►
From YouTube: 2020-10-13 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
Hey
guys,
can
you
hear
me
yeah,
yes,
who's
who's,
x,
y
y-I-r-a,.
B
A
B
A
A
A
Josh,
okay,
so
if
you
could
take
like
one
minute
to
to
sign
up
your
name,
I'll
put
the
link.
A
A
I
think
we
can
start.
This
should
be
a
really
short
meeting.
I
think
mark
and
karen
are
on
vacation
karen,
like,
unfortunately,
I
got
dragged
into
this
meeting.
Probably
we
want
to
finish
as
quick
as
possible.
Johannes
is
a
is
running
other
business,
so
he
he's
not
going
to
join
this
time
and
it'll
just
be
a
few
of
us
and
we
can
start
so.
I
have
I
have
one
topic.
I
want
to
update
on
the
open,
telemetry
overall
status.
A
So,
as
you
guys
know,
the
open
country
spec
is
trying
to
get
like
the
tracing
part
down
as
soon
as
possible,
and
we
believe
the
all
the
blocking
issues
are
resolved.
There's
some
changes,
for
example
the
status
quo.
We
need
to
update
to
reflect
the
latest
change,
but
all
the
major
issues
are
already
covered
and
if
you
look
at
the
link
I
put
in
the
in
the
meeting
agenda,
let
me
share
my.
A
Screen
so
this
one
has
a
matrix,
and
I
I've
done
the
initial
version
like
just
to
fill
in
how
compliant
we
are
from
this
c,
plus
plus
sdk
to
the
spec.
Something
has
changed
since
then,
so
I
need
to
update
that
and
the
goal
is
to
focus
on
the
first
release
candidate
in
the
next
couple
weeks
and
like
in
this
past
past
project.
A
We
don't
have
the
commitment
to
ship
the
ga
release
by
any
time
this
year,
but
still
I
want
to
try
if
we
can
get
the
tracing
part
done
by
end
of
the
year
released
as
a
preview
or
alpha
version
of
the
package
and
the
release
candidate
tool
for
the
open
time
overall
product
would
be
covering
the
baggage
and
other
features.
So
at
least
for
now.
We
don't
have
to
worry
about
that.
The
the
focus
should
be
get
the
tracing
part
down
and
align
with
the
the
spec
100
and
get
the
initial
release
by
another
year.
A
Questions
yeah,
so
I
I
think
on
that
part,
the
work
should
should
not
be
too
too
heavy.
So,
besides
that,
I
I'm
going
to
spend
time
working
with
other
approvers
to
make
progress
on
the
logging
part
in
karen
mark.
They
have
done
a
lot
of
like
a
job
on
the
initial
investigation,
and
I
think
this
week
we
should
be
able
to
move
things
even
like
more
like
forward.
C
A
A
A
A
D
I'm
just
here
to
watch
okay,
I'm
not
in
the
workspace
yeah.
E
So
I
had
to
find
the
window,
okay,
so
I've.
I
had
two
things.
I
wanted
to
kind
of
ask
status
of
so
there's
a
there's,
a
pull
request
to
compile
using
the
standard
library
and
get
rid
of
no
std,
and
so
this
this
question
is
kind
of
related
to
the
second
question,
but
I
was
kind
of
curious
what
the
take
on
that
pr
is,
if
anyone's
had
a
chance
to
review
it.
E
If
I
should
go,
look
through
it,
I
I
mean
I
looked
through
it
and
it
seems
like
a
reasonable
thing.
I
just
wanted
to
know
if
that's
like,
like
a
super
crazy
stretch
like
we,
don't
want
to
have
a
way
to
compile
this
using
standard,
live
or
not.
F
Josh,
first
of
all,
I
would
love
to
get
that
change
in
because
it's
not
taking
away
anything.
It's
just
giving
a
freedom
to
statically,
compile
a
header-only
implementation
which
uses
standard
library
rather
than
ocd,
and
I
see
a
number
of
use
cases
where,
like
internally,
for
example,
with
my
microsoft
customers,
where
it
is
of
a
great
benefit,
performance-wise,
stability-wise
and
trust-wise,
because
we
use
something
that
is
kind
of
a
bit
more
trusted
than
our
own
custom.
F
Backboards
right
and
the
other
thing
I
was
thinking,
is
about
upsell,
google's
upsell
library
as
yet
another
option.
So
our
motivation
here
is
some
project
that
is
already
relying
on.
Stdlib
should
use
that
some
project
that
uses
upsell
classes
may
use
that
and
for
the
abi
compat
like
dynamically
loadable
library,
we
would
probably
go
with
our
own
backboards
with
exception
that
I
I
spoke
to
email
about
it.
F
We
would
most
likely
need
to
patch
and
replace
our
current
variant
implementation
with
up
sales
variant,
because
upsells
vary
and
it's
kind
of
better
it.
It
works
better
across
a
larger
number
of
compilers
and
if
you
can
take
a
look
at
the
pr
like
it's
a
work
in
progress
pr,
please
let
me
know
your
feedback.
E
Yeah,
I
am
I
I
mean
I
took
a
what
a
cursory
look
at
it
and
I
can.
I
can
look
at
it
more
in
deep
or
more
deeply,
but
I
I
like
the
direction
I
just
was
curious:
what
kind
of
the
status
in
terms
of
adoption
you
know
what
we're
thinking,
because
that
does
introduce
the
whole.
You
have
to
make
sure
that
the
thing
compiles
with
a
standard
library
and
without
and
etc.
You
know,
first,
if
you
can
take
a.
F
Look
at
like
of
my
fork
versus
this.
I
kinda
tried
to
cover
this
by
spinning
two
sets
of
belts
and
tests.
It
wasn't
my
back
burner
frankly
and
the
main
reason:
why
is
the
only
customer
I
currently
work
with?
They
can
use
the
header
only
implementation.
I
didn't
have
to
push
hard
to
make
it
happen,
but
I
think
we
should
set
a
goal
to
make
it
happen
like
by
the
end
of
this
year.
If
you
guys
agree.
A
A
That
means,
if
someone
is
taking
the
open
time
to
api,
and
is
that,
for
example,
on
a
linux
machine
and
they
have
a
different
sdk
from
new
relic,
which
is
not
open
source.
They
should
be
able
to
use
that
or
they
have
a
like
plugin,
which
is
using
the
api
model.
They
should
be
able
to
use
that
binary
without
asking
you
really
hey
like
this.
One
is
not
working
because
I'm
using
this
particular
compiler
or
particular
runtime,
and
that's
why
we
have
the
like.
A
When
you
look
at
the
the
current
product,
the
api
package
is
header
only
and
we
don't
take
dependency
on
the
specific
standard
library
like
we
like.
We
even
allow
like
no
exception
and
no
standard
library
dependency
and
in
order
to
make
it
easy,
we
do
have
a
lot
of
apis
that
will
take
a
a
complex
type
like
a
list
of
something
or
a
map
or
something,
but
we
decided
not
to
take
dependency
on
the
standard
like
the
stl
containers
because
they
are
not
guaranteed
to
be
abi
compatible.
So
we
kind
of
forked
our
own
version.
A
A
Meanwhile,
many
like
real
cases,
we're
saying
customers
will
say
like
this
is
hard
to
use,
because
when
we
use
your
like
api,
we
already
have
the
data
structure,
which
is
an
stl
map,
and
is
that
possible
that
we
can
pass
that
indirectly
or
like
like
I'm,
not
confident
with
your
non-standard
version
implementation?
Because
I
have
a
compliance
reason.
I
have
to
use
my
like
the
version
that
my
organization
has
chosen,
but
then
there
was
a
debate
between
multiple,
like
approvers
and
maintainers
at
that
time.
A
One
particular
reason
is
in
this
case
we're
actually
moving
away
from
our
original
goal,
because
if
this
customer
decided
okay
we're
going
to
turn
on
the
compile-time
flag
and
use
this
use
my
standard
version
of
stl,
then
later
they
cannot
use
the
plugin
from
the
other
compatible
apis
because
they
might
be
compiled
in
a
different
compound
flag.
But
the
the
contour
argument
from
from
max-
I
I
think,
is
if
you
make
different
compilation
flags,
the
thing
is
already
broken.
A
A
The
other
argument
from
the
like
not
not
going
this
past
is
to
consider
the
test
and
and
the
maintenance
matrix,
because
we're
adding
another
dimension,
which
is
a
little
bit
costly,
but
it
seems
like
like
from
most
of
the
the
contributors
in
this
product
that
should
be
okay.
Thing
like
most
of
the
folks
should
be
okay
to
pay
for
that
cost,
just
to
get
the
benefit
guys.
F
I
can
help
with
the
the
setup
of
the
infra
to
add
the
extra
validation
of
another
variant
right
yeah.
If
I
see
benefit,
if
there's
a
price
to
play,
I
am
ready
to
do
the.
A
Yeah,
so
me
too,
like
speaking
for
microsoft,
I
I'm
I'm
also
seeing
the
value,
so
if
I
I'm
going
to
pay
for
the
cost
I'm
willing
to
pay,
but
as
a
maintainer
I
will
like
I
wear
the
the
cnc
I've
had.
So
I
cannot
just
speak
like
microsoft.
Has
to
do
this
so
at
that
time
there
was
no
of
this
reason
why
we
need
to
prefer
this,
but
if
we
have
more
like
waste
and
people
are
saying
this
is
a
good
direction,
we
should
pay
for
that.
I'm
definitely
happy
to
pursue
that
direction.
A
F
But
we
are
not
taking
away
the
other
portable
and
compatible
flavor
understand
yeah.
A
I
understand
yeah,
so
the
only
debate
I
think
at
that
time
was
like
from
meal
and-
and
he
believes
adding
this
additional
thing
is
not
something
he
would
need
and
that
gives
actual
maintenance
cost,
and
I
I
think
he
has
a
reason.
A
F
The
in
general,
I
was
thinking
like
about
how
we
run
all
these
google
tests
in
google
benchmarks,
some
sort
of
kind
of
matrix
comparison
to
make
sure
that
if
we
double
the
flavors,
we
still
need
to
run
export
the
result
and
say
hey
check
by
default.
Everything's
passing
something's,
failing
hey.
This
is
the
difference.
F
This
is
where
we
found
that
our
no
std
port
is
incompatible
or
behaves
differently
than
the
standard
library,
and
it
is
actually
a
good
cross
reference,
because
if
our
non-standard
library
claims
to
support
a
certain
feature
of
a
container
or
something,
but
it
doesn't
actually
behave
exactly
the
way
how
the
standard
library
does
then
well,
we
actually
caught
something.
We
need
to
fix
it,
because
people
would
have
some
expectations.
E
Yeah
I
I
have.
I
have
two
questions.
I
guess,
if
I'm
using
this
option
right,
am
I
a
static
only
compiled
version
of
the
binary,
where
I
don't
do
dynamic
loading
right?
Am
I
forcing
myself
to
do
that
and
like
I'm
okay
with
that,
if
that's
what
this
means,
but
that
also
means
that
the
idea
of
people
who
want
to
static
only
link
is
a
is
a
supported
use
case,
as
is
people
who
want
to
dynamically
load
in
pre-compiled.
F
Yes,
and
I
would
even
say
that
we
are
talking
static
and
dynamic,
I
would
say
even
more,
perhaps
in
some
scenarios,
since
we
provide
api
as
a
separate
surface
and
sdk
as
a
separate
piece,
some
folks
might
say:
hey
we
actually
want
a
header
on.
So
let's
get
another
dimension.
I
don't
I
don't
say
we
have
to
commit
to
that
this
semester,
but
I
see
this
as
a
viable
third
option.
F
Where
I
say
hey,
I
want
this
open,
telemetry
compatible
dpi
and
I
have
a
full
header,
only
implementation
that
works
with
that.
So,
like
I
see
that
a
usable
scenario.
F
And
from
the
library
perspective
I
I
have
seen
like,
I
was
trying
to
to
make
it
work
with
the
standard
the
default
across
all
os
there
were
quirks
even
with
that,
like
variant
is
extremely
quirky
on
mac,
it's
just
even
the
the
apple's
clang
is
different
from
the
main
line
along
the
m
clang.
In
this
regard,
so
there
was
quite
a
bit
of
fun,
but
I
think
once
we
get
it
up
and
running,
maintaining
it
and
keep
running
is
gonna,
be
not
a
significant
experience.
A
Yeah,
I
I
think
so,
unless
we
decided
to
add
yet
another
non-standard
like
spec
like
we
need
to
support
something
from
there.
F
There's
another
hack,
oh
well,
bad
work,
so
imagine
that
we
have
that
static
library
flavor
and
we
decided
to
go
with
standard
library
for
that
rather
than
with
the
non-standard,
I
think
the
compiler
itself,
or
we
may
have
a
quite
a
bit
of
optimizations
because
right
now
we
do
all
this
transform
like
from
map
through
key
value,
iterable
going
back
and
wrapping
it
in
sdk
again,
and
it's
like
it's
it's
a
waste
right.
So
if
you're
building
a
super
hyper
thing
or
somebody
who's
looking
at
it,
why
do
they
do
it?
F
And
the
answer
is
we
do
it
for
abi
combat?
Then
somebody
says:
hey,
I'm
still
statically
linking
the
whole
thing.
Why
do
I
have
to
worry
about
it?
Why
do
I
have
to
pay
for
perth
penalty
for
a
combat
feature
that
I
don't
benefit
from
right?
So
that's
where
we
can
say
hey.
F
E
That's
kind
of
what
I
was
looking
for
personally
yeah,
like
the
the
ability
to
kind
of
go
a
direction
where,
if
I
decide
to
statically
link,
I
can
get
optimizations
and
make
statically
link
kind
of
a
first
class
feature,
and
then
abi
is
a
thing
that
you
can
also
opt
into.
But
you
you
paid
the
penalty
of
abi,
but
you
could
find
it.
We
either
need
to
support
both
or
anyway
yeah,
yeah
and.
F
Just
this
is
exact
thinking
of
most
of
my
customers
right
now
like
when
I
work
with
like
big
star
microsoft,
customers
like
office,
that's
the
exact
same
thing.
F
Fun
fact:
a
really
relevant
fun
fact.
I
was
working
on
like
mobile
platform
telemetry
for
some
of
our
internal
customers
recently,
and
I
looked
at
how
android
is
handling
this
like
android,
has
the
user
mode
and
decay
and
stl
and
a
platform
mode,
ndk
and
stl,
so
guess
what
their
answer
is.
These
are
incompatible,
go
away,
don't
interlink
between
the
two.
If
you
want
to
build
the
user
mode
up,
use
this
tool
chain-
and
this
c
plus
plus
implementation.
A
F
Can
do
that
extra
push
and
I
think
emil
would
be
in
favor
of
replacing
variant
with
upsell
variant
we
can
borrow.
We
can
give
attribution
to
google
for
that,
and
this
also
gives
us
support
for
slightly
older
visual
studio,
2015
compiler,
because,
whatever
our
variant
that
we
have
sorry,
I
couldn't
make
it
work
with
15..
From
that
perspective,
I
think
I
trust
more.
The
up
sales
implementation
of
the
variant
so
I'll
refresh
my
pr
with
this
as
well.
A
Okay,
so
so
max
probably
do
this
like
we
have
a
week
to
collect
feedback
and
there's
a
like
like
enough
like
good,
to
go
like
thumb
up
on
that
in
the
next
sick
meeting,
we
can
make
a
call
and
see
if
there's
a
like
anyone
try
to
stop
and
see.
This
is
crazy.
If
not,
then
we're
ready
to
to
go.
F
A
Pop-Up
that
you
have
on
your
screen,
okay,
josh
like
shall
we
move
to
the
next
topic
or
still
yeah.
E
Yeah
this
is
this
is
somewhat
related,
so
you
know,
I
think,
we're
planning
to
mostly
statically
link
the
library,
but
I
want
to
make
sure
when
I
make
contributions,
I'm
not
breaking
the
abi
or
if
we
were
to
release
like
a
1.0
and
then
want
to
make
sure
we
don't
break
the
apa,
like
I'm
kind
of
curious
what
the
abi
plug-in
guarantees
are
from
1.0
to
like
1.1
or
1.01
or
bug
fixes.
E
E
Can
I
just
make
code
and
assume
that
there's
going
to
be
enough
tests
there
that
it'll
catch
it?
If
I
make
a
mistake,
or
do
I
need
to
like
go-
spend
some
more
time,
understanding
what
I
could
do
that
could
break
avi.
F
Maybe
if
we
can
use
some
of
these
and
historically
compare
like
you,
know,
symbols
and
all
that
as
well
as
as
you
mentioned,
the
actual,
like
google
benchmark
test,
google
google
test
that
instruments
like
loading
of
a
process
with
dll-
that's
great.
I
want
to
add
something
to
this:
it's
not
just
the
api,
like.
I
noticed
some
interesting
quirk
with
visual
studio,
for
example,
if
you're
statically
building
a
few
libraries,
you
actually
have
to
use
the
linker,
which
is
the
latest
of
all.
F
So
if
you
are
build
a
with
15
b
with
17
and
c
with
19,
you
actually
have
to
use
the
19
linker
to
link
all
of
that,
and
I
hate
it
when
I
was
testing
different
tool
chains
and
I
was
trying
to
reuse
the
google
test
and
google
benchmark
build.
I
was
like
what
my
symbols
are
not
there
like.
Some
of
this
is
also
relevant.
A
Okay,
so
so
josh.
Currently
I
would
say
the
coverage
is
not
very
comprehensive
and
I
I
remember
in
the
history
folks
have
done
something
that
broke
the
abi
compat
and
was
not
caught
by
the
ci.
I
I
would
say
at
this
stage
you
could
pass
the
ci.
Don't
worry
about
breaking
the
other
thing.
That's
not
covered
and
when
we
add
more,
like
coverage
test
will
will
improve
that
part.
A
E
F
As
of
note,
you
can
actually
break
anything
right
now
because
we
are
not
officially
released,
we
don't
have
stable
edi
yet,
like
I
think
the
goal
is
once
we
say
hey,
this
is
version
one,
that's
where
we
have
to
worry
and
that's
before
that
we
need
to
have
all
these
good
tests
to
verify
the
api.
A
Yeah
so
for
now
we
have
a
pretty
low
bar,
as
stated
in
the
very
bottom
here,
so
we
don't
guarantee
any
backward
compatibility
until
we
have
the
ga
release
and
for
after
ga
release,
we
follow
the
overall
open,
telemetry
approach,
where
we
say
we
follow
the
semantic
version.
Two
so
same
version
tool
like
has
a
strict
definition.
If,
like
we
have
1.1
release,
it
will
be
compatible
with
1.0
and
2.0
will
be
a
major
version
upgrade
we
can
put
breaking
changes
and
in
other
reports
I
believe,
for
example,
in
open
telemetry.net.
A
E
Okay,
that's
good
to
know
thanks
all
right,
so
so
we'll
we
will
have
api
testing
when
it
becomes
relevant
yeah,
but
move
fast
and
break
things.
E
Depending
I
don't
know
what
the
guarantee
is
from
the
sdk
standpoint,
but
the
the
way
java
does
binary
compatibility.
I
think
that
likely
it
will
be
true
that
you
can
use
plug-ins
appropriately
as
long
as
you
don't
call
any
of
the
new
methods
you
don't
break.
C
Okay
for
python,
I
think,
if
up
after
okay
there's
a
lot
of
dependencies,
python
version
right,
I
think
different
python
version
doesn't
guarantee
binary
compatibility.
The
bytecode
format
is
different.
A
Actually,
actually,
python
has
been
stable
for
quite
a
while.
So
if
your
python
is
read,
I
should
be
fine,
but
that's
another
topic,
and
normally
people
use
the
dot
py
instead
of
dot
pyc
so
like.
I
would
be
surprised
if
people
don't
copy
the
python
script,
but
just
copy
the
compile
the
byte
code.
Okay,
that's
another
topic,
yeah,
okay,
so
we
can
move
the
next
one.
C
Okay,
yeah
for
the
I
think
we
discussed
these
ones
in
maybe
months
ago
and
yeah
at
that
time.
I
thought
this.
This
doesn't
have
have
so
much
effects
but
seems
yeah.
This
effect
our
this
industry's
many
convenient
stripes
in
the
new
language
back
in
the
last
week.
I'd
like
to
that
we
can
use
a
generic
lambda,
which
is
only
available
in
c
plus
plus
14..
I'm
wondering
I
think
I
remember.
C
A
F
F
Judging
by
like,
if
we
focus
on
who
are
we
delivering
this,
for
I
can
comfortably
tell
that
for
most
of
the
microsoft
customers,
c,
plus
plus
14
is
likely
there
because
we
would
be
most
likely
targeting
visual
studio,
17
and
above,
but
we
need
to
get
feedback
from
unix
guide
like
a
new
rally,
light
step,
server,
monitoring
legacy
server
monitoring,
especially
since
red
hat
enterprise.
Linux
is
going
to
be
supported
like
until
2024
25.
F
That's
where
this
requirement
is
coming.
F
A
Yeah,
so
so
far
so
far,
we
we
support
c,
plus
plus
11
and
above
and
I
I
think
I'll
I'll
do
my
job
to
clarify
that
by
putting
something
similar
in
the
start
page.
This
is
what
we
support
and-
and
I
I'll
need
to
close,
that
discussion
thread
because
it's
already
concluded
and
if
there's
need,
we
got
to
change
that.
It
should
be
a
separate
issue
or
pr
and
we
need
to
discuss,
but
for
now,
okay
class
class,
11.,
okay,.
F
And
the
the
other
thing
is:
imagine
all
these
fancy
things
like
web
assembly.
I
know
that
maybe
most
likely
they
are
already
running
way
ahead
of
us,
but
there's
a
whole
vast
variety
of
compilers.
Like
I
don't
know
a
nintendo
switch.
Now
I'm
talking
about
real
things
like
minecraft,
for
example.
Playstation
are
we
sure
this
stuff
is
going
to
have
the
latest
language
standard?
It
certainly
has
c
plus
fossil
and.
F
A
Open
the
link
again
to
the
chat
window
and
they
like.
I
would
appreciate
if
you
could
do
a
quick
introduction,
just
like
like,
where
coming
from
and
the
workflow
company
and
what
they
talking.
Looking
for
from
this
community.
D
A
But
oh
yeah,
so
by
putting
your
name,
it
means
you
join
this
meeting,
it
doesn't
mean
you
have
committed
any
time.
The
commitment
is
following
the
open
telemetry
community
page.
So
let
me
send
you
that,
and
it
has
an
explicit
process
by
joining
the
meeting,
it
doesn't
imply
anything.
D
No,
I
just
I
would
prefer
to
be
in
this
in
my
personal
capacity
that
that's
all
I
meant
so
yeah,
I'm
andy,
I'm
from
citadel,
so
the
hedge
fund
we've
been
looking
into
using
open
telemetry
for
tracing
of
some
of
our
c
plus
plus
services,
which
is
why
I'm
here
so
I'm
just
interested
to
learn.
A
D
D
I
see
oh
yeah,
but
I'm
yeah
we're
yeah
we're
it's
just
something
that
I've
been
looking
into
this
in
my
personal
free
time.
It's
not
like.
A
D
Yeah
but
it's
it's
just
yeah,
so
I'm
I've
been
looking
into
open,
so
we've
been
using
open
tracing
for
some
of
for
to
look
into
the
observability
of
some
of
our
systems,
and
I
understand
that
the
industry
seems
to
be
converting
all
open
telemetry.
So
I'm
just
here
to
see
how
the
project
is
going
and
then
maybe
to
make
a
decision
as
to
what
to
do
in
the
future.
A
So
the
the
goal
is
to
use
open,
telemetry
to
replace
open
tracing.
The
state
will
be
different
depending
on
the
each
like
language
status
and
the
goal
is
to
have
the
open
tracing.
Shame
on
top
of
open
telemetry
and
deprecate
open
tracing
eventually,
but
I
think
the
c
plus
password
project
will
probably
take
a
bit
longer.
D
Yeah,
so
I
think
I
might
be
in
this
meeting
just
a
few
more
times,
because
I
I'd
love
to
listen
to
any
progress
and
then
see
where
the
community
is
going,
but
I
hope
I
probably
won't
be
able
to
contribute
too
much
so
sorry
about
the
any
confusions
there.
A
D
A
A
Okay,
I
I
think
that
got
all
the
public
word
and
look
at
the
pr
it
seems
like
we
got
all
the
all.
The
recent
pr
is
like
making
steady
progress.
A
This
one
is
approved,
there's
a
you're,
not
not
sure
what
happened,
but
let's
take
a
look.
If
it's
now
like
it's
not
related
to
this
one
will
I'll
emerge
that
and
for
this
one
I
think
lalit
is
not
here,
but
he
got
a
feedback
from
johannes
there's
a
way
that
we
can
change
the
the
signature
to
make
it
more
efficient,
and
that
is
working
on
that.
I
think.
F
For
the
for
the
first
one,
I
just
submitted
a
an
example
how
we
can
flatten
it
in
flatten,
that
is
in
respect
in
response
to
logging,
rfc
spec.
Just
just
to
know,
I
needed
a
place
to
put
a
markdown
with
some
code
in
it,
and
I
thought
it's
gonna
be
a
good
resource.
We
can
eventually
like
close
it
for
now.
I
guess
karen
and
mark
can
use
that
as
a
reference.
E
Okay,
cool
thing:
the
check
that
failed
on
the
thread,
safety,
one-
was
the
code
coverage,
but
I'm
not
really
sure
how
to
it's
like
saying
that
the
the
test
actually
has
less
code
coverage
than
it
did
previously.
That's
the
thing
that's
read,
so
I
kind
of
don't
understand
what
it's
trying
to
tell
me.
A
A
E
E
So
a
lot
of
those
were
like
no
standard
files,
and
things
is
that
just
we
know
not
to
touch
those
or
what's
going
on.
F
I
think
that
I
hate
the
same
issue
as
well
different
os.
I
have
different
a
version
of
bank
formatter
and
they
follow
the
same
rule
differently.
Yeah.
What
I
ended
up
thinking
is,
I
ended
up
installing
exactly
that
same
version
of
the
tool
that
is
running
in
our
ci,
and
after
that
I
was
getting
a
small
deltas
and
just
pushed
exactly
what
what
the
it.
It's
probably
a
bit
of
a
process
issue.