably
a
good
idea
to
have
them
alternate
week
to
week,
whether
it's
early
and
late,
unless
we
want
to
make
and
nikita
you
might
have
an
opinion
on
this-
make
the
thursday,
whatever
the
thursday,
I
guess,
the
sorry,
the
tuesday
late
night
meeting
like
the
official
java,
the
official
other
java
meeting
time.
Or
do
we
want
to
find
a
fourth
meeting
time.
That's
friendly
to
honor
ugg.
B
Well,
it's
going
to
be
really
hard
to
make
it
unified.
I
mean
the
10
pm.
One
is
the
only
one
where,
if
west
coast
is
willing
to
show
up
at
10
pm
that
we
can
get
everybody
in
and
I'm
only
willing
to
do
one
of
those
a
week.
B
Yeah,
I
guess
it's
1am
for
you,
then
right
yeah,
that's
that
doesn't
work
at
all.
So
I
think
we
should
have
two
meetings
right
now.
There
are
a
total
of
four
four
official
java
plus
instrumentation
meetings
in
the
week.
D
B
That's
well,
I
show
up
to
that
one
too.
So
that's
like
six
six
pm
pacific
time
on
thursdays
and
actually
zoe
from
australia
has
been
showing
up
to
that
as
well.
So
we've
had
we've
actually
had
four
people
there,
so
it's
actually
getting.
B
B
No,
I
can
all
right,
so
I
think
if
we
could
change,
I'm
gonna
just
put
a
note
here,
not.
B
There
so
we're
gonna
wednesday
morning,
and
I
would
if
we
can
do
9
a.m,
that
9
a.m,
pacific.
That
would
be
fantastic.
E
H
B
E
Okay,
so
shall
we
create
the
event
now
by
the
way,
if
you
can
yeah
that'd
be
great
okay,
so
what
should
we
call
it
like.
B
What
is
the
current
one
called?
I
think
we
should.
I
think
we
just
moved
the
friday
one
to
wednesday
morning.
So
that's
what
I
think
we
should
do.
D
Got
it
just
wanted
to
make
sure.
B
Yeah,
I
think
I
mean
I
will
show
up
to
both
just
to
have
some
continuity
and
I
don't
know
trotsky's,
usually
there
too
so
we'll
have
continuity,
at
least,
but
don't
be.
I
wouldn't
say
we
should.
You
should
feel
like
you
need
to
go
to
both
and
I
think,
like
trask,
you've,
been
doing
a
really
good
job
about
covering
the
thursday
morning
meeting
in
the
thursday
evening.
One-
and
I
think,
we'll
probably
just
add
the
like
the
wednesday
morning-
update
as
well
to
that
to
the
thursday
evening.
One.
G
So
yeah,
I
think
that's
a
good
yeah.
Let's
see
how
that
goes.
I
mean
the
only
thing
that
I
regret
is
that,
like
honorable,
it
can't
be
present
for
like
these
sort
of
decision
making
morning
meetings,
but
we
we
do
chat
online
with
him
a
lot
tuesday.
I
mean
a
couple
times
a
week
in
person
and
then
on
slack
all
the.
B
All
right,
oh
welcome
david.
We
have
a
new,
a
new
person,
hello.
B
Hey
everyone
all
right,
yeah,
actually
david,
if
you
could
introduce
yourself.
Let
us
know
why
you're
here,
what
you're
up
to.
F
G
B
I
I'm
going
to
merge
on
monday
if
nobody
screams
and
yells,
so
I
think
carlos
has
taken
a
look
at
it.
I
don't
think
bogdan
is
here.
Oh
bogdan
is
here
hello,
bogdan
hello.
I
see
your
name.
Oh,
I
haven't
seen
your
face.
Welcome.
B
Cool
anyway,
last
call
for
comments
on
that.
It's
been
open
for
a
long
time.
It's
been
quiescent
for
a
week,
so
please
I
don't
know
bargain.
If
you
have,
I
think
you
looked
at
it
at
least
a
little
bit
a
week
or
so
ago.
J
1631
yeah,
okay,
I
will
take
another
look.
This.
B
Okay
and
then
two
kind
of
biggish
items
to
talk
about.
E
Actually,
john,
since
bogdan
and
me
were
supposed
to
leave
to
do
some
triaging.
Can
we
talk
about
the
context?
One?
I
think
that's
the
important
one.
B
Okay,
I
think
the
moved
out
job
eight
is
also
important,
but
context.
One
may
be
more
important,
but
yeah.
E
J
B
Yeah,
so
I
think,
as
of
yesterday,
dynatrace
has
agreed
that
we
can
move
to
java
8.
They
may
be
keeping
a
private
fork
of
a
java
7
version.
J
B
Yeah.
This
is
why
I
kind
of
wanted
to
have
an
offer
like
a
quick
off-cycle
release
to
get
us
into
compilation
of
java
8.
So
then
we
can
then
start
actually
doing
java
8
work
immediately
after
that,
so
it's
like
starting
next
week
like
starting
at
the
end
of
next
week
or
mid
next
week
after
the
release,
we
can
start
considering
java
8
interfaces,
etc.
J
Does
java
8
interfaces
come
with
the
same
problem
as
abstract
classes
or
or
other
things,
and
here
I'm
I
mean
because,
for
example,
if
you
you,
you
can
add
new
methods
to
interfaces
by
providing
the
default
implementation.
Correct,
yes,
but
but
for,
for
example,
if
there
are
static
methods
to
to
do
something,
I
I
know
in
java,
you
can
now
have
static
methods
on
the
interface
correct.
H
One
I'm
not
quite
following
the
difference
between
I
mean
we
have
to
bridge
everything
anyway.
It
doesn't,
I
think,
the
the
stuff
with
more
of
the
discussion
that
we're
having
around
spring
and
dependency
injection
okay,
but
I
think
that
these
static.
J
Yeah
but
but
the
the
problem
is,
for
example,
or
maybe
it's
not
a
problem,
I
don't
know
if
you
have
of
interface
and
you
have
a
create
method
on
it
or
whatever
you
call
it
a
builder
in
a
build
like
a
factory
static
method.
J
If
that,
if
that
adds
a
new
overload
for
that
function,
is
that
gonna
cause
you
a
problem
or
are
you
gonna
shade
anyway?
The
interface.
G
We
will
need
to
add
bridging
capability
for
that
in
the
agent,
so
the
agent
1.0
won't
work
with
the
sdk.
Will
you
need
a
you
always
need
the
latest
agent.
B
J
Yeah
yeah
anyway,
the
re.
The
thing
that
I
want
to
make
sure
is
java.
8
comes
with
a
lot
of
features,
especially
around
interfaces,
and
I
want
to
make
sure
that
we
hear
the
requirements
from
auto
instrumentation.
For
example,
don't
use
static
methods
on
interfaces
or
or
things
like
that.
If,
if
you
have
a
set
of
things
that
you
would
prefer
us
to
not
do
or
to
do,
let
us
know
in
advance,
so
we
can.
We
can
respect
that.
H
G
Yeah
thanks,
I
think,
we're
good,
because
the
byte
code,
we
can
do
bytecode
instrumentation
of
all
those
things.
It's
basically
like
you
said
that
sort
of,
like
abstract
method,
stuff
that
existed
already,
but
we
will
we've
been
pulling
in
the
snapshots
pretty
quickly
as
they've
been
published
from
the
sdk.
K
F
G
Helped
with
on
getting
you
know,
quick
feedback
back
to
the
sdk
folks.
B
Cool
any
other
concerns,
so
my
action
item
then,
is
maybe
today
I'll
put
in
a
pr
to
well.
I
don't
know
I've
time
today,
maybe
I'll
see
if
I
can
put
in
a
pr
today
to
change
the
compilation
to
java
8..
Although
I
don't
know
we
have
to
also
figure
out
animal
sniffer
settings,
that's
the
other
thing
that's
going
to
be
annoying.
J
By
the
way,
don't
just
don't
just
do
java
compilation
try
to
change
any
random
class
somewhere
or
add
a
random
class
that
not
necessary
you
expose
that
uses
a
java
8
feature
just
to
ensure
that
that
the
compiler
is
not
doing
some
crazy.
The
bytecode
manipulation
the
bytecode
generated
is
actually
java.
7
compatible.
B
J
J
B
Yeah,
that's
and
we
can
even
we
can
even
be
a
a
package
access
class.
It
doesn't
have
to
be
public.
J
B
Yep
yeah
yeah,
so
android
api
level
is
the
other
question.
If
we
target
24,
then
24
is,
I
believe,
fully
compatible
with
java
8,
which
seems
like
a
pretty
good
option
david.
I
don't
know
if
you
have
any
android
contacts
inside
the
google
monolith
monster,
who
could
who
are
interested
in
android
support
for
open,
telemetry
or
could
at
least
be
an
advisor,
but
this
has
definitely
been
a
we've.
Had
a
lack
of
android
support
or
interest
in
the
project
and
we've
been
working
hard
to
support
android
without
a
lot
of
knowledge.
J
Cool
which
one
is
24.
D
J
So
24
is
nougat,
which
is
7-0
android
okay,
so
we
we
probably
gonna,
miss
three
percent
of
android.
If
we
do
that
devices.
So
I
I
looked
at
the
the.
C
J
What
I
was
trying
to
say
is
android
out
of
all
the
devices,
only
three
percent
are
on
version
less
than
24.
B
G
But
the
problem
with
that
analogy
is
that
in
the
java
7
case,
that's
we're
targeting
well
in
the
android
case.
G
But
I
I
think
that
you
know
we
can
I'm
comfortable
with
going
to
a
more
recent
version
and,
like
nikita
says,
because
we
don't
have
any
you
know
con.
I
mean
it's
hard
to
hold
the
project
back
for
something
that
you
know
nobody
has
expressed
interest
in
and
we
can
always
if
we
do
get
android
interest
and
older
api
level
interest
later
we
can
always
do
you
know
the
guava
thing
of
having
a
dash
android,
artifact.
B
B
G
Yeah
you
could
look
at
okay
http
and
see
what
api
level
they
are
targeting.
That
might
be
a
a
good.
I
know
honorary,
and
I
had
looked
at
that
one
in
one
meeting,
but
I
don't
remember
what
the
level
was.
B
Android
instrumentation
that
they
were
concerned
with
was
the
externals
from
http
calls
yeah
all
right.
Well,
I
think
we
have
a
plan
we
will
adjust
as
needed.
If
something
changes
we
can,
I
don't
think
we're
making
any.
We
haven't
yet
made
any
drastic
decisions,
they're
going
to
break
the
world
too
badly.
So
I
think
we
should
move
forward
and
react
if
somebody
gets.
B
Grumpy
all
right,
so,
let's
get
a
little
bit
of
discussion,
so
anarag
has
taken
on
the
task
of
essentially
taking
the
grpc
context,
implementation
and
stripping
it
down
slightly
and
putting
it
into
our
code
base,
removing
features
that
we
don't
need,
but
essentially
keeping
everything
else
for
the
most
part
intact
in
this
draft
pr.
B
B
It
won't
will
not
will
no
longer
require
a
grpc
named,
even
if
it's
only
a
name
really
name
only
dependency
to
be
pulled
in,
but
it
definitely
has
all
the
issues
that
we
know
about
with
the
fact
that
we're
going
to
have
our
own
context
and
it'll
be
different
than
everybody
else's.
C
J
I
don't
think
it's
a
giant.
I
nikita,
I
don't
think
is
a
technical
decision,
it's
more
like
do.
We
want
to
share
this
with
others
and
do
that
or
we
say
screw
it
if
a
third
party
library
wants
to
directly
depend
on
open,
telemetry
and
also
uses
grpc.
For
example,
let's
assume
one
of
the
google
client
libraries
for
spanner
or
something
like
that,
we'll
ask
essentially
we'll
ask
them
to
have
to
propagate
two
contacts
as
an
example.
G
So
that's
not
that's
not
true,
though,
because
nikita's
pr
includes
storage
spi,
so
you.
B
G
Write
sorry,
thank
you
on
rxpr
includes
storage
spi
so
that
his
the
new
context
mechanism
can
be
direct.
Can
basically,
you
can
write
a
little
storage
implementation
that
stores
it
into
the
grpc
context.
G
So
we
could
provide
that
as
a
jar
that
has
you
know
the
spi
manifest
services
file
and
then
people
who
want
grpc
want
to
use.
Grpc
only
can
use
that.
E
J
So
I
I
don't
know
anyway,
that's
one
example
and
one
concern
for
me
and
the
other
one
is
even
if
we
use
grpc.
If
people
are
using
reactor,
they
have
to
replicate
that
we
started
an
email
with
the
reactor.
Folks,
I
don't
think
they
responded.
Trust.
Can
you?
Can
you
send
me
your
email?
I
was
not
able
to
find
your
email
for
whatever
reason.
C
G
Yes,
mpsc
or
whatever
the
micro
profile,
yeah
yeah
yeah,
so
I
mean
yeah,
that's
why
I
like
the
you
know,
honorable
is
knowledgeable
of
you
know
different
contexts.
He's
worked
with.
You
know
that
he's
worked
on
armeria
for
a
long
time,
so
I
know
that
he's
building
this
with
that
use
case
in
mind
of.
G
Interop-
and
we
could
definitely
ask
for
you
know
to
prove
that
out
with
interop,
you
know
showing
interop
with
different
frameworks.
You
know
asking
for
interop
or
show
us
interop.
What
interop
looks
like
for
project
reactor
for
army
area
for
grpc,
for
maybe
micro
profile
pick
those
four
you
know
I
can
definitely
I
could
do
the
project
reactor
one
I
know
honorable
would
could
do
the
the
army
area.
G
You
know
bogdan
honorary
actually
also
has
worked
with
grpc
a
lot,
so
he
could
probably
do
that
if
you
don't
have
cycles,
the
mpsc
is
that
I
don't
know
if
I'm
even
getting
that
name
right,
but
the
microprofile
one.
I
didn't
look
at.
I
think
sergey
looked
at
that
a
bit
carlos
yeah,
that's
correct,
so
maybe
he
could
try
to
prototype.
What
interop
looks
like
for
micro
profile
using
this
pr.
J
B
F
J
G
Yeah
and
the
the
reason
is
because,
if
we
stored
other
contacts
inside
of
us,
then
when
they
we
wouldn't
really
the
other
libraries
would
basically
have
to
change
their
storage,
but
by
us
storing
inside
of
someone
else.
Inside
of
the
other
context,
we
basically
piggyback
on
the
on
their
context
propagation.
So
if
they're
already
propagating
the
grpc
context,
the
next
time
you
know
open
telemetry
goes
to
pull
the
current
span.
We've
magically
propagated,
along
with
the
grpc
context,.
G
The
swapping
is
spi,
so
you
need
a
jar
file.
You
know
we
can
have
an
extension
jar
file
that
people
need
to
add
to
their
dependency
dependencies.
That
has
the.
J
J
J
B
So
what
is
the?
What
does
the
auto
instrumentation
do
right
now
in
this
case,
where
someone
is
using
both
jrbc
and
arm
area
and
reactor
or
whatever,
with
all
the
so
you
know,
because
users
will
throw
together
every
library
they
can
find
into
one
giant
app,
that
does
all
sorts
of
crazy
stuff
right.
So
what
does.
C
C
B
G
J
B
Yeah
yeah,
I
know
I
was
just
trying
to
make
sure
that
this
wasn't
making
things
worse
or
was
or
was
potentially
making
it
better.
But
I
don't.
I
don't
think
it
is.
J
No,
I
don't
think
it
does
make
it
better
or
worse.
I
think
the
the
problem
is
when
people
are
using
manual
propagation
and
then
they
rely
on
on
third-party
libraries
to
do
the
same
whenever
they
are
used
and
that's
where,
where
you
have
problems
so
so,
if
if
there
is
no
auto
instrumentation
involved,
that's
where
a
lot
of
the
problems
will
come
with
yet
another
context
without
the
instrumentation.
J
B
Oh
yeah,
I
know
I've
been
on
the
working
on
instrumentation
at
new
relic.
Let
me
tell
you
having
to
people
to
interrupt
when
they're
trying
to
do
their
own
things
with
stuff.
We
didn't
support
in
auto
instrumentation
it.
It
was
probably
our
number
one
support
nightmare
and
that
had
nothing
to
do
with
open
telemetry.
Obviously,
but
anyway,.
J
E
Yeah,
in
my
case,
I
could
say
two
two
things:
the
first
one
is
that
I
would
like
to
keep
this
clean.
I
think
that
jrpc
context
had
a
clean
interface,
so
you
skipping
things
that
we
need
now
we
can
add
them
later.
The
second
thing
is
grpc
context.
Had
a
clean
interface
well
compared
to
others.
You
know
like
like,
for
example,
for
example
like
now
context
in
the
prototype,
it
has
a
lot
of
methods.
It
has
like
wrappers
for
runnable,
for
callable
and
for
adoption.
B
E
E
Stuff,
too,
you
don't.
G
E
E
The
second
thing
is
that-
and
probably
this
is
just
throwing
some
ideas
at
the
air-
is
that
since
we
are
defining
your
on
our
own
artifact,
I
wonder
if
there's
any
value
about
publishing
it
under
a
different
organization
like
we
did
talk
about
before
we
were
considering
putting
grpc
context
into
its
own.
E
You
know
repo,
I
think,
there's
a
lot
of
value
into
that,
but
I
understand
there
could
be
a
lot
of
work
as
well.
This
could
be
like
the
factor.
You
know
hopefully
context
propagation
artifacts,
but
probably
probably
too
hard
to
do.
B
I
think
it's
more
if,
if
somebody
wanted
to
own
that
and
make
it
like
push
it
through.
B
I
wouldn't
be
opposed
to
that,
but
I
think
at
this
point
we
haven't
had
the
we
haven't
had
the
the
gumption
to
get
it
actually
over
the
finish
line
right.
G
G
True,
no
you're
right
the
actual
context
you
wouldn't
you
would
still
be
dealing
with
that,
but
we
hide
no.
No,
we
do
we
do
we
have
to
pass
that
around
yeah.
G
So
I
take
that
back
sort
of
I
mean
you
still
have
that
separate
context.
That's
an
interesting
idea,
though
of
is
the
context
I
don't
remember
defined
as
an
interface
here.
J
B
J
I
mean
yes,
that
can
be
a
dream
in
order
to
do,
I
do
believe,
even
though
people
don't
believe
in
cancellation
and
deadline.
I
think
there
are
two
of
the
main
things
for
a
context
that
we
need
to
ensure
that
we
can
implement
those
on
top
of
this.
J
I
don't
know-
maybe
maybe
you
are
not
that
familiar
or
you
don't
use
these
two
concepts,
but
you
can
see
like
go,
has
them
and
they
are
very
useful
and
a
bunch
of
other
places
have
I
I've
seen
them
very,
very
useful
and-
and
I
think
we
should
at
least
prove
that
those
can
be
implemented
on
top
of
our
context,
not
the
way
how
grpc
did,
which
is
embedding
them
together
and
stuff,
but
making
sure
that
we
can
implement
them.
On
top
of
the
context.
J
Yeah,
I
I
did
some
prototyping,
but
I
haven't
finalized
the
apis
for
that,
but
before
before
making
it
like
its
own
artifact,
I
think
we
should
investigate
or
continue
that
work
a
bit.
J
G
K
G
I
I
got
a
little,
I
mean
just
from
trying
to
navigate
all
the
different
contexts
and
sort
of
gauging
appetite
for
a
centralized
context.
G
I
didn't
get
a
very
good
feedback,
very
warm
feedback
in
terms
of
that
being
a
something
that
was
going
to
work
or
without.
Like
I
mean
basically,
I
think
we
need
a
a
a
java
enhancement
proposal
to,
and
that
would-
and
I
think
you
know
that
would
probably
take
years
to
for
everybody
to
hash
out
their
differences
on.
You
know
whether
they
want
typed
keys,
whether
they
want
the
thread:
local
storage,
whether
they
want
deadlines,
so
it
just.
G
It
feels
very
complicated
to
me
to
it
feels
like
a
huge
work
stream
to
try
to
make
a
context
that
would
gain
that
kind
of
traction
to
be
worth
having
it
separate
versus
just
having
you
know,
focus
on
what
we
need
for
net
focus
on
what
we
need
and
put
it
in
open,
telemetry.
B
B
So
when
project
loom
lands,
there
will
be
a
context
in
java,
like
a
first
class
context
in
java.
B
Yeah
project
loom
is
super
cool
they're
doing
some
really
interesting
things
around
the
concurrency
model
that
you
use
for
actually
running
things,
and
it
does
include
a
context.
C
C
B
Here,
but
it's
it
is.
It
is
interesting.
B
We
were
just
chatting
we
just
project
loom
was
brought
up
as
like
hey,
we
have
to
support
project
loom
propagation,
you
know
in
java,
21
or
whatever.
Probably
I
was
thinking
21
is
when
it
will
show
up
or
24.,
but
at
that
point
there
is,
there
will
be
a
context
for
propagation,
a
standard
context
for
propagation.
In
java.
B
Oh,
I
haven't
looked
at
it
in
a
long
time,
and
it
also
is,
I
mean,
there's
there
are
a
long
time
from
being
done
with
that,
so
we
can
take
a
look
at
it
and
they
will
change.
C
B
Yeah,
I'm
sure
of
that
yeah
I
can.
C
B
Can
probably
take
a
look
at
that
at
some
point
in
the
next
week,
depending
on
what
my
schedule
looks
like
next
week
see
what
they
their
context
looks
like
these
days.
B
Looked
at
it
in
a
year
a
year
ago,
when
I
looked
at
it,
I
saw
it
they
had
one
there,
because
they
have
this
idea
of
structured
concurrency.
Where
you,
you,
basically
can
only
spawn
tasks
in
a
essentially
a
try.
Try
with
the
resources
type
model,
where
the
context
will
is
auto
closable
and
gets
closed
at
the
end
of
the
or
detached
at
the
end
of
the
operation.
B
Anyway,
I'll
take
a
look
and
see
what
their
see,
what
their
interface
or
what
their
api
looks
like
at
the
moment.
K
B
B
Okay,
well,
I
think
that
is
probably
going
to
wrap
it
up.
I
think
we
we've
made
a
decision
makes
me
very
happy.
I
will
document
that
in
the
very
long
running
issue
and
put
some
comments
on
honorable
draft
around
interfaces.