►
From YouTube: 2022-07-14 meeting
Description
Open Telemetry Meeting 1's Personal Meeting Room
A
A
Cool,
let
me
share
my
screen
real
quick.
A
Okay,
I
guess
we'll
get
started,
so
it
doesn't
look
like
there's
any
topics
right
now.
What
we
can
do
is
go
over
to
the
metric.
Stable
board,
see
what
needs
to
be
done.
A
I
don't
believe
diego
is
here
today
correct.
That's
fine!.
A
I
believe
there
are
some
more
tasks,
especially
related
to
docs.
I
will
be
picking
up
some
this
week.
A
Is
anyone
else
interested
in
kind
of
taking
ownership
of
some
of
that
it
would
help
alleviate
some
workload
from
the
ao.
We
are
trying
to
get
magic
steeple
up
soon,
so.
C
A
Sure
yep
sounds
good
thanks,
okay
cool,
not
much
update
for
that.
Then,
let's
take
a
look
at
issue
open
issue.
D
D
Has
some
idea
about
how
to
deal
with
that?
I
mean
how
to
start
the
mystic
instrumentation
and
stop
it
in
case
of
http
request,
but
he
doesn't
have
any
idea
about
how
to
do
with
that
in
the
web
socket.
So
I
thought,
maybe
you
guys
can
guide
us.
C
C
So
see
the
tracing
work
also
is
broken.
You
will
see
like
there
is
also
a
couple
of
issues
where
people
have
raised
because,
like
it's
creating
a
lot
of
traces
and
then
they
appear
they
they're
kind
of
broken,
and
then
there
is
no
relation
between
them.
So
I
think
that
that
needs
to
be
handled
separately,
so
they
can
go
ahead
and
create
it
with
the
and
the
web
sockets.
D
Sure
do
we
need
to
raise
separate
issue
for
http
and
website,
because.
A
Oh
I'm
assuming
that
right
now,
our
message
to
users
is
that
websockets
are
just
not
supported
for
ascii
instrumentation.
So
as
long
as
we
have
that
clear
message
that
then.
A
We
don't
need
to
support
it
from
metrics
either.
However,
as
zukon
said,
there
are
a
lot
of
like
issues
related
to
that
right
now.
So
to
answer
your
question,
thank
you.
Maybe
we
can
have
it
as
like
a
yeah
like
a
feature
like
a
separate
featured
ticket,
but
I
wouldn't
block
the
the
current
issue
based
off
of
that.
Okay.
D
C
Yeah,
I
was
going
to
say
the
same
thing
because,
like
there
are
like
around
four
to
five
issues
that
talk
about
the
sockets
but
like
how
to
handle
it
better
like
today,
it's
not
really
useful
so
that
we
will.
We
will
take
it
up
in
separate
issue
that
needs
discussion
that
needs.
You
know
some
specification
guidance
around
how
to
handle
that,
because
it's
not
just
you
know
you
send
a
request
and
then
the
request
is
response
coming
back
completed
right
so
yeah,
that's
a
separate,
separate
thing.
A
Hey
thank
you
would
be.
Would
you
be
all
right
if
you
created
the
feature
ticket
and
link
the
original
issue,
as
well
as
the
one
that
sukan
just
linked,
yeah
yeah
awesome
thanks.
A
Cool
all
right,
I
am
going
through
the
prs.
A
Instrumentation
flask,
we
do
have
a
lot
of
yeah
we're
all
on
point
on
the
reviews.
So
it's
pretty
it's
pretty
good.
Okay,.
A
Moving
right
along
to
prs,
I
did
put
this
up
again,
so
I
did
talk
to
trask.
I
did
say
that
we
kind
of
did
want,
because
I
know
you
kind
of
mentioned
this
two
days
ago.
Would
it
be
better
if,
like
we
waited
until
the
metric
stable
release?
So
then
we
don't
have
to
like
always
resolve
all
of
these
conflicts,
and
then
we
can
like
focus
on
like
that,
instead
of
just
like
leaving,
because
I
don't
think
like
like
every
time
like
trask
needs
to
update
this.
A
It's
like
a
like
there's
a
lot
of
like
churn
and
overhead,
and
then
we
gotta
review
it
again.
So
what
do
you
think
like
wait
until
the
repo
is
in
a
stable
state
and
then
start
working
on
it
again?.
C
Yeah,
that's
a
very
point
because
he
has
been.
You
know
doing
this
like
rebasing
this
thing
for
quite
some
time
and
then
we
are
not
getting
it
much
yeah.
So
why
I
suggested
this
is
like
you
know.
We
do
not
have
the
like
two
most
prolific
contributors
like
the
matrix
people
running
here.
So
what
I?
C
What
I
had
in
mind
is
like
it's
going
to
take
some
longer
time
to
release
the
metrics
stable
right
because,
like
they
will
not
be
here
until
20
and
then
even
if
they
come
back,
we
will
you
know
we
will
take
some
more
time.
So
that's
why
I
thought
yeah.
But
again
I
I'm
fine
with
getting
the
stable
release
done
and
then
you
know
go
ahead
with
this
yeah
either
way
it
works.
For
me,
yeah.
A
All
right.
Is
there
any
more
pr's
or
issues
that
people
want
to
bring
up.
B
I
guess
I
wanted
to
bring
up
mine
about
the
resource
json,
oh
yeah,
because
it
it
seems
yeah.
It
seems.
B
Yeah
there
it
is
the
the
way
like
resource
json
works.
Is
it
includes
fields
even
if
they're,
empty
or
or
null,
which
I
I
think
is
like
best,
because
it
leaves
you
know
it
actually
provides
the
most
accurate
information
anyways,
but
the
I
think
it's
format,
resources,
which
is
the
method
above
it
that's
currently
set
to
not
include
null
or
empty,
and
so
in
order
to
use
resource
json.
B
I
had
to
like
actually
like
edit
it
after
the
fact
to
not
change,
basically
to
not
actually
like
break
it,
to
not
actually
have
to
change
tests
and
stuff.
So
I
guess
I
wanted
to
bring
this
up
because
it
seems,
like
some
people
also
thought
that
it
would
be
good
to
print
that
value
regardless
and
so
like
I
can.
I
can
do
that
easily,
but
it
would
require
like
changing
tests
and
and
therefore
you
know,
therefore,
the
output
of
of
open
telemetry
spans
so
or
at
least
the
format
attributes.
C
Yeah,
as
I
mentioned
in
the
comments,
we
do
not
allow
the
non
values,
so
they
will
never
be
there,
but
there
will
be
zero
values
like
zero
on
the
empty
string.
In
the
things
like
that,
we
should
print
them.
We
should
not
skip
them.
What
is
there
like?
What
the
behavior
that
you
are
mentioning
is
currently
buggy.
We
should
like
not
printing
them,
because
you
know
it's
empty.
C
Our
zero
value
is
a
buggy,
because
the
key
is
like
it's
essentially
we're
saying
that
the
key
this
key
and
the
value
never
existed
in
this
detail,
even
if
it
existed,
because
it
has
some
zero
value.
So
we
should
change
that
and
then
we
should
also
update
the
tests.
A
Okay,
also,
if
you
do
like,
if
v
it'll
evaluate
whether
or
not
v
is
truthy
or
not
so,
values
like
zero
would
also
be
ignored,
and
you
really
don't
want
that.
But
that's
a
separate
issue.
That's
a
separate
kind
of
bug
from
also
this
like,
even
if
it
was
empty
right
like
we
still
want
to
print
that
right.
B
Yeah,
okay,
all
right
I'll,
just
remove
that
piece
and
then
change
the
tests.
A
You
are
correct
that
it
is
a
breaking
change,
though,
but
like
we
were
kind
of
taking
the
stance
that,
like
it
wasn't,
it
was
a
bug
in.
A
C
C
So
I
was
going
to
ask
like
if
we
should
create
the
the
pending
like
which
we
haven't
implemented
so
far
and
then
reach
them.
If
we
want
them
to
be
included
in
the
table,
it
is.
C
Because,
like
what
the
project
we
have
is
now
what
we
think
in
the
repo,
but
these
are
from
the
specification
point
of
view
we
should
be
doing
like
the
whether
do
we
want
to
support
this
thing,
or
can
we
postpone
it
after
the
stabilities,
for
example,
there's
a
we
don't
have
we
don't
have
the
http
matrix
exporter
that
can
that's
not
something
that's
blocking
this
table
release.
They
can
you
know
we
can
do
it
later
as
well,
but
there
is
something
like
example
or
support.
C
We
did
not
want
to
include
it
for
the
rcc
and
then
even
the
prototype
implementation
we
have,
but
is
this?
Is
it
something
that
we
wanted
to
be
included
in
the
stable
release
or
not.
A
For
the
spec
compliance
matrix
is
that
remember
we
did
the
same
exercise
for
traces
except
there
are.
There
are
a
lot
of
things
that
were
like.
They
said
that
if
it
exists
in
the
spec
compliance
matrix,
it
means
we
want
it
for
stable.
Is
that
the
same
for
metrics?
Do
you
think.
C
A
A
C
That's
a
separate
component
right,
that's
probably
should
be
included
in
the
stable,
but
let's
say
if
you're
talking
about
the
exporter
like,
if
we
have
one
exporter,
that's
enough
for
the
stable
release.
You
know
the
other
one
can
be
added
later,
but
there
are
some
things
which
are
entirely
missing,
so
bring
like
discussing
whether
to
be
needed
for
the
stable
release
or
not.
C
Okay.
We
should
also
do
that.
I
believe,
since
two
other
maintainers
are
not
here,
we
can.
We
can
do
this
when
they
are
here
or
we
can
create
the
issues
now
and
then
add
it
to
the
project.
C
A
Okay,
yeah
I'll,
probably
ask
josh
both
josh
to
see
if
first,
we
have
to
figure
out
whether
or
not
this
is
a
comprehensive
list
and
then
after,
if
we
get
confirmation
on
that,
or
at
least
like
a
subset
of
this,
then
we
can
create
issues
for
each
of
them
and
determine
like
which
ones
are
they
different,
stable?
Which
ones
are
not.
C
A
Cool
yeah:
it
would
help
to
get
diegos
and
aaron's
input
on
this
too.
So.
C
Okay,
like
if
we
complete
our
like
the
python
matrix
table
project
in
our
report,
doesn't
necessarily
mean
that
we
are
ready
for
this
table
so.
A
A
All
right,
if
that's
the
case
short
meeting
this
week,
I'll
see
you
guys
next
week.