►
From YouTube: 2021-12-15 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).
A
A
A
B
A
A
B
C
All
right
does
anybody,
anybody
is
anybody
else,
expect
anybody
else
or
shall
we
get
rolling.
A
I
don't
know
about.
I
forgot
this
real
name,
and
I
don't
even
know
how
to
pronounce
this
get
that
name.
Grunette.
C
I
haven't
heard
from
him
this
week,
maybe
he's
not
planning
on
attending.
I
know
that's
actually
one
of
the
topics
on
my
agenda
today.
I
know
we're
gonna
have
a
bunch
of
what's
the
right
word
holidays
and
paid
time
off
over
the
next
couple
weeks,
and
I
figured
we
should
talk
about
that
in
our
meeting
too
I'm
technically
on
paid
time
off
today,
but
I
just
wanted
to
come
and
see
your
smiling
faces.
C
C
A
A
Feelings,
that's
too
much!
I'm
sorry.
C
A
And
there's
still
a
problem
that
you
cannot
really
fulfill
it
because
you
don't
have
the
laws
in
place.
It's
a
real
problem
for
for
our
side
as
well,
because
you
know,
there's
a
general
was
a
general
agreement
between
usa
and
the
eu.
Considering
on
this
topic,
when
there
have
been
lawsuits
with
this
invalid
and
very
so
for
european
countries,
it's
still
even
you
know
if
you
offer
gde
pr
compliance,
it's
still
kind
of
a
gray
thing
to
accept,
because
at
the
end
of
the
day
we
cannot
get
you
to
court
or
something.
B
C
B
C
A
You
can't
even
block
ip
addresses.
You
know,
that's
really
really
problem
with
a
lot
of
people,
don't
think
of,
but
that's
what
I
p
addresses
with
the
time
stamp
are
personal
information,
because
you
could
figure
out
who
that
person
wants.
It's
super
complicated,
so.
C
That's
true
all
right,
let's
get
rolling
on
our
open
discussions
brad.
Would
you
like
to
go
first
talking
about
global
versus
logger,
aware
trait.
B
B
Yeah
yeah,
I
hope
some
people
know
tomorrow,
you've
had
a
good
look
through
them
yeah.
I
think
it's
time
to
choose
personally
like
they
present
much
the
same
api.
Personally,
I
think
that
the
global
tracer
holder,
one
is
just
a
lot
simpler,
like
that.
The
api,
particularly
for
the
sdk,
doesn't
have
to
really
change
it.
All.
You
know
a
couple
of
other
sort
of
minor
issues
I
could
think
of
with
logger,
aware
trade,
the
difference
between
logging
and
application,
and,
first
looking
for
open,
telemetry.
B
An
application
sort
of
bootstrapping
and
creating
an
open,
telemetry
thing
so
as
if
it's
creating
you
know
an
exporter
or
a
span
converter
and
sees
that
it
is
influence,
logger,
aware
interface,
then,
is
it
going
to
inject?
A
Well,
you
you
can
still,
if
you're
talking,
for
example,
about
something
or
something,
but
you
can
still
configure
this
behavior,
so
it
can,
there
might
be
a
default
which
you
can
turn
on
and
off,
but
then
you
can,
you
know
narrow
it
down.
Only
have
this
do
this
injection
for
a
certain
type
of
classes
or
with
when
they
implement
another
interface,
and
things
like
that,
so
you
can.
You
can
have
very
fine
grained
control
if
you
want
to
have
also.
I
wouldn't
worry
about
too
much
what
applications
do
at
this
point.
A
So,
but
you
probably
know
my
stand,
I
don't
like
global
things,
so
I
guess
you
and
bob
decide.
I
will.
I
don't
have
to
have
everything
my
way
so,
but
I
still
have
my
opinion.
So
one
thing
I
see
is
it's
a
problem
which
maybe
would
solve
this
in
the
future.
We
have
an
open
ticket
which
give
me
something
to
find
and
bobbed
it
like
the
last
command
and
give
me
one
second
yeah.
It's
I
put
it
here
in
the
rest
of
the
chat.
Let's
take
this
one.
A
So
this
is
a
little
bit.
I'm
complaining
about
that.
We
don't
have
a
global
access
point
and
things
like
that,
like
other
sdks,
have
and
having
something
like
this
is
just
the
correct
one.
I
think
would
actually
solve
a
little
bit
of
some
of
the
problems
we
actually
have.
If
there
would
be
a
global
global
state.
You
know
what
I
mean
so
one
one
access
point
which
can
reference
all
the
global
stuff.
We
have
so
one
global
object
holder.
Then
I
wouldn't
have
less
of
a
problem
with
all
the
scroll-over
stuff.
A
So,
but
for
me
now
it's
basically,
you
could
put
the
logger
into
a
super
global.
You
know
it's
not
much
different.
I
know
I'm
exaggerating
a
little
bit
to
get
more
points,
but
but
I'm
fine,
whatever
pop
you
decide
and
I
will
keep
my
mouth
shut,
I
can
only
and
we're
dealing
with
software.
You
know
if
we
want
to
change
it
later.
A
A
Well,
but
you
will
you
will
if
we
have
agreed
disagreeing
positions,
then
you
are
the
one
or.
D
A
There
are
disadvantages
of
using
global
state,
so
it's
hard
to
reason
about
you
never
know
actually
in
what
this
state
something
global
is
because
the
access
point,
this
you
know,
the
benefits
of
it
being
accessible
from
all
from
everywhere
is
also
its
downsides.
You
don't
know
which
parts
are
changing
anything
about
it
and
there
are
a
lot
of
things,
but
at
the
end
of
the
day,
we
actually
have
to
look
how
this
plays
out.
A
A
C
The
other
thing
that
we
have
to
remember
is,
I
don't
know
if
y'all
have
you
heard
the
term
signal
get
used
yet,
in
the
context
of
this
project,
a6.
A
A
C
It's
very
complicated,
but
like
the
idea
of
a
signal
is
tracing,
is
a
signal.
Metrics,
as
a
signal
logging
is,
is
another
signal
that
we'll
do
in
the
future
and
then
any
additional
quote
unquote
that
google's
the
word
signal
very
ambiguously.
Any
additional
signal
that
may
come
into
the
open,
telemetry
library
in
the
future
could
fall
into
its
own
classification.
C
The
reason
that
I
bring
this
up
is
because
eventually
we're
going
to
want
to
work
on
the
logging
signal,
and
I
think
that
it's
important
to
think
about
we
potentially
might
want
to
use
our
own
dog
food
at
some
point
in
that
case
and
be
able
to
use
a
different.
You
know
inter-operate
between
our
signals
and
I'm
not
sure
which
way
would
be
better
to
interoperate
signals,
whether
it
is
to
use
the
logger
aware
trade
or
the
global
context
variable.
C
A
Well,
that's
actually
a
valid
point.
We
often
tunnel
vision
on
tracing
only
while
we
have
three
other
quote-unquote
signals
or
whatever
you
might
want
to
call
it,
which
we
actually
have
to
think
of
as
well
so
but
yeah.
I
think
this
in
general,
open
telemetry
and
all
we,
the
stuff
we
have
to
implement
is
it's
kind
of
complex
and
it's
a
little
bit
too
much
for
one
person's
brain
to
hold
all
the
information
and
all
this
stuff.
A
You
know
to
be
present
all
the
time,
so
I
think
yeah
we
just
should
go
for
one
of
the
implementations
and
then
the
good
thing
and
the
change
that
brett
did
was
so
that
now
it's
a
trade
and
so
it's
a
little
bit
isolated.
So
if
you
want
to
change
the
global
logger
holder
into
something
else,
we
don't
have
to
change
the
reference
to
it
everywhere,
but
just
to
change
the
trade.
So
this
this
helps
me
a
little
bit
with
my
concerns.
A
C
C
C
So
that's
something
that
you
mentioned,
that
you
might
want
to
implement
in
the
future.
Are
you
saying
that
that's
important,
but
not
urgent,
and
that
you
will
eventually
get
to
it?
I
think
that's
that's
what
we
came
up
with
well.
A
It's
more
like
a
it's
a
development
environment
which
offers
kubernetes
or
but
you
can
simply
it
kind
of
works
already
out
of
the
box.
You
can
point
it
to
the
repo
and
it
takes
the
compose
file
and
tries
to
do
something
with
it.
You
know
turn
it
into
kubernetes
objects,
yeah.
A
Well,
not
really!
No,
I
think
it
doesn't
create
manifest.
It
just
tries
to
create
deployments
or
something
like
this
and
okay,
but
you
can
do
something
better
with
this,
so
the
idea
is
just
to
have
some.
You
know,
like
the
collectors
and
all
the
stuff
we
need
to
in
a
in
a
place
which
is
not
on
a
local
computer
and
it's
free
and
it
comes
with
you
know,
the
the
quotas
are
really
really
good
for
something
that
is
free.
A
So
if
you
were
to
have
to
pay
for
cluster
like
this
would
be,
I
don't
know
aws
at
least
thousand
bucks
a
month
or
something
like
this,
so
the
idea
is
just
to
offer.
I
would
create
some
little
charts
that
I
do
this
all
the
time
you
know
it's
just
and
then
there
would
be
a
very
small
configuration
for
octet
or
whatever
it's
called,
and
if
we
people
want
to
use
it,
then
they
can
just
do,
but
it's
not
there
to
actually
replace
docker
compose.
But
I,
for
my
my
part,
I
don't
use.
A
I
move
to
other
container
runtimes
than
docker,
so
for
me,
docker
is
kind
of
a
vendor.
I
actually
just
start
the
docker
engine
for
this
project
here,
but
that's
really
my
personal
problem,
the
other
one
for
octato
is
basically
just
to
you
know,
have
people
something
out
of
their
local
box.
You
know
which,
where
they
can
run
stuff
or
try
out
stuff,
so
maybe
some
things
which
use
a
little
bit
more
resources
or
have
have
a
collector,
endpoint
or
back
end
running
all
the
time
stuff
like
this
and.
D
A
What's
the
word,
I'm
looking
for
service
mesh,
so
in
service
mesh
is
kind
of
really
hard
to
simulate
with
juggle
compose,
and
but
it's
often
you
know,
this
is
actually
the
use
case
for
open
tele-energy,
but
because
it's
a
cncf
project,
so
at
the
end
it
should
work
in
kubernetes
right.
A
C
I
also
I
meant
to
mention-
I
also
asked
about
this
with
the
other
open
telemetry
maintainers,
to
see
if
any
of
the
other
ones
had
considered
using
it
or
if
there
are
viable
alternatives
or
whatever
so
I'll,
be
curious
to
see
what
they
say.
I
haven't
heard
back.
I
asked
a
couple
minutes
ago.
I
haven't
heard
back
yet,
but
I'll
be
interested
to
see
what
they
say.
A
Yeah,
so
it's
basically
just
an
option
for
people
to
use
which
should
be
hassle
free.
You
know
they
just
run
this
up
and
it
should
work
and
if
they
want
to
look
into
how
kubernetes
works
or
these
things,
then
they
can.
You
know,
inspect
this
stuff,
but
there
should
be
no
need
for
any
knowledge.
You
know.
So
I
will
take
care
of
this
and
make
it
very
simple
and
not
not
over
engineering
too,
over
engineered
so
and
yeah.
It's
basically
just
an
option
for
people
to
use
nothing
which
should
be
used.
C
That's
very,
very
true,
all
right.
The
next
thing
we
have
on
our
agenda
was
the
php
8.1
compatibility
pr
number
489,
I'm
posting
in
the
chat.
Now
this
looks
like
it's
more
or
less
outside
of
our
control
now,
but.
A
It's
not
really
so.
There
is
one
thing,
this
ffi
context
which
is
causing
the
tests
to
crash
after
51.
So-
and
this
is,
do
you
know
actually
everybody
what
this
is
doing.
C
A
C
A
A
To
then
allow
context
switching
into
fiber,
so
fiber
is
basically
like
a
little
bit
like
a
core
routine.
If
you
know
what
this
is,
so
it
has
a
different
execution
context,
but
this
is
solves
actually
and
really
in
edge
case
so
context,
switching
in
inside
of
a
file,
that's
what
it
does
and
it
does
only
it
works
only
if
you
have
ffi
installed,
which
is
the
case
for
my
local
php
or
I
don't
know.
A
I
think
it's
not
in
our
development
images
and
it
only
works
for
ff
for
php
8.1
or
you.
If
you
have
the
extension
not
styled,
which
only
works
for
php,
eight,
zero
and
yeah.
The
thing
is
it:
doesn't?
This
error
doesn't
occur
when
you
execute
it
normally,
but
it
does
occur
in
the
test
because
there's
something
failing.
A
If
you
look
at
the
code,
then,
if
you
pre
it's
a
little
bit
hard
to
explain
if
you
pre-load
the
stuff
which
is
used
by
ffi,
then
the
try
catch
part
there.
It
doesn't
occur
so
and
basically
it's
trying
to
assign
a
namespace
to
an
object
and
this
so
the
object
is
not
serializable
and
it's
kind
of
a
little
bit
messy
at
this
point.
But
I
couldn't
work
out
a
way
how
to
fix
this.
Without
doing
you
know
rendering
all
the
stuff
avoid.
A
So
I
don't
know
about
this,
so
it's
a
little
bit
kind
of
a
hacky
approach
to
solving
something
which
is
really
like
an
edge
case.
So
my
idea,
maybe
we
move
this
to
some
experimental
or
make
it
some
additional
thing
or
turn
this
off
by
default,
and
people
have
to
turn
on.
D
A
C
A
Well,
I
tried
a
lot
played
around
with
this
code
and
what
it
actually
does,
because
I
didn't
look
too
much
into
it
before,
but
now
I
understand
basically
what
it
does.
A
A
C
That's
all
right.
C
A
Well,
I've
been
trying
to
fix
this,
I
for
the
tests
and
to
work
and
generally
so
I
didn't.
I
had
not
much
success
yet,
but
there
might
be
a
way,
but
yeah
we
still
would
be
good.
If
you
guys
take
a
look
into
it
and
then
you
can
maybe
think
if
we
want
to
support
this,
you
know
per
the
default
or
as
a
default
and
what
to
do
actually
because
yeah
at
the
end
of
the
day,
it's
an
edge
case
and
we
put
as
well
say,
don't
use
context
switches
in
fibers.
So.
A
Yeah
just
take
a
look
into
what
it's
actually
doing
then
maybe
have
a
better
idea.
If
you
want
to
support
this
and
generally
ffi
is
kind
of
it's
a
kind
of
a
nice
thing
but
yeah,
it's
still
marked
the
extension
itself.
It's
still
marked
as
experimental
and
if
you
take
a
look
at
the
php
site,
so
it's
very,
very
big
disclaimers,
so
don't
use
in
production
stuff
like
this,
and
I
don't
don't
know
if
you
want
to
ship
something
like
this
as
part
of
our.
B
A
Yeah
we
can
so
I
don't
want
to
get
rid
of
it,
so
we
could
still
move
it
into
a
separate
package
or
something
like
this.
You
know,
and
if
people
actually
really
really
need
this
solution,
then
we
would
have
it
but
yeah
just
take
a
little
look
guys
and
then
maybe
have
a
better
idea
what
it
does.
A
I'm
working
on
a
workaround,
so
I
well,
we
could
easily
just
get
rid
of
it,
but
I
don't
want
to
be
this
untested.
You
know
there
and
yeah
give
me
a
few
add
some
other
things
I
was
working.
I
was
just
fiddling
around
a
little
bit
with
this,
so
give
me
a
few
days.
Then
we
have
a
solution.
Now
we
can
proceed
with
this
without
removing
everything
right
now.
A
C
Next
on
this
was
your
your
note
about
the
issue
about
missing
retries
on
exporter
errors.
A
It's
basically
there
to
take
care
of
retry
login,
and
it
does
I
tried
so
if,
if
there's
nothing,
it
can
push
to,
it
will
try
again
try
again
and
then
at
some
point
complain.
So
why
do
we
have
to
implement
retry
logic
on
our
side
then?
Or
yes,
and
did
you
read
my
comments
about
this
is
really
hard
to
implement
to
make
it
to
make
everything?
A
You
know
everybody
happy
that,
so
we
simply
cannot
turn
it
on
by
default,
because
you
know
apache
will
probably
crash
at
this
sometime
if
people
are
still
using
catches.
C
C
A
The
other
solution,
though
I
mentioned
that
do
you
know
how
zero
and
q
works,
which
I
mentioned
on
the
so
it's
basically,
we
could,
you
know
somewhere
in
the
future,
offer
like
a
ex
like
a
package
or
something
crunchy
which
actually
uses
euromq,
and
this
is
a
99.9
failsafe,
because
it's
just
buffers
the
messages
you
know
and
if
there's
no
endpoint
or
no
peer,
then
it
will
just
wait
until
it
can
push
the
message
and
we
could
you
know
if
somebody
is
interested
in
implementing
like
a
little
demon
which
just
you
know
where
people
can
offload
something
like
this.
A
We
can
still
offer
this
in
the
future,
but
for
now
I
think,
if
we
want
to
implement
retail
logic,
then
yeah
should
be
configurable
and
turned
off.
A
Network
issue:
you
know
the
network
will
fail,
and
this
is
zero
and
q
is
safe
against
this,
because
it
will
buffer
the
locally.
The
message
you
know
and
yeah
so
use
rabbit
mq
in
the
same
way.
C
I
agree
with
you
I
I
couldn't
agree
with
you
more.
I
guess
I
should
say
that,
like
this
shouldn't,
this
should
not
be
at
the
top
of
our
purview
right
now,
because
the
collector
does
this
very
elegantly,
and
that
should
be
the
happy
path
for
implementation
for
our
customers
of
the
open,
telemetry
php
library.
So,
from
my
perspective,
this
seems
like
a
slam
dunk
to
me,
but.
A
A
C
Should
I
was
going
to
think
about
that
a
lot
we
should
probably
figure
out
a
way
in
our
issues
to
rank
them
on
like
a
scale
of
1
to
10
of
important
and
urgent,
because
there
are
a
lot
of
things
like
this
that
we
talk
about
and
they're
like.
Oh
maybe
this
is
like
a
one
out
of
ten
an
important
scale
or
when
I
attend
an
urgent
scale,
but
that's
that's,
maybe
something
for
another
day.
A
Yeah,
that's
another
thing.
It
was
last
week
we
talked
about.
You
know
this
discussion
board
and
basically
I
think
it's
a
good
idea
of
opening
this,
because
there's
some
things
which
are
just
ideas,
you
know
and
being
them
as
tickets.
As
I
see
tickets
are
something
you
can
solve.
You
know
like,
like
a
user
story
which
you
get
from
a
product
owner
or
a
use
case
where
you
have
a
at
least
more
or
less,
can
figure
out
when
something
is
solved.
A
A
Could
have
an
idea
actually
looked
into
this,
creating
like
a
extension,
you
know
if
you
have
a
if
you
know
how
saphir
works,
it's
actually
not
so
super
complicated,
but
that's
not
something
important
and
you
know
that
people
come
to
could
look
into
it,
get
have
their
own
ideas
and
babies
with
somebody
who's
really
interested
in
implementing.
They
could
pick
up
the
idea
and
you
know,
implement
it,
but
that's
not
worth
something
well,
not
unpolished
ideas.
You
know,
maybe
are
better
suited
to
be
like
in
this
discussion
board
or
something
like
this.
A
Yes,
at
least
at
least
there's
the
button
so
and
the
same
button
that
works
on
my
private,
where
I
tried
it,
let's
see.
C
A
A
If
this
depends
on
the
country,
so
I'm
not
sure
it's
difficult
to
talk
with
people
from
different
countries
and
you're
speaking
countries.
Sometimes
you'll,
probably
know
better
than
me.
C
A
Yeah,
let's
I
have
a
different
topic
down
the
line,
which
is
good,
splits
and
package.
So
let's
just
discuss
unless
there's
something
else
well,
this
would
be
my
topic
for
the
secrets
you
know,
but
unless
somebody
else
has.
A
A
A
Look
on
the
on
the
packages
website.
I
think
this
is
the
link
which
brad
posted
on
the
topic
respective
issue.
So
then
we
can
just
follow.
C
A
B
In
theory,
we
can
yeah,
you
know,
with
a
secret
push
a
request
out
to
packages
at
the
end
of
our
ci
ro.
A
A
Yeah,
so
it's
not
super
complicated,
the
only
thing
we
have
to
well.
We
can
solve
this
with
the
packages.
You
know
when
we
host
them
on
this
other
repo
organization,
which
I
created
in
which
you
will
get
invites
for
this,
then,
if
everything
is
set
up,
I
don't
want
to
manage
this
on
my
own.
So
that's
not
the
my
intention
at
all.
It's
just
yeah
I
want
to
implement
stuff.
A
C
Cool
the
next
thing
on
the
list
was
out
of
office
for
the
rest
of
the
year
are,
do
any
of
you
all
have
vacation
set
up
over
the
next
two
weeks.
I
know
we
have
christmas
and
new
years
and
all
sorts
of
different
celebrations
or
whatever.
A
Oh
definitely,
I
don't
know
how
much
leaf
days
do
you
get
if
you
want
to
rough
normal,
so
we
get
up
to
30,
and
this
is
weekdays
not
including
weekends,
so
yeah
in
germany,
you
know,
especially
but
other
northern,
but
you
know
that's
why
people
think
and
germans
you
know-
are
good
at
working,
so
no,
we
just
want
to
get
stuff
done.
A
So
we
can
have
our
free
time,
because
free
time
is
more
important
for
us,
your
english
people,
your
english
speaker,
you
don't
even
have
a
word
for
the
time
after
work
and
the
german
word
is
firearm,
which
basically
means
party
evening.
So
free
free
time
is
very
important
for
what
party
it's
it's
tough
to
translate.
Often
english
is
the
very
ambiguous
language
compared
to
german.
So
sometimes
it's
really
hard
to
translate
something.
A
English
and
you
know
it
doesn't
really
like
party-
doesn't
mean
like
party
like
you
know
the
golden
club
or
something
like
this,
but
more
to,
but
basically
that's
yeah
our
word
for
the
time
after
work.
So.
A
Yeah,
it's
not
actually
it's
a.
I
think
that
that's
how
it
used
to
in
english-
that's
more
like
usually
used
like
how
can
I
explain
that
if
a
bully
gets
sorrow,
you
know
somebody
who's,
usually
creating
sorrow
for
other
people.
So
that's.
If
you
see
my
comment
on
the
logging
stuff,
that's
basically
when
I
saw
this
log4j
problems
because
I've
I
have
so
many
silly
remarks
over
the
years
from
java
developments
against
php.
You
know
so
many
unqualified
ones
so,
and
I
thought
so
yeah
at
least
that
wouldn't
hurt
me
php's.
C
Well,
and
the
reason
I'm
mentioning
this-
I
am
friday-
is
my
last
day
of
work
for
the
year.
So
I'm
technically
not
working
all
the
last
two
weeks
of
december
and
my
availability
is
going
to
be
kind
of
intermittent.
I'm
not
yet
certain
whether
or
not
I'm
going
to
be
able
to
show
up
to
the
sig
meetings
both
times,
because
I
have
family
obligations
and
we
have
lots
of
travel
and
celebration
and
other
stuff.
So
we
can
still
hold
the
sig
meeting.
But
I
can't
promise
that
I'll
be
there.
A
C
A
C
I'm
still
going
to
be
checking
pr's
and
doing
doing
all
sorts.
I
might
have
more
time
for
open
telemetry
php
than
I
normally
do,
because
of
not
having
to
work
and
probably
getting
bored
with
my
family,
but
I
just
wanted
to
put
out
that
standard
disclaimer
that,
like
I'm,
not
gonna,
be
necessarily
super
available.
All
the
time.
A
B
A
That's
fine,
we'll
see
what
happens
so
20
yeah,
I
mean
do
you.
Where
do
you?
Actually?
When
do
you
celebrate
christmas
in
australia,
which
date
still
the
20th
year?
25Th?
Yes?
Well,
that's
not
the
case
for
a
lot
of
other
countries.
So,
for
example,
you
know
we
said
we
celebrate
on
on
christmas
eve.
So
a
lot
of
like
I
tell
italy
as
well,
so
we
get
we
get
our
apostles
or
gifts
one
day
earlier.
C
And
then
our
our
neighbors
to
the
north,
the
canadians
celebrate
boxing
day
on
the
26th,
which
is
we
celebrate
that
one
too.
Oh.
A
A
I
can
tell
you
because
of
british
family
or
in-laws,
so
it's
basically,
I
don't
know
well,
I
cannot
really
explain
it.
I
just
know
that
they
celebrate
it's
just,
but
for
us
in
germany,
it's
like
the
same.
We
actually
celebrate
all
the
three
days.
You
know
one
first
with
the
close
family
and
then
you
can
probably
go
to
the
larger
family
of
one
side
or
the
other
side.
So
we
are
busy
all
three
days
usually.
C
A
C
A
Yeah
so
the
whole
glitz
glitz
glitz
bit.
No,
no,
it's
good!
That's
very
nice!
How
do
you
pronounce
freud
in
english
food
for
sigmund
freud
right,
sigmund
freud?
This
actually
is
women.
Well,
I
looked
a
lot
of
a
lot
of
solutions
how
to
solve
this
and
also
solve
other
problems,
because
I'm
automation,
nerds.
You
know
we
have
in
the
future
so
and
it's
fought
a
little
bit
into
the
stuff.
We
know
need
those
web
hooks
and
to
actually
synchronize
different
repos
with
with
our
main
ones.
A
So
we
have
to
actually
tell
another
repo
of
their
updates
so
and
I
looked
at
a
lot
of
different
implementations
and
the
one
which
is
the
easiest
to
set
up
which
is
like
this
get
split
as
a
service
which
would
require
a
lot
of
bureaucracy
from
open
telemetry,
because
we
have
to
ask
again
for
this
bot
to
work
there
and
then
it's
it's
easy
to
set
up,
but
it's
not
really,
there's
no
api
or
something
where
you
can
actually
change
something.
A
I
have
to
go
into
there
and
do
it
manually,
it's
easy,
but
so
it's
not
really
feasible
for
our
use
case,
and
I
love
looked
into
two
others
which
a
little
bit
more,
but
I
think
which
the
thing
we
will
probably
end
up
with
is
like
basically
mirroring
the
whole
repositories
and
then
having
you
know,
then
just
synchronizing
this
with
one
web
hook,
that's
rather
easy
than
to
set
up
everything
in
the
main
repository,
and
then
you
know
what
I
mean
so
I've
created
in
this
other
there's
two
clones
and
then,
if
we
just
synchronize
the
our
main
repositories-
or
you
know,
update
the
the
clones
each
time,
something
is
merged
into
the
main
ones.
A
Then
we
can
easily
take
care
at
this
point
there
how
to
do
all
the
stuff
around
this
and
some
you
know,
introduce
some
automation
and
things
like
this,
then
to
when
we
have
to
deal
with
open,
telemetry
bureaucracy.
So
which
is
you
know?
I
don't
want
to
say
this
is
a
bad
thing,
but
everything
has
a
process
cons.
You
know,
I
understand
why
it's
implement
that
way,
but
for
all
use
cases
it
makes
it.
Of
course,
a
little
bit
more
difficult.
C
A
Yeah
so
yeah,
I
will
probably
in
the
next
days
I'm
very
close
to
having
like
a
one
solution
which
I
want
to
try,
and
the
first
thing
I
will
probably
do
is
just
set
it
up
that
I
have
to
manually
synchronize,
like
you
do
with
on
github.
With
this
button
you
know
and
see
how
it
works
and
then,
because
for
this
one
to
work,
then
we
have
to
decide
anybody
has
to
give.
A
One
of
us
has
to
offer
his
token
for
this
team
to
work,
and
I
don't
know
if
we
yeah,
which
one,
if
you
trust
me
to
do
nothing
harmful,
because
this
token
has
to
be
as
repository
access
and
all
the
stuff.
You
could
potentially
do
bad
things
for
them,
so
maybe
we're
gonna
be
read
early
access
to
the
open.
A
Well,
no,
because
you
have
to
push
yeah
well,
I
will
try
to
figure
out
this
if
we
can
actually
work
around
this
to
have
at
least
privileges
as
possible
and
so
yeah,
I'm
not.
I've
played
around
with
a
lot
of
different
solutions
and
ideas.
So
I'm
not
really
settled
and
I
have
to
actually
have
this
test
of
organization
where
we
have
to
set
up
for
different
solutions,
and
I
would
try
to
which
actually
works
and
stuff
like
this.
A
So
it's
not
yeah,
as
I
said
before,
it's
not
easy
to
make
things
easier
sometimes,
and
I
want
this
to
be
easy
so
far,
so
the
goal
is
to
once
this
is
actually
in
place
to
be
as
hassle
free
for
us
as
possible.
You
know
so
at
the
end
of
the
day,
even
if
there's
a
new
package,
it
will
maybe
create
the
repository
for
us,
so
we
just
have
to
so.
A
A
We
have
those
benchmarks
at
the
moment,
but
we
have
no
really
not
really
a
place
to
run
them
and
we
could
run
them
in
github
actions,
but
that's
not
reliable
at
all,
because
we
don't
know
what
container
we
don't
know
nothing
about
what
it's
running
into
and
it
could
be.
You
know
sometimes
it's
very
slow.
If
you
look
at
the
build
times,
you
know
of
the
images,
it's
very
clear
that
those.
A
Yes,
so
that
it
doesn't
have
always
the
same
performance
so,
but
so
I
was
thinking
of
a
solution
for
this
and
there's
a
very
nice
solution
for
us,
because
we
can
just
ask
people
for
stuff
for
free
and
I
will
post
you
the
link
and
what
I
would
do
is
since
we
are
cncf
project,
our
subs
up
project.
We
can
ask
for
resources
from
the
community
infrastructure
lab,
which
is
in
which
has
equine
equinix
metal,
and
we
can
just
ask
for
one
or
two
or
three
servers
which
we
could
use.
A
So
we
could
use
the
thing
which
I
was
thinking
about.
Then,
if
we
have
the
service,
so
maybe
we
or
I
can
create
a
little
application
which
meets
our
own
dog
food
which
helps
us.
You
know
solving
some
of
our
problems
and
uses
a
telemetry
itself,
so
we
can
have
like
enter
enters
and
you
know
a
solution
for
our
problems
and
a
test
in
the
same
way.
You
know
what
I
mean,
so
we
could
run
some
because
at
the
moment
we
are
also
the
problem
that
we
have
all
these
exporters.
A
But
we
are
not
really
testing
if
they
actually
enter
and
work.
You
know
they
might
comply
to
and
that's
why
we
have
this
bug
fix
a
few
days
ago,
where
actually
somebody
said,
oh
there's
actually
no
working
with
vega,
so
what
you're
doing
and
we
actually
never
tested.
So
we
could
tell
use
some.
A
You
know
some
end-to-end
tests
or
acceptance
tests
or
stuff
like
this,
and
it's
basically
free
I've
had
a
look
so
other
six.
So,
for
example,
for
java
and
they're
using
this
as
well
and
for
our
code,
that's
actually
not
really
hosted
on
the
on
the
open,
telemetry
project
itself.
So
if
you
have
nothing
against
it,
I
will
just
ask
for
for
some
I
will
try
to
be
greedy
and
ask
for
one
in
one
server
in
australia,
one
in
america
and
one
in
europe.
A
C
It
looks
like
it
looks
like
there
have
been
other
people
that
have
asked
for
open
telemetry
resources
in
that
cluster.
I
looked
at
some
of
the
other
issues,
so
I
think
that
shouldn't
that
sounds
great
to
me.
I'm
bothered
I.
A
Then
I
will
open
an
issue
there
and
sell
it
like.
I
have
a
lot
of
use
cases.
You
know
I
can
make
up
10
different
reasons
why
this
would
be
beneficial
and
so
okay,
then
I
will
just
go
for
it
and
we
can
see
how
it
works
out
and
we
have
something
we
can
actually
where
we,
maybe
even
we
can
house.
You
know
like
a
demo
website
or
stuff
like
this,
so
I
think
this
will
help
in
a
lot
of
places
if
we
just
have
some
server
where
we
can
host
stuff
independently.
C
A
All
agree
with
that,
then
I
will
try
to
make
it
happen,
and
then
I
will
invite
you
guys.
I've
looked
into
this,
so
I
will
just
give
them.
I
have
to
give
them
your
email,
addresses
and
stuff
like
this.
I've
looked
in
the
history
all
this.
How
people
asked
for
stuff
there
so
to
make
sure
they
will
say?
Yes.
C
A
A
C
Yeah,
I
think,
I'll
all
that's
on
my
list
of
things
to
do
I'll,
go
and
see
if
I
can
add
stuff
to
one
password
and
share
it
with
you.
A
A
This
was
actually
I
was
mentioning
this
when
we
talked
about
this
thing.
You
know
brett.
A
But
I
I
wasn't,
I
didn't
know
where
I
was
seeing
this
so
I
was
researching
a
little
bit
because
I
knew
I
saw
we
could
get
free
stuff.
So
I
just
had
to
find
it
again
so
and
it
seems
like
another
problem.
So
what
else
can
they
say?
And
then
and
if
they
say
yes,
then
it
would
be
very
stuff.
A
Okay,
then,
the
last
point
which
I
added
it's
actually
funny-
that
my
initials
got
turned
into
a
trademark
thing.
So.
A
If
you
oh,
no,
I
have
my
own
icons
on
slack,
so
tm
is
always
there
so
because
I'm
still
kind
of
fiddling
around
with
the
exporter
and
rework
or
refactoring,
and
so-
and
I
was
thinking
about
abstracting
this
as
much
as
possible.
So
you
know
so
we
have
a
transport
which
is
http,
and
then
we
have
the
serialization
which
may
be
grpc
and
all
this
different
concerns
basically
or
responsibilities,
and
to
make
this
you
know
best
maintainable.
A
And
reuse
codes
and
that's
very
difficult.
I
know
this
task
that
it's
even
more
difficult
than
having
different
database
endpoints,
because
there
are
more
different
requirements
for
all
the
stuff
than
if
you're
dealing
with
database.
So
it's
not
an
easy
task,
but
I
think.
A
For
well
I
create
a
pr
for
this
in
the
next
states
and
then
you
can
say
what
you
like,
what
you
don't
like
and
but
one
point
was
basically
abstracting
away
like
the
things
like
grpc
or
thrift
or
http.
So
this
is
just
like
one
kind
of
component
which
you
add
to
exporter,
and
then
it
can
work
with
the
soil.
No
matter
what
this
exponent
component
is
called,
so
the
we
can,
there
could
be
a
different
name
spot
so
that
we
can
solve
stuff
like
this.
You
know
and
abstracted
away
so
long
story.
A
So
to
look
into
this,
I
created
for
actually
for
zipkin,
I
used
the
grpc
code
and
to
I
worked
fiddled
around
with
create
a
grpc
client
or
just
to
see
how
it
is.
If
we
have
different
grpc
implementations,
you
know
for
code,
what's
what's
what
we
actually
need
to
make
this
happen
in
abstracted
ways,
so
maybe
I
I
will
probably
create
a
pr
for
a
grpc
and
exporter
then,
but
that's
that's
just
a
side
effect
that
was
just
you
know
having
something
to
test
with.
A
But
what
I
found
is
that
we
have
to
move
the
grpc
code
to
a
subdirection,
so
the
hotel,
because
the
zip
can
one
is
they
would
collide
a
little
bit
if
we
have
to
create
packages
out
of
this.
So
the
auto
code
have
to
go
just
basically
in
a
sub
package
and
that's
probably
with
grunette
is
working
on
the
thrift
stuff
at
the
moment.
A
So
we
have
to
probably
do
the
same
for
the
thrift
I
was
fiddling
around
with
first
after
herself,
but
I
don't
want
to
implement
anything
that
he's
working
on
so
yeah.
So
basically,
and
then
the
the
good
or
the
side
effect
is
that
we
can
use
this
generated
code
and
publishes
separate
packages
as
well.
A
I
will
take
care
of
this,
so
they
have
to
move
a
different
subdirectory
to
to
have
separation,
because
between
all
this
generated
code
and
one
additional
thing
I
have
because
I
do
for
this-
I
looked
a
lot
in
this
prototype
of
and
well
I
mean
it's
messy
by
made
by
nature,
but
one
of
my
ideas,
if
we
just
run
after
generation
just
run
the
coding
style
fixer
over-
maybe
some
stuff
which
doesn't
you
know,
change
behavior,
but
just
makes
it
a
little
bit
more
readable
because
it's
sometimes
yeah
it's
a
little
bit
hard
to
get
through
or
what
all
of
this
is
doing.
A
It's
really
hard
to
to
know
anyway,
but
do
you
think
this
will
create
any
harm?
Or
do
you
have
any
concerns?
A
Let's
put
this:
let's
put
this
on
hold
for
now.
Until
we
have,
we
have
the
servers
in
place
and
have
end-to-end
tests,
and
then
we
can
actually
test
if
we
break
something
about
all
this
about
this
yeah
yeah.
B
A
There
shouldn't
be
a
problem,
but
I'm
not
sure
because
this
and
okay,
let's
just
set
up
a
test
where
we
can
actually
see
if
it
shouldn't,
but
where
we
can
be
100
sure
that
it
doesn't
break
anything.
A
Okey
doke
anything
else:
oh
yeah
did
we
have
we
didn't
look
into
any
open,
pull
requests
or
something
like
this?
Did
we
anything
else.
C
A
Being
a
few
or
two
five,
whatever
you
like
yeah,
I
think
we
yeah,
then
maybe
you
can
tell
me
again
what
you
do
for
the
release
and
then
we
can
automatically
in
the
future,
and
one
thing
I
was
I
was,
you
probably
have
seen
I
added
some
more
labels
to
stuff
in
the
issues
so
to
make
it.
You
know.
A
And
prettier
one
thing
is:
I
was
stumbling
a
little
bit
on
the
description
about
help
wandered
because
it
says
extra
attention
is
needed.
C
They
all
need
extra,
I
don't
know,
I
think,
that's
more.
The
the
original
intent
of
that
label
was
people
like
how
do
I
say
this
the
right
way
that
if
you
are
it's,
it
was
sort
of
to
distinguish
between
and
a
good
for
new,
a
good
for
new
users
tag
and,
like
just
a
general
we
nobody's
working
on
this.
We
need
some
help
for
it,
but.
A
Yeah,
that's
usually
when
I
see
tech
like
this,
my
my
impression
is
this
is
not
well
not
somebody
asking
for
help
so
help
me
help
me,
but
you
know
somebody
could
you
know
we
we
want
to
have
help
implementing.
You
know
somebody
could
take
this.
A
So
so
I
was
stumbling
a
little
bit
on
the
description:
yeah,
okay,
but
that's
what's
the
intended
use,
because
that's
what
I
used
for
okay
right
and
then
I
added
something
a
label
as
well
for
refactoring,
so
just
to
point
out
that
something's
not
intended
to
change
behavior
and
just
making
things
prettier
faster,
better
maintainable
or
whatever.
A
Another
thing
is.
There
are
a
lot
of
these
old
user
research
fees,
feedback,
improvements,
issues
and
some,
I
think,
a
little
bit
outdated,
because
the.
A
Yeah
things
changed
over
the
time
and
one
I
closed,
because
it
was
really
a
little
bit
all
over
the
place
with
the
exceptions
and
stuff
like
this,
you
know
I
I
explained
to
I
don't
know:
what's
name
is
dekalavi
why
I
did
this,
but
this
was
basically
was
solved
in
a
lot
of
other
ways
and
we
were
wanting
to
move
away
from
this
and
it
was
a
little
bit
much
too
much
symphony
tunnel
focused.
You
know,
and
I
have
created
other
tickets
for
this
so
and
sometimes
yeah.
A
I
think
for
some
tickets
there,
actually,
the
yeah,
for
example
the
one
on
brady
actually
commented
on
this
around
short
sample.
We
could
probably
close
this.
I
think,
because
we
don't
expose
this.
The
problem
is
we
have
this
in
a
lot
of
examples
so
use
this
should
sample
method,
but
it's
actually
not
intended
to
be
used
anymore,
and
I
agree
totally
with
the
concerns
in
there
when
I
first
first
saw
this,
I
I
was
really
you
know
baffled
by.
A
I
didn't
know
what
to
do
as
well,
but
since
now
it's
actually
hidden
and
should
not
be
used
by
developers
or
people
using
the
library.
I
think
we
could
close
this
and
rather
change
the
examples
to
not
use
this
anymore.
Somebody
else
complained
about
it
as
well
as
on
about
me
using
this
in
the
symphony
demo
demo
application,
where
I
actually
used
it
by
intent,
because
right
yeah,
I
think
we
can,
for
example,
close
this
one,
no
yeah
yeah
we
could
and
yeah.
A
There
are
some
other
ones,
so
I'm
not
really
sure
always
some
some
issues
are
still
assigned
to
people
which
I
don't
know
if
they're
still
working
on
this,
so
you
know,
I
think
you
asked
you
did
like
a
around
asking
people
if
they
still
working
this
in
august,
but
yeah
since
I.
C
Think
it's
then,
I
think
it.
I
think
if
something
hasn't
been
worked
on
and
we
should
probably
determine
a
time
frame
if
something
hasn't
been
worked
on
x
amount
of
months,
we
can
undersign
it
from
somebody
yeah.
So,
like
you
know
two
three
six
months,
I
don't
really
care
what
it
is.
I'm
just
yeah
me
neither.
A
A
There
should
not
be
too
much
administration
for
us
with
something
like
this
yeah.
So
we
can.
I
think
we
can
assume
when
people
didn't
work
on
this
for
quite
a
while
no
feedback
and
if
we
asked
them
there's
no,
no
answer
to
that
that
they're
not
working
on.
C
A
D
B
D
We
like
there
are
two
possible
implementations,
so
if
we
decide
to
go
ahead
with
either
one
so
now
we
have
a
logging
api
created
right.
So
after
that,
once
we
have
the
logging
api
created
now,
I
wanted
to
initially
work
on
the
issue
298,
which
is
to
create
automatic
error
handler.
So
what
would
be
the
next
steps
like?
First,
we
have
the
api
created
and
then
we
have
logging
instrumented
in
the
in
the
different
classes.
A
Oh,
this
is
the
self-diagnostic
okay.
Can
you
ask
your
question
in
the
story
I
was.
Can
you
ask
a
question
again,
please.
D
Yeah
sure
so,
first
we
have
the
link.
I
when
I
started
looking
into
this
issue,
I
was
directed
to
like
the
present
pr
about
the
logging
api
creation.
So
now
the
logging
api
is
created.
Suppose
it's
like
we
move
forward
with
that
and
we
instrument
the
logging
api
in
our
different
classes.
So
what
are
the
next
steps?
What
is
what
is
the
bridge
between
this
and
the
handler
like
what
does
handler
mean
in
context
of
hotel,
php.
B
It's
a
really
good
question.
I
suppose
the
first
first
thing
to
point
out,
though,
is
the
semantics
of
like
should
right
in
may,
so
we
should
log
the
error
and
then
may
to
me
means
like
it's.
It's
an
optional.
B
You
know
a
stretch
goal,
but
but
assuming
that
we
did
want
to
go
ahead
with
allowing
callbacks.
A
What
do
we
do?
Yeah?
I
guess
it's
not
really
100
defined,
because
that
may
be
very
language
programming
language
specific.
So
so
the
thing
which
we
could
offer
is
some
way
of
exposing
a
callback
in
addition
to
to
having
a
logger.
You
know
so
that
this
callback
is
called
them
so
event,
logging
events,
yeah,
I'm
a
big
fan,
yeah,
I'm
you're
triggering
me
right
now
there.
A
That
was
exactly
what
would
I
was
going
for
that
this
is
something
which
we
could
do,
but
yeah
some,
some
yeah
events
are
probably
the
best
so
that
we
can
couldn't
have
downstream
projects
or
hooking
just
into
the
system.
And
then
we
don't
have
too
hard
to
implement
anything
any
solution
around
us.
So
events
are
actually
the
best
best
way
to
do
this,
because
we
can
really
decouple
from
implementations.
A
You
know
and
but
yeah,
basically
offering
something
like
that
that
there's
a
like
an
event.
I
don't
know
if
there's
a
event,
dispatcher,
that's
a
psr
specification.
If.
A
Which
could
be
used
and
then
people
could
just
hook
into
this
or
yeah
anything
around
this,
but
this
to
cut
the
long
story
short.
There
is
no
real
from,
I
think
from
the
specification
itself.
There's
no
real.
This
is
not
really
specified,
so
we
can
implement
what
we
want,
or
you
can
just
you
know,
create
something
they
can
also
like
use
you've.
Seen
with
what
brett
has
done,
you
can
also
create
two
different
implementations
of
three
and
then
everywhere.
A
If
you
want
to
have
feedback
or
something
you
know,
you
don't
have
to
come
up
with
the
best
solution
right
so
in
this
project.
Nobody
can
because
it's
very,
very
complex
and
you
have
to
think
of
a
lot
of
things
and
which
somebody
maybe
ignores
or
doesn't
think
about,
but
the
other
person
has
more
in
the
view,
so
there's
no
no
best
solution
for
this,
like
mostly
in
programming,
but
in
the
specific
case
we
can
just
see
and
then
maybe
which
works
out
and
with
other
parts
and
stuff
like
this.
D
B
B
A
Yeah
something
like
this
so,
for
example,
do
you
know
how
symphony
works
come
on.
D
A
I
don't
know
I
think
laravel
doesn't
have
like
an
event
system,
so
it
doesn't
matter.
It
was
just
an
example,
but
in
symphony
a
lot
of
things
are
handled
with.
Well,
I
think
it's
kind
of
like
because
all
frameworks
for
laravel
itself,
I
use
the
this.
A
You
could
saw
it
call
it
life
cycle
events,
you
know
in
the
middleware
and
simply
uses
a
little
bit
more
so
can
hook.
There
are
events
created
which
you
can
simply
hook
into,
and
so,
but
I
can
actually
I
can
post
you
an
example
if
you're
interested
so
certain
things
we
you
could
look
into
more
to
solving
this.
If
you
want
to
so
I
don't
so
it's
not
a
must.
You
don't
have
to
follow
my
advice,
it's
simply
an
offer
because.
A
Yeah,
I
can
give
you
a
lot
of
because
I've
dealt
with
I'm
one
of
the
php
event
pioneers,
and
I
can
prove
this
if
you
look
into
my
github
repository,
so
I've
dealt
with
how
events
event
stuff
in
php
is
a
long
long,
long
time
so
I've.
A
B
Yeah
just
to
talk
about
498
really
quickly.
That
was
that
how
to
do
config,
which
got
very.
B
Given
that
there's
a
there's
now
a
big
issue
on
open
telemetry
for
how
to
do
sdk
config,
I
wonder
whether
this
should
just
go
back
on
hold
and
whether
that
well-defined
config,
if
they
can
come
up
with
it,
might
just
solve
this
problem
for
us,
particularly
if
there's
a
prototype
for
it,
and
we
auto-generate
the
code
like
it
might
just.
A
Well,
yeah,
that's
a
little
bit
of
a
problem.
I
mean
they.
They
said
it's
planned
for
next
year,
right
so
yeah,
but
yes,
next
year
could
be
12
months
away,
and
so
we
don't
know
where
this
actually
will
land
but
yeah.
The
problem
is
a
little
bit
that
then
we
have
to
change
all
the
stuff
again.
It's
since
we're
not
that
many
people,
and
maybe
more
people
next
year,
who
knows,
but
we
cannot
just
plan
with
having
a
lot
of
people
and
changing
things
around
so
might
be.
A
We
put
this
on
hold.
There's
still
I
mean
yeah,
there's
other
things,
the
the
the
other
pointers
that
we
have
to
when
we
change
the
configuration.
I
have
to
change
the
configuration
of
the
symphony
bundle
as
well,
which
I
have
no
problem
with,
but
I
don't
want
to
change
it.
Two
or
three
times.
D
B
A
Yeah,
so
I
happily
follow
follow
the
sdk
with
this
and
because
I
can
basically
get
rid
of
a
lot
of
code
on
there
which
handles
all
the
country
refrigeration
stuff
in
there
and
yeah.
But
I
won't
don't
want
to
do
this
like
more
than
once,
or
at
least
not
for
changing
everything.
So.
A
I
may
think
we
can
still
fiddle
around,
or
maybe
we
open
one.
You
know
in
the
discussion
board
like
an
idea
thing
where
we
can
just
pull
stuff.
If
you
think
oh
I've
seen
this
or
maybe
this
is
a
good
idea
and
then
maybe
it
will
grow
over
time
when
so
we
come
up
with
the
even
better
solution
or
you
know,
sometimes
I'm
a
friend
a
friend
of
when
I
have
time
you
know
not
on
work
this.
I
don't.
A
You
can
often
not
do
something
like
this,
because
you
have
to
implement
features
and
that's
some
something.
For
example,
I
hate
about
scrum
that
it's
very
creates
often
a
lot
of
of
problems,
but
I
find
from
creative,
complicated
things.
A
It's
often,
you
know
put
it
away
for
some
time
and
then
the
brain
or
the
subconscious
comes
up
with
better
ideas
or
you
you
remember
something
which
oh
you
could
actually
do
this,
and
this
this
will
solve
all
the
stuff,
so
maybe
by
next
year
beginning
we
could
tackle
this
again
and
then
maybe
we
have
better
ideas
or
a
very
clearer
path.
How
we
actually
can
solve
this.
C
Sounds
good
to
me
great
now
you
are
now
you're
our
leading
understander
of
conte
of
configuration,
bread
so.
D
A
And
still,
I'm
not
really
satisfied
with
what
I
did
there.
I
just
wanted
to
get
this.
You
know
done
so,
but
I
mean
this
is
configurations.
Always
is
it's
complicated
because
you
have
to
especially
for
a
use
case,
which
we
will,
because
you
have
to
think
of
a
lot
of
things.
You
know
what
people
could
want
to
configure.
What
is
too
much
configuration?
What
is
the
little.
B
B
Over
to
the
issue
on
specification
and
just
see
if
I
can
contribute
anything.
A
Yeah
actually
make
them
work
like
who
you
want
we'll
make
this
specification
yeah
get
there
first
for
once,
yeah
yeah,
actually
a
nice
idea.
If
you
want
to
do
this,
would
be
nice
if
we
could,
because
sometimes
I
think,
a
lot
of
parts
of
the
specification
well,
first
and
foremost,
it's
very
we
can
clearly
see
that's
mostly
written
by
english
natives,
because
it's
something
that's
really
hard
to
understand.
A
If
and
my
english
is
not
that
bad,
it's
not
me
fishing
for
a
compliment,
but
sometimes
I
have
to
read
things
three
or
four
times
and,
and
I
have
to
say
english
is
not
the
best
language
to
write,
specifications
and
then
german's
way
better
suited
because
it's
way
less
ambiguous
than
english.
So
it's
sometimes
a
little
bit
difficult,
but
also,
I
think,
sometimes
a
little
bit
driven
by
java
people
or
in
gold.
So
a
lot
of
things,
solutions
that
are
very
you
know,
because
the
way.
D
A
C
They're,
ambiguous
they're,
ambiguous
on
purpose
because
they
know
that
things
might
not
fit
into
every
language,
but
it's
I
agree
with
you.
I've
been
playing
that
game
since
we
started.
I
started
working
on
this
project.
It's
you
play
specification,
catch
up
with
languages
that
have
other
idioms
built
into.
A
Yeah,
but
you
could
still
you
know,
we
could
implement
this
in
a
really
different
way.
If
you
notice,
for
example,
our
messaging,
so
enterprise
integration
patterns
happens.
So
that's
where
you
actually
the
problem.
The
sdk
is
all
it's
designed
like
an
application.
You
know
where
you
have
different
tiers,
which
hook
into
each
other
and
they
can
kind
of
coupled
and
but
basically
we're
dealing
with
messages.
You
know
if
you
just
have
like
a
flow
specification.
You
know
where
things
you
know
how
red
node
red
works
so
like
float.
A
Yeah,
because
actually
I,
like
I
love
node
red,
so
you
could
solve
this
so
much
easier.
Things
like
this
was
something
like
flow
programming,
because
you
don't
have
to
go
through
all
these
tears
and
stuffs
and
you
can
just
isolate
use
cases.
You
know
and
then
have
a
messy
pass
messages
around.
So
basically
just
that's
how
go
works
but
it's
or
functional
programming
in
general,
but
yeah.
That's
that's
not
really
how
the
spec
is
written,
because
I
think
it
has
to
work
in
java
as
well.
So
that's
the
main
things.
A
B
A
Yeah,
I
I
had
to
learn
how
to
be
political,
correct
when
dealing
with,
especially
americans.
I
have
to
say:
well,
we,
our
german
political
pregnancies
are
a
little
bit
different.
It's
not
so
much
about
semantics,
and
so
I
don't
even
understand
why
you
guys
write
him.
His.
C
Oh
yeah,
I
short,
I
can
shortly
explain
we
have
he
delivering
your
pronouns
to
others,
makes
people
feel
comfortable
in
context
where
you
don't
use
normal
pronouns.
I
have
co-workers
that
use
they
them
as
their
pronouns.
So
I
wouldn't
say
I
wouldn't
say:
brett
brett
he's
a
great
guy,
I
would
say
brett
they
are
a
great
guy
or
a
great
person
or
whatever
it's
it's
for
gender
ambiguity.
It's
used
for
gender
ambiguity
most
of
the
time.
So
there
are
certain
people
that.
D
C
A
C
Necessarily
because
some
people
like
to
it's
again,
it's
it's
very,
it's
very
person
dependent,
but
often
they
you
can
they.
Somebody
will
decide
that
they
want
to
use
she
they
or
she.
You
know
like
in
singular.
A
A
You
have
to
see
from
our
point:
you
don't
even
have
the
problems
that
we
do
have
in
german,
because
you
use
use
all
this,
the
generic
masculine,
so
you
know
there's
a
developer,
you
don't
have
a
developers,
you
don't
have
female
versions
of
names
so
which
we
every
every
word
in
german,
has
a
female,
not
a
chair,
but
you
know
something
like
a
role.
You
know,
there's
a
word
for
female
developer,
there's
a
word
for
for
male
developer
and
stuff
like
this.
A
So
this
this
would
be
just
luxury
issues
for
us,
and
people
would
fight
at
the
end,
so
so
we're
trying
to
find
more
more
like
a
relaxed
yeah.
Well,
so
mostly
people
try
to
have
more
like
a
relaxed
approach
here,
because
at
the
end
of
the
day
you
cannot
have.
You
cannot
make
everybody
happy.
You
know
what
I
mean.
A
This
is
a
problem
a
little
bit
with
political
correctness,
and
so,
for
example,
I
have
people,
you
know
who
are
from
minorities
and
they
want
me
to
use
certain
ways
which
political
correct,
which
you
know
the
words
may
start
with
n
or
something
like
this,
because
that's
what
they
want
to
be
and
which
is
not
the
end
which
we
use
in
america.
You
know
you
have
to,
but
and
then
people
who
don't
know
anybody
of
this
minority
time.
You
cannot
tell
this
because
they
feel
uncomf
comfortable.
A
You
know,
hearing
these
words
and
but
the
people,
the
people
actually
who
are
from
this
minority,
want
me
to
use
vino
because
they
and
they
take
these
words
and
own
it.
You
know
you
cannot
insult
me
with
this
word
because
we
own
it
so
try
to,
and
I
feel
often
it's
a
little
bit
like
people
having
too
much
buttons
for
others
to
press.
You
know
to
feel
uncomfortable.
A
We
rather
it
especially
in
berlin.
Our
style
is
more,
you
cannot
try,
we
don't.
I
give
you
buttons
to
press
to
make
us
uncomfortable,
you
know
and
when
it's
a
choice,
so
I
we
don't
feel
offended
easily
because
that's
just
feeling
offended
it's
not
something.
Nice.
A
Yeah,
I've
noticed
this
and
it's
really
for
me.
I've
discussed
this
with
a
lot
of
americans
and
often
they
don't
understand
my
point
of
view,
but
it's
sometimes
really
weird
to
look
for
to
see
from
the
outside,
because
it's
very
semantic
driven
and
very
while
on
the
one
side
it
tries
to
be
open,
it's
still
very
centered
on
english
language
or
the
usa.
C
Yeah,
it's
a
different
and
then
yeah
not
to
get
not
to
go
too
deep
in
the
weeds
people,
get
stuck
on
political
and
social,
political
and
social
issues
and
classify
certain
types
of
people
using
a
using
like
their
political
affiliation,
with,
like
other
things,.
A
Yeah,
it
just
gets
that's
the
worst,
that's
the
worst
thing.
I've
seen
with
you
know,
dealing
with
americans
that
people
get
too
politicized
because
they
you
just
have
two
options
in
the
usa.
You
know
this
doesn't
happen
too
much
here,
because
you
don't
call
anybody.
A
liberal,
first
of
all
liberal
means
something
very
different
in
most
parts
of
the
world.
It's
actually
trump
would
be
more
liberal.
You
know
than
actually
the
democrats,
but
we
don't
have
this.
You
know
there's
two.
A
C
C
You
have
to
send
that
to
me
in
in
slack,
because
I
want
to
see
how
to
read
it.
Yeah.
A
Oh
one
thing
I
want
to
ask:
how
do
you
pronounce
your
surname
in
english.
C
A
C
There's
a
there's:
a
guy
named
straw,
the
nikansky
strikanski.
C
Is
a
polar
it's
in
in
polish,
it's
actually
with
an
eye
most
of
the
time,
but
it
got
switched
to
one
move.
A
Yeah,
I
think
it's
on
the
firstness,
that
is,
that
can't
speak,
so
I'm
living
you
know,
100
kilometers
away
from
poland
and
a
lot
of
actually
there's
so
many
germans
with
polish
names
and
the
other
way
around.
So
it's
kind
of
next
to
all
the
things.
So
I
was
just
interested
because
I
it's
always
kind
of
sometimes
a
little
bit
weird,
how
you
know
european
names
are
pronounced
and
in
english.
So
that's
true.
A
No
problem:
everybody
have
a
nice
day
evening,
thanks
a
lot
good
job,
bye.