►
From YouTube: 2021-11-18 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
A
D
I'll
give
it
a
minute
brian
might
show
up,
he
usually
does.
If
you
have
any
agenda
items
you
want
to
discuss,
you
can
put
them
on
the
in
the
google
doc.
D
I
feel
like
there
was
something
I
wanted
to
put
in
there,
but
right
now,
there's
nothing
there.
I
know
that
there's
the.
B
A
I
was,
I
was
hoping
to
not
actually
bully
brian
into
working
on
open,
telemetry
telemetry
with
me.
B
A
Dan
is
the
member
of
my
team,
who
found
the
bug
and
added,
like
2
million
days
to
the
length
of
the
span
or
something.
E
A
So
we've
only
been
tracing
for
less
than
a
month.
F
I
maintain
the
spandex
library,
so
I've
been
using
that
for
a
long
time
with
datadog.
I
don't
remember
how
long
lots
of
years
20
18
2016.
I
forget.
D
Yeah
something
like
that
for
me
as
well.
I
started
with
the
open
census
library,
but
now
I
don't
actually
maintain
anything
that
is
instrumented
or
traces.
I
just
work
on
the
tracing
libraries,
no
actual.
D
A
I
maintain
the
stripe
open
or
like
stripe,
connect
omni
off
library
for
like
ruby
on
rails,
I
haven't
ridden
rails
in
like
half
a
decade.
Now
I
got
a
dm
the
other
day
of
someone
who's
like
our
whole
business
runs
on
this.
Please
update
it.
I'm,
like
I
mean
sorry
dude,
maybe.
D
D
I
guess
brian
won't
be
making
it
today.
I
pulled
up
the
pull
request
myself
to
take
a
look
for
the
removes
explicit
setting
of
start
time
and
duration.
Let
me
put
a
link
in
the
docs
just
to
have
notes
from
the
meeting.
D
D
So
the
only
thing
I
see
is
the
duration
attribute.
D
Like
what
I
couldn't
tell
yeah,
this
looks
fine
to
me.
B
D
B
D
D
I
haven't
made
a
release
of
this,
so
I
don't
actually
know
if
I
can
so
I'll
have,
but
I
will
be
sure
I'll
ping
brian
after
this
and
I'm
sure
we
can
get
a
release
out
pretty
quick.
D
A
D
Oh
yeah
yeah,
I'm
not
sure,
but
I
mean
I
know
what
he
meant
by
that,
but
in
the
sense
that
it's
tied
to
open,
telemetry,
api
rc3,
but
okay.
A
D
So
yeah
should
yeah
get
this
out.
I
don't
think
I
can
release
it,
but
I
should
probably
get
permissions
for
that.
If
I
can't
and
we
need
to
move
it
over
to
the
contrib
repo
get
everything
out
of
open,
telemetry
beam
but
yeah,
I'm
sure
we'll
get
this
released
today.
D
G
D
Good
to
hear
more
usage
being
done,
that's
how
bugs
get
found.
A
A
Exporting
to
first
well,
we
go
agent,
collector
jaeger.
A
A
A
D
E
D
It's
also
huge
if
I
can
tell
people
there's
who
use
elixir,
who
are
using
splunk
apn,
because
right
now,
the
there
there's
splunk
customers
that
use
erlang
and
elixir,
I'm
sure,
but
they
don't
know
about
them,
because
there's
no
official
support-
and
I
don't
know
if
any
are
using
the
apm
there
probably
are,
but
the
most
are
just
you
know
blogs.
A
We
need
to
get
a
little
bit
more
buy-in
to
get
into
the
apm
side
of
things
from
folks,
but
you
know.
A
D
F
I
was
going
to
ask
with
your
open
telemetry
telemetry
thing:
are
you
using
that
to
like
use
the
telemetry.span
thing
throughout
your
your
actual
code
base,
or
is
this
like
integrating
into
different
libraries
that
don't
have
their
own
dedicated
things.
A
We
have
a
library
that
takes
some
of
like
the
three
main
things
you
expect
a
software
system
to
do.
It
takes
like
attempted
calculations,
right
calculations
and
read
calculations
and
spans
or
emits
telemetry
events
for
those,
as
well
as
doing
spans
now
for
each
of
those
types
of
events.
A
A
In
between
that
and
contributions,
I
made
to
quantum
earlier
this
week.
It'll
also
be
in
our
like
cron
job-like
dtls
too.
A
D
I
just
looked
at
github
and
I
I
knew
the
fair
enough,
but
now
I
used
to
work
in
that
area
because
I
worked
for
orbits
and
then
another
company
down
the
street.
That's.
E
A
D
A
A
Here's
here's
our
hello
bronski.
B
A
D
Yeah
that'll
get
released
today,
so
you
can
pull
it
in
and
the
only
update
I
really
have
for
this
meeting
about
open
telemetry
early
elixir
is
we're
still
kind
of
poking
away
at
a
couple
things
for
ga.
I
think
the
last
thing
is
custom
id
generators.
I
need
to
look
at.
We
were
doing
planning
this
week.
D
It
worked
so
I
haven't
gotten
to
the
fred's
review
on
it,
but
that
should
be
merged
in
pretty
soon
and
then
I
think
pretty
much
it's
just
putting
being
able
to
put
limits
on
the
number
of
spans
and
the
size
of
attributes
and
then
will
be
good
for
going
1.0,
so
stable
api.
D
So
please
look
at
the
api
as
much
as
possible
right
now
and
let
us
know
if
there's
any
anything
confusing
or
it
could
be
better-
that
we
can
change
before
it's
going
to
be
stable
and
we
can't
change
it.
G
G
Also
well,
thank
you,
for
you
know
caring.
A
Appreciate
it
and
like
having
access
to
this
sort
of
meeting
where
you
know
the
maintainer
of
spandex,
the
maintainer
of
the
open,
telemetry
stuff
like
this
is
this
is
really
helpful
and
hopefully
we'll
be
able
to
be
useful
participants
inside
of
this
microcosm
of
the
community.
G
D
D
Yeah,
why
I
liked
the
open,
telemetry
project
and
kicked
off
the
erlang
live
because,
like
open
tracing,
the
problem
always
was,
it
was
never,
it
was
never
it
never
specified
enough
and
it
never
specified
like
propagation
or
the
exporter
and
some
other
stuff.
So
if
you,
the
company,
vendor,
supported
open
tracing,
that
really
didn't
mean
anything,
you
still
had
to
go
write.
You
know
spandex.
D
D
Eventually,
splunk,
we
will
probably
have
a
early
elixir
distro
of
open
telemetry
that,
like
makes
it
easier,
so
it
won't.
Unlike
the
open
tracing
days.
This
will
just
mean
that
it
you
pull
this
library
in
and
it
does.
You
know
some
setup
for
you
to
make
it
easier
instead
of
having
to
write
the
same
configuration
files
every
time,
but
it
won't
be
actually
changing
any
functionality
or
protocols
or
anything
like
that.
Just
be
a
nice
helper
library,
essentially
awesome.
F
I
have
a
quick
update
that
I
I'm
realizing.
I
mentioned
in
the
observability
meeting
last
time
about
working
on
getting
finch
hooks
to
support
context.
Propagation
finch
is
a
http,
client,
library
and
elixir,
and
it's
it's
gone
through
a
few
iterations
and
it
seems
like
we'll,
probably
end
up
getting
what
we
want
there
to
have
like
a
request
transformer
hook,
that
every
outbound
request
will
go
through
this
callback
function.
F
If
you
have
it
set
up
and
that'll,
allow
you
to
inject
the
tracing
headers
that
you
want
just
via
configuration,
instead
of
needing
to
like
wrap
the
library
or
inject
your
headers.
Every
time
manually,
so
that'll
be
cool.
It's
it's
kind
of
separate
from
the
middlewares
that
something
like
tesla
has.
F
Where
you
know
it's
not
like
you
don't
need
to
like
declare
a
middleware.
It's
just
a
callback
function
that
you
get
the
request
and
mutate
it.
So
that's
what
we're
doing
there
and
then
the
plan
is
that
we'll
probably
write
a
little
hotel,
finch
bridge
library
once
that
hook.
Is
there
that
you
can
just
say
like
use
this
hook.
You
know
use
this
library
as
your
callback
for
mutating
it
and
then
the
hotel,
finch
library
will
just
do
the
context.
Injection
thing.
Yes,
so
that's
the
plan,
hopefully.
A
Kristen,
you
also
work
on
the
grpc
box
library
yeah,
we're
currently
using
a
lot
of
the
grpc
elixir
grpc
from
tony
six
one.
Two,
all
right.
A
Let's
is,
is
grpc
box,
something
that
the
elixir
folks
might
also
like,
or
is
this.
D
A
D
I
know
at
one
point
I've
heard
from
a
couple
people
that
they
were
using
grpc
bucks
from
elixir
because
they
had
some
issues,
but
I
mean
that
was
a
year
year
ago,
it's
like
yeah
and
grpc
box
has
a
number
of
has
its
own
issues,
some
performance
issues
and
stuff
that,
because
I
don't
have
the
time
to
work
on
it,
but
now
I
should
be
having
the
time
to
work
on
it,
because
open
telemetry
exporter
relies
on
it.
D
D
And
certainly
would
like
to
make
everything
plugable
like
right
now.
The
exporter
just
uses
jrpc
box
and
httpc
inets.
That
should
eventually
be
able
to
be
so.
You
don't
have
to
pull
in
two
grpc
libraries
or
in
multiple
http
clients,
but
that's
something
for
the
future.
Yet.
D
So
yeah,
I
think
these
tests
passed.
So
oh,
that's
good
with
that
pull
request,
I'll
talk
to
brian
and
get
that
release
today.
Yeah.
That's
all
for
me
about
the
small
update
about
where
we're
going,
we're
closing
in
on
1.0,
going
through
the
technical
review
and
yeah.
D
A
I
think
we're
so
far
quite
happy.
There
there's,
I
think,
some
more
work.
One
of
our
other
colleagues
is
doing
on
with
respect
to
expanding
the
sorts
of
things
we
bring
in
on
our
collector
to,
for
example,
kubernetes
attributes,
but
that's
probably
maybe
I'll
send
him
on
the
path
over
to
the
collector
meetup
if
they,
if
they
have
a
sega
meeting,
specifically
for
the
collector.
A
Have
the
damon
the
damon
set
of
the
agent
okay,
that
already,
but
it's
something
like
adding
the
pod
ip
or
something
to
get
it
to
daniel.
You
looked
at
the
flow
request
last
I
might
be
flandering
a
little
bit.
C
D
Yeah
I
mean
so
we
definitely
one
that
you
should
have
support
for
that
in
the
open,
telemetry
library,
the
rolling
elixir
library.
But,
of
course
you
have,
if
you're
already
running
the
collector,
then
it
can
do
it
as
well,
but
yeah
we
need,
we
should
get
it
added
anyway.
D
It
can
possibly
make
it
easier
if
it's
just
in
the
in
the
library
itself.
D
Awesome
all
right,
I'm
merging
this
pull
request
and
yeah
all
right
cool
yeah.
Nobody
has
anything
else,
called
get
30
minutes
back
and
see
everybody
on
slack
cool.
Thank
you.