►
From YouTube: 2022-06-29 meeting
Description
cncf-opentelemetry meeting-2's Personal Meeting Room
B
This
is,
let's
see
this
is
seattle
right
here
and
then
this
is
kind
of
the
this
is
pewd,
keep
doing
the
wrong
one.
This
is
puget
sound
and
then
this
is
bainbridge
island.
Where
I
live.
B
B
A
The
first
of
july
is
a
canadian
holiday.
I
guess
we
get
that.
D
A
Hi,
everyone
there's
nothing
in
the
agenda.
I
guess
it's
going
to
be
a
continuation
of
what
we
were
talking
about
last
week
right
so.
A
D
Well,
yes,
and
no
so
yes
in
the
sense
of
we,
we
discuss
different,
like
patterns
and
in
the
java,
sig
and
and
what
options
might
be
available
and
no
in
the
sense
of
like
I
didn't
dive
deep
into
the
history
of
the
specification
and
figure
out
why
that
was
the
the
noble
notable
thing
you
know
we,
I
guess
rediscovered
or
are
about
spans.
D
Is
that
so
the
there's,
a
different
interface,
that's
passed
to
the
processors
than
is
passed
to
the
exporter.
So
there's
a
readable
span
or
a
read
and
writable
span.
That's
sent
to
the
the
span,
processors
and
it's
that
the
exporters
have
a
span.
Data
interface,
which
is
which
is
different.
And
so
you
know
it's
up
to
the
processors
to
convert
a
read
or
bull
span
to
expand
data,
which
is
the
fully
immutable
version
of
the
interface.
D
The
processors
get
access
to
the
writable
version
yeah.
So
it's
it's
it's
a
bit
nuanced,
because
the
span
processor
has
access
to
the
writable
version
on
start
of
the
span
and
the
and
only
a
readable
version
at
the
end
of
the
span.
So
by
the
time
the
span
ends.
D
The
this
fan
is
immutable,
but
you
know
they
they
do
have
access
to
a
writable
version
at
the
start,
and
so
I
think
we
could
have
a
similar
pattern
for
logs
where
log
emitters
have
access
to
a
well
there's
still
there's
still
like
a
bit
of
a
tricky
piece
there.
So
log
emitters
have
access
to
some
sort
of
interface
that
allows
them
to
mutate
the
log
data
and
then
the
final
log
processor
in
the
chain
is
responsible
for
calling
two
log
data
and
passing
that
to
the
exporters
yeah.
A
Okay,
so
I
guess,
let's
give
it
a
bit
more
time.
I
guess
if
you
want
to
have
a
deeper
look
into
what
the
what
is
happening
on
the
tracing
side,
and
maybe
we
can
discuss
it
next
week.
If
you
want-
and
you
have
more
information
in
the.
A
Okay,
the
vo.
D
A
I
guess
so
I
don't
see
any
blockers
there
to
be
honest,
anyone
blocking
we.
We
have
a
few
approvals
already,
so
I
personally
want
to
re-read
it
one
more
time
and
then
I
will
approve
it.
I
don't
know
if
we
need
to
discuss
anything
specifically.
A
A
So
please
do
please
do
review
guys.
You
you
in
this
room
are
the
most
knowledgeable.
I
guess
about
the
intent
of
that
app.
It
doesn't
matter
whether
you're
an
approver
or
not
on
the
spec,
your
approval
still
counts.
It
means
that
one
more
knowledgeable
person
did
read
it
and
thinks
that
it's
a
good
idea.
So
please
do
approve
if
you
agree
with
it,
and
I
will
do
that
myself
as
well.
E
A
Either
way,
I
think
it's
fine,
we
we
do
need
some
sort
of
qualification
there,
because
you're
so
right
there
were.
There
was
some
confusion
around
what
what
is
a
domain.
How
is
it
different
from
the
scope?
A
Maybe,
let's,
let's
do
it
when,
when
the
all
type
is
merged-
let's,
let's
maybe
focus
our
our
efforts
on
doing
that.
I
I'll
try
to
review
my
do
my
final
review
today
and
then
improve
unless
I
finally
find
something
that
somehow
I
disagree
with
and
then
I
think
in
that
case
we
will
have
formally
we
we
already
have
formerly
four.
E
B
A
But
yeah,
maybe
we
can
give
it
a
bit
more
time
and
also
there
is.
I
don't
know
if
there
is
any
open
comments
remaining.
So
if,
if
there
is
anything
that
is
still
unresolved,
any
comments
please
actively
resolve
them
like.
A
If
there
is
anything
to
be
answered,
please
answer:
if
there
is
anything
remaining
there,
if
you're
waiting
for
the
for,
I
guess
a
response
from
the
person
who
created
the
comment,
make
sure
to
kind
of
drive
the
discussion
so
that
it
doesn't
remain
open
right
so
because
the
rule
is
that
yeah
all
the
conversations.
E
Be
resolved
yeah
now
that
I
have
removed
a
lot
of
the
you
know,
the
interface
details
itself,
the
api
interface.
Most
of
the
comments
are,
you
know,
they're
gone.
I
think
they,
I
will
revisit
them.
You
know
during
the
spec
changes
yeah.
So
right
now
for
the
current
contents,
the
the
one
that
needs
clarification
is
the
domain.
E
A
I'm
going
to
be
out
tomorrow
and
on
friday
I'll
try
to
give
my
approval
today,
and
maybe
someone
else
can
merge
it
afterwards.
If
not,
then
I
can
do
it
on
monday,
probably
it
it
maybe
makes
sense
to
keep
it
open
for
a
bit
more
because
you
did
make
make
some
significant
changes.
I
guess
by
removing
the
beats.
A
C
Yeah,
the
there's,
I
think
I've
got
one
comment
with
one
thumbs
up
that
sort
of
says,
basically
that
I'm
suggesting
that
all
other
signals
need
to
be
nested
and
that's
you
know,
the
wording
is
should
support,
it
doesn't
say,
must
support.
C
So
there
is
one
change
I
do
need
to
make
and
that
I
did
have
javascript
is
supporting
nested
attributes,
but
on
rereading
they
don't
so
I
need
to
make
that
change,
but
I
was
waiting
for
other
comments
to
come
in
before
I
push
out
an
update,
but
I'll
keep
bringing
it
up
at
the
spec
meeting
and
trying
to
drive
and
add
comments.
A
C
Yeah
yeah,
it's
a
case
like
we
have
it
for
logs,
like
the
oh
chap
for
spans
we're
going
to
be
defining
events
as
having
them.
Ideally,
we
want
at
least
span
events
to
support
them,
so
we
can
define
the
event
once
and
it
gets
used
in
both
places,
but
yeah
we'll
just
go
see
where
we
lead.