►
From YouTube: 2022-03-31 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
B
903
so
yeah
feel
free
to
add
your
name
to
the
attendees,
probably
going
to
be
a
short
meeting
this
week.
If
they
believe
da
goes
out.
B
Yeah
there
has
been
a
lot
of
work.
I
believe
same
kid.
In
ashtosh,
I've
been
working
on
the
contrib
repo,
I
believe
specifically
pertaining
to
a
lot
of
the.
B
What
is
it
adding
custom
request
response
headers
correct
me:
if
I'm
wrong
guys
yeah?
How
is
that
going
for
you
guys
it's
a
smooth
process.
A
B
Right,
yeah
awesome
thanks
thanks
for
your
contributions,.
C
B
Yeah
awesome,
I
I
think
I
I
think
there
was
one
thing
that
like
was
kind
of
missed
with
the
the
previous
instrumentations
that
have
already
been
merged,
but
I
mean,
I
think,
you've
already
attested
to
this,
but
it
would
be
great
if
we
could
have
like
examples
of
the
usage
in
the
in
the
usage
docs
for
the
corresponding
instrumentation
files.
That'd
be
great.
B
Okay
cool,
so
I
guess
not
many
new
topics.
I
think
it's
not
time
for
a
new
release
yet
usually
released
on
the
first
or
second
tuesday
of
every
month.
B
Some
of
the
metrics
bug
fixes,
as
well
as
the
exception
handling
for
logs.
It
does
look
like
we
have
a
pr
up
right
now.
Let's
take
a
look
at
that.
Let
me
share
my
screen.
Real
quick.
B
All
right,
okay,
cool.
So
this
is
one
of
the
interpretation
pr's
that
we
were
referring
to
yeah.
D
Basically,
here
the
question
was
in
whiskey
and
instrumentation:
there
is
a
difference
like
whiskey,
instrumentation.
D
That
is
added
by
the
user
and
still
he
would
be
able
to
capture
the
request
headers.
That
might
be
the
case
here
and-
and
there
is
one
more
thing
that
I
saw
you
when,
when
we
are
actually
capturing
the
response-
headers,
those
those
actually
preserve
the
casing,
as
well
as
the
iphones.
D
B
Right
so
from
correct
me,
if
I'm
wrong,
my
understanding
is
them
so
they're
they're
in
like
for
whiz
key
itself
right,
the
request
headers
are
already
normalized,
whereas
ascii
they
are
not
set.
The
point.
B
B
So,
what's
the
issue
with
this,
don't
we
like
in
our
instrumentations,
make
it
so
that
we
convert
both
of
them
to
a
film
some
similar
format.
D
We
are
able
to
capture
it
if
the
name
is
given
correctly,
so
the
case
was
actually
when
user
gives
a
wrong
name
like
let's
say
he.
He
sends
a
request
by
where
the
header
is
custom,
hyphen,
request
and
python
header,
but.
B
I
see
and
I
guess
this
behavior
will
be
different
if
they
use
whiskey,
vs
ascii.
B
A
So
we
would,
we
would
expect
the
like
the
user,
who
is
like.
Let's
say
I
have
a
ascii
based
framework
that
I
want
to
capture
certain
headers.
I
would
already
be
familiar
with
like
what
the
kind
of
head
of
armaments
right,
so
we
can.
Can
we
expect
the
users
to
you
know
configure
properly
for
the
based,
like
which
kind
of
framework
they're
using
like
if
they're
using
something
like
class?
A
D
Yeah
yeah
we
can,
we
can
document
the
behavior,
but
the
thing
is
yeah
that
will
work,
but.
A
Like
in
custom,
hyphen
request,
hyphen
header,
I
mean
just
document
that
expect
like
assume
that
people
like
the
users
will
be
giving
in
that
format,
but
if
they
somehow
give
it
in
the
underscore
format,
it
still
works
for
the
whisky.
Like
document
the
behavior,
where
I
encourage
people
to
give
it
in
the
hyphen
comment.
Also,
if
we
document
the
like,
it
would
still
be
possible
to
capture
the
request
headers
in
whiskey
based
frameworks
using
underscore
that
should
be
500.
B
Okay,
cool
yeah,
that
makes
sense
it's.
I
don't
think
it's
much
of
a
comparison
between
whiskey
nasty,
it's
like
as
long
as
users
of
whiskey
know
that,
like
they
should
know
the
behavior
of
you
know
headers
that
go
in.
So
if
you
as
only
document
this
and
like
suggest
them
to
use
it
the
right
way,
I
think
it
was
fine.
B
Okay,
also,
if
you
could
add
kind
of
like
a
comment
in
which,
on
the
decision
that
we
talked
about
in
the
in
your
pr
that'd,
be
great.
B
B
Okay,
I
get
it
okay,
it's
just
to
make
it
more
explicit
in
the
spec
right
got
it
cool
yeah,
we'll
try
to
get
viewers
on
this.
I'm
sure
that
people
in
the
spec
community
will
will
have
we'll
be
able
to
help
you
out
here
too
so,
yep
nice.
B
All
right,
well,
okay,
cool,
looks
like
aaron's
in
the
call
anything
on
the
metrics
project
board
that
we
want
to
bring
up
this
week.
E
Oops,
hey:
what's
up
dude
much,
I
didn't
really
get
to
work
on
any
stuff
that
I
was
assigned.
I
saw
diego
opened
a
lot
of
issues
for
spec
compliance,
stuff.
E
Yeah
does
anybody
else
have
have
any
updates
or
put
around
to
work
on
any
of
the.
B
The
stuff,
not
on
my
side,
no.
B
E
One
thing
I
was
going
to
say
was:
I
think
we
had
like
an
announcement
in
the
slack
channel,
but
it
might
be
good
to
make
an
announcement
somewhere
with
a
broader
audience,
to
try
to
get
people
to
try
out
the
experimental,
sdk
yeah,
maybe
either
like
in
our
readme
or
even
potentially
like.
E
Maybe
we
could
do
a
blog
post
with
the
other
experimental
sdk
language.
Sdks,
like
I
think
go,
has
something
I
think
java's
had
something
for
a
while.
Maybe
we
can
get
more
people
to
try
it.
B
Yeah,
I
might
want
to
ask
them
like
if,
if
they
did
any
like
marketing
or
something
specifically
like
yeah
like
java,
and
go
not
only
for
that
for
also
for
logs,
as
well
also
for
logging,
sdk
and
they're
good
yeah,.
B
E
Okay,
cool.
I
know
we
also
mentioned,
maybe
adding
some
instrumentation
to
maybe
like
the
ascii
and
whiskey
plugins
symmetric
instrumentation.
We
should
obviously
be
in
a
lot
of
people.
Don't
have
a
reader
provider
instantiated,
so
I
don't
think
we
have
any
bugs
to
track
that
or
anything
and,
like
I
mentioned
last
week,
the
the
one
tricky
bit
there
is.
If
we
make
any
api
breaking
changes
we
need
to,
we
just
need
to
be
really
careful
because
the
I
don't
think
the
api
version
is
pinned
in
the
segmentations.
E
Right
so
yeah,
I
don't
know
if
anybody
wants
to
work
on
that
there.
I
think
there's
like
really
few
semantic
conventions
right
now
for
for
http
metrics
with
like
the
http
duration,
and
I
think,
there's
a
one
for
the
number
of
concurrent
or
the
number
of
like
in-flight
active
requests.
E
And
then
most
of
the
labels
are
the
same
as
in
trace,
there's
just
fewer
of
them,
so
it
should
be
pretty
straightforward
to
add.
It
would
mostly
just
be
making
sure
the
backward
compatibility
thing
is
handled,
maybe
really
defensively,
adding
a
try
catch
or
something.
So
I
don't
know.
Is
anybody
interested.
B
Right
and
also
the
like
everything
in
the
you
know,
the
the
stable
and
unstable
for
specs
for
instrumentations
applies
to
this
as
well.
So.
B
A
B
Next
time
in
the
sig
and
when
everyone's
here
and
people
can
always,
you
know,
assign
themselves
or
comment
on
the
issue
so
yeah.
I
think
it
is
also
important
to
like
just
try
to
get
these
spec
compliance
stuff
as
well,
so
that
kind
of
should
be
prioritized,
given
our
limited
amount
of
people
who
are
working
on
metrics.
So.
E
Yeah,
I've
looked
through
a
few
of
them,
some
of
them
I
feel
like
we
probably
don't,
have
anything
to
do,
and
then
some
of
them
are
hopefully
things.
B
Cool
nice
all
right,
any
other
issues,
prs
topics
that
people
want
to
talk
about.
B
All
right:
well,
if
not,
you
can
have
40
minutes
back
and
I'll
see
you
guys
next
week.