►
From YouTube: 2022-03-24 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
B
Yeah
I
I
was,
I
was
here
sitting
at
the
same
place
and
something
about
thursday
afternoons.
Something
goes
wrong,
you
know
3
45
or
so,
and
I
can't
I
know
I
wouldn't
be
able
to
dedicate
any
attention.
So
I
didn't
join
as
I
was
dealing
with
other
problems.
B
A
Totally
yeah,
that
makes
sense.
I
yeah
there's
a
lot
of
meetings
that
I
kind
of
am
half
involved
in
sometimes.
B
But
yeah,
I
usually
I
try
to
block
out
thursdays
for
a
long
time.
We
had
thursday
afternoons,
we
had
declared
were
engineering,
quiet,
hours,
yeah,
and
so
I
I
continued
so
they
stopped
that
they
declared
failure
like
people
just
weren't
honoring
it,
but
there's
too
many
people
too
many
competing
things.
B
A
B
A
I
just
I
make
my
phone
go
on
vacation,
that's
usually
a
helpful
thing.
C
A
Oh,
that's.
A
good
idea
goal
is
just
go
on
vacation
when
your
password's
getting
reset.
You
know
because
it's
time
it
out
and
yeah.
B
I
remain
concerned
with
the
way
that
slack
has
three
or
four
different
ways
that
you
can
attempt
to
either
isolate
yourself
or
try
to
indicate
to
others
that
I'm
not
here
or
I'm
not
available,
but
what
it
still
fails
to
do
is
to
bounce
back.
Like
any.
You
know
when
somebody
attempts
to
write
to
you.
It
doesn't
do
any
job
of
signaling
this.
No,
I
I
won't
even
send
the
message
for
you.
Do
you
realize
what
you're,
what
you're
trying
to
barge
into
here?
It's
like
people,
don't
look
at
the
status
icon.
B
C
So
what
that
sounds
like
to
me
is:
there
needs
to
be
proper
feedback
to
the
user.
Writing
it.
So
a
bot!
That's
like
you
said
you
just
sent
the
message
to
somebody
who's
out
of
office
or
you
know
unavailable.
C
So
so,
if
you
I
was
going
to
say,
if
you,
if
somebody
set
themselves
to
like
do
not
disturb
or
notifications
off,
it
will
send
the
message
and
then
give
you
a
little
response
that
says:
hey
this
person
has
notifications
off.
You
can
notify
him
if
it's
an
emergency,
but
probably
not
best
too.
B
Still
not
enough
apple
messages,
at
least
I
guess
it's
on
ios
has
a
thing
where
I
usually
notice
that
the
person's
driving
you
know
it's
like
a
feature
you
can
opt
into.
So
some
people
have
this
turned
on
automatically,
where,
if
you
write
to
them
in
imessage
it'll,
say
I'm
driving
right
now.
I
can't
I
can't
respond
to
your
message
or
something
that
comes
back
immediately.
It's
very
helpful
for
letting
you
like,
oh
back
off,
that
sounds
important.
A
Yeah,
maybe
that's
just
the
answer.
Is
that
bot
that
aaron's
talking
about
has
to
respond
with
something
dramatic
like
aaron,
can't
talk
right
now,
he's
doing
heart
surgery
or
something
like
that.
A
A
Cool,
but
let's
jump
into
it
here,
I
think
we're
yeah
getting
into
three
minutes.
Past
got
a
fair
amount
of
people
showing
up
looks
like
everyone
I
see
is
on
the
attendees
list.
If
you
have
something
to
talk
about,
please
edit
the
agenda
and
we
can
jump
right
in
so
I
think
most
people
on
the
call
are
into
this,
but
just
kind
of
recap.
Yesterday
we
released
1.6
and
v028
of
the
028
of
the
metrics
experimental
metrics
api
specifically
and
some
sdk
refactors
to
support
it.
A
By
that
I
mean,
like
a
shim,
we've
been
talking
about
this
for
a
little
while.
So
hopefully
it's
not
too
big
a
surprise
for
anybody.
A
It
does
mean
there
are
quite
a
lot
of
breaking
changes
if
you
are
using
the
existing
metrics
api
package
after
seeing
aaron
go
through
the
contrib
repo,
I
don't
know
if
it's
actually
that
dramatic,
it
actually
looked
pretty
I
mean
outside
of
the
metrics
test
stuff,
but
outside
of
that
it
looked
pretty
straightforward,
so
yeah,
I
don't
know
if
it's
going
to
be
too
big
a
deal,
but
it's
worth
pointing
out
that
that
has
happened.
I
think
it's
worth
kind
of
like
advertising
it
as
well.
A
If
anybody
internal
to
your
company
uses
it
just
to
kind
of
get
some
user
feedback,
I
know
we
have
some
opinions
about
this
and
we
have
already
got
some
user
feedback,
but
the
more
the
merrier,
I
think,
is
probably
ideal.
I
know
internal
to
splunk.
We
have
customers
that
have
wanted
to
look
at
it
and
so
I've
given
them
the.
I
don't
know
a
very
light
green
light
on
that
one
saying:
there's
probably
issues
and
we
want
feedback.
It's
not
going
to
be
100
included
so
yeah.
It's
also.
A
Maybe
I
should
have
started
with
this.
A
good
point
to
you
know
celebrate.
We
did
actually
achieve
something
so
yeah
yeah,
definitely
happy
to
say
that
we
got
this
out
so
really
excited
thanks.
Everyone
from
the
call
or
anyone
not
another
call
listening
to
it
afterwards
contribute
to
this.
It's
definitely
valuable
cool.
A
So
that
said,
the
contribution
needs
to
follow
up.
There
are
two
prs
that
I
really
want
to
prioritize
people.
If
you
have
time
to
go,
take
a
look
at
one
of
them
I
think,
can
get
merged,
but
I
would
like
to
not
do
it
in
this
order.
This
one,
I
think,
can
get
merged
now,
but
I
would
rather
do
this
because
it's
two
people
going
at
something
with.
I
think
some
impedance
mismatch
here,
mostly
my
faults,
but
I
think
this
is
probably
the
good
one
to
lead
with.
A
We
need
reviews
on
this.
This
is,
I
think,
the
bulk
of
the
work
for
the
metrics
package
upgrade,
and
so
I
would
say
like
if
you,
if
you
have
a
place
to
start
this,
would
be
it
code.
Patch
yeah
everything
looks
passing.
So
please
go
take
a
look
at
this
one.
I'd
like
to
merge
this
one
after
that,
and
this
will
get
the
full
upgrade
path
for
going
to
the
new
releases
from
hotel.
A
Important
to
note
is
that
aaron
in
producing
this
pr
has
removed
the
metrics
test
package
usage
and
so
with
that
some
of
the
next
steps
we'll
kind
of
touch
on
those
in
just
a
little
bit,
but
that
one
of
the
things
we'll
do
after
the
release
is
to
address
that
and
maybe
do
a
patch
or
a
whole
feature
release
after
the
fact
anything
else.
You
want
to
say
aaron
before
I
jump
into
some
other
things.
We
want
to
touch
on
for
the
release.
C
You'll
notice,
most
of
it,
is
just
slightly
changing
of
the
of
the
either
of
the
verbs
that
are
used
like
the
methods
that
are
used
or
in
some
cases,
reorderings,
because
you
have
to
create
your
instruments
before
and
then
the
other
bulk
change
that
happens
here
is
because
the
metric
test
is
not
available
in
the
api
anymore.
C
That
was
kind
of
an
anemic
meter
provider
that
was
kind
of
one-off
and
had
could
have
potential
funky
problems
with
it.
Just
like
we
had
with
the
trace
test
package
way
back
in
the
day
that
was
removed.
I
think
the
better
solution
that
I'm
pushing
for
is
to
put
an
in-memory
thing.
I
I
left
it
commented
out
specifically
so
that
we
could
come
back
to
it.
C
I
don't
want
this
to
sit
around
for
too
long,
and
it
would
be
a
lot
more
hassle
to
pull
it
all
out
and
then
try
and
put
it
back
in
rather
than
just
uncommenting
large
swaths.
So
I'm
not
happy
with
it,
but
it's
it's
better
than
than
sitting
on
it
for
another
month.
A
Yeah,
I'm
pretty
happy
with
it.
So
if
you
need
some
motivation,
I
honestly,
like
I
think
aaron
said
like
it's
incremental
and
it
gets
us
to
the
next
step
so
yeah.
I
would
definitely
recommend
we
go.
Take
a
look
at
it
cool.
That
said,
the
only
other
thing
that
was
asked
to
be
included
in
this
next
release
for
contrib
is
a
remote
sampling
for
this
aws
x-ray,
detector
or
a
sampler,
and
so
this
is
a
behemoth.
We've
got
some
reviews
on
it
and
we're
working
on
the
feedback
on
it.
A
I,
I
would
probably
say
if
you
have
an
infinite
amount
of
time,
go
take
a
look
at
this.
If
you
don't,
maybe
skip
this
one,
because
we
already
have
enough
reviews
and
it's
it's
a
it's
a
big
one
and
enough
reviews
to
get
this
once
everything's
addressed.
We
should
be
able
to
merge
it
like.
I
said,
though,
if
you're
really
interested,
I'm
not
going
to
discourage
you
in
any
way,
but
that
being
said,
this
gives
an
opportunity.
A
We
have
some
samplers
that
would
be
really
cool
to
include
in
our
our
next
release
and
the
other
one
that
has
been
proposed.
A
long
time
ago
was
this
consistency,
probability
sampler.
This
has
to
do
with
the
probability
sampling
sig.
That
was
started,
probably
about
the
same
time
that
this
was
proposed.
A
A
That
doesn't
happen
that
that
just
doesn't
happen.
So
yeah
I'd,
probably
just
advertise
this
one
as
being
a
priority.
If
you
have
some
time.
A
Maybe
this
afternoon,
but
probably
tomorrow
at
the
earliest,
and
so
I
think
that's
something
to
keep
in
mind.
I
know
people
are
gonna,
be
headed
out
of
town
tomorrow,
so
it's
not
ideal,
but
from
a
maintainer's
perspective
having
nothing
out
by
next.
Actually,
if
we
get
these
two
pr's
merged,
dependable
won't
blow
us
up.
So
I
guess
it
won't
be
the
end
of
the
world.
A
If
this
goes
up
next
week,
however,
having
the
metrics
from
otel
not
consistent
with
the
contrib
is
a
point
of
frustration
for
users,
so
we
do
want
to
try
to
prioritize
this.
A
We've
talked
about
in
the
past
and
just
wanted
to
like
reiterate
what
those
are,
one
of
which
is
adding
the
metric
recorder
for
use
with
the
actual
sdk
that
aaron
kind
of
alluded
to
there's
an
issue
tracking
this
currently
next
would
then,
once
that's,
been
released,
fix
the
contrib
metrics
test,
as
identified
in
the
issue
for
this
pr
here
so
yeah,
that's
kind
of
like
the
next
step,
half
probably
also
with
a
little
release,
go
ahead,
steve.
B
C
Aside
from
the
breaking
change
in
metrics,
it
should
be
100
compatible
to
use
a
newer
sck
with
an
older
contrib,
because
we
haven't
removed
anything.
B
Was
going
to
say,
this
has
been
more
of
a
user
than
a
developer
at
this
point
of
the
library
I
tend
to
not
even
I
don't
adopt
the
new
sdk
until
the
contrib
is
available,
because
all
my
use
of
it
involves
a
few
contrib
libraries
so
like.
If
you
you
know,
if
you're
hoping
for
feedback
sooner
ahead
of
the
contribution
just
from
the
perspective
of
consumers,
I
think
it's
I've
held
off.
Maybe
I
should
start
upgrading
the
sdk
sooner.
A
No,
I
think,
that's
good
feedback
you're,
not
alone,.
A
E
Feedback
that
we
got
from
jurassic,
who
was
trying
to
even
prior
to
the
1.6
release,
trying
to
use
the
the
main
you
know
the
head
of
the
the
core
repo
and
was
running
into
issues
with
contrib,
because
it's
not
kept
in
sync.
I
think
the
issues
he
had
there
was
specifically
trying
to
use
the
new
metrics
sdk
with
the
hotel,
http
instrumentation,
which
includes
metrics.
E
A
Yeah,
but
I
think
that's
good
motivation
to
try
to
get
this
contributory
lease
out,
and
you
know
a
good
motivation
again
to
say
that,
like
taking
out
the
metric
test,
stuff
is
good
and
it's
a
good
step
to
just
get
us
out
the
door,
so
I'm
all
motivated
by
that
cool.
It
also.
E
Kind
of
raises
the
question
of:
should
we
change
our
development
process
to
be
more
like
what
the
collector
does
in
keeping
the
contrib
repo
closer
in
sync?
Updating
it
closer
in
time
letting
keep
keeping
it
so
that
you
could
build
off
of
the
head
of
each
pre-release.
A
So
how
does
the
collector
contrib
depend
on
code
that
hasn't
been
released
from
the
main
repo?
They
just
do
like
a
point
release.
They
do
pseudo
versions
in
their
go
mod.
E
I
don't
think
the
multi-module
handles
it.
What
the
collector
does
is
basically
deletes
all
of
the
gosum
files
does
a
go
mod,
get
to
main
on
the
core
repo
and
the
or
the
core
mod
and
then
gomod
tidy,
which
fills
it
in
with
the
current
pseudo
version.
It's
a
bit
easier
there
because
there's
a
single
version.
E
A
A
E
I
think
the
the
one
advantage
that
may
have
is,
as
we
start
to
iterate
on
the
metrics
sdk,
it
can
help
ensure
that
we've
got
the
tests
there
and
you
know,
as
long
as
and
also
for
changing
the
metrics
api
to
deal
with
feedback
right.
We
we
keep
velocity
on
people
being
able
to
use
the
most
recent
version
and
give
feedback
at
the
most
recent
version.
E
A
Cool,
I
will
take
a
look
at
that.
I
think
that's
something
we
can
add
to
our
releasing
documentation
there
cool
out
of
these
kind
of
stream
of
conscious,
but
I
don't
think
this
goes
next.
Why
don't?
We
talk
a
little
bit
about
the
metrics
sdk
next
steps
and
kind
of
what
the
plan
is
there.
A
I
don't
know
I'd
rather
not
be
just
a
single
person
as
always,
so
maybe
we
can
kind
of
outline
next
steps
for
where
we're
going
and
how
we
can
build
up
that
strategy.
Aaron,
maybe
I'll
hand
it
off
to
you
to
see
what
your
thoughts
on
this
are.
I
guess
you
have
more
context.
C
I
have
a
little
more
context
and
he
I
will
report
that
he's
been
heads
down
in
this
most
of
this
week.
There's
been
a
few
meetings
with
our
planning.
Week
ended
up
being
this
these
last
two
weeks.
C
We
are
hopefully
days
away
from
having
an
into
incomplete
prototype
of
the
sdk,
and
I
think
with
that,
with
a
prototype
in
that
shape.
We
can
then
take
it
as
tasks
to
finish
out
things
like
finish
what
we
need
for
testing
for
discussion
on
this,
the.
C
C
That
being
said,
the
it's
the
sdk
is
a
giant
gordian
knot
and
unraveling.
It
is.
C
Challenging
at
to
say
the
least,
so
I
will
take
steps
to
see
if
we
can
get
even
like
a
work
in
progress.
C
Version
out
so
that
people
can
see
it
and
start
to
start
looking
through
it
just
so
that
we're
not
all
surprised
by
it.
That
being
said,
I'm
not
sure
how
to
really
break
apart
the
the
different
components
of
the
of
the
sdk,
because
there's
a
lot
of
like
really
interdependent,
linked
things.
A
Yeah,
that's
a
good
question.
I
would
imagine
it's
just
gonna
have
to
be.
A
I
mean
honestly,
like
I-
I
see
it
as
like
if
we
could
get
something
that
looks
really
similar
to
what
is
done
in
the
globals
package
that
you
released
aaron
as
like
a
foundation
for
the
sdk
and
then
build
out
tasks
for
each
one
of
the
instruments
to
actually
implement
the
logic.
That
might
be
helpful.
The
only
downside
there
is
that,
like
you
know
this
collection
issues
and
timing
issues
that
the
global
package
obviously
doesn't
have
to
handle
because
it's
not
doing
any
operations.
So
maybe
those
aren't
perfect
sorry
go
ahead.
C
The
the
real
place
to
break
it
apart
is
sort
of
like
there's
handling
of
collection,
of
asynchronous
state
versus
synchronous
state
right,
like
those
two
things
need
to
be
collected
and
then
somehow
combined
into
one
global
state,
and
that
needs
to
have
the
views
portion,
which
is
you
know,
filtering
and
re-aggregating
data
and
then
finally
passed
off
to
the
collection
or
or
or
or
what
not,
the
collection
engine
which
then
can
actually
like
interpret
all
of
that
and
read
through
to
provide
it
to
an
exporter
like
those
are
the
kind
of
big
components
the
you
know,
the
the
duplication
of
of
code.
C
In
that
you
know,
a
a
sync
counter
looks
very
similar
to
a
sync
up
down
counter
which
looks
very
similar
to
a
sync
float
64
counter,
like
that's
that's
mostly
handled
by
generics.
At
this
point.
A
Yeah,
okay,
so
I
heard
a
few
things
there
it
sounded
like
you
had
actually
already
broken
down
some
of
the
logic.
I
don't
know
how
isolated
and
or
interdependent
some
of
those
are,
but
is
that
some
way
we
could
break
that
down
where
we
can
get
some
sort
of
still
some
structure.
A
So
sort
of
skeleton
structure,
then
there's
some
sort
of
we
need
to
resolve
ace
and
sync
collection.
Handle
views
pass
off
to
a
collection,
machinism
that
passes
to
exports.
I
mean:
are
these
some
of
the
big
things
that
you
yeah.
C
Those
are
kind
of
the
bigger
components
and
then
like
there's
gonna
have
to
be
wherever
we
draw
the
boundary
lines
on
those
like
the
one
of
the
problems.
Right
now
is
they're
sort
of
bleeding
into
each
other
yeah,
both
in
the
prototype
and
and
in
our
current
sdk
like
they're,
the
the
collection
mechanism,
kind
of
pervades
everything,
and
it
happens
to
be
both
like
things-
can
push
and
things
can
pull,
and
that
makes
everything
every
component
a
lot
more
complex.
C
So
I'm
sorry
I
I
I
know
josh
has
been
working
in
his
repository.
I
have
kind
of
an
out-of-date
copy.
C
I
will
work
with
him
to
get
just
a
work
in
progress
in
front
of
us
so
that
we
can
start
seeing
the
the
issues
and
then
I'll
take
it
on
tasks
over
this
next
week
to
kind
of
document.
Our
big,
like
diagram
level,
components
that
we
can
start
having
real
real
structured
debates
or
discussions
around
those
okay.
A
Diagram
design,
structure,
cool
and
then
I'm
guessing
you
that'll
be
like
a
google
doc
or
something
you
can
just
share.
Okay,
yeah,
google,
doc
yeah
cool.
A
Next
week,
we'll
just
block
off
some
time
for
reviewing
this,
as
well
as
we'll
be
talking
in
a
sec
cool.
A
Oh,
what
what
did
I
get
to
see?
Sorry,
not
a
virtual
intelligence
right,
yeah.
A
A
I
guess
I
could
scroll
over
users
can't
see
everybody,
okay
with
that.
There's
this
somewhat
unrelated
to
the
sig,
but
also
pretty
related
to
the
sink
issue.
Here
they
kind
of
want
to,
like
maybe
have
some
high
level
discussion
about.
The
open
selection.
Proto-Go
is
something
that
we
depend
on
for
the
grpc
exporter
of
otlp
and
technically
the
http
one
as
well.
But
that's
that's,
I
guess
relevant
to
this
discussion.
A
So
in
the
upgrade
path
we
haven't
actually
upgraded
the
docker
container,
but
the
docker
container
for
generating
photo
buffs
for
otel
has
updated
and
in
the
process
the
dependencies
that
are
generated
by
the
workflow
also
will
update
these
two
latest
versions,
most
importantly
of
which
is
this
dependency.
It
goes
to
a
v2
and
in
the
process
it
resolves
a
security
patch
for
this.
I
think
a
json
library
here,
yeah
or
yaml,
not
json,
which
that's
more
important.
I
thought
that
was
json.
We
definitely
use
yaml.
A
So
maybe
that's
more
important
that
we
get
this
resolved.
But
the
problem
is
is
in
the
upgrade
strategy.
It
depends
on
go
1.7
specifically
coming
in
anthony
pulled
out
from
this
dependency.
A
A
I
wonder
if
maybe
we
can
get
a
new
container
generated
or
if
we
just
say
that
we
support
only
117,
because
that's
the
direction
that
we're
going
and
then
in
like
a
month
or
two
we
actually
start
using
it,
but
that
would
take
a
timeline
for
something
that
maybe
we
need
a
patch
for
security
vulnerability.
I
wanted
to
get
some
opinions
on
this.
E
C
D
A
E
I
mean
I'm
I'm
on
board
with
jettisoning
116
in
favor
of
a
security
patch,
but
I
was
also
the
one
who
thought
we
should
drop
116
support
as
soon
as
eight
one
eighteen
one
came
out
business
part
of
the
policy.
I
think
we
we
ended
up
settling
on
keeping
n
minus
two
for
a
while
longer
and
and
that's
really
what
would
be
holding
us
up
right
as
soon
as
118.1
comes
out,
the
collector
is
going
to
drop
once
well.
E
C
And
the
collector's
a
little
bit
of
a
different
beast,
because
it's
mainly
meant
for
running
as
a
single
binary
right.
So
they
get
to
somewhat
choose
what
version.
C
C
E
C
E
I
think
if
we,
if
we
were
to
go
down
that
path,
we
absolutely
need
to
get
the
next
contrib
release
out
supporting
116
and
say
here's
here's
the
end
of
the
line
for
116..
We
have
to
move
to
117
for
security
reasons.
This
is
as
far
as
you
can
go
in
116.
C
E
Yeah
yeah,
so
like
one
one
point,
six
point
one
we
would
then
ship
do
we
do
that
as
a
practice?
Do
we
do
that
as
one
point
seven,
because
we're
requiring
the
new
version?
I
mean,
I,
I
guess
it's
from
semver
it's.
It
is
a
patch
right,
we're
not
adding
a
new
feature.
C
E
A
Okay,
I'm
gonna
take
notes
and
I'll
take
the
action
item
because
I
open
up
the
issue
as
well
to
update
this,
to,
I
think,
follow
up
one
with
an
issue
tracking
this
in
the
hotel
repo,
because
we
want
to
make
sure
it's
communicated
and
documented,
which
we,
you
know
have
said
in
the
past.
We
need
to
do
a
better
job,
so
I'm
trying
to
do
that
and
I'll
take
the
action
so
make
sure
we
have
something
tracking
that
and
then
move
this
forward.
A
Okay,
awesome
a
I
instead
of
ac
cool
cool,
then
the
only
thing
I
had
on
the
agenda,
which,
if
I
guess
no
one
else
added
anything,
was
this
build
tooling
refact.
It
needs
another
pair
of
eyes,
not
a
high
priority
compared
to
everything
else
we
talked
about
today,
but
it
includes
a
lot
of
really
nice
things
that
help
split
speed
up
the
build,
especially
locally.
You
can
run
make
parallelized
for
a
lot
of
tasks.
So
would
love
another
review
on
that
one,
but
again,
like
I
said,
not
critical.
A
A
Okay,
well,
then,
we
can
jump
onto
my
favorite
section.
Go
ahead.
B
Steve
and
one
thing
I
wanted
to
mention-
maybe
you
could
consider
this
a
user
story.
I
had
I've
gone
through
the
revised
a
set.
One
of
my
favorite
topics
is
the
detectors
for
the
resource
detectors,
and
I
activated
the
one
that
I
think
is
called
with
process
which
is
sort
of
a
tree
of
a
bunch
of
individual
process
detectors
and
one
of
the
things
that
includes
is
the
the
arguments.
The
program
arguments
and
I
discovered
then
that
there
were
several
secrets.
B
Strangely,
you
know
they're
first
bound
as
environment
variables
and
kubernetes
I'd
say
and
then
you
plug
them
into
the
command
line.
Kubernetes
does
a
nice
job
of
hiding
them
when
you
inspect
things,
but
here
they
are
so.
I
went
in
and
I
broke
that
that
with
process
tree
down
and
knocked
out
that
entry,
I
don't
know
if
it
warrants
something
in
the
documentation,
though
to
suggest
maybe
we
should
be
careful.
A
That's
a
really
good
call
out.
I
definitely
documented
that
in
our
user
stories
section
here.
A
B
Yeah,
because
I
you
know
at
first,
I
thought-
oh,
that's
cool,
because
in
some
cases
I
want
to
know
well
what
were
the
command
line
arguments
you
know
like?
Do
you
change
this
cache
size
or
whatever,
with
command
line
flag?
That
might
be
good
to
know,
but
then
again
you
could
argue
that
maybe
we
should
just
include
that
kind
of
detail
after
we
parse
the
command
line
in
the
program.
B
If
we
really
care
about
that
rather
than
trying
to
track
like
in
the
same
way,
we
would
not
publish
the
secret
values
in
our
instrumentation,
you
could
kind
of
turn
around
and
say
we
should
default
all
right.
You
know
we
should.
We
should
not
be
looking
for
those
command
line,
arguments
the
same
way
that
maybe
dumping
the
environment
table
and
your
resource
would
also
be
dangerous.
C
B
Right
so
I
don't
know,
I'm
assuming
that
that
that
all
comes
down
from
the
spec
as
far
as
what
the
you
know,
what
should
be
included
there,
I
don't
know
anyway,
it
was
I
obviously
that
was
a
maybe
yeah.
I
think
that's
that's
more
of.
B
E
B
Right
or
like,
if
even
if
we
add
starting
options
to
it,
it
gets
kind
of
sit
like
with
process.
But
with
this
included
or
not,
you
know
at
that
point
you
probably
don't
need
the
the
sort
of
rapper
macro
like
that.
You
might
as
well
just
me
force
people
to
collect
them
individually,
because
that's
what
I
did
I
went
and
I
replaced
that
one
with
like
six
others,
because
I
think
there's
it
aggregates
something
like
seven
of
them.
E
A
Yeah,
I
also
wonder
how
that
detector
is
created
if
it
allows
us
to
pass
options
to
it.
Maybe
we
could
say
like
exclude
this
argument
or
something
like
that,
but
I'm
quickly
looking
through
the
code,
it's
the
resource
package
just
hard
to
find
anything.
So
I
can't
find
it,
but
I
I
don't
think
if
that's
the
case,
but
I've
got
an
action
item
to
take
a
look,
maybe
adding
a
filter
but
at
the
very
least
documenting
that
this
is
probably
something
to
be
careful
about.
B
A
A
It
seems
that
you've
moved
us
on
to
the
user
story
section,
which
is
where
I
was
going
anyway.
So
if
anybody
else
has
any
other
cool
projects
or
uses
of
open
telemetry
I'd
love
to
hear
about
them,.
A
I
did
notice
the
josh,
the
github
feed
thing
came
out
with
some
new
things.
I
noticed
josh
mcdonald
aaron
is
working
on
some
cool
iot
device
stuff
with
with
go
up
until
you
go.
I
don't
know,
maybe
you
can
come
and
talk
about
that
next
week,
but
I
took
a
look
at
it
and
it
was.
It
was
more
than
just
a
two
minute
glance,
so.
C
Yeah,
so
he
I,
I
can't
give
all
of
the
details,
but
he's
gotten
a
number
of
like
weather
instrument
stations,
I
believe,
and
the
monitors
all
speak
mqtt,
but
he's
working
on
some
kind
of
adapter
into
hotel
for
monitoring
his
home.
C
For
it
it
sounds
really
cool
and
I
I
can
understand
about
half
of
what's
going
on
there,
but
I'll
have
him
I'll
see
if
next
time
he's
available
here
to
to
actually
just
give
a
little
burp?
It
doesn't
even
have
to
be
like
a
full
presentation.
A
Yeah,
I
would
love
to
hear
what
he's
doing
with
this,
because
I
was
super
interested
in
it
once
I
started.
I
definitely
got
nerd
snipes
as
well,
but
there's
some
really
cool
things,
especially
the
conversions
from
mqtt
to
otlp.
So
there's
like
this,
I
think
a
really
neat
use
case
so
yeah.
I
would
very
excited
to
hear
his
use
case
on
this
I'll
post
this
in
the
doc
as
well
in
case
anybody
else
wants
to
check
it
out.
A
Cool,
I
think,
with
that
we
have
addressed
everything
so
yeah.
We
can
probably
hear
20
minutes
early.
I
definitely
want
to
value
everyone's
time.
So
thanks
everyone
for
joining,
appreciate
the
contributions
to
the
project
and
helping
progress
this
and
going
forward.
So
it's
all
valued.
Martin,
I
see
you
and
you
don't
know
if
there's
something
you
wanted
to
add.
D
Yeah
it's
just
this
month.
I
just
want
to
say
my
name
is
martin
and
I'm
just
coming
into
the
community.
So
you'll
see
me
trying
to
do
stuff
out
there
and
I'll
probably
be
like
bambi
on
ice
for
a
while
and
I'll
probably
get
in
your
way
and
annoy
you,
but
I'm
gonna
try
and
help
so
yeah.
If
you
have
any
suggestions,
any
areas,
you
want
me
to
help
in
just
go
with
it.
D
A
All
though,
awesome
yeah
welcome,
martin
sorry,
I
totally
forgot.
I
we
probably
should
add
that
board
to
the
list
of
introducing
new
users
as
well.
I
always
say
this
reviews
are
always
helpful
for
new
users,
surprisingly
more
helpful
than
I
think
most
new
users
realize.
So
if
you
wanted
to
just
review
openpr's
or
other
issues,
I
think
are
good
to
have
feedback
from
the
community
at
large
and
then,
if
you
really
are
interested
in
building
out
something,
we
have
a
help
wanted
tag.
A
I
don't
know
how
up
to
date.
That
is
still,
but
it's
definitely
one
to
look
at
in
the
in.
D
A
The
hotel
and
the
contributory
vote
so
yeah
glad
to
have
you
here
cool
well,
thanks
again,
everyone
I
will
plan
on
seeing
you
all
next
week
and
virtually
otherwise
bye.
Thank
you.
Bye.