►
From YouTube: 2021-12-14 meeting
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).
A
A
A
Yeah,
so
I
don't
think
there
was
a
lot
of
progress
in
the
spec,
so
I
mean
there
were
like
some
pr's,
which
attempted
to
make
the
baggage
part
removed
from
the
view,
so
just
to
make
sure,
like
view
is
like
not
complicated
with
the
cross
signal
thing
package,
so
it
says
they
are
still
attempting
to
release
the
metric
sdk
spec
as
table
by
end
of
this
month.
A
A
So
it's
relatively
small,
but
that
is
something
which
we
need
to
do
so
I'll,
be
like
keeping
a
track
of
that
if,
like
spec,
is
getting
merged
or
not,
and
based
on
that
I'll.
Do
some
updates
and
quite
like
it's
very
unlikely
that
we'll
do
a
release
on
before
12
30..
A
It
should
be
like
sometime
in
the
first
week
of
next
year,
so
yeah
yeah,
we
should
be
able
to
like
have
many
more
than
one
people
reviewing
it,
because
I
don't
want
to
really
do
a
release
in
pr
like
when
no
one
else
is
available,
so
I
think,
like
riley's,
on
vacation,
like
utkar
she's
on
vacation,
maybe
robert
was
already
on
vacation.
A
He
mentioned
that
he's
already
on
vacation,
so
at
least
like
michael
and
you
I
will
be
gone
next
week,
so
we'll
just
like
rejoin
on
early
next
year
and
then
figure
out
the
rest
of
the
issues
and
do
the
release.
A
There
are
like
few
pr's,
which
I
haven't
had
a
chance
to
look
at
it
on
call
this
week
primary
on
call.
So
I
might
I
mean
it
should
be
like
relatively
free
neck
tomorrow
onwards.
I
have
mostly
like
free
day
tomorrow,
so
I
try
to
get
to
the
remaining
pr's.
I
think
there
are
a
couple
of
them.
A
One
is
the
otlp
exporter
options,
thing
two
pr's
and
then
there
is
another
about
explicit
layout
changes
for
metric
point,
and
there
is
a
third
one
actually
from
one
of
my
teammates
about
adding
min
and
max.
A
A
It's
not
yet
the
otp
exporter
won't
be
having
it.
So,
even
if
you
do
it,
it's
just
a
change
in
the
core
sdk
and
maybe
like
the
console
exporter,
it
can
display
it
as
well.
I
don't
think
prometheus
cares
about
minion
mac,
so
it's
only
applicable.
I
mean
it's
only
relevant
when
actual
otlp
spec
gets
merged.
That
spec
I
checked
like
last
week.
It
have
not
been
much
so
maybe
I
can
quickly
check
the
yeah.
It's
still
not
much
so
without
this
we
don't
oh
it's
approved
by
many.
A
I
don't
know
why
it's
not
much
yeah.
There
are
some
questions
about
like
what
do
we
do
for
min
and
max
if
it
is
cumulative
or
something
here,
okay,
they're
trying
to
add,
like
some
experimental
features?
Okay,
so
it's
just
like
mostly
in
principle,
people
agree.
I
can
see
like
george
and
other
josh,
also
approving,
tyler,
approving
and
maybe
like
book
then
and
jack
had
some
concerns
each
other
so
yeah
we
just
wait
for
it
to
be
merged
before
we
do
anything
so
yeah
sounds
good
yeah.
Oh
michael
is
here
so
hello,
michael.
A
A
little
bit
of
a
break,
I
see
yeah,
so
we
won't
be
having
the
next
two
ones:
the
21
and
28
one
I'll
put
a
note
in
the
stack
as
well.
I
don't
know
how
to
go
and
cancel
the
calendar
thing
so
I'll
see
if
I
can
figure
it
out,
but
otherwise
I
just
use
the
slack
channel
to
announce
that
we
won't
be
having
the
submitting
for
the
next
two
weeks
and
we'll
reconvene
on
the
giant
first
week.
A
A
So
the
latest
update
as
of
today
morning
is
they're
still
trying
to
make
it
happen,
the
like
technical
committee
and
who
is
owning
the
spec
now
trying
to
make
it
happen
by
end
of
this
month,
and
for
that
they
are
like
cutting
some
features
from
the
stability
guarantee
and
those
happen
to
be
the
features
which
we
don't
have
anyway,
like
doing
baggage
in
view,
and
example
are
so.
A
These
are
features
which
we
don't
have
anyway,
so
we'll
have
will
not
be
impacted,
even
if
those
things
are
like
removed
from
the
stack
but
either
way.
Given
that
there
is
a
general
lack
of
approvers
and
maintainers
in
the
next
two
weeks,
we
will
just
do
the
release
on
giant
first
week,
even
if
the
specs
do
get
stable
in
the
next
few
weeks.
It's
very
unlikely.
I
don't
think
that
spec
is
going
to
be
stable,
because
there
are
still
open
issues
and
there
are
like
several
open
pr's.
A
A
We
want
this
to
be
merged
before
we
can
release
this
table.
So
it
looks
like
there
are
like
a
lot
of
open
questions,
so
it's
very
unlikely
that
this
would
happen
end
of
this
year,
so
we
just
come
back
early
next
year
and
see
where
things
are
and
based
on
that
we
either
do
a
1.2
stable
release
or
do
a
1.0
release
scan
day
2,
depending
on
how
stacks
are,
at
that
day,.
A
Let
me
open
that
actual
issue
yeah.
So
this
is
an
issue
which
is
being
tracked
in
the
dot
net.
Runtime
required
both
general
things
for
observability,
basically
open
elementary,
and
I
have
linked
all
the
issues
which
we
as
in
open
elementary.net.
We
cares
and
we
basically
ask
p0
for
all
those
things,
so
the
sampler
should
be
allowed
to
modify
traced,
and
this
is
a
p0,
because
this
spec
exists
for
a
long
time.
A
So
it
is
a
p0
and
then
there
is
schema
url,
which
is
harder
to
spec
few
months
back
except
p0
for
us,
because
this
is
required
for
us
to
have
any
instrumentation
library
sustainable.
So
we
want
to
treat
it
as
p0,
so
I
mean
this
is
basically
our
ask
it's
to
be
seen
whether
or
not
team
agrees
to
that
they
will
come
back
from
vacation
early
january
and
do
the
triaging.
There
is
one
more
thing
which
we
might
we
requested.
A
I
say
p0
observable
and
I've
done
counter,
and
these
are
again
like
out
of
the
spec
already.
So
these
three
are
our
top
three
asks
and
then
there
are
like
other
things
like
http
client,
doing
the
legacy
thing
like
they're
still
using
the
diagnostic
source,
where
we
want
them
to
migrate
to
activity
source.
So
as
of
now
it
was
created
as
p2,
but
I
know
that
this
will
be
like
needed
to
be
treated
as
a
high
priority,
because
there
is
a
working
group
called
open
elementary
instrumentation
sync.
A
They
are
proposing
some
changes
to
the
the
semantic
inventions
for
instrumentations
and
some
of
them
do
require.
Let
me
just
open
that
see
if
we
can
find
it.
So
there
are
some
proposals
in
order
to
implement
that
we
need.
We
need
the
http
client
and
sp
net
core
to
migrate
to
the
activity
source.
We
cannot
because
we
currently
have
a
workaround
like
the
legacy
thing.
We
do
that
hacky
work
around
so
that
won't
be
possible
once
the
new
spec
got
merged
and
that
spec
is
somewhere
here.
A
It's
maybe
got
closed
due
to
inactivity
yeah.
Actually,
this
one.
A
A
And
if
this
gets
merged
we
would
be
forced
to
like
use
the
http
client
to
migrate
to
activity
source,
because
whenever
there
is
a
retry,
we
need
to
add
the
links
to
the
newly
created
activity
and
that
link
should
be
to
the
previous
one.
And
we
can
only
do
that
if
you
use
the
activity
source
api
because
xod
ap,
which
allows
us
to
attach
a
link
to
an
activity.
So
so
this
would
likely
become
a
higher
priority
than
p2.
A
Where,
like,
when
there
is
a
retry,
there
will
be
some
links
created,
which
should
be
yeah.
I
mean
I
didn't
read
it
fully,
but
like
the
person
who
opened
this
pr
he's
trying
to
protect
this
in
our
repo,
like
open
elementary
town,
and
he
basically
said
that
it's.
I
asked
him
to
make
that
pr
like
non-draft
or
he
basically
shared
it
with
me.
A
D
A
A
So
basically,
it's
like
there
are
some
effort
to
make
it
stable,
which
means
we'll
be
unblocked
from
releasing
our
instrumentation
library
sustainable,
but
before
we
do
that,
we
need
to
let
go
and
make
sure
that
we
are
able
to
follow
the
spec,
and
so
basically
I
put
this
issue
just
to
him
that,
like
if
these
things
become
stable,
we'll
be
still
stuck
in
non-stable
version,
because
we
how
to
figure
out
how
to
like
work
around
these
limitations.
D
A
It's
too
hard
to
predict
what
would
be
the
thing,
and
I
mean,
based
on
what
I
understand
from
the
working
group
they're,
trying
to
like
focus
on
http,
so
there
are
like
many
semantic
conventions,
but
they're,
just
trying
to
like
put
a
very
research
focus
on
one
thing.
So
that's
why
I
put
http
client
here.
It
might
be
the
case
for
like
grpc
and
asp.net
core
as
well,
how
it
might-
and
you
probably
know,
like
asp.net
core
partially
moved
from
the
legacy
thing
to
activity
source.
A
I
think
maybe
I
shared
it
like
sometime.
I
can
hear
so
this
is
asp.net
core
and
they
have
something.
So
you
can
see
like
they're,
using
activity
source
and
somewhere
here
they
are
moving
to
use
the
activity,
source,
dot,
stat
activity.
I
think
they're
doing
like
create
first
and
then
start,
but
either
way
once
yeah
this
one.
A
So
once
they
do
this,
we
our
samplers,
all
the
things
would
like
be
natively
called
without
doing
any
work
from
us,
but
there
were
like
some
issues
which
I
think
is
open
here
I
mean
as
an
issue,
but
unfortunately
no
one
is
actually
working
on
it,
so
you
have
to
figure
out
how
to
really
work
around
that.
A
So
we
expect
like
at
the
minimum
asp.net
core
and
http
client.
These
are
the
two
things
which
come
from
the
overall
net
ecosystem
to
move
to
activity.
So
so
that's
what
is
being
tracked
here
so
the
specific
action
item
would
I
mean
the
specific
priority
would
depend
on
some
spec
if
it
gets
merged
or
quite
likely,
it
will
be
merged,
and
then
we
need
to
like
increase
pressure
to
see
if
we
can
put
it
as
ip
one
thing
yeah.
So
my
only
ask
is
like
review
this
item.
A
There
are
things
which
are
like
no
directly
ready
to
open
telemetry,
but
if
you
think
that
there
is
something
which
we
really
want-
and
it's
missing,
please
put
it
as
a
comment
here
or
in
our
repo,
and
we
can
link
it
here
as
well.
A
I
also
added,
like
some
other
things
I
saw
like
bound
instruments
metric
things.
These
are
things
which
were
originally
part
of
the
metric
spec,
but
it
got
removed.
So
it's
a
possibility
that
it
will
be
added
to
the
spec
bank,
and
so
I
basically
say
there
is
a
spec
dependency.
If
the
spec
comes,
then
we
need
to
do
that
in
the
document,
so
yeah.
A
So
this
is
my
attempt
at
capturing
all
the
us
from
open
elementary,
I'm
just
releasing
it
into
the
dot
net
team's
visibility,
so
anything
which
we
are
missing
feel
free
to
add
it
so
that
we
like
we'll,
have
a
nice
story
when
like.7
releases
we
we
saw
like
a
lot
of
things
in
six
like
we
have
the
propagators.
We
have
status
metrics
but
yeah.
There
are
like
still
few
things
which
you
need
to
get
before.
We
can
really
say
it
or
not,
is
like
natively
the
open
elementary
api,
which
we
can
use.
A
D
About
these
options
prs
that
alan
and
I
have
done.
A
A
You
know
I
did
not
review
them.
In
fact,
I
I
did
not
review
that
via
the
comparison,
and
I
also
did
not
review
the
layout
changes
to
the
struct,
because
that
affects
the
max
as
well
so
yeah
yeah,
I
hopefully
I
I
would
be
doing
that
remaining
days
of
the
week,
because
I'm
on
call
this
week
so
depending
on
my
actual
load
per
cloud,
I
might
have
like
more
free
time
or
less
free
time.
So
I'll.
D
Try
to
at
least
review
one.
If
you
get
a
chance
to
look
at
the
options,
maybe
we
can
find
some
time
for
the
three
of
us
to
chat
only
because
so
we
have
kind
of
similar
things.
We
need
some
other
opinions,
so
we
can
decide
on
a
direction.
I
don't
want
to
do
a
whole
bunch
of
work
on
it
until
we
kind
of
have
everybody
in
sync
with
how
we
want
to
approach
it.
A
A
Seven,
this
one.
D
D
So
the
builder
thing
that
I
did
basically
moves
this
shared
stuff
into
sdk
in
a
way
that
all
exporters
can
reuse
it.
Whereas
what
we'll
have
to
do
with
this
design
is
every
exporter
will
need
basically
the
same
class,
so
that's
kind
of
the
fundamental
decision
we
have
to
make.
Do
we
want
to
keep
this
stuff
in
each
exporter
or
do
we
want
to
try
to
reuse
it?
That's
sort
of
the
kind
of
whatever
we
need
to
decide.
We
want
to
pull
yeah,
so
this
specific.
A
Thing
about
like
things
which
are
part
of
the
options
class
are
not
really
supposed
to
be.
Part
of
this
has
come
up
before
and
we
think
I
linked
that
issue
earlier
like
this
has
nothing
to
do
with
otp
exporter
itself.
It's
really
just.
I
think
I
have
an
issue,
it's
kind
of
like
a
ruler
play
for
all
yeah,
all
right
so
like
did,
we
have
an
issue,
maybe
like
allen,
do
you
look
like
having
an
issue,
or
did
I
create
one.
A
This
issue,
even
before,
like
in
the
tracing
also
we
have
this
thing
like
otlp
exporter
options,
but
it's
not
really
the
exporter,
it's
just
the
pipeline.
They
are
controlling,
like
which
processor,
to
use
the
batching
one
or
simple.
So
whatever
we
decided
as
long
as
we
can
do
it
consistently.
I
would
be
like
happy
to
support,
but
anyway,
like
at
least,
let
me
like
get
some
time
to
review
the
builder
one,
because
if
that
is
extracting
into
a
the
common
sdk
please
and
before
the
exporters
can
leverage
that.
A
Then
that
would
like
solve
at
least
two
big
issues
in
one
shot
like
we
will
be
solving
the
this
issue
and
the
other
like
ugliness
issue,
where
we
have
too
many
things
in
the
export
option,
so
yeah
I'll
give
it
another.
I
mean,
give
it
a
first
rate
sometime
this
week
and
yeah.
If
it
it's
needed,
I
can
like
ping
ellen
and
michael,
like
offline,
I
mean
in
chat
and
see
if
we
can
do
it
quick
meeting
before
end
of
the
year
to
make
some
progress.
D
C
B
Deeper
later
once
cj
you've
had
a
chance,
but
I
think
that
the
the
other
thing
that
is,
on
my
mind,
is
that
I'm
I'm
kind
of
liking,
I'm
liking
the
builder
pattern,
but
I'm
still
kind
of
also
liking
the
idea
of
pulling
the
options
out
of
the
of
the
main,
like
otlp
exporter,
options,
class,
the
the
the
unrelated
options
like
the
metric
reader
options
and
the
reason
the
main
reason.
Why
is
because
it's
really
just
these
extension
methods
that
are
just
kind
of
special
in
that
they
they
do.
B
Two
things
to
the
pipeline
rather
than
just
one,
and
I'm
wondering
if,
like
the
the
builder
thing
that
you
have
michael,
is
useful
in
the
context
of
the
extension
methods,
but
then
for
the
more
advanced
use
cases
where
people
may
not
be
using
those
extension
methods.
I
think
it
would
be
nice
to
have
some
cleaned
up
options,
classes
and
I'm
wondering
if
there's
a
different
way
to
go
about
centralizing.
Some
of
the
common
stuff
like
like
you
said.
B
I
totally
agree
with
you
that
the
metric,
though
tlp
metric
reader
options,
class
that
I
currently
have
maybe
could
be
elsewhere.
B
Class
into
sdk
and
rename
it
yeah
the
the
thing
that
was
on
my
mind
there
and
there's
probably
a
way
to
solve
this-
is
that
the
metric
reader
options,
if
I
were
to
just
move
that
straight
up
into
the
sdk,
I
I
think
we'd
want
different
defaults
for
different
when
used
in
the
context
of
different
x
borders.
D
I'm
trying
to
push
for
a
solution
that
doesn't
you
know,
force
it
down
on
our
users,
not
saying
we
won't
ultimately
do
that,
but
that's
sort
of
what
I
was
exploring
with
the
builder
is:
if
you,
if
we
preserve
a
single
options
class
that
has
the
kitchen
sink,
which
gives
you
that
kind
of
advanced
option
right,
the
user
can
always
new
up
an
exporter
called
add
processor
like
they
can
have
full
control
if
they
really
want
it
and
they
have
the
options
class
there.
That
has
everything.
D
If
they
want
to
own
it,
then
the
builder
was
just
sort
of
a
happy
path.
Api
that
gives
us
a
way
to
hide
all
that
complexity.
You
know
with
little
helper
methods
like
configure
pipeline
or
whatever
the
heck.
I
called
it,
and
it
also
gives
you
a
way
it
sort
of
works.
Kind
of
like
how
the
deferred
processor
works
is,
you
can
always
build
extension
message
methods
that
register
actions
that
will
be
fired,
so
people
can
extend
it
really
without
cost.
So
that's
sort
of
where
my
thinking
was
now.
D
The
complexity
got
kind
of
out
of
hand.
There's
a
lot
of
generics
going
on
it's
a
little
bit
nasty,
I'm
not
100
in
love
with
how
it
came
out,
but
that's
sort
of
what
my
thinking
was,
which
was
really
based
on.
Can
I
do
this
without
obsoleting
anything
which
may
or
may
not
be
valid?
It
was
just
sort
of
what
I
was
attempting.
A
A
Okay,
yeah.
Let
me
spend
some
time
on
it
and
like
share
my
opinion
as
well,
and
as
we
decided,
we
can
probably
have
a
short
meeting
before
like
we
really
go
for
vacation
cool.
Thank
you
any
other
peers
or
anything
which,
like
anyone,
wants
to
discuss
quickly
or
because
I
am
like
only
like
finding
limited
amount
of
time.
So
I
was
only
looking
at
things
which
are
like
really
small,
but
I
think
yeah
this
gear
is,
should
be
done.
A
So
maybe
I'll
do
a
I
mean
if
any
of
you
are
having
time
and
if
you
think
that
it's
already
feel
free
to
hit
the
merge
button
here,
something
which
we
yeah.
A
So
we
should
be
able
to,
and
then
there
are
like
few
things
which
yeah
we
have
to
come
back.
There
is
this
activity
status
thing
but
vishwesh
who
worked
on
this
he's
on
vacation?
Maybe
he'll
come
back
next
week.
So
let's
see
what
what
is
the
status?
Is
it
waiting
for
reviewers
or
he's
waiting
for
something
else?.
B
Yeah
I
think,
last
week
I
said
I
was
going
to
look
into
the
the
grpc.
C
B
I
think
that's
probably
the
thing
that
I'll
be
doing
kind
of
over
the
holidays.
The
I'm
curious,
though,
if
because
I've
certainly
run
into
customers
who
would
benefit
from
some
retry
logic,
where
they're
losing
data
today
using
the
otlp
exporter,
I'm
curious
if
others
have
encountered
that.
A
So
that's
the
strategy
which
I'm
aware
of
so
at
least
none
of
the
customers
whom
I
worked
with
complained
about
this
thing,
but
they
do
complain
about
lack
of
free,
try
in
our
custom
exporter
but
yeah.
It's
a
separate
topic.
A
Do
you
know
if
other
languages
have
it
the
retreat
capability,
like.
B
A
A
All
right
yeah,
nothing
else
for
the
agenda,
so
we
will
try
to
meet
some
time
before
we
go
on
vacation
to
discuss
the
options
thing,
but
that
I
said
we'll
see
next
year
and
yeah
happy
vacation.
Two
four
is
taking
vacation.