►
From YouTube: 2022-02-23 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).
B
A
C
Well,
I
don't
want
that
to
be
recorded
on
youtube.
No.
C
C
B
No
problem,
oh,
is
in
a
different
meeting.
No,
this.
A
The
box
lincoln
and
it
wouldn't
let
us
it
wouldn't
let
us
in
I
wasn't
authorized
to
start
the
meeting.
I
think
oh.
C
The
larger
the
one
before
right,
perhaps
it
was
oh
yes,
how
would
you
say
this?
The
second
last
one.
C
C
Yeah
we're
just
waiting
if
somebody
else
wants
to
show
up
in
a
few
minutes
and
then
we're
happy
to
start.
A
In
congratulations
on
getting
your
first.
I
think
that
was
your
first
merge
requesting
pull
request.
Yes,.
D
I
think
both
of
them
are
most.
I
guess.
B
Did
you
have
two?
There
was
the
resource
info
one
I,
how
much.
C
C
C
Sorry,
I
didn't
have
much
sleep
this
night,
so
I'm
a
little
bit
confused
in
our
brain
today.
I
try
to
not
make
it
too
obvious
yeah,
the
other
one
I
did
so
yeah.
Okay,
then
it's
five
five
past
half
past.
C
In
german,
we
have
different
ways
to
express
this,
which
I
don't
want
to
go
into
any
detail
into
it.
Okay,
then,
let's
just
start
by
having
a
look
at
what
actually
open
to
request
at
the
moment,
and
I
think
we
have
two
one
at
the
moment
in
the
main
repository.
C
The
first
one
is
from
mark
okay,
this
is
the
issue
with
just
as
few.
C
Yeah,
I
think
it's
just
gonna
go
stale
yeah,
it's
they're!
Rather
the
two
of
your
changes.
I
will
do
once
grown.
It
is
done
with
this
thrift,
exporter,
stuff
and
so
on.
I
will
anyway
go
over
refactor
all
the
exporter
stuff
so,
and
I
actually
wanted
to
make
this
all
configurable
and
stuff
like
this,
and
we
can
take
a
look
at
this.
That's
not
so
not
something
we
couldn't
reproduce
and
if
he,
if
he
wants
to
still
work
on
this
thing,
it's
fine
right.
C
Yes,
okay,
then
just
leave
that,
like
that
and
the
other
one
is
the
grunettes
one.
So
he
got
back
to
our
sunslick
today
and
said
he
was
a
little
busy.
That's
fine!
C
Let's
probably
educate
the
thing
he's
working
on
there
so
we'll
we
will
just
wait
less
time
to
work
on
this
and
we
can
resolve
any
issues
that
might
be
there
yeah
and
once,
however,
once
this
is
done,
then
we
can
take
a
look
or
I
I'm
still
have
some
work
done
on
this
refactoring
or
the
exporter
stuff,
make
it
a
little
bit
more
maintainable,
more
abstracted
so
that
we
have
actually
stuff
with
just
the
same
thing
and
dedicated
classes
and
stuff.
C
Like
this,
I'm
happy
to
work
on
this,
although
I
had
other
stuff
I
was
working
on,
so
it's
not
like.
I
was
bored
or
anything,
I'm
still
not.
So
I
still
have
a
lot
of
things
on
my
table.
C
Okey
doke,
there's
two
other
things
I
wanted
to
look
into
regarding
issues.
C
C
Let
me
have
a
look.
They
must
be
old.
Yes,
one
is
yeah,
they
both
think
are
coupled.
So
if
you
go
into
the
issues
on
the
last
page,
there's
actually
the
last
two
issues
which
are
open
at
the
moment,
one's
called
on
called
exception
and
child
span,
the
other
one
is
finish:
all
spam
features.
Yes,
do
you
find
this
so
otherwise.
D
C
Give
a
link-
one
of
them
actually
is
at
least
as
it
stands.
For
now,
it's
not
evil.
Even
we
cannot
even
implement
because
that's
not
how
it
works,
and
this
is
actually
the
uncaused
exceptions
and
child
spencer.
C
If
I
correctly
understand
and
sees
it
looks
like
we
should
close
bands
whenever
there's
an
exception,
but
there's
no
way
to
do
this
because
we
cannot,
you
know,
create
a
catch
block
on
all
the
client
code,
which
is
using
our
chords.
You
know
magically
you
can
close
stuff
like
this,
so
it's
just
technically
not
possible.
I
think,
but
I
have
sometimes
troubles
understanding
all
the
issues
which
are
from
this
old
user
research
feedback
thingy,
and
so
I
actually
wanted
to
ask
about
this.
A
C
One
and
a
half
years
old,
a
lot
of
those
tickets,
so
there
are
probably
this
is
something
we
could
may
do
if
we
have
auto
instrumentation
where
you
can,
you
know
then
create
a
catch
block
around
stuff.
That
is
auto
instrumented,
but
I
don't
think
well.
This
cannot
be
part
of
the
sdk
or
something
it's
just.
Not
it's
just
not
how
php
works
and
adopt
any
other
language
could
do
something
like
this
or
maybe
yeah.
Maybe
you
could
do
in
javascript,
where
you
can
patch.
You
know
core
stuff.
C
Some
trickeries
might
be
possible
there.
Okay,
then
the
other
one
I
wanted
to
look
into.
It's
kind
of
the
finished
all
span
features
yeah.
That
might
be
if
it's
not
related
to
of
the
catch
that
might
be
feasible
to
do,
but
our
we
will
just
wait
for
about
to
come
back
soon,
because
maybe
he
has
a
little
bit
more
inside
right.
A
Yeah,
I'm
just
reading
the
just
reading
that
issue.
Now
I
haven't
seen
that
one
before
so
far
back.
I
think
I've
never
paged
all
the
way
to
the
end.
C
Yeah,
it's
related,
but
this
one
well,
we
should
talk
about
if
this
actually
is
doable
with
respecting
the
specs
and
so
on,
but
I
think
you
know
finnish
all
spans
would
be
something
which
at
least
what
we
could
be
able
to
implement.
So
that's
not
like
magically
kept
catching
exceptions.
A
No,
no,
but
as
a
method,
potentially
yes,
maybe
even
like
a
flag
to
like
yeah
shut
down
functionality.
A
I
see
I
see
another
issue
there,
though,
which
is
locating
all
of
the
spans,
because
that
was
one
of
the
very
first
things
that
I
tried
to
do,
which
was
you
know
if
you've
got
a
stack
of
spans,
such
as
in
that
example,
spam.
One
call
spam
two
and
spam:
two
is
your
current
span?
How
do
you
get
back
programmatically
to
spam,
one
from
span
two,
and
I
don't
think
that
we
have.
C
C
Actually,
I
don't
think
I
would
take
into
maybe
a
look
into
some
other
languages
how
they
do
it,
though.
That's
not
always
the
best
solution,
because
php
or
all
languages
are
different.
So
but
yeah,
that's,
I
think,
that's
quite
well.
At
least
the
topic
I
had
in
my
mind
also
when
I
started
working
on
on
portal
stuff,
so
how
you
actually
would
go
about
something
like
that.
C
A
C
A
C
D
Any
more
regarding
this,
I
have
one
question
regarding
the
in
the
nested
in
the
ticket:
there's
a
nested
span
right.
Could
it
mean
that
span
two
is
a
children
of
span,
one.
D
So,
okay,
so
with
the
recent
code
changes
I
mean
from
0.0.5,
I
am
not
sure
how
to
create
a
child
span.
So
before
0.0.5
we
had
this
laravel
setup
demo.
I
think
a
laravel
demo
on
our
github
page
right
and
there
I
think
we
had.
We
had
in
the
laravel
application,
of
example,
of
creating
a
child
span,
but
I'm
not
sure
how
to.
C
Create
no
that's
totally
outdated
and
actually
we're
we're
discussing
something
regarding
this.
But
to
get
to
your
point,
that's
actually
quite
easy.
So
for
the
first
span
you
just
call
activate
on
the
spam
and
when
you
create
another
span
it
will
automatically
be
the
child.
C
A
C
Some
might
be
outdated,
but
that's
if
there
are
quite
some
old
ones.
So
there
is
actually
a
ticket
to
update
all
the
examples
because,
for
example,
the
sampling
decision
you
all
will
find
often
especially
also
in
the
in
the
symphony
and
laravel
examples,
but
also
in
the
that's
something
we
don't
need
anymore
and
so
don't
we
know
that
those
have
to
be
updated,
but
you
should
actually
find
this
activate
thingy
there.
C
Otherwise,
let
me
quickly
check
if
I
could.
You
quickly
find
an
example.
A
Right,
I
think
so.
I
already
got
the
code
open.
D
Yes,
sorry,
no,
no,
ask
your
customer!
So
no!
I
was
just
saying
that
the
laravel
application
was
a
good
one,
because
if
we
wanted
to
make
some
change
and
test
it
out
locally
right
that
laravel
application
would
be
good
to.
D
C
We
have
some
problems
about
the
solution
which
is
presented
there
as
as
well
this
one
as
the
symphony
one.
It's
not
really
integration,
because
it's
using
a
global
state-
and
this
is
something
we
really
don't
want
to
promote-
actually
to
for
ex
it's
a
little
bit
difficult
so
from
the
writing
and
all
the
stuff
it's
more
on
targeted,
for
I
say,
junior
developers
or
people
who
are
not
so
familiar
with
open
telemetry.
C
But
the
solution
is,
I
wouldn't
say,
dang.
Well,
you
could
even
call
it
dangerous
because
putting
so
first
of
all,
it's
tinkering
with
the
front
controller.
That's
index
php!
There's
something
can
do,
but
actually
you
shouldn't
do
any
hardcoded
stuff
in
there
and
then,
but
I'm
referencing,
then
tracer
or
something
global
is
really
not
something
you
should
do
and
we
would
rather
avoid
you
know
picking
people's
stuff
up
like
this
and
thinking
that's
a
good
way,
because
it's
represented
on
on
the
reboot
understand
what
I
mean.
D
C
So
if
you're
interested
in
rewriting
the
laravel
application,
that's
better
and
then
that
then
be
our
guest,
so
we
would
be
happy.
So
arava
is
not
really
my
expertise
and
I
guess
neither
what's
our
bread.
So
if
that's
something
you
would
industries,
we
would
be
actually
happy
so
and
there's
also.
We
had
a
ticket
where
somebody
once
wanted
to
contribute
some
laravel
integration
thing.
It
wasn't
really.
It
was
a
little
bit
half-baked.
You
will
still
find
that.
So
at
the
moment
we
have
some
symphony.
C
Bundle,
that's
what's
called
symphony
so,
but
we
actually,
at
the
end
of
the
day
we
want
to,
you
know,
have
all
the
frameworks
and
everything
supported.
So
it's
just
you
know
a
matter
of
finding
people
who
want
to
work
on
this
stuff.
I
may
I
maybe
in
the
future,
will
work
on
laravel
on
my
job.
So
then
maybe
I
would
look
into
it,
but
at
the
moment.
C
So,
but
just
to
be
aware,
it's
we
think
it's
a
good
way.
You
know
it's
the
easy
way
to
get
started,
but
the
problem
is
more
that
people
take
it
as
an
example.
How
to
do
things.
You
know-
and
this
is
yeah-
we
don't
want
to
give
false
advice.
That's
a
problem.
A
Okay,
I
did
to
show
and
find
an
example.
In
fact,
our
getting
started,
one
seems
to.
C
A
Even
the
very
very
first
one
just
getting
started
has
you
know
child
spans.
C
Yes
and
yeah,
it
might
be
it's
like.
Like
always,
you
know
software
developers
like
to
create
software
and
they
most
of
the
time
don't
like
create
auto-right,
documentation
or
examples,
or
something
like
that.
That's
why
yeah
we
are
lacking
a
little
bit
in
this
regard.
So
we
know
this,
and
but
also
the
point
is
in.
C
C
Since
I
actually
created
this
nearly
half
a
year
ago
and
things
in
the
sdk
change,
I
was
expecting
this,
but
then
I
had
to
to
make
changes
in
the
symphony
and
the
contrary,
repo
in
the
symphony
bundle
itself
and
to
make
this
change
to
pick
up
the
changes
from
the
sdk
and
also
to
fix
all
the
aws
stuff
which
brad
and
bob
knew.
I
was
not
really
excited
about
because
they
had
to
suffer
from
my
grumpiness
in
my
own
maintenance
chat.
Because
of
that
it's
fine
and
I
just
created
the
best.
C
I
had
to
vent
a
little
bit,
because
this
was
like
the
fifth
time
thing
I
had
to
fix
just
and
create
this
demo.
You
know
I
I
didn't
plan
for
this
to
take
a
few
days
and
I
have
to
fix
things
and
anyway
it's
fixed
now
and
in
the
future
we
will
have
if
we
get
closer
to
and
better
now
the
stable
reviews.
We
won't
have
less
problems
with
it,
and
maybe
I
have
to
work
with
aws
too
in
my
next
job,
so
that
will
be.
C
Anyway,
I'm
only
good
one
thing
I
wanted
to
mention
brett,
because,
while
working
on
the
whole
symphony
stuff,
first
of
all
had
to
understand
what
actually
I
did
there
a
few
months
ago,
which
this
wasn't
that
easy,
because
actually
I
had
to
tackle
this
was
actually
one
of
my.
I
think
my
first
shot
at
auto
as
well,
and
I
had
to
solve
a
lot
of
problems
or
things
we
didn't
have
solved
at
the
moment
at
the
time.
C
You
know,
for
example,
configuration
and
working
with
environment
variables
all
the
stuff,
so
I
still
have
to
change
a
little
bit.
Things
are
on,
but
one
of
the
good
things
I
found
now
is
basically
it's
the
only
kind
of
it's.
I
wouldn't
call
it
a
reference
implementation,
but
the
only
real
demo
which
takes
all
of
the
stuff
at
least
the
tracing
into
account.
You
know
so
once
I
was
going
back
into
there
and
to
change
some
things
I
found
some,
you
know
flaws.
C
We
maybe
have
in
the
sdk
and
stuff
like
this,
so
I
will
create
some
issues
and
I
also
think
we
should
pick
up
the
configuration
things
we
have
been
working
on
in
the
past.
A
Yeah,
I
was
thinking
about
the
other
day
because
yeah
they
were
okay,
yeah.
Well
only
that
there's
been
absolutely
no
action
that
I've
seen
on
the
on
the
open,
telemetry
spec
side.
There
was
there
was
a
proposal
and
it
just
hasn't.
C
Moved
yeah
and
the
thing
for
me,
so
I
had
to
solve
a
lot
of
this
stuff
there
and
actually
I
created
like
this
custom
factory
for
the
exporters,
because
basically
I
needed
one
in
the
symphony
because
I
had
to
pray.
We
have
to
create
all
the
services
there,
so
I
need
things,
but
actually
I
want
to
be
things
in
the
something
to
be
much
more
simple.
C
So
what
about?
I
have
a
suggestion,
so
maybe
we
can
find
in
the
next
couple
of
days.
We
can
talk
about
this
and
select,
but
I
wanted
to
just
talk
about
it
in
general.
We
want
to
talk
about
now
that
we
can't
find
a
meeting
where
we
have
a
call
or
something
and
think
a
little
bit
around
your
stuff
or
make
do
some
brainstorming
and
yeah
yeah.
Okay,
then
yes,
and
we
don't
have
to
find
a
date
now,
but
let's
just
talk
in
slack
in
this
and
then
give
them.
C
C
You
know
check
ping
pong
or
even
worse,
you
know,
discuss
things
on
github,
so
it
would
be
good
as
we
open,
but
I
think
to
get
started.
It
would
be
better
if
you
just
you
know,
put
our
heads
together
and
just
brainstorm
about
some
stuffs
and
some
sort
of
stuff
to
get
started
right.
Okay,
yes,
coolio!
Then
let's
do
that.
A
C
C
We
will
find
something.
That's
this
will
be
suitable
for
all
the
stuff
to
configure,
because
it's
it's
quite
it's
complicated.
I
learned
this
when
doing
the
symphony
stuff,
so
taking
care
of
all
this,
and
it's
not
like
I'm
super.
C
C
Okay.
This
was
the
one
check
check
next
thing:
oh
yeah,
the
other
thing
I
learned
so
at
the
moment
the
country.
C
Repository
we
have
a
different
versioning,
so
we
only
have
version.
C
Another
number
one
at
the
moment
and
and
yeah
we
should
come
up
in
the
future
with
a
better.
Actually
I
don't
I'm
not
sure
if
we
should
tie
the
versioning
directly
to
the
to
the
main
repository.
C
However,
well
one
part
which
I
we've
still
said,
I'm
in
the
process
of
setting
up
is
the
you
know,
handling
or
automation,
stuff,
with
pull
requests
and
to
actually
set
up
an
entire
test.
C
So
I
think
one
of
the
things
we
could
do
there
is
actually
on
a
pull
request,
run
the
checks
or
the
tests
of
the
of
the
contract.
Repository
against
the
pull
request.
D
C
Not
to
make
it
mandatory
well
that
won't
work,
but
to
at
least
know,
then
that
we
have
there's
something
we
have
to
change
in
the
control
repository
you
know,
and
so
that
we
know
we
have
to
create
some
issues
for
that,
but
we
cannot
mandatory
to
pass
because
that
will
never
work
so
otherwise
we
will
need
first
to
make
a
change
to
the
cultural
repository
which
won't
work,
because
there
is
a
dependency
and
there
is
a
case.
So
the
only
thing
is
probably
it
will
help
us
to
find
out.
C
C
You
probably
get,
I
don't
know
brett.
If
you
read
all
this
stupid
stuff,
I
will
write
in
more
chats.
C
I
don't
envy
you
but
well
sometimes
I
guess
my
stuff
is
fine
cool.
Then
oh
next
thing
is,
did
you
get
any
feedback
or
from
the
prototype.
A
I
think
it's
I
check
up
on
that.
Every
couple
of
days,
I'm
really
just
waiting
for
someone
to
to
stamp
out
a
new
release.
I
might
because
it's
been
26
days
now,
so
I
just
I
just
missed
the
last
minor
release
by
a
couple
of
days
so
but
but
we
do
know
that
that
bob
has
a
good
contact,
so
that
might
be
it
might
be
good
things.
C
Just
to
kind
of
ask
yeah
the
the
other
thing
is
what
I
was
thinking
since
this
always
created
creates
this
red
cross.
You
know,
input
requests,
and
so
actually
we
thought
of
this
like
a
temporary
solution,
but
now
it's
I
think,
it's
way
over
a
month,
and
since
we
have
this,
so
maybe
we
can
just
maybe
find
a
way
to.
C
Let
me
think
again
to
configure
the
php
any
so
that
we
don't
expose
deprecated
notices
in
the
test
so
that
those
won't
actually
show
up.
A
Yeah,
I
think
that's
easy
enough.
I
think
that's
a
php
unit
option
as
well
from.
C
Memory
yeah:
is
it
possible
to
just
create
this
for
one
of
the
I
mean
it's
probably
possible,
but
can
you
think
of
an
easy
way
to
do
that
for
the
just
for
this
one
matrix
item,
because
on
the
other,
actually
we
want
to
have
this
this.
We
want
to
have
notice,
notices
to
be.
C
But
well
maybe
you
don't
have
to
have
an
answer
if
you
come
up
with
anything,
so
maybe
I
wouldn't
know
anything
easy
out
of
my
mind,
but
now
you
know
that
you
can.
Maybe
you
have
an
idea
or
something
like
that.
You
always.
I
know
you
always
go
to
stuff
like
this.
So
that's
wrong.
C
Next,
one
yeah
only
only
one
thing:
we
talked
a
little
bit
in
in
tennis
chat,
so
we
have
the
style
boards.
We
created
bob
set
up
last
week
to
help
us
with
stale
issues
and
yeah.
He
just
used.
We
talked
about
this.
We
just
use
like
the
standards
and
complex
things
like,
so
we
want
to
find
a
little
bit
more.
C
C
C
C
A
C
Would
be
quite
easy
and
then
assigning
it
would
be
another
thing,
and
maybe
we
would
do
the
same
for
something
like
looking
for
help
and
stuff
like
this.
You
know
so
because
that's
actually
means
we
are
looking
for
people
to
maybe
fix
this
stuff.
So
maybe.
C
Right
because
yeah
I
had
some,
I
had
some
fun.
You
know
swearing
at
the
boat
to
not
making
clothes
issues
and
chinese
stuff
so
too,
but
not
make
it
too
obvious
to.
C
If
you
want
to
learn
some
chinese
or
berliners
feel
free
to
see
my
comments,
but
they
actually
they're
just
gibberish
just
to
make
them
mozzarella
sticks
right,
yeah.
I
just
ask
in
chinese
if
you
can
read
and
stuff
like
this,
so
it's
just
you
know.
I
just
had
some
fun
too.
C
I
think
it's
a
it's
about.
I
think
it's
not
against
the
cncf
regularities
too.
Oh,
you
have
to
be
nice
to
it.
Yeah.
I
don't
think
you
have
to
be
nice
to
bots,
so
I
hope
not
otherwise.
I
have
to
remove
all
the
stuff.
No,
I
I
didn't
use
just
some
funny
funny
stuff.
I
I
was
telling
him
so
okay,
then
we
can
all
think
about
some.
I
think
at
the
moment
we
are
good.
So
you
won't.
You
won't
actually
close
some
important
issues,
and
otherwise
I
can
just
work
on
them.
D
Okay,
so
first
thing:
while
I
was
refactoring
this
environment.php
right,
basically,
we
had
to
move
the
get
env
out
from
that
file,
calling
the
get
env
function
rather
than
calling
get
env
function.
We
are
using
the
accessor
class
so
in
the
source
code
I
found
two
other
files.
Where
get
env
is
used,
one
is
span,
processor,
factory
and
another
is
send
observer,
fiber
dot
php,
I'm
not
sure
whether
whether
we
need
to
refactor
that.
C
C
D
Sure
I
can
create
an
issue
I
will
attach
both
of
the
files
there
as
well.
You
can
have
a
look
into
that
whether
the
refactoring
is
actually
required
or
not.
Well,
if
you
say
so,.
C
C
I
was
just
kidding
so
actually
we
are
kind
of
responsible
for
the
technical
decisions,
but
that
doesn't
mean
that
we
are
the
only
ones
who
can
create
issues
or
important
stuff.
That's
not
about
it.
We
are
more
the
servants
of
this
of
this
project
than
yeah.
You
can
just
open
an
issue
and
you
can
tag
us
from
look
into
it,
but
yeah
anyway,
otherwise
feel
free.
If
you
find
something
or
you
think
a
feature
would
be
viable
for
the
for
the
library.
C
But
usually
it's
not
because
we
don't
none
of
us
get
paid
for
doing
working
on
telemetry,
so
we
can
only
do
so
much
things.
You
know
there's
a
lot
of
stuff
which
is
missing
and
we
know
we
don't
have
issues
for,
and
sometimes
we
talk
about
things
and
yeah.
This
should
be
changed
and
then
forget
to
create
issues.
So,
let's
we're
far
from
being
perfect
for
this
stuff.
C
Very
fun
so
we're
happy
for,
if
anything,
also,
if
you
find
something
in
the
specification
we
didn't
have,
we
don't
have
issues
for
or
we
have
not
implemented
so
feel
free,
we're
happy
for
people
to
create
issues
yeah.
We
don't
have
any
issue
with
that.
D
Yeah
sorry,
the
next
thing
I
wanted
to
talk
about
is
service
name
specification.
So
basically
I'll
post.
D
D
So
basically,
this
the
default
value
for
this
service
dot
name
is
supposed
to
be
either
the
service
name
that
we
set
in
environment
variable
or
through
resource
attribute.
But
if
it
is
not
found,
then
it
is
unknown
service
followed
by
the
process,
executable
name.
So
I
I
believe
that
in
our
php
hotel
php
this
process,
executable
name
part-
has
been
not
implemented.
I
saw
it
in
the
python
hotel
agent
and
they
have
implemented
it.
A
Yeah,
I
mean
that's
what
the
spec
says.
So
that's
what
we
should
try
to
do
and
you've
worked
on
resource
resources.
D
We
already
have
a
process-
detector
yeah,
but
in
in
that
detector
we
we
are
detecting
everything
except
this
process,
executable.
C
Yeah
sure
I'm
not
really
because
sure
what's
meant
with
that
specifically.
C
A
C
Well,
the
heart
should
be,
yeah
could
be
different
and
then
also
if
it's
an
apache
module,
there
should
be
no
pitch
well,
it's
using
the
it
depends.
So
it's
all
also
there's
a
little
on
something
in
the
implementation
of
actually
the
the
run
time
and
in
the
moment
is
it's,
for
example,
using
the
ss
api
right.
C
Yep
and
that's
not
always
it's
kind
of
tricky
in
case
of
php,
but
I
would
not
always
call
that
the
x
executable,
for
example
a
school
or
something,
then
I
am
not
not
the
executive
sort
of
the
the
runtime.
I
would
call
something
like
through
the
runtime
you
know
and
actually
what
they
s
as
a
pi
shows
there
is
the
cli
or
key
or
whatever
you
want
to
call
it.
So.
C
It
really
depends
on
you
know
the
definition,
what
you
call
the
runtime
right,
so
maybe
there's
also
the
php
runtime
at
the
I
can
maybe,
if
I
find
it,
which
defines
it
a
little
bit.
I
think
I
made
a
com
comment
on
this
when
we,
when
we
were
looking
at
the
pr
back
then
so
I
will
seconds
so.
This
is
a
link
which
actually
defines
runtimes
a
little
bit
different.
You
can
have
a
look
into
it
because
they're
very
different
ways.
C
Nowadays,
you
know
back
in
the
days
15
years
ago
or
something
the
apache
wasn't
really
about.
20
years
ago,
apache
was
more
or
less
the
only
way
how
you
can
execute
php,
while
they're
still
a
binary,
but
that's
what
people
did
and
then
there
was
light
hp,
all
the
other
stuff,
but
nowadays
you
have
like
a
lot
of
different
ways
on
how
to
run
php.
So
some
more
exotics
are
less
exotic,
but
at
least
there
are
a
lot
so
yeah.
Maybe
we
could
also
think
about
how
we
define
what
the
runtime
is.
D
A
Just
reading
that
the
documentation
in
the
spec
it
just
talks
about
whatever
proc
status
in
linux,
says
yeah,
at
least
as
a
cross
reference
for
what
it
should
be.
A
Yeah
also
I'm
reading
the
process.
Runtime
resources,
part
of
the
spec
which
which
kashan
just
linked
before.
C
Actually
the
thing
is,
this
is
yeah.
This
is
the
semantic
conventions
specification
right
and
so
you
can
take.
You
can
take
the
default
from
their
unknown
service.
It's
not
defined
in
the
in
the
environment,
variable
association
as
a
default,
because
otherwise
I
would
have
put
it
into
into
the
constants
there.
But
again
maybe
well,
you
know
the
content,
because
the
defaults
for
the
constants
is
just
something
I
put
there.
You
know
to
have
everything
in
those
interfaces
so
which
we
can
reference,
but
there's
no.
C
We
can
actually
put
put
the
default
for
unknown
service
into
there
as
well.
You
know
it's
not.
This
is
not
part
of
the
spec
to
have
this
interface
there,
but
what
was
just
I
just
implemented
all
this
stuff,
so
we
can
have
a
clean
reference.
You
know,
if
it's
a
better
way,
if
you
put
it
there,
we
still
have
it
once
in
a
single
place.
Then
that's
totally
fine,
so
we
can
actually
put
in
there
everything
you
want.
So
I
just
I
mainly
put
everything
in
there
which
I
could
find
in
the
specification.
C
D
So
that
environment
variable
would
help
us
the
name
of
the
environment.
Variable
is
hotel,
php
detectors,
and
that
will
help
us
enable
or
disable
certain
subsets
of
detectors
for
resource
info.
D
I
I
just
have
doubt
regarding
the
implementation,
from
the
name
of
that
variable,
whatever
values
we
set
right,
those
values
we
want.
Those
detectors
we
want
to
enable
is
that
correct.
C
Yes
but
you're
correct
and
having
a
little
bit
thought
about
this,
because
it's
actually
tackling
something
which
we
don't
have
a
place
yet
and
which
gets
a
little
bit
much
brighter
and
I
talked
about
the
configuration.
So
I'm
not
quite
sure
if
we
really
can
implement
this
at
the
moment,
because
we
at
the
moment
we
don't
have
a
single
entry
point.
C
You
know,
if
you
look
at
a
lot
of
the
other
other
implementations,
then
they
often
have
like
you
can
set
up
something
the
global
tracer
or
something
like
this,
and
maybe
we
can
internet
breath.
Can
you
take
a
grip?
C
I
sure
you're
not
will
not
fall
from
the
chair?
Okay,
sorry,
okay!
Maybe
we
can
create
like
a
global,
tracer
and
stuff
like
this.
C
Just
to
give
you
an
idea
kishan,
so
normally
I
would
always
you
know,
argue
against
global
stuff.
So
that's
usually
I'm
the
one
arguing
against
this.
That's
why
I
asked
where
to
take
your
groups
from
I
knew
what's
going
yeah,
but
I
think
in
this
regard
and
it's
we
have
to
always
be
worried
about.
I
mean
some
global
states,
but
in
this
regard,
if
you
have
one
single,
you
know
like
the
global
thing
we
have,
then
it
will
be
reasonable
because
it's
not
like
global
stuff
is
always
bad.
C
So
if
you're
dealing
with
something
like
a
process
which
is
always
global
or
some-
you
know
anything
like
this,
then
it's
actually
in
this
case
the
better
way
to
have
something
like
this.
So
you
don't
have
an
application
think.
So
I'm
always
I'm
just
a
little
bit
worried
about
some
other
lower
stuff
which
may
lead
them
into
global
scope.
So
in
general
we
should
avoid
it.
But
lot
of
other
languages
have
something
like
that,
but
that
doesn't
mean
that
we
have
it
in
php,
because
the
language
work
differently.
C
You
know,
for
example,
in
javascript,
it's
less
much
issue
to
have
global
objects,
because
there
are
global
objects
in
the
first
place,
because
that's
an
object,
oriented
language,
you
know,
so
you
have
always
every
object
and
stuff
like
this
and
there's
a
different
ways
to
reference
all
this
stuff,
so
yeah,
but
maybe
we
should
overthink
and
we
can
brainstorm
a
little
bit
about
this.
C
C
No,
no!
I
was
thinking
actually
at
the
lower
level
where
in
the
where
actually
the
standard
providers
are
created.
This
is
in
the
res
sources.
At
the
moment,
right
yeah,
where
the
resource
the
resources
are
created
there
you
will
be
find
and
it's
called.
Oh
don't
let
me
lie,
get
defaults
or
something
the
method
is
closed.
B
A
C
C
C
D
It's
a
good
information,
but
my
question
was:
if
we
were
to
implement
the
variable
right,
whatever
values
we
put
in
there,
those
detectors
we
want
to
enable
that
would,
by
definition
mean
that
by
default,
none
of
the
detectors
are
enabled
that
that
is
my
understanding.
Is
that
correct
I
mean
by
default.
D
Let's
say
it
would
be
good
to
have
all
the
detectors
enabled
by
default,
but
then
suppose,
if
user
or
yes,
if
user
wants
to
disable
one
particular
detector,
then
that
user
has
to
write
all
the
other
detectors
name
in
that
variable.
A
C
Okay,
yeah,
the
only
thing,
so
let
me
think
about
one
thing:
so
the
resource,
no
it's
just
created
once.
Okay,
no
okay,
forgot
happening
a
about
performance.
I
will
look
into
this,
so
is
it
actually,
the
results
at
the
moment
is
trade
once
once.
Yes,.
C
Yes,
okay,
I'm
sorry!
I'm
literally,
I
had
to
think
about
a
lot
of
different
things
so
that
I'm
a
little
bit
confused
about
this
at
the
moment
yeah,
but
I
think
all
the
one
well
at
least
the
ones
which
are
the
defaults
at
the
moment.
C
C
And
there's
a
reason
because
we
have
those,
because
it
was
quite
even
though
you
see
a
lot
of
environment
variables
added
in
the
last
days
and
a
couple
of
weeks
we
were
told
when
we
wanted
to
add
one
that
there
should
be
more
once
more
added.
So
yeah,
that's
what
we
create
our
own
ones
here
and
instead
of
going
to
all
the
hassle
specifications.
A
Yeah,
yes,
yes,
I've
even
noticed
some
of
the
other
language
implementations
just
creating
environment
variables
in
the
global
space
anyway,
yeah
yeah,
they
don't
even
use.
C
The
the
language
thing,
but
let's,
let's
just
be
safe,
so
as
we
create
this,
then
if
for
some
reason
some
variable
like
this
is
created,
we
don't
have
to
change
this.
Let's
just
can
you
have
both
options
or
no
don't
have
to
deprecate
something
as
fast.
C
C
Nope,
okay,
gentlemen,
then
talk
thanks
for
this
interesting
meeting
and
we
will
last
for
at
least
maybe
talk
next
week
in
the
meeting.
Otherwise
we
will
talk
on
slap
or
we
will
right.
So
we
will
then
keeshan
have
a
nice
sleep
great
to
have
a
nice
morning
and
chimo
will
have
a
nice
evening.