►
From YouTube: 2022-07-19 meeting
Description
cncf-opentelemetry meeting-2's Personal Meeting Room
A
That
was
because
who
car
should
take
off
sick
and
seizures
out
so
I,
don't
like
I,
don't
know
who's
next
in
line.
C
C
C
So
I've
come
to
decide
since
I
just
got
this
new
Macbook
M1
I
can't
run.net
core
3.1
on
it
anyways,
because
that
was
prior
to
arm
support.
So
I
think
we
should
just
drop
it.
C
If
you
guys
have
anything
in
addition
to
this,
this
is
all
I
got.
I.
Think
CJ
said
that
he's
itching
to
do
a
release
this
week
tomorrow,
possibly
I,
think
the
main
thing
that's
like
the
biggest
highlight
of
that
is
the
log
stuff
that
Michael
you've
mostly
been
doing
yeah.
The
pooling,
thank
you
well
is
it
going
to
include
well,
has
any
of
the
log
emitter
stuff
been
landed,
or
is
that
all
in
that
one,
those
PR's
that
you're
breaking
up.
B
C
I
think
that
that's
like
the
main
thing
I
mean
I
did
I
did
a
small
thing
that
I
just
merged
in
today
with
it
with
a
that
additional
overload
of
the
record
exception,
API,
enabling
users
to
add
additional
tags
on
the
activity
event
that
gets
created
for
the
exception.
That's
a
small
thing,
so
people
have
been
asking
for
it.
So
I
think
that'll
be
cool.
C
Yeah
I
still
do
have
one
PR
out
there,
I.
Don't
think
people
have
had
a
chance
to
review
it
that
if
you
have
an
opportunity
to
review.
B
C
This
one
it's
a
wonky
one
I
I-
think
in
some
sense
it
might
be
hinged
on
this
discussion
of
of
whether
we're
dropping
support
for
Diagnostic
Source,
six
I
guess
in
the
sense
that
it
totally
bypasses
the
allocation
free
enumeration
right
now.
C
Currently,
as
the
pr
stands
in
order
to
support
that
I
was
thinking,
you
know
just
take
the
perfect
for
the
people
who
are
configuring,
these
limits
for
now
and
then
once
we
merge
in
the.net7
Branch,
you
know
we
would
obviously
move
to
the
enumerator
functions,
but
I
guess
if
there's
the
possibility
that
we're
gonna
support
diagnostic
Source
six,
which
is
a
big,
maybe
I,
don't
know
I.
D
C
D
B
C
Right
right,
yeah
and
I
think
that
if
we
were
to
go
that
path
and
again
it's
a
big,
maybe
it
would
probably
be
only
compiled
I
imagine
into
like
oh
I,
don't
even
know
what
targets
we'd
have
but
like
we,
maybe
the
nut
standard,
20,
Target
or.
C
Which
I
guess
just
brings?
The
last
thing
is,
though,.
D
C
You
open
Michael,
just
put
in
that
here
in
case
folks,
do
watch
this
Sig
meeting.
Recording
I
I
think
this
is
a
pretty
important
issue
for
folks
to
comment
on
if
it
at
all
concerns
them.
C
I
guess:
I
hadn't,
like
fully
digested
I,
mean
I
knew
this
was
coming,
but
I
I
personally
hadn't
fully
digested
the
the
potential
impact
of
this
until
I
really
started
thinking
about
it
last
week,
and
we
were
we've
been
talking
about
it
since
then,
but
yeah
I
mean
basically
open
Telemetry.
If
you
try
to
pull
in
come
at
seven,
when
we
have
a
stable
release,
targeting
it
open,
telemetry
will
not
compile
in
applications
so
that.
C
People
I
guess
I'm
still
looking
for
folks
who,
who,
like
actual
end
users
who
who
intend
to
speak
with
some,
hopefully
this
week,
just
to
get
their
Impressions.
You
know,
though,
a
lot
of
the
end
users
I
speak
with
that
are
open,
Telemetry
users
given
that
they're
open
Telemetry
users,
they're
kind
of
on
the
they're,
already
kind
of
on
the
bleeding
edge
of
things.
So
it
may
be
all
right.
C
I,
don't
know,
I'd
be
very
curious
if
there's
any
insights
from
like
Microsoft,
like
app
insights,
type
of
teams
or
whatever,
but
sounded
like
she
just
said.
There's
there
may
or
may
not
be.
A
Foreign,
just
a
just
a
very
small
question,
just
to
fill
the
silence
I'm
working
on
the
the
test
coverage,
improvements
and
I
just
had
open
question,
but
what's
the
best
way
to
contribute,
those
I
could
do
like
so,
for
example,
I
found
a
class
that
only
had
like
50
code
coverage,
so
I
can
commit
a
new
PR
to
just
address
like
this
one
class,
but
I
feel
like
if
I
do
one
class
at
a
time,
I'm,
never
gonna
get
done
so
I
was
trying
to
kind
of
bundle
it
up
in
like
10
or
12
classes
and
submit
them
all
at
once,
but
I
also
don't
want
to
overload
reviewers
with
an
unreasonably
large
PR,
so
just
kind
of
curious
for
some
conversational
feedback
like
what
are
your
guys
Impressions
on
that?
C
A
So
would
you
prefer
to
see
more
small
ones
like
this
and
then
the
risk
is
that
well,
there
could
be
10
or
12
PR's
open
or
like
just
one.
You
know
bigger
PR
with
a
bunch
of
small
test
classes,.
C
A
Like
that,
I
don't
think
it's
gonna
really
break
up
across
themes.
I
can
try
to
break
it
up,
based
on
like
namespaces.
It's
like
right
now,
I'm
looking
at
the
base,
the
open,
Telemetry
library
and
the
the
code
Club
website
sometimes
doesn't
cooperate
with
me,
but
I'm
digging
around
through
just
classes.
So
it's
oh,
where
did
I
put
my
notebook
things
I'm
looking
at
this
week
and
it's
a
goal
of
having
a
PR
by
the
end
of
the
week.
A
I'm
looking
at
pooled
lists
and
sampling
result,
I,
think
sampling
result,
isn't
a
tracing,
namespace
and
pulled
list.
I
think
is
in
the
internal
and
they
both
had
low
test
coverage.
Those
classes
aren't
really
related
to
each
other,
except
that
they're
in
the
same
Library.
C
I
mean
if
they
get
really
hairy
and
and
large,
you
know
I'd,
maybe
keep
it
in
the
back
your
mind
how
you
might
split
them
up,
but
like
I
I,
get
it
like.
This
word
doesn't
always
split
up
very
nicely,
so
you
know
it
is.
It
is,
and
is
it
mostly,
is
it
mostly
like
adding
additional
test
coverage,
or
is
it
a
lot
of
like
refactoring
his
existing
test
coverage.
A
All
new
tests
I'm
not
like,
there's
nothing
to
refactor
it's
yeah
gotcha,
like
some.
Some
of
these
smaller
helper
classes
or
base
classes,
looks
like
they
got
tested
just
like
as
a
consequence
of
running
other
tests,
but
they
never
had
like
an
explicit
test
class
targeting
themselves.
Sure.
C
Is
like
trying
to
figure
out?
Oh.
Does
this
test
test
the
same
thing?
Is
it
used
to
type
of
thing
so
yeah
new
tests?
B
I
think
your
your
idea
to
keep
the
pr
small
is
good.
If
it's
sitting
out
there,
no
one
is
reviewing.
It
don't
be
afraid
to
just
like
Ping
me
around
on
slack
and
say:
hey:
can
you
guys
review
this?
It's
ready!
That's
typically
like.
If
I
see
a
PR
come
up,
that's
super
interesting
I'll
go
and
look
at
it,
but
generally
I
wait
till
people
are
reviewing
it
or
someone
has
asked
me
or
it's
I
see
like
it's
being
ready
to
be
merged
and
I'll
go
and
look
at
it.
A
Yeah
and
Alan
did
catch
this
one
in
particular
like
pretty
early
and
got
me
unblocked
on
one
of
the
the
the
validation
sea
ice
and
then
I
had
Pink
Siegel
for
review
where
he
took
off
for
training,
and
my
new
car
show
review
and
he
took
off
today
so
I'm,
just
kind
of
like
finding
anybody.
I
can
and.
C
B
B
And
then,
in
the
case,
like
I
had
a
super
big
one,
it
was
like
hundreds
of
files
over
the
last,
like
three
four
weeks,
I've
been
breaking
that
up
into
a
bunch
of
smaller
ones,
just
because
it's
it's
much
easier
to
steal.
You
know
a
few
moments
of
time
from
people,
as
opposed
to
like
a
whole
day
going
through
some
massive
things.
A
B
It
helps
to
to
have
the
big
one
to
kind
of
show
like
here's
the
whole
Direction
I'm
going
and
get
people
to
kind
of
look
at
it
and
get
some.
You
know
high
level
feedback
and
then
split
it
into
smaller
ones.
With
tests,
that's
they're
more
self-contained
for
like
deep
review.
It
was
just
I
think
in
that
case
it
was
easier
to
get
it
in
smaller
strategically,
even
though
it
was
more
work.
I
had
to
keep
like
merging
all
those
changes
back
to
the
Big
Branch
right.
A
D
I
have
a
quick
question.
Sorry,
if
yeah
I
was
just
a
couple
of
minutes
late
did
we
say
we
are
going
to
merge.net7
Branch
into
main,
or
we
are
waiting
for
the
issue
to
be
resolved.
I
I
didn't
catch
that.
C
We
haven't
talked
about
that
yet
again
this
week,
though
I
think
that
I
think
it's
I
mean
I
personally
would
like
to
see
emerged
but
yeah
this
this.
The
discussion
for
dropping
three
one
five
support
is
a
discussion
that
I
think
we
should
just
settle
on.
But
hopefully
we
can
do
that
this
week.
I
I'm
interested
in
CJ's
input
particularly
also
I,
think
that,
from
the
standpoint
of
the
I'm
actually
mostly
interested
in
seizures
input
on
these
suggestions
from.
C
C
Well,
actually,
no
it's:
where
is
it
some
suggestions,
I
think
to
the.net
team,
basically
saying
like
hey,
you
know
what,
if
we
were
to,
maybe
just
do
a
warning.
I
think
you
had
mentioned
this
before
Michael.
C
You
know
change
it
to
a
warning
rather
than
an
error,
or
if
it
is
an
error,
you
know
making
it
more
descriptive,
actually
stay
saying
that
you
can
override
this
Behavior
I
think
that
if
we
were
to
do
this,
if
this
is
the
path
that
we
did
like
make,
this
same
kind
of
error
message,
which
is
one
of
the
one
of
the
proposals,
had
our
own
poison
pill
to
the
targets
file.
C
If
we
were
to
do
that,
then
I
would
like
to
see
that
kind
of
like
work
around
articulated
in
it.
But
anyways
to
your
question
about
the.net7,
Branch,
I
I
think
it
would
be
nice
to
have
an
answer
to
this,
even
if
it's
not
the
final
answer,
but
like
just
an
answer
for
when
we
do
that
next
release
with
the.net
7
Branch
merged
in,
what's
going
to
be
our
story.
A
B
I
agree
with
that
I'm
I'm
skeptical
of
it.netball
they're
not
going
to
go
back
into
the
packages
and
re-add
the
support.
They're,
probably
not
going
to
want
to
support
out
of
support,
run
time,
so
it'll
just
be.
Will
they
tweak
the
tooling
or
the
error
becomes
a
warning,
but
from
our
perspective
we've
already
removed.
B
C
C
Yeah
I
suspect
it's
kind
of
like
a
the
ship
is
sailed
type
of
thing,
but
I
do
think
that
I
I
do
hope.
At
least
that.
D
C
The
the
verbiage
of
that
warning,
or
that
error
I
guess
is,
is
improved
both
on
our
part,
because
I
think
we
should
do
as
you
suggested
at
our
own,
but
also
on
the.net
teams,
part
so
fingers
crossed
that's
a
thing.
I,
don't
know
if
folks
saw
this,
but
I
did
add
a
some
data
that
I'd
collected
from
the
perspective
of
New
Relic.
There's
still.
This
is
not
open.
C
Telemetry
users
by
the
way
like
this
is
this
is
people
that
use
new
relics,
.net
agent,
but
50
of
the
users
are
running
dot
net
core
three
one.
So
I
don't
know!
That's
it's
even
though
this
poison
pill
was
introduced
with
this
ship
ing,
of.net6
I
think
it's
going
to
be
ever
more
prevalent
now
with
the
shipping
of.net7
just
because
so
many
more
libraries.
A
C
Not
said.net6
was
not
actually
that
far
behind
so
I
found
that
kind
of
like
a
a
nice,
a
nice
thing.
C
But
refresh
the
actually,
while
we
were
talking
about.net
seven,
the
issue
you
opened
I
think
is
an
is
an
interesting
one
for
us
to
mull
over
in
the
in
the
near
term.
Here.
C
Yeah,
it's
this
one,
so
yeah
with
asp.net
core
and
HTTP
client
libraries
using
activity
Source
like
that's
that's
cool,
but
as
it
stands
now
with
these
changes,
the
instrumentation
Library
name
and
version
I
think
right
are
going
to
be
the
name
and
version
of
the
of
the
actual
libraries
themselves,
which
are
still
not
necessarily
the
things
that
are
producing
the
Telemetry.
C
D
Yeah
so,
as
far
as
like
I
understand,
I
think
like
one
of
the
points
and
you
made,
that
is
the
plan
going
to
be
like
the
the
libraries
adding
the
tags
onto
activity
so
that,
like
we,
don't
really
need
the
instrumentation
Library
I
think
that
that
was
one
of
the
things
that
you
wanted.
Some
information
on
yeah
as
far
as
I
know,
I,
don't
think.
D
There's
such
a
plan
there,
but
I'll
I'll,
like
double
check
with
see
Joe
I,
don't
know
like
plans
you
if
you
know
something
about
that,
but
like
I
had
initial
conversation,
it's
a
joke
because,
like
then,
it
involves
the
complexity
of
handling
the
different
versions
of
the
semantic
conventions
which
dotnet
may
or
may
not
handle
like
it
can
change
right,
so
yeah.
D
But
this
this
is
an
interesting
thing,
because,
right
now
we
we
have
the
activity
source
that
we
attach
with
the
instrumentation
library
and
version
which
we
won't
and
the
version
I
I,
don't
think
they
are
populating
right
now
like
it,
doesn't
really
tell
the
HTTP
client
version
as
well
or
the
ESPN
network
version
as
well.
D
So
maybe
like
would
that
help?
If
we
can
have
the
versions
of
the
the
libraries
I
don't
know
like
I
mean
it's
still
not
going
to
help,
because
we
can
have
a
different
version
from
the
instrumentation
Library
I'm,
just
like
trying
to
think.
If
what
could
be
the
alternative
solution
here
that
we
like.
If,
if
we
need
some
help
from
the.net
team
we
can
or
like
from
the
instrumentation
Library
HTTP
client
and
the
SB
Network
team,
we
can
ask
them
so
yeah.
D
C
I
was
just
gonna
say
it's
probably
not
a
hugely
pressing,
pressing
issue
at
this
point
in
time,
namely
because
I'm
in
the
semantic
conventions,
I,
don't
know
when
they're
going
to
be
stable,
I
mean
it
doesn't
seem
like
it's
going
to
be
anytime
soon,
so
it
in
that
regard.
You
know
it
makes
sense
to
me
that,
like
the.net
team
is
not
gonna,
take
on
the
work
of
of
decorating
the
Telemetry
themselves
on
based
off
of
unstable
semantic
conventions,
but
longer
term
I
think
it's
the
I.
C
D
C
Yeah
instantiating
a
tracer,
giving
it
a
name,
a
version
and
then
the
schema
URL
is
supposed
to
be.
The
thing
that
can
be
interpreted
by
you
know
back
ends
that
are
receiving
the
Telemetry.
At
least
this
is
one
use
case
for
a
schema
URL.
You
know
a
back
end
being
able
to
use
that
to
understand
the
structure
of
the
Telemetry
from
that
from
that
Tracer
and
potentially
convert
from
one
version
of
the
schema
to
another,
so
obviously.net
not
supporting
us
today.
C
One
two
like
the
specification
on
schemas,
is
still
very
much,
not
settled
I,
don't
know
how
pressing
this
is,
but
I
think
it
at
some
point
in
time.
If,
if
this
repository
or
the
instrumentation,
that
generates
the
actual
like
attributes
for
asp.net,
core
and
HTTP
client
is
always
managed
by
open
telemetry.net
repository
that
I
think
that
it
does
make
sense
at
some
point
to
figure
out
what
we
actually
want
to
populate
these
fields
with,
because
I
think
it
would
be
most
accurate
to
populate
it.
C
D
Yep
makes
sense
to
me,
so
we
are
looking
at
like
two
things
right
like
the
one
is
the
schema
URL
and
the
other
one
is
the
version
that
is
the
instrumentation
Library
version
that
we
should
be
able
to
populate
or
pass
the
information
onto
the
backends.
D
All
right,
yeah,
cool,
yeah
I'll,
see
like
I'll,
try
to
chat
with
sijo
and
see
like
what's
the
plan
if
he
he
has
any
knowledge
about
that
already.
If
there
is
anything
in
the
pipeline
from
the.net
side
or
if
that
is
something
that
we
should
start
having
conversation
about.