►
From YouTube: 2020-10-16 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).
B
Yeah
likewise
I
was
visiting
my
I
had
some
family
issues
to
attend
to
and
then
I've
had
like
a
whole
week.
So
I
feel
like
I,
I
blipped
out
of
the
spec
process
last
week
a
little
bit.
I
think
the
main
thing
with
sampling,
though,
is
ludmila,
has
left.
B
Microsoft
is
my
understanding
and
she
was
the
person
championing
the
the
pr
to
get
a
standardized
sampling
priority
in
so
that
pr
is
still
outstanding
and
it's
got
some
feedback,
but
maybe
has
some
legs,
so
I
feel
like
someone
needs
to
take
over
championing
that
I
was
hoping
pete
olsen
from
honeycomb
was
gonna
be
on
here,
but
my
plan
is
to
to
reach
out
to
him
and
just
see
if
pinecone
wants
to
take
that
over
the
finish
line,
so
that
that's
that's
probably
my
most
majorist
of
of
updates.
B
Just.
I
think
that
that
pr
is
important
and
I
think,
there's
I
think,
let's
see-
let's
just
look
at
what's
in
progress
here.
If
I
share
my
screen.
B
B
So
I
would
say
this
is
number
one
for
on
the
plugin
side
to
kind
of
get
this
resolved
looks
like
they've
decided
it's
backwards,
compatible,
so
they've
sorted
that
out,
but
this
is
my.
I
I've
been
poking,
actually
people
at
lightstep
to
try
to
make
a
sampling
plug
in
and
try
out
this
trace
state.
That's
just
my
only
outstanding
concern
is
I've
yet
to
see
someone
from
the
people
who
care
come
back
and
be
like
here's.
B
A
B
Yep-
and
that
was
the
old
tracer
thing
so
yeah
I
think
this
is
this
is
probably
again
it's
maybe
not
necessary
for
ga,
but
it
would
be
nice
to
not
just
let
this
die
on
the
vine.
B
So
that's
my
update
for
today
cool
now
that
I'm
getting
back
in
the
swing
of
things
next
week,
yeah
I'll
try
to
reach
out
to
honeycomb,
but
of
course,
if
either
you
are
interested
in
championing
this,
that's
also
totally
fine.
A
We're
it's
not
really
relevant
to
what
we
do
so
like
we
just
let
people
do
whatever
they
want
to
do
in
terms
of
instrumentation
and
like
we
don't
have
any
skin
in
the
game
either
way
like
we
have
no
agents,
we
have
no
apis.
We
have
no,
it's
just
jaeger,
so.
B
Yeah,
I
do
think
at
light
step
I
have
seen
people
start
to
be
like.
Oh,
we
could
use
sampling
priority
to
get
more
accurate
counts
of
you
know
spans
and
drop
spans,
and
things
like
that
so
they're
starting
to
realize
like
oh.
If
we
had
access
to
that,
maybe
we
would
use
it
for
some
handy
things.
A
Okay,
I
would
like
to
figure
something
out:
that's
sort
of
related
to
that
we're
building
like
a
metric
aggregator
for
traces.
A
Yes,
yeah
exactly
and
obviously
like
sampling
is
going
to
affect
the
accuracy
of
the
aggregates
yeah
and
so
we'd
like
to
do
it
like
pre-sampling
metric
creation.
A
A
B
It's
it's
feasible.
I've
certainly
seen
people
pull
it
off,
but
yeah.
That
is
a
tricky
problem
to
get
right,
especially
if
you're
trying
to
do
it
without
storing
all
of
the
traces
first
right,
yeah.
A
A
Yeah
I
mean
we
can
do
it
that
way
too,
but
it
it's
actually
kind
of
nicer
to
be
able
to
do
this
in
open
telemetry,
and
then
you
could
potentially
send
the
metrics
to
one
or
more
other
systems
like
if
you're
just
doing
metrics
or
you
want
to
send
them
to
prometheus
or
you
want
to
you
know,
do
whatever
you
want
to
do
with
them.
B
Yeah
and
to
solve
that
that
sampling
issue
would
really
help
to
scale
this,
so
that
would
be
a
great
addition
to
open
telemetry.
In
my
opinion,.
A
B
A
The
problem
is
my
the
main
engineer
that
works
on
that
and
is
a
committer
in
yeager
is
in
australia
and,
like
all
these
calls
time
zones
are
like
the
worst,
because
they're
all
pacific,
you
know
focused.
B
Well,
we
can
certainly
make
an
australia-friendly
time
to
talk
about
this
stuff,
absolutely.
A
B
There's
a
similar
discussion
in
the
australia
time
zone
on
tuesday,
the
tuesday
spec
meeting
coming
up
by
the
way
they're
going
to
be
talking
hashing
out
issues
with
context
that
I
think
some
people
from
asia
had
so
so
that
meeting
is
the
like
next
door
topic.
So
you
might
be
interested
in
that.
I
don't
know.
B
A
Yeah
and
it
doesn't
show
up
in.
A
A
Yeah,
let
me
chat
with
him
yeah
and.
B
A
B
But
it
probably
won't
there
won't
be
time
to
like
dig
into
sampling,
but
I
do
think
we
could
organize,
like
an
australia
friendly
time
to
to
talk
about
sampling
the
engineers
I
know
who
are
interested
on
it
are
in.
You
know,
california,
so
certainly.
B
That
I
was
saying
that
the
meeting
is
happening.
I
think
this
next
week,
there's
there's
already
a
scheduled
discussion
around
digging
into
context
which
is
like
next
door
to
sampling.
I
don't
know
which
part
of
the
pipeline
those
guys
are
concerned
about,
but
but
there's,
but
we
could
also
add
an
afternoon
sampling
meeting
on
friday
or
maybe,
let's
just
or
something
and
we'll
see.