►
From YouTube: 2022-08-03 meeting
Description
cncf-opentelemetry meeting-2's Personal Meeting Room
C
A
Okay,
let's,
let's
get
started,
if
you
have
any
topic
you
want
to
discuss,
please
add
it
to
the
list.
A
I
don't
really
have
any
big
updates
today
or
any
new
topics,
but
I
thought
you
know
I
thought
maybe
giving
a
quick
status
update
would
be
useful.
What
do
you
think.
A
Just
overview
of
where
we
are
so
in
my
mind,
we
have,
we
have
a
few
things
in
in
progress
and
correct
me
if
I'm
wrong,
but
I
think
we
have
four
major
streams
of
work.
A
One
is
one
is
the
work
on
on
the
api
for
for
events
and
logs,
the
other
one
was
defining
semantic
conventions,
and
then
we
have
a
discussion
about
sessions
and
have
you
started
the
work
on
the
optimized,
sdk
yeah?
So
for
the
events
events
api
there
is
there
is
the
santosh
has
the
spec
pr
in
progress.
I
think
that's
that's
moving
along
and
there
is
I.
I
worked
on
a
draft
implementation
of
the
of
the
api
in
in
javascript.
A
So
that's
there's
an
pr
now
in
the
javascript
javascript
repo,
it's
in
draft
mode,
I'm
waiting,
santosh,
I'm
waiting
for
the
api
is
back
to
be
merged
until
and
and
then
then
I'll
make
the
then
I'll
make
the
the
pr.
C
A
But
I
I
kind
of
got
got
ahead
of.
I
think
I
got
a
little
bit
ahead
of
it
like
typically
like
the
spec,
the
spec
will
be
merged,
merged
issue
will
be
created
and
then
then
you
create.
You
know
the
pr
like
that's
for
that
issue.
A
I
you
know
I
started
working
on
the
api
as
to
help
us
maybe
like
with
have
example
implementation,
but
I
think
it's
pretty
straightforward.
So
I'm
just
at
this
point
waiting
for
the
spec
to
be
finalized.
Okay,.
C
Can
you
send
me
the
link
to
the
pr
I
I
want
to
look
at
some
sample
usage
of
your
api.
You
know
maybe
in
the
tests
in
the
form
of
tests,
I
want
to
link
it
to
my
pr
for,
for
reference.
A
A
So
that's.
I
think
that
stream
is
pretty
straightforward.
I
think
we're
just
waiting
for
that
spec.
The
second.
A
For
the
semantic
conventions
there
is,
we
worked
on
the
two
navigation
and
resource
timing,
resource
timing,
specs,
and
I
think
also
I'm
here-
I'm
waiting
for
the
spec,
because,
because
the
the
spec
for
the
api
includes
the
event
event
name
space
so
before
I
open
like
additional
prs,
like
with
semantic
conventions,
I
I
want
that
to
be
defined.
C
You
can
actually
open
a
pr
against
your
own
repo
in
your
personal
reports.
That
way,
you
know
we
can.
This
group
can
comment.
A
So
I
I
did
so
let
me
just.
C
A
Okay,
so
and
then
then
we
have
the
sessions
discussion
and
that's
currently
blocked.
I
think
because
of
the
discussion
on
the
ephemeral
resources.
A
So
I
think
I
think
we
need
with
that
ted's
absence.
I
think
we
need
someone
from
tc
to
help
us
with
this.
D
D
D
It
really
is
a
case
of
like
this.
First
pr
is
just
the
script
to
merge
the
history.
I
have
a
second
script,
which
I
I
reference
some
of
the
stuff
in
my
notes,
for
that
one
about
how
to
get
the
into
the
form
that
we
want
so
yeah.
It's
going
to
raise
it
with
daniel,
because
I
need
I've
got
you
and
daniel
listed
as
a
reviewer
for
this,
but
we
need
to
there's
some
names
in
there.
We
just
need
to
agree
on.
D
A
The
the
only
other
thing
that
I
can
think
of
in
this
list
is
you
know.
Most
of
this
has
is
focused
on
browser.
There
is,
there
can
be
a
separate,
separate,
maybe
topic
for
mobile,
and
I
know
that
we
have
some
folks
in
this
in
this
group
that
are
interested
in
mobile
and
santosh,
and
I
talked
about
it
this
little
bit.
A
C
On
the
topic
of
mobile
sirath,
I
have
a
question
for
you,
so
earlier
john
used
to
work
on
the
android
sdk
and
he
had
mentioned
that
splunk
is
open
to
contributing
the
android
sdk
to
open
telemetry,
but
they
don't
have
resources
themselves
to
do
it.
C
So
I
want
to
revisit
that
conversation
with
whoever
is
the
current
maintainer.
You
know
in
in
couple
ways
you
know
one
is
to
to
to
more
to
port
that
code
to
open
telemetry.
It
basically
involves
separating
splunk,
specific
stuff
out.
C
Firstly,
you
know
we
need
to,
you,
know,
agree
on
it
and
figure
out
any
formalities
with
the
hotel
dc
and
and
then
secondly,
you
know
those
folks.
You
know
it
would
be
great
if
you
know
they
can
join
once
in
a
while.
You
know
this
meeting.
B
Yeah,
let
me
take
that
as
an
action
item
and
back
to
you
next
week.
Okay,.
C
For
ios,
so
there
is
already
a
group
and.
C
B
Okay
and
then
have
we
also,
I
know,
might
be
jumping
the
gun,
but
like
dark
and
flutter.
C
So
yesterday
I
think
somebody
posted
a
adult
implementation,
yeah
in
in
the
slack
channel.
I
I
don't
know
which
company
he's
from,
but
I
think
we
could
ask
them
again.
It's
the
same
process
right.
I
think
we
need
to
figure
out.
We
need
to
have
you
know
these
newer
language
implementations
to
be
just
part
of
hotel
and
not
unless
you
know
they
are
open
source
anyways,
but
you
know
they
should
just
open
it
open
the
reports
in
the
open,
telemetry
project.
You
know
github
namespace,.
B
E
B
Work
but
but
having
said
that,
I
think
we
need
to
have
that
conversation,
whether
it's
splunk
sort
of
upstreaming,
the
android
implementation
or,
if
it's
or
we'll,
keep
our
doing
open,
telemetry
dart.
So
we
can't
find
a
solution
here.
C
Yeah
actually
at
this
point
we
have
literally
zero
representation
from
mobile
side
yeah.
So
any
one
audition
would
be
a
you
know
great
progress
now
any
anybody
they
can
join
like,
for
example,
there
was.
There
is
a
discussion
in
the
on
the
events
api
pr
to
to
basically
reserve
the
event
domain
for
the
different
domains
and
for
the
browser
you
know
we
are
going
with
browser
for
the
mobile.
C
You
know
it's
recommended
that
you
know
we
use
device
and
there's
some
contours
here
on
that,
so
it
will
be
so
we
we
don't
want
to
take
decisions
on
it
unless
there
are
people
from
mobile.
You
know
that
say
something
about
it.
C
Okay,
that's
good
to
know,
but
do
you
deal
with
android
ios
about.
E
So,
basically,
I'm
not
actively
maintaining
it,
but
dealing
with
the
teams
that
maintain
ios,
android
and
also
back
native
flutter
and
submarine.
C
C
Okay,
so
stephen
would
would
you
like,
would
your
team
be
available
to
help
with
porting
the
android
sdk
to
open
telemetry.
E
I
would
need
to
ask
so
I
have
to
clarify
that
from
resources,
and
I
will
note
this
as
action
item
and
I
can
let
you
know
next
week,
then,
okay,
other
than
that,
I
would
highly
appreciate
the
peering
partner.
So
you
know
for
discussions
because
otherwise
it's
to
die,
not
race,
biased.
I
have
to
admit.
C
No,
no,
there
are
folks
from
our
side
as
well
from
from
cisco.
It's
just
that
they're
they're
new,
so
so
I
they
wouldn't
be
able
to
lead
it
so,
but
we
will
be
able
to
participate
for
sure.
That's
great
yeah.
B
A
D
Okay,
so
I
guess
that
that
moves
us
to
the
next
one,
which
is
rand's
document,
of
course,
he's
not
here,
so
I
can
talk
to
it.
So
the
intent
of
this
stock
is
to
effectively
bootstrap
the
process
of
starting
to
identify.
D
What
are
all
the
events
that
exist
in
the
rum
space
that
we
want
to
try
and
define
as
part
of
this
this
group,
so
the
the
first
table
is
really
just
identifying
the
domain
name
of
the
event
and
then
listing
by
vendor
if
they
have
an
equivalent,
what
they
call
it.
D
That's
really
what
this
first
table
does
and
then
from
there
on
getting
each
vendor
just
to
start
defining
what
are
the
fields
for
their
particular
event,
so
that
we
can
then
look
at
them
and
go
okay.
Well,
a
if
it's
a
common
event
across
or
all
vendors,
then
it's
probably
definitely
a
a
potential
for
to
define
an
hotel
and
then
b.
You
know,
let's
find
the
common
ground.
D
I
think
we
do
need
someone
from
elastic
ram.
Had
a
quick
look
at
the
elastic
common
schema
yesterday,
but
he
couldn't
find
a
list
of
events
so
while
they
have
all
their
fields
defined,
he
was
really
trying
this
from
an
event
perspective.
D
He
does
intend
to
drive
this
across
there
so
called
into
a
meeting
this
morning.
So
yeah
we've
talked
about
this
a
few
times
and
we've
made
no
traction.
So
this
is
an
attempt
to
try
and
get
some
traction.
A
So
we
did
a
little
bit
of
this
work
in
the
past.
D
Yeah
in
they
did
the
events
api
one.
I
think
it
was
oh
this
one,
okay
yeah,
I
think
john
worked
on
originally.
C
No,
no
martin
me
and
quinn.
We
provided
info
in
one
document
that
martin
had.
A
Yeah
we
had
this,
so
it's
linked
in
this
so
yeah
in
this
in
in
the
top
here,
symmetrical
budgets
work
in
progress
and
yeah.
There
are
two
links
up
here
and
maybe
I'll
add
this
address
document
here
too.
D
Yeah
but
yeah,
this
is
coming
from
a
slightly
different
perspective
rather
than
starting
to
define
it.
It's
less,
let's
start
to
identify
them.
First,.
A
Yeah,
so
we
we
did
a
little
bit
a
little
bit
of
this
in
this
document.
In
this
spreadsheet
there
is
there
like.
There
are
some
attributes
that
I
added
from
you
neurologic
perspective.
I
think
santosh
steam
added
for
their
team
and
for
aws.
A
A
C
But
talking
about
the
contents
nev,
if
you
think
so
like
prior
to
open
telemetry,
each
of
us
used
to
have
only
events
right,
we
used
to
call
everything
as
event
yep
and
now
we
have
spans
and
events,
and
so
some
of
the
objects
that
we
used
to
call
as
events
and
you
know
are
now
you
know
spans
like,
for
example,
you
know
a
span
representing
ajax
call
right.
It
used
to
be
an
event,
and
you
know
now
it's
a
span.
C
So
do
you
want
us
to
put
everything
here
and
not
just
basically
put
both
the
different
span
types
and
the
event
types.
C
A
B
A
Are
these?
Are
these
attributes
like
event,
event,
attributes
or
or
resource
attributes,.
D
Ram's
not
calling
them
out
at
the
moment
he's
just
identifying
the
field
names.
Okay,
so
looking
at
these,
though
these
are
all
like
they're
they're
event
ones.
So
for
the
for
the
particular
page
view
event
yeah,
that's
just
the
list
of
some
of
the
stuff.
We
collect,
that's,
not
even
all
of
what
we
collected.
D
A
D
Yeah,
so
that's
really
where
the
the
the
the
whole
point
of
trying
to
list
it
all
together
is
it's.
In
one
place
we
can
start
having
that
discussion.
Yeah
yeah.
A
Totally
yeah
like,
for
example
like
we
have
at
neural,
like
we
have
a
page
view
so
very
similar,
but
I
was
thinking
that,
like
the
navigation
event,
would
basically
give
you
all
that.
B
B
D
And
there
is
going
to
be
a
common
set
and
there's
going
to
be
vendor-specific
sets
so
yeah.
We
just
need
to
identify
them
and
figure
out
okay.
Well,
we
can
define
an
hotel
and
it
has
this
meaning
and
then
we
go
and
do
our
own
mappings
or
whatever.
We
need
to
do.
Okay,.
A
Okay
yeah,
so
we
have
so
we
have
some
some
overlap
here
in
this
work.
Yeah
and
like
with
this
dark
as
well,
and
I
think
we
also
had
in
that
original
plan
doc,
we
had
a
list
of
events.
B
D
Yeah
that
sort
of
all
started
to
bring
together,
I
think,
ram,
wanted
the
the
first
table
to
effectively
start
from
the
vendor
perspective,
so
that
so
that
we
can
identify
what
our
common
ones
are,
because
really
any
of
the
common
events
are
the
ones
we
really
want
to
concentrate
on.
Okay,.
A
D
A
Okay,
going
back
to
this,
am
I
missing
something
like
from
this
list
like?
Does
that
resonate
with
everyone?
That's
where
we
are.
A
A
I
wanted
to
ask
kirby:
have:
is
there
any
update
on
the
protobuf.
B
I
was
on
call
last
week,
so
I
haven't
had
much
time
hoping
to
pick
it
up
before
the
next
sick
meeting.
So
should.
A
D
Though,
just
fyi
that's
linked
to
my
that
initial
merge
pr.
So
just
so,
we
had
a
link
there.
Okay,
so
I've
got
a
bunch
of
comments
which
I
added,
because
when
I
make
pr's,
I
like
to
comment
on
them
actually
now
the
spec
one
okay.
Now
this
is
the
the
nested
attribute
spec.
D
Okay,
I
updated
it.
Yesterday
there
was
a
table
talking
in
the
meeting
yesterday
the
heading
was
optional.
I
had
the
yes
no,
which
made
it
seem
like.
It
was
wrong
around
so
I'll
change
that
to
support
it,
didn't
really
want
to
call
it
supported,
because
by
saying
spans,
don't
support
nested
attributes
I
to
me
it
feels
like
it's
closing
the
door
rather
than
leaving
it
open,
but
following
tigran's
advice
of
just
trying
to
bite
it
off
in
smaller
chunks,
just
so,
we
can
get
it
in.
That's
what
I've
been
doing.
D
Okay,
really
at
the
end
of
the
day,
if
this
pr
doesn't
get
merged,
it
doesn't
really
affect
us,
because
logs
already
supports
nested
acreage.
So
I
was
just
trying
to
clarify
it.
So
the
potentially
open
the
door
or
crack
for
defining
span
events
more
like
log
events,
so
we
can
unify
them.
B
A
Okay,
one
quick
thing:
I'm
not
going
to
be
here
next
week,
actually.
A
So
if
for
the
discussion
with
ram
but
and
fields
and
attributes,
maybe
you
can
we
can
schedule
it
for
the
following
following
week.
D
C
Alternately,
I
think
we
could
put
since
some
of
us
have
already
put
the
data
in
some
data
in
martin's
excel
sheet.
You
know
the
rest
can
do
the
same
there,
and
then
you
know
if
there
are
any
additional
fields
required
additional
form
required
for
the
data.
You
know
you
could
add
those
fields.
C
D
Yeah,
he
actually
just
pinged
me
said
he
couldn't
make
it
today,
but
he's
available
next
tuesday
and
he's
suggesting
that,
yes,
we
can
do
it
on
slack,
so
yeah.