►
From YouTube: 2020-08-31 C/C++ SIG
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).
C
We
can
wait
for
a
couple
of
minutes.
I
I
think
this
time
it
should
be
a
quiet
meeting,
because
I
I
don't
expect
a
lot
of
interns.
Most
of
the
interns
should
be.
D
A
Location
was
great,
so
I
was
for
four
days
without
phone
and
computer,
so
it
was
very.
E
C
Okay,
I
I
think
we
can
start
I.
I
hope
this
meeting
will
be
very
short
because
most
of
the
interns
have
done
their
project
and
should
be
ready
to
like
prepare
for
their
school
stuff
measurement.
Talk-
and
I
do
have
some
topic
about
like
how
we
can
make
this
product
more
formal,
because
we
we
do
want
to
get
this
to
a
more
like
production
by
the
state.
C
This
okay,
so
so
the
first
topic
I
I
want
to
see
like
if
folks
are
okay
with
having
an
alternative
time,
so
in
c
sharp
and
the
gold
language
it
seems
like
folks,
have
been
practicing
for
a
while
and
the
open
telemetry
python
we
used
to
working
that
way,
but
now
we
moved
back
because
most
of
the
community
members
are
in
the
same
or
similar
time,
though,
for
c
plus
plus
the
reason
I'm
looking
for
this
change
is
because
I
do
have
a
developer
in
the
team
who's
in
the
ist
time
zone,
so
8
am,
would
be
probably
7,
30
p.m
for
him.
C
Otherwise
it's
like
3
p.m.
It's
like
like
2,
30
a.m,
which
is
impossible
so
I
like.
Hopefully
I
can
eventually
make
him
a
like
approval
and
maintainer
for
this
product,
so
he
will
be
spending
most
of
his
time
dedicated
on
this
product
and
he
used
to
be
a
like
c
plus,
plus,
like
low-level
developer,
working
on
like
android,
ios
or
stuff,
and
have
experience
on
the
telemetry
sdks,
so
just
want
to
see
like
they
folks
think
atm
is
too
early
or
we
can
probably
accommodate
the
time.
A
C
C
I
know
like
8
a.m.
It's
a
bit
early!
So
thanks
for
the
like
accommodation
here,
so
we
can
try
and
literally
figure
it's
not
going
to
work
well,
we
can
always
adjust
the
time.
I
think.
Initially
we
were
talking
this
with
a
meal
because
he's
in
australia
and
looks
like
the
time
was
not
working
for
him
quite
well
like
we
hardly
see
him.
Hopefully
this
would
work
better
for
him.
C
Okay,
next
topic,
eric
on
the
prometheus.
E
Okay,
cool,
so
just
some
updates
on
the
exporter.
I
got
my
the
collector
pr
was
merged
and
I
know
you
merged
that
without
review
riley.
So
I
don't
know
what
exactly
is
going
to
happen
with
that
like
is
that
going
to
get
reviewed
later,
or
is
that
just
going
to
stay,
as
is
I'm
not
really
too
sure,
what's
happening
there?
But
this
is
my
last
week
here.
So
I'm
not
sure
if,
if
reviews
did
happen
later,
I'm
not
sure
if
I
could
do
anything
about
that.
C
Yeah,
so
so
quick
update
for
everyone
here,
so
I
I
probably
won't
like
I
was
busy
and
probably
won't
be
busy
this
week,
because
I
I
got
another
like
on-call
rotation
this
week
so
yeah
for
this.
Permit
this,
like
the
series
of
prometheus
pr,
I
think
number
one
thing
is
to
make
sure
things
are
like
working
end
to
end.
So
you
got
the
ci.
Everything
is
green
and
based
on
the
time
so
either
like
johannes
or
me
or
like
someone
else
from
the
community
can
help
to
reveal
the
code.
C
C
If
you
want
to
touch
the
other
parts
of
the
ripple,
I
think,
as
long
as
the
change
is
small
and
reasonable,
we
can
take
apart
and
review
them,
but
in
general
I
wouldn't
put
like
permit
this
pr
review
as
my
priority,
because
I
have
other
higher
priority
things
and
in
this
case
like
to
unblock
you,
I
think
it
makes
sense
to
get
those
pr
merged.
As
long
as
everything
is
green
and
it's
not
like
changing
the
other
part
does
that
make
sense,
yeah.
C
Okay,
that
makes
a
lot
of
sense
yeah
regarding
the
the
status,
so
I
I
think
the
current
code
I'm
trying
to
move
fast
instead
of
trying
to
get
something
at
the
production
quality
at
the
very
beginning.
So
this
has
been
an
experience.
We
learned
from
other
recall
as
well.
I
want
to
quickly
try
and
see
what
we
can
get
and
then
get
all
the
learnings
instead
of
trying
to
move
slowly
but
build
something
stable
and
later
realize.
Oh,
something
is
not
working
and
we
have
to
rebuild
the
entire
stuff.
C
So
with
that,
I
think
the
current
metrics,
the
entire
matrix
part,
is
not
having
a
production
called
housing
like
you
can
see
all
those
like
global
logs
and
the
memory
allocations.
Those
things
I
think
it's
fine
just
to
look
at
the
learning
and
also
metrics
like
spec
itself,
is
changing
so
hopefully
like
by
end
of
this
week.
C
We
can
have
the
open,
telemetry
tracing
spec
lockdown,
so
the
goal
is
to
split
the
spec
into
tracing
contacts,
semantic
convention,
metrics
and
logs,
and
focus
on
tracing
semantic
convention
and
context
and
try
to
mark
them
as
stable
as
soon
as
possible.
And
after
that
we'll
come
back
to
matrix.
So
please
expect
there
will
be
breaking
changes
in
the
metrics
part
in
api
and
in
the
sdk.
In
this
case,
I
I
think
I
wouldn't
bother
of
spending
too
much
time
trying
to
polish
this
and
later
realize.
C
We
have
to
redo
that
again,
so
take
the
learnings
and
get
the
code
working,
and
I
think
that
that
should
be
a
good
balance.
Okay,
yeah.
That
makes
sense
to
me
and
regarding.
C
Question
how
how
do
we
plan
to
repeat
those
pr's?
I
I
think
we're
not
going
to
come
back
and
revisit
those
pr
and
put
comments
once
they
are
merged.
We'll
probably
have
some
like
dedicated
like
developer
from
either
microsoft
or
your
rally.
Cars.
Some
other,
like
google
folks
mentioned
they're
planning
to
add
some
like
full-time
employees
on
this
project,
probably
in
the
next
couple
of
weeks.
So
we
expect
like
developers
to
join
this
community
and
take
the
responsibility
to
refactor
some
of
the
existing
code.
E
Okay,
yeah,
that
that
makes
sense
to
me:
that's
perfectly
fine,
so
I
guess
for
now
I'll
figure
on
I'll
finish,
working
on
the
ci
issues,
I'm
having
in
the
last
pr
that
we
haven't
merged
yet
the
prometheus
exporter
class,
and
then
once
I
get
those
figured
out,
then
I
guess
this
gets
to
go
from
there
and
then
last
thing
to
briefly
touch
on.
Oh
one,
more
thing
before
I
talk
about
this,
but
thank
you
again.
Riley
for
merging
the
label
name
change
pr.
I
know
we.
E
You
just
mentioned
that
it's
not
like
changing
the
the
sdk
isn't
like
something
we
should
be
like
doing,
but
it
was
again
like
you
said
it
was
a
relatively
minor
change
and
as
reasonable
change
as
well.
So
there
wasn't
any
like
breaking
thing
with
that
pr.
So
thank
you
again
for
merging
that,
and
then
last
thing
is
that
I
just
filed
four
different
prs
for
just
documentation
things.
These
aren't
like
I
want
to
say
these
are
very
high
priority.
E
I
would
just
say
that
these
are
good
resources
for
any
new
developers
who
want
to
start
working
to
use.
E
So
I
have
a
documentation
for
the
prometheus
exporter
design
document
a
readme
for
the
prometheus
exporter
as
well,
and
then
we
also
have
an
example
for
the
exporter
and
then
one
last
thing
is
a
google
test
demo
that
I
filed
an
issue
for
I
think,
a
month
and
a
half
ago,
and
it's
just
like
a
like
an
informational
demo
on
how
to
use
google
test
and
how
to
test
in
the
context
of
open
telemetry,
and
so
I
think,
it'd
be
pretty
helpful
again
for
new
developers
to
reference.
E
But
I
just
have
these
filed
and
you
can
take
a
look
at
them
whenever
you
have
time,
but
again
these.
I
don't
think
these
are
like
extremely
high
priority
either.
So
these
are
just
for
reference
and
then
the
last
thing
I
want
to
say
is
that
this
is
probably
my
last
c
plus
plus
sig
meetings.
So
I
guess
I
just
want
to
say
thank
you
to
all
of
you.
You
guys
have
been
super
helpful
for
me
throughout
my
internship.
E
C
Contribution
and
look
forward
for
the
future,
like
collaboration
as
well,
you
can
you're
always
welcome
to
like
come
to
this
report
and
like
how
to
make
some
progress
on
the
on
the
tracing
or
the
matrix
part.
Thank
you
thanks.
A
lot.
C
Okay,
it
seems
like
we
have
a
short
meeting
this
time,
so
I
want
to
quickly
have
an
update
on
the
overall
project
status
so
currently
for
the
like.
The
overall
open
telemetry
project,
people
are
working
very
hard
on
trying
to
get
the
open,
telemetry
spec,
at
least
the
the
tracing
part
and
the
context
part
and
the
semantic
convention
for
tracing
those
parts
try
to
get
like
a
stabilization
as
early
as
possible.
So
what
I
heard
earlier
this
morning,
there's
a
maintainer's
meeting
at
8.
00
a.m:
9
a.m
early.
C
C
So
what
I
I
will
do
is
I
I
will
take
a
look
at
the
difference
between
our
current
sdk
and
the
spec
changes.
For
example,
some
of
the
sampler,
like
parent
or
else,
are
renamed
as
parent-based
there's
a
like
a
probability.
Sampler
got
renamed
to
like
parent
trace
id
based,
sampler
or
something
so
we
have
just
done
the
all.
C
The
changes
in
the
open,
timely.net
I'll
follow
the
similar
approach
to
file
some
issues
to
track
and
maintain
that
in
the
class
class
project
and
and
the
overall
goal
is
to
get
five
languages,
I
believe
it
is
a
python
goal
and
javascript
and
donet
and
and
java
ga
by
november
time
frame
so
c,
plus
plus
we
realized
it's
there's
a
physical
limitation,
so
hopefully
we
can
get
to
a
preview
version,
something
like
alpha
or
we
don't
have
alpha
like
official
release
today.
So
I
hope
before
november.
C
At
some
point
we
can
get
alpha
release
just
to
cover
the
tracing
part.
Try
to
make
that
as
close
to
the
latest
spec
as
possible,
and
at
that
time
the
spec
on
the
tracing
part
should
be
already
finalized.
So
whatever
we
have
should
be
like
a
stable
api.
It's
just
like
probably
the
sdk
is
not
stressed
highest
than
those
things,
so
we
cannot
announce
that
at
the
ga
status
and
for
matrix
part
we'll
we'll
just
follow
the
other
sdk.
C
So
my
current
guesses
will
probably
spend
some
time
in
the
open
telemetry.net
to
catch
up.
The
matrix,
part
and
c
passport
will
follow
that.
Historically,
we
have
been
following
the
open
time
trip
python
sdk,
because
python
is
made
the
first
movement,
so
it
used
to
be
the
first
release.
The
beta
version
of
the
open,
telemetry
ick,
and
things
are
at
least
at
that
time
were
in
a
very
good
shape,
but
recently
some
of
them
the
python
maintainers
and
the
provers
change.
C
C
So
this
is
a
an
issue
and
on
the
other
side,
we're
making
better
progress
because
we
put
more
developers.
So
you
see
us
as
we
try
to
get
the
same
amount
of
investment
and
I
think,
moving
forward.
Probably
we
need
to
follow
down
that
more
closely
because
donald
is
not
catching
up
with
python
and
inside
of
the
like,
complex
and
and
sampler,
and
also
the
processing
pipeline
is,
is
now
much
closer
to
the
specs.
C
So,
probably
like
better
from
the
spec
perspective,
comparing
to
python
that's
my
update,
so
I
want
to
quickly
get
the
feedback
from
from
folks
here,
like
from
your
side,
like
any
potential
dependency,
that
you
would
expect
open
time
to
c
plus
class
to
arrive
at
a
certain
maturity.
For
example,
you
want
to
have
that
va,
probably
like
early
next
year
or
some
of
the
things
that
you
think
might
be
challenging
for
you,
for
example,
like
you
just
realized
we're
not
into
ga
by
november,
just
like
project
management
timeline
feedback
from
you
guys.
B
B
You
know
in.
I
just
threw
out
that.
I
believe
that
we
should
do
the
traces
piece
first
and
get
that
all
the
way
through.
But
that's
you
know
if
we
were
to
do
that,
you
know.
Maybe
could
we
mark
that
somehow
on
a
project
board
and
say
here's
what
we
need
and
here's
the
the
work
you
know
to
be
done?
You
know
to
get
us
through
to
that
to
that
milestone.
A
I
mean
that's
something
that
I
actually
wanted
to
suggest
that
the
I
mean
I
can
do
that.
I
can,
and
maybe,
if
it's
closed
it
together,
I
can
like
add
a
project
to
like
on
the
github
projects
board
for
just
the
traces
part,
and
I
tried
like
to
identify
what
currently
is
what
they
still
are
lacking
from
the
spec
and
parts
of
implementations.
A
That
are
a
major
because
I
think
the
context
part
needs
some
kind
of
a
overhaul
at
least
parts
of
fit,
and
I
could
create
issues
for
that
and
add
that
to
a
perfect
board.
A
And
then
we
would
get
an
idea
like
what
is
actually
still
missing
and
also
riley,
then
could
add
issues
to
there
like
based
on
the
spec
finalization
yeah
yeah.
But
I
was
just
thinking
about
this
matrix
because
I
started
matrix
that
that
riley
tries
to
brought
up
and
that
could
serve
as
a
great
template
for
trust,
creating
issues
for
stuff
that
you're
still
missing.
C
Yeah
so
so
just
give
a
quick
update,
so
the
spec
now
has
a
matrix,
and
I
I
think
currently
we
focus
on
the
traces
part
because
it's
almost
complete,
I
think,
like
folks,
got
action
item
to
do
later
today
to
make
it
complete.
And
after
that
we
should
be
able
to
see
like
where
we
are
and
like
once
we
figure
out
the
status.
We
can
mark
the
like
items
that
we
need
to
do
just
for
traces
and
for
matrix
resource
those
parts.
C
I
think
countries
low
priority,
so
we
can
probably
forget
about
them
for
now
and
the
context
propagation
error
handling
those
parts
are
important,
but
I
I
think
it's
not
a
blocking
issue
for
us,
so
it's
some
internal
renaming
most
of
the
folks,
wouldn't
use
those
like
context,
propagation
api.
C
C
Yeah,
so
I
I
can
take
a
stab
just
to
go
through
so
so
how
about
this
I'll
I'll?
Send
the
initial
pr
to
this
back
just
to
update
what
we
have
for
the
c
plus
plus
project?
And
now
I
add
you
and
I
wonder,
like
like:
what's
your
github,
alias
to
toddy,.
B
C
C
Yeah,
so
I'm
I'm
not
not
a
hundred
percent
sure
where
I
understand
what
what
you
mentioned
fully
but
I'll
share
some
of
the
the
tagging
we
have.
So
this
is
not
invented
by
opentelemetry.net
right,
applied
to
all
the
open
time
trip
projects.
So
we
have
this
area
it's
based
on,
like
which
folder
we
have,
and
then
we
have
some
very
generic
thing
like
whether
we
we
need
this
for
ga
or
not
for
ga
and
different
priorities.
C
So
currently
in
dot,
we
don't
have
the
specific
area
whether
it's
tracing
or
matrix.
I
think
I've
seen
like
in
the
spec.
We
have
more
tags.
B
Riley,
if
you
click
at
the
top
there,
if
you
went
back
to
that
repo
and
click
on
projects
and
see
what
it
said
or
not,
sorry,
not
here,
but
yeah.net,
and
actually
the
javascript
repo
is
pretty
it's
pretty
good,
like
it
actually
breaks
out
this
rather
than
having
just
one
project
for
ga
tracker
has
project
boards
for
traces
and
metrics,
etc.
B
C
Yeah,
so
in
the
open
time
tricks
back,
you
can
see
we
have.
We
have
a
better
like
view
of
the
projects,
I
think
there's
probably
don't
have
access.
There's
the
opening.
B
B
C
A
I
I
just
pasted
a
link
for
the
javascript
repository
I
mentioned,
and
I
think
it's
helpful
for
people
to
get
an
overview.
Like
jody,
I
mean
she's
more
in
the
product
management
role.
I
think
it's
helpful
at
an
overview
of
where
we
stand,
for
example,
these
traces
also
like
in
talking
to
our
customers
and
maybe
showing
them
yeah.
A
A
And
just
for
what
you
said
before,
I
also
share
your
estimate
that
we
might
have
something
l
for
a
beta
in
autumn,
but
that
the
final
3a
for
traces
might
be
a
next
year.
I
think
that
is
realistic.
C
F
D
Just
did
some
changes
in
my
setup
here
so,
depending
on
destination
or
exporter
or
system,
we're
gonna
use
for
the
data
store,
like
some
customers
might
go
with
the
jlp
exporter,
google
cloud
new
rail
lake,
light
step
whatever
and
in
a
few
scenarios
I'm
looking
at
tw
and
our
internal
microsoft
agent,
which
will
capture
the
info
gathered
through
opentv
api
surface.
D
So
I'm
thinking
in
terms
of
readiness
like
for
each
for
each
destination
should
we
have
some
sort
of
matrix
that
says:
hey
we're
done
and
it
works
and
to
end
with
this
destination,
and
perhaps,
like
you
know,
as
a
result,
a
readme,
a
markdown
document
says
how
you'd
instrument
to
route
your
telemetry
to
lightstep
or
neural
or
googlebot,
or
how
to
emit
etw
on
windows
platform
to
etw
listeners.
D
Should
we
have
something
like
this
because
before,
for
example,
google
cloud
exporter
is
ready,
I
might
have
something
for
cw
path.
For
example,.
C
Yeah
so
max
take
a
look
at
this
like
here.
These
are
the
exporters
recommended
by
the
open,
telemetry
project
to
put
under
the
specific
sdk
repo.
Anything
else
will
go
to
the
registry,
so
you
can
click
this
and
different
companies
they're
free
to
register
their
own
exporter
libraries.
So
I
think
you
can
see
something
from
lifestyle.
I
guess
oh
there's
already
a
new
relic.
This
is
contributed
by
island
from
new
relic.
I
believe-
and
there
should
be
some
like
light
stab
exporters
as
well.
D
D
D
Let
me
rephrase
my
question
for
etw
and
windows,
where
atw
is
kind
of
event,
racing
for
windows
and
the
standard,
pretty
much
mechanism
recommended
by
the
platform.
Should
we
have
something
as
part
of
this
repo?
D
C
D
For
example,
console
exporter
well
console
is
a
great
tool
right,
but
it's
mentioned
as
one
of
the
first
things.
I'd
say
that
event
tracing
on
windows
is
the
first
thing
on
windows.
C
E
A
C
Right
yeah,
so
so
there's
a
compliance
matrix
and
it
also
defines
what
things
should
we
put
here
so
I'll
search
for
ever
yeah?
So
you
see,
these
are
the
recommendations,
if
you
won't
put
something
else,
then
send
the
pr
to
the
spec
and
solve
that
problem
first.
So
this
is
the
general
thing
like
how
you
can
like.