►
From YouTube: 2022-06-23 meeting
Description
OpenTelemetry Prometheus WG
A
B
B
A
This
your
first
visit
over
here.
A
Great
well
welcome.
We
get
lots
of
different
folks
showing
up,
and
so
I
figure
we
can.
I
threw
the
meeting
notes
into
the
chat
window.
If
you
want
to
go
in
there,
hi
morgan.
A
B
Yeah,
I
think
one
one
thing
that's
been
kind
of
on
my
mind
recently.
B
I
guess
for
a
while
is
I
think
I
I
personally
at
least
kind
of
have
a
hard
time
when
I'm
like
reading
parts
of
this
fact
or
like
trying
to
figure
out
like
how
something
should
work
or
what
it
needs
to
be
implemented
and,
and
I've
mostly
been
trying
to
help
on
help
out
on
the
php
library
a
little
bit
and
so
yeah
also
trying
to
figure
out
how
things
need
to
be
implemented.
And
but
I
think,
a
lot
of
the
time
I'll
like
read
this
spec
and
be
like.
Oh
okay.
B
I
think
that
makes
sense,
but
I
don't
really
know
like
all
the
like
backstory
behind
it
or
like
kind
of
like
why
it's
like
maybe
a
certain
way,
and
I
think
in
the
past,
I've
tried
to
like
look
through
like
get
history
and
like
look
at
old
prs
and
like
try
to
follow
the
conversation
that
way
and
but
yeah.
B
I'm
wondering
if
there's
maybe
like
a
better
way
for
for
me
in
general,
to
like
get
kind
of
more
context
as
to
like,
like
how
how
the
spec
got
door
was
perhaps
and
yeah
like
for
a
really
concrete
example.
Recently
and
php
doesn't
have
an
api
for
and
setting
a
spam
into
into
that
context,
and-
and
I
think
the
spec
recommends
that,
and
so
you
know
like
I
added
one
in
and
but
then
I
was
like.
B
Oh,
I
don't
really
know
what
use
cases
this
actually
has,
and
then
you
know
I
was
trying
to
dig
into
that
trying
to
understand
how
other
languages
did
it
tried
to
figure
out
where
this
book
came
from
and
yeah.
It
was
just
kind
of
like
yeah
a
lot
of
like
I
don't
really
know
what
I'm
looking
for,
but
I
guess
I'll
figure
it
out
and
so
yeah.
I
think-
and
I
think
also
from
like
a
user's
point
of
view.
B
If
that
makes
sense
like
you
know,
and
maybe
have
some
really
really
really
high
level
model
of
like
oh,
this
is
what
I
see
a
jager
honeycomb
and
like
that
kind
of
corresponds
to
what
I'm
kind
of
putting
out
then
but
yeah.
I
don't
I
guess.
As
of
yet
yeah,
I
don't
really
have
a
different
mental
model
for
like
what's
happening
machinery
wise
kind
of
under
the
hood,
then
so
yeah.
I
think
those
are
kind
of
my
two
big
things.
Let's
say.
A
C
B
B
I
think
my
biggest
challenge
is
still
just
around,
like
you
know,
like
the
mental
model
of
like
me,
emitting
a
log
line
in
like
a
node
app
is
like
a
lot
more
simple,
I
think
than
kind
of
thinking
about
how
to
use
like
yeah,
yeah,
open,
columnar
tree
for
tracer,
and
I
think
what
I've
found
helpful
for
other
tools
in
the
past
is
like,
like
I'll
make
an
analogy
to
like
testing
tools
or,
like
you
know,
for
a
while.
B
I
didn't
really
know
like
anything,
about
testing
the
javascript
space
and,
like
all
the
like
magic.
I
think
it's
like
you
know
just,
and
so
you
know
sign
on
we're
doing
we're
just
kind
of
like
wow.
This
is
amazing.
I
have
no
idea
how
this
works
and
yeah,
I
think.
Since
then,
I've
managed
to
get
kind
of
like
a
somewhat
better
understanding
of
how
these
things
are
working
under
the
hood,
which
I
think
helps
them.
You
know
trying
to
to
use
them
as
well,
but
yeah
for
me
right
now
we're
at
the
tomorrow
tree.
B
B
B
I
think
I
didn't
even
really
understand,
like
anything
around
the
like
schemas
for
semantic
convention
and
stuff
and
then
feel
like
a
section
about
that,
and
I
was
like
oh
that's
what
these
are
for.
Like
so
yeah,
that's
really
really
helpful
and
yeah.
I
think
there's
still
probably
plenty
plenty
of
stuff
kind
of
like
fuzzy
on
but
yeah.
That
was
that's
been
good.
So
far,
though,
gained
some
momentum.
A
Very
complex
project,
and
also
just
a
ton
of
different
spots,
I
don't
think
anyone
probably
has
a
great
mental
model
of
everything.
I
heard
that
commonly
from
folks
that
I've
worked
with
I've.
Just
not
it
takes
a
lot
of
time
and
a
lot
of
engagement
to
get
there.
So
I
appreciate
that
feedback,
since
this
is
definitely
the
first
time
that
I'm
aware
of,
like
an
end
user
persona.
A
But
the
lack
of,
maybe
what's
it
called
a
decision
register,
is,
I
think,
what
you're
running
into
or
like
a
corporate
pro
project
would
have
written
down
decisions
so
that
new
folks
can
kind
of
understand
some
of
the
backstory.
I
don't
know
if
that
would
be
sounds
like
it
would
be
helpful,
but
I'm
not
sure
if
that's
the
right
solution.
B
Yeah
for
sure-
and
I
think
like
like
by
kind
of
analogy,
I
think
like
I-
haven't-
really
brought
open
telemetry
up
too
much
at
work
and
just
because,
like
I
feel
like
I
yeah
like
I,
I
don't
like
if
I
don't
even
know
how
to
kind
of
talk
about
it
correctly
and
you
know
overall,
like
I
feel
weird
like
trying
to
tell
other
folks
about
it
and,
like
I
actually
have
a
hard
enough
time.
It's
just
trying
to
explain
how
like
murals
works,
a
lot
a
lot
of
the
time
to
other
folks
so
yeah.
B
I
think
it's
like
yeah.
I
think
until
I
get
to
the
point
where
I
would
feel
uncomfortable,
like
kind
of
teaching
others
about
it
at
some
level
and
I
think
I'm
never
going
to
really
feel
too
comfortable
because
yeah
yeah
and
I
think
it's
like
a
similar
thing
where,
like
most
of
our
engineers,
aren't
like
super
invested
in
open,
telemetry
or
telemetry
stuff
in
general.
It's
just
like
one
small
aspect
of
everything
they
do
and
so
yeah.
D
Hi,
so
I'm
gonna
pause
on
some
of
my
questions
related
to
sadiq's
dialogue,
specifically
because
I
wanna
know
in
terms
of
in
terms
of
things
that
we
can
help
out
with,
rather
than
just
kind
of
come
to
the
table.
D
Actually,
this
does
relate
to
stakes
league.
There
are
people
who
are
cncf
ambassadors.
I
don't
know
how
many
of
them
are
fully
up
to
speed
with
the
hotel
stuff
right,
that's
probably
a
smaller
than
kubernetes
a
group
of
people,
but
you
know
like
if
you
don't
know
how
to
speak
to
something:
that's
okay,
because
we
all
start
somewhere
but
probably
reaching
out.
If
you
did
plan
some
kind
of
internal
meetup
having
one
of
the
cncf
ambassadors
and
just
scheduling
that
out,
I'm
sure
somebody
would
be
willing
to
help.
D
D
Don't
don't
worry
about
that?
The
the
other?
If
I
may
also
ask,
while
I'm
here
about
this,
trying
to
understand,
the
backstory
is
what's
in
the
document,
so
I'm
sorry,
I
was
a
little
bit
late
and
I
probably
didn't
catch
that
part
of
the
conversation,
but
it
sure
brought
that
up
as
part
of
the
dialogue
too.
D
B
D
D
Yeah,
so
really
the
the
only
reason
well,
first
off,
I'm
glad
to
be
here,
I'm
glad
to
see
all
your
faces.
I
really
enjoyed
open
telemetry
community
day
on
monday
and
got
to
see
shar
in
person,
and
we
missed
you
morgan
and
I
don't
think
I've
met
rin
yet
but
hiran
I'm
I'm
here
to
help
right
not
just
to
ask
questions
so
char.
Is
there
anything
that
the
end
user
group
is
really
focusing
heavily
on?
I
noticed
you've
got
next
priorities.
A
So
I
would
say
the
heavy
focuses
in
general
right
now.
There
are
three
of
them.
One
is
developing
and
continuing
to
promote
the
surveys
so
that
we
can
do
end
user
research,
the
creation
of
community
spaces
for
end
users,
and
so
that's
going
to
be
that,
like
lean
coffee
style
discussion
group,
that's
launching
next
month
and
the
slack
chat
private
slack
channel
and
then
the
third
thing
is
kind
of
track
of
work.
D
D
I
do
a
lot
of
work
with
the
ux
design
team
here,
which
requires
us
to
go
out
and
ask
and
have
honest
interviews
that
aren't
trying
to
put
somebody
in
a
place
they're
actually
trying
to
open
up
the
space
for
people
to
to
say
more.
So
I'd
be
more
than
happy
to
help.
If
I
can
with
defining
that,
but
also
doing
the
work,
the
busy
work
of
potentially
going
out
to
people-
and
you
know
if
we've
got
the
list
of
people
that
need
that
we
need
to
go
out
to
so.
A
Yeah
absolutely
there's
lots
of
opportunities
to
contribute,
and
so
paul,
that's
really
kind
of
on
that.
Like
more
I'd,
say
synchronous,
structured
interview,
I
I
can
so
yesterday
I
put
in
a
pr
in
hotel
community
to
like
put
some
of
these
working
group
content
because
a
lot
of
it's
google
docs
and
if
you
don't
know,
then
you're,
not
it's
very
non-discoverable
and
so
trying
to
push
some
of
that
stuff
out
to
to
github.
So,
hopefully
that
pr
people
are
cool
with
it
and
it
gets
accepted.
A
A
Yeah,
I
hope
it
will
be
discoverable
for
end
users,
if
that's
where
they
kind
of
like
start
and
so
on
that
subject
specifically,
that
leads
me
into
the
next
kind
of
discussion
point
here:
the
structured
interview
guide,
so
we
had
our
first
end
user
interview,
scheduled
for
immediately
after
this
session,
but
unfortunately
they're
having
to
push
that
out
a
little
bit
and
so
in
preparation.
A
For
this,
my
question
was
well
what
what
kind
of
questions
do
we
want
to
ask
and
trend
over
time
in
these
structured
interviews,
or
do
we
want
them
to
be
kind
of
like
end
user
led,
and
so
I've
got
like
some?
A
I
heavily
got
inspired
by
atlassian's
user
interview
guide
framework
that
they
put
out
there,
and
so
we've
got
something,
but
I
think
we
should
build
it
out
and
say
and
be
a
little
bit
more
intentional
about.
Are
there
specific
areas
that
we
want
to
ask
people
much
like
what
we're
doing
with
the
surveys
and
finding
those
trendable
themes
over
time,
but
you
get
so
much
more
rich
responses
when
you're
talking
synchronously.
D
Yeah,
the
more
far
more
qualitative
oftentimes,
no
matter
how?
Well
you
write
the
questions,
some
people
just
don't
grok
it
the
way
that.
D
D
Question
so
those,
but
those
qualitatives
are
good
not
only
to
help
people
actually
answer
it
in
a
meaningful
way,
but
also
to
get
a
sense
for
what
might
we
need
to
adjust
about
the
survey
itself?
What
are
we
not
you
know
what
what
areas
are
we
not
asking
that?
Maybe
we
should
so.
A
Absolutely
and
I
just
dropped
a
link
to
the
kind
of
the
discussion.
It's
pretty
empty
discussion
guide
framework
that
I
recommended
that
I
lifted
from
atlassian,
and
so
I
think
that
would
be
a
great
thing
for
this
group
to
kind
of
collaborate
on
what
are
some
of
those
questions
that
we
should
see.
A
Also
leaving
you
know,
the
ability
for
the
great
thing
about
interviews
is
they
might
go
in
a
direction
you
didn't
expect
them
to
yeah,
and
so
how
do
we
balance
the
flexibility
to
take
the
conversation,
while
also
capturing
a
few
key
pieces
of
information
that
we
think
are
generally
going
to
be?
What
we
want
to
do?
The
other
benefit
of
having
something
like
this
documented
is
for
end
users,
who
are
like,
oh
I'm,
a
little
nervous
to
get
in
front
of
people
and
just
answer
random
questions.
A
So
anybody
have
thoughts
that
do
we
want
to.
You
know
maybe
spend
the
next
10
minutes
chatting
about
a
discussion
guide.
A
Okay,
cool
I'll
watch,
the
clock.
The
link
is
in
the
zoom
chat,
warm
up
questions.
A
I
think
you
know
those
are
very
typical
product
e
questions.
Morgan,
I
feel
like
this
is
your
your
your
expertise.
D
Sure
would
it
make
sense
to
actually
have
this
in
this
doctor
and
another
one,
this
being
sort
of
criteria
for
selecting
the
interviewees
right
like
sort
of
self-select
out?
If
no
welcome
to
the
the
interview,
if
we
got
to
the
wrong
person,
let's
not
waste
time,
you
know
sort
of
a
simple
description
of
this.
This
is
the
type
of
role
or
set
of
responsibilities
that
we're
looking
for
is
that
inside
this
one
or
outside.
A
So
I've
got
a
role
in
responsibility,
but
really
it's
on
the
interviewee
or
sorry
a
little
I
mean
I'll,
send
it
throw
it
in
zoom
chat
right
now,
so
it's
got
a
little
bit
of
rules
and
responsibilities
here
so
far
the
plan
has
been
that
someone
from
the
community
reaches
out
to
because
we've
got
so
many
vendors
working
in
the
project
reaches
out
to
one
of
their
end
users
and
is
like
hey.
I
think
your
story
would
be
really
interesting
for
the
community.
A
We
would
like
to
share
that
outside
of
you
know
the
walls
of
vendorism.
Would
you
be
interested,
and
so
at
least
that's
how?
I
think
they've
been
structured
so
far,
but
I
think
it
you
know,
ideally
over
the
course
of
time.
I
want
to
throw
like
a
google
form
out.
That's
like
hey.
Do
you
want
to
talk
to
someone
in
depth
fill
out
this
form
and
we
can.
We
can
engage
with
you,
and
so
it
would
be
instead
of
a
pull
mechanism
and
be
like
a
push
mechanism.
Yeah.
D
Outbound
inbound
yeah
yeah,
I
guess
what
I
mean
is
criteria
about
the
the
candidate
like.
Let's
put
it
this
way,
if
we
were
to
just
say
the
only
people
we
want
to
talk
to
are
people
who
even
tried
to
use
open
telemetry
before
wait.
A
second
we'd
have
to
like
roll
back
a
little
and
be
like.
Well,
we
want
to
know
we
want.
We
might
want
to
interview
people
who
haven't
used
open
telemetry
before
but
are
in
what
have
you
know?
D
Their
work
has
characteristics
that
you
know
they're
in
the
space
of
monitoring
observability
telemetry
operation.
I
mean
it's
not
not
to
start
naming
out
position
titles,
but
to
kind
of
get
a
simple,
maybe
three
or
four
characteristics
that
are
either
or
not
not
all
of
these
characteristics.
But
if
you,
if
any
of
your
work,
looks
like
this,
we
want
to
talk
to
you
right,
that's
kind
of
what
I'm
I'm
trying
to
figure
out
if
we
need
crisp
identification
on
morgan.
What
do
you
think.
C
I
think
it's
so
we've
only
just
started
the
more
in-depth
user
interviews,
and
so
these
ones
are
mostly
being
selected
by
people
who
are
already
known
to
a
lot
of
participants
in
the
community.
So
for
those
ones
like
it's,
not
that
we
have
any
like
strong
rules
about
it,
but
like
I'm
not
as
concerned
about
the
attendees,
but
you
make
a
good
point
right
once
we
eventually
want
this
to
be
a
broader
invitation
should
not
even
eventually,
I
think
we
want
this
sooner
rather
than
later,
to
be
a
broader
invitation.
D
I'll
just
put
it
in
brackets
yeah
underneath
the
end
user
interview.
Should
this
include
selection
criteria,
perfect.
A
A
A
An
interesting
question
for,
like
the
customer
interview
guide,
should
we
ask
people
where
yes,.
A
C
Yeah
like
what
what
what
telemetry
you're
capturing,
where
you're
sending
it,
I
think
would
be,
would
both
be
really
good,
because
the
the
interview
can
take
a
very
different
direction.
If
we
ask,
even
if
we
just
ask
what
you're
capturing
and
they're
like.
Oh,
we
capture
everything
we
capture
logs,
right
and
we'd,
be
like
like.
Oh
that's,
interesting,
there's
more
out
there
than
that,
as
opposed
to
a
customer
who's
already
capturing
like
the
gamut
of
spans
and
traces
and
logs
and
everything
else,
and
sending
those
to
various
locations.
A
F
A
G
A
F
We
have
could
we
think
about
it,
maybe
like
a
flow
diagram
of
questions
like
if
we
think
the
person
fit.
Maybe
maybe
we
need
to
develop
customer
profiles,
but
if
we
think
the
person
fits
this
profile
like
advanced
adopter
of
open,
telemetry,
beginning
adopter
of
open
telemetry
go
with
these
questions.
C
A
A
Yes,
so
when
I
wrote
up
some
stuff
research
efforts
to
gather
high
fidelity
information
from
end
users
through
the
medium
asynchronous
interviews,
learning
about
their
challenges,
preferences
and
behaviors
regarding
open
telemetry,
the
intent
of
this
is
to
help
test
our
assumptions.
Categorize
end
users,
hey
there's
that
classification
and
describe
their
needs
so
that
the
community
can
react
to
their
feedback.
D
That
they
have
an
easy
path
to
actually
putting
the
open
telemetry
into
their
code.
I
mean
whether
you
have
vendor
stuff
already
in
there
or
not
anytime,
you
touch
the
code,
it's
what
did
you
do?
Somebody
else
has
to
review
it.
Then
it's
like
wait.
You
can't
just
sneak
this
in
there.
Well,
of
course,
it's
not
a
sneak.
So
you
reverse
engineer
from
that
question.
That
might
be
an
intro
question
that
then
you
can
reverse
engineer
into
more
what
what
are
some
organizational
difficulties
preventing
you
from
embed?
D
You
know
injecting
better
telemetry
into
your
apps
and
services
code.
You
know
what
are
things
that
are
actually
preventing
them.
What
was
the
first
part
of
what
I
just
said.
C
D
Yeah,
maybe
maybe
how?
How
is
your,
how
is
the
code
that
you
interact
with
instrumented
or
logged
currently.
A
So
I
I
have
kind
of
an
assumption
that
I
want
to
prove
is
that
open
telemetry
works
for
a
variety
of
use.
Cases
in
production
right
now
today,
like
I
think,
all
of
the
words
about
stability
or
ga
or
experimental
means
very
different
things
to
different
end
users,
depending
on
their
risk.
Tolerance,
and
my
the
assumption
that
I
want
to
test
is
for.
A
C
And
that
that's
that's
the
assumption.
I
agree.
That's
like
similar
one
that
I
want
tested
is
like,
like
particularly
for
instrumenting,
like
back-end
services
back
in
infrastructure,
like
for
things
like
rum
scenarios,
we're
a
little
less
mature
but
like
where.
Where
are
you
finding
the
gaps
today?
Because
if
we
did
this
research
like
a
year
ago
right
when
tracing
was
going
to
or
like
was
that
a
year
year
and
a
half
ago,
I
would
have
been
like.
Oh
it's
a
lack
of
instrumentation
right.
C
That's
the
problem
everyone's
going
to
have
they're
going
to
use
the
java
agent,
but
they're
going
to
be
using
some
library
we
don't
have
instrumentation
for,
and
so
it's
not
going
to
work.
We've
added
a
lot
of
instrumentation
since
then,
and
so
my
base
assumption
is
that
in
most
cases
it
just
works,
and
so
I
I
want
to
find
out
where
end
users
actually
perceive
those
gaps,
because
I
no
longer
have
a
good
sense
of
where
they
are.
D
So
I
created
a
quick
section
way
down
the
bottom.
That
was
just
totally
separate
from
the
nice
structure
that
we've
got
here.
I
think
the
rest
of
the
documents
well
structured,
but
I'm
starting
to
name
out
this.
I
like
I
think
rin
mentioned
it.
First-
is
working
backwards
from
hypotheses
right,
so
the
one
that
I
expressed
was
you
know
people
have
a
hard
time
making
the
case
to
add
telemetry
code
to
their
new
work.
D
How
can
how
can
we
possibly
accept
this?
If
it's
not
certified
certified,
you
know
certified
in
what
which
way
well
somebody's
properly
signing
these
binaries.
C
A
D
D
Just
because,
just
because
ocelot
fixed
something
for
me
and
I
can
play
with
it
in
an
incubation
zone
situation
doesn't
mean
that
all
the
organizations
have
that
luxury
of
just
pulling
a
great
branch
from
some
random
person
who
happens
to
be
awesome
by
the
way.
So
so
let
me
let
me
write
that
one
down.
H
C
For
me,
a
lot
of
this
is
is
really
just
about
the
exploration
right
so
like
like.
I
haven't,
been
adding
too
many
questions,
but
I
think
it's
because,
like
like
char
like
in
many
cases,
this
can
go
many
different
directions
and
at
least
for
the
first
batch,
I'm
I'm
fairly.
Okay
with
that
yeah
right,
like
I,
I
it's
entirely
possible.
C
We're
gonna
learn
things
that
we've
never
considered
in
some
of
these
user
interviews,
just
because
at
least
so
many
of
us,
like
myself
and
char,
and
I'm
sure
many
others
on
this
call,
and
so
many
other
calls
work
at
vendors
and
so
we're
not
spending
most
of
our
days
using
open,
telemetry
or
spending
our
days
extending
it,
which
can
make
us
at
times
a
bit
disconnected.
F
Right,
it
seems
to
me,
like
a
lot
of
our
hypotheses,
have
to
do
with
you're,
just
starting
to
add
telemetry
or
you're
in
yeah,
I'm
afraid
we
don't
have
good
hypotheses
for
I've
added
at
least
some
telemetry,
I'm
sending
a
lot
of
telemetry,
I'm
already
solving
problems
with
telemetry
yep.
F
But
if
we're
gonna
focus
on
beginners,
I
would
also
add
like:
should
we
be
thinking
about
how
easy
has
it
been
for
them
to
get
started
with
open,
telemetry
understand
the
documentation?
Some
of
that
stuff
should
be
table
stakes,
but
if
we're
interviewing
a
bunch
of
beginners,
then
that
seems
valuable
information
to
collect
and.
C
Well,
shopify,
they
contribute
heavily
to
open
telemetry
like
they're,
very
familiar
with
it,
and
I
think
probably
the
first
three
or
four
sets
of
these,
I
suspect,
are
going
to
be
if
anything,
the
opposite
of
beginners,
just
because
they've
been
selected
by
people
in
the
community
and
they're,
naturally
choosing
people
they
know
and
so
for
them
like.
I
doubt
there's
going
to
be
too
many
questions
about
documentation
because
they're
going
to
be
the
same
people
like
well,
I
you
know
I
wrote
most
of
this.
C
So
of
course
I
know
how
it
works,
but
so
that
that
goes
back
to
selection,
right
and
so
sorry,
to
your
point
in
paul's
point
like
once,
we
have
the
selection
criteria,
then
we
can
focus
more
on
either
beginners
or
advanced
questions
or
create
like
flow
charts
for
both.
E
C
C
Exactly
like,
when
I
think
of
our
more
advanced
users
within
like
our
splunk
customer
base,
use
open
telemetry,
their
questions
are
not
about
like
documentation
getting
started
because
they've
long
passed
those
hurdles,
it's
like.
Oh,
can
you
bring
in
this
very
specific
new
type
of
processing
to
the
collector?
For
me
right
like
how
do
I,
how
do
I
do
like
custom
write,
excess
performantly
in
log
processing
in
the
collector
right,
it's
very
sort
of
specific
things,
or
discussions
about
performance,
yeah.
B
C
C
Some
we've,
at
least
for
the
first
batch,
who
I
think
mostly
ted,
has
been
in
contact
with,
or
maybe
shari
you
have.
There
were
a
few
customers
or
customer
users
of
open
telemetry
who'd
requested
that
they're
not
recorded
just
because
they
weren't
sure
about
their.
H
C
C
D
Yep,
okay,
okay,
so
one
thing
I
threw
into
the
meaty
question
is
really
more
like
future.
Thinking
is
what
prevents
you
from
implementing
telemetry
better
or
what
there
are
variety
of
ways
to
say
that
but
like
if
you're
starting
at
zero,
what's
preventing
you
from
taking
it
first,
if
you're
already
like
knee-deep
in
it,
then
what's
preventing
you
from
implementing?
Oh
we're
good,
we
don't
need
anything.
You
know,
that's
that's
the
collins
of
the
world.
One
of
the
speakers
at
ollyfest,
colin
over
at
cloudflare,
was
talking
about.
D
Like
hey,
we
tried
a
little
open
telemetry,
but
we
ended
up
doing
this
cleo
door
thing
because
it's
better
for
our
purposes.
It
helps
in
many
different
ways
and
yeah.
We
still
use
open
telemetry.
It's
like
this
is
where
we
really
focused
and
it
was
kind
of
like
oh
okay,
so
you
kind
of
moved
on
from
just
thinking.
You
can
improve
this
situation
and
now
you've
done
your
own
thing
so
that
that's
a
perfectly
fine
answer
too.
It's
just
capturing
those
by
asking
the
question.
What's
next,
for
you
right
what?
D
D
D
Up
and
you
know
so
like
that's
a
useful,
it's
not
a
demographic.
It's
it's.
I'm
sure,
there's
some
oh
graphic
to
describe
this,
but
it's
like
the
the
organizational
path
right.
The
organizational
structure
to
decision.
D
Would
we
also
sure
do
you
think
we'd
also
want
to
include
for
especially
for
the
people
that
have
implemented
it
already
right,
not
going
from
zero
to
one,
but
maybe
the
question
of
what
something
towards?
D
C
A
That's
been
a
very
strong
theme,
so
I'm
still
at
open
source
summit
of
the
open
source
surveys
that
say
I'm
a
user
versus
I'm
a
contributor
and
looking
at
the
large
delta
between
people
who
are
contributing
versus
people
who
are
taking
and
they
went
into
like
economic
theory.
It
was
pretty
fascinating.
I
can
highlight
that
once
the
recordings
come
out
the
cost
of
open
source,
if
you
will-
and
so
I.
A
I
think
there's
a
lot
of
reasons
for
that.
Maybe
the
question
is:
would
you
like
more
information
about
how
to
become
part
of
the
community
or
something.
A
C
Yeah
it'll
be
interesting,
like
there's
some
companies
and
user
firms
whose
mentality
is
like
no
I'm
either
using
this
because
I'm
a
like
new,
relic
or
splunk
or
whomever
customer,
and
they
happen
to
be
telling
me
to
use
it
there's.
And
so.
Thus
I
don't
care
there'll,
be
someone
say
no,
I'm
into
the
mission
of
open
telemetry.
I
just
don't
feel
the
need
to
contribute
back,
hey,
fair
enough.
It's
free,
open
source
software
go,
do
what
you
want,
or
or
even
further
along
those
lines
who
say:
oh
well,
it
already
does
everything
I
want.
C
A
Excellent,
well,
I
think
this
I
I
really
like
have
having
built
out
some
of
these
warm-up
questions,
because
I
feel
like
that
will
get
more
into
whatever
direction.
The
interview
takes
a
little
more
quickly
than
not
having
them.
So
thank
you
definitely
continue,
throwing
ideas
into
this
and
we'll
revisit
it.
A
I
think
the
first
well
obviously,
the
first
interview
that
was
originally
scheduled
for
15
minutes
from
now
wouldn't
have
gone
with
any
guide
at
all,
so
it
would
have
been
very
ad
hoc
and
so
we'll
see
how
a
couple
of
these
go
and
continue
to
visit.
This
conversation.
A
Yeah
awesome,
okay,
so
back
to
the
agenda:
okay,
lots
of
words,
okay,
so
installing
config
survey,
I
heard
a
couple
of
requests
while
I've
been
out
and
about
that
we
want
deeper
information
as
a
community
about
the
installation
and
configuration
experience.
A
We've
got
a
super
basic
survey
put
together
and
I
will
drop
that
link
into
the
splatter
into
the
zoom
chat,
and
so,
if
folks
would
just
check
this
out,
if
you
have
suggestions
on
how
to
bring
it
a
little
bit
more
deep,
that
would
be
awesome
and
we've
got
a
couple
of
sections
yeah
that
I
will
copy
over
from
the
first
one,
because
we
we
published
it
and
we've
had
23
responses
and
we've
had
no
other
responses.
D
Well,
considering
that
out
of
the
results,
well,
I
just
got
an
email
the
other
day
from
the
attending
the
hotel
days
community
in
austin.
It
was
like
you
know,
for
your
chance
to
receive
something
like
50.
I
was
like
that's
interesting.
I
didn't
know
they
even
bothered
to
do
that
kind
of
stuff,
but
like
duck
is
like
nobody
does
survey.
D
Rarely
does
anybody
do
surveys
without
even
just
the
chance
for
a
small
menial
amount
of
money
or
more
right,
so
it
would
be
cool
to
to
know
like
if
we're
by
morgan
be
cool
to
to
have
a
quick
discussion
on
what
channels
do
we
have
to
promote
the
thing,
but
I'm
sure
you've
already
got
that
situation
solved.
Just
me
coming
up
to
speed
and
then
the
other
one
is
when
we
promote
it.
D
What
do
we
put
in
there
to
provide
confidence
that
it's
people
are
even
going
to
care
right,
whether
it's
a
chance
for
a
hoosier?
What's
this
or
you
know,
hear
your
voice,
you
know
somebody
will
always
reach
out
to
you.
If
you
answer
you
know
or
something
like
that,
I
feel
heard
you
know
so.
A
Yeah
I
hadn't
considered
any
type
of
incentive
survey
incentives,
but
I'm
sure
we've
got
more
than
enough
vendor
sponsorship
that
they
could
kick
in
some
kind
of
incentive.
That's
more
tangible.
D
Yeah,
if
we
do
that,
it's
it's
the
usual
sort
of
marketing
thing.
It's
like
you
know
you
can
kick
out
of
a
random
set
of
somebodies
or
whatnots
of
who
actually
completed
the
thing
first,
second
third
place
to
make
sure
it's
published
and
shared
somewhere
like
a
tweet
somewhere.
Just
like
hey
thanks
to
these
folks
who
won
some
stuff
because
they
had
a
they
cared.
D
You
know
they
did
the
thing
just
so
people
know
that,
like
we,
where
we
actually
did,
do
it
we're
not
just
you
know
fake
bait,
so,
but
I
also
think
that
the
the
the
option
to
have
a
conversation
with
somebody
real
afterwards,
though
it
it
does
cost
time-
is
a
motivational
incentive
for
people
for
certain
people
and
I've
seen
that
actually
work.
A
Select
what
yeah
select,
what
incentive
you
would
want
to
receive
for
having
completed
the
survey.
D
No,
that's
it!
I
love
the
stickers
from
from
earlier
this
week,.
A
And
it's
actually
not
incredible:
it's
just
super
time
intensive,
it's
not
too
expensive.
I
learned
how
to
to
mail
stickers
like
in
an
envelope
around
the
world.
Oh
really
cool.
I
did
yeah.
I
I've
been
using
stickers
as
bribes
at
work
for
a
while
now
yeah,
so,
okay
paul,
I'm
gonna,
think
on
that.
That
is
yeah,
sometimes
forget
that,
like
end
users,
this
is
not
their
day-to-day.
So
we
need
to
show
appreciation
in
more
than
just
making
sure
that
the
surveys
are
structured
well,.
D
A
D
If
you've
already
got
a
target
audience
of
like
peop,
you
know
people
who've,
used
it
more
more
dedicated
to
this
thing
and
you're
already
going
that
as
wave
one.
Might
that
comment
that
I
made
might
might
actually
apply
to
more
like
wave
two
or
something
like
that?
You
know
yeah,
so.
A
Awesome.
Okay,
so
I
just
wanted
to
throw
a
quick
update,
the
end
user
collaboration
spaces.
So
I'm
still
targeting
the
discussion
group,
the
synchronous
discussion
group
to
launch
on
july
14th.
None
of
this
is
published
and
stuff
until
I
would
like
to
because
I
want
the
discussion
group
to
be
under.
Oh
man,
I'm
gonna
have
to
listen
to
this
word,
a
lot
cadm
house
rules,
and
so
I
wanted
that
published
on
github
and
not
just
you
know,
like
a
google
document
before
I
do
it.
A
So
I
link
to
github
instead
of
linking
to
a
google
doc,
I'm
trying
to
actively
deprecate,
and
so
sometime
soon
that
discussion
group
is
gonna,
get
announced
and
it's
being
promoted
on
the
open,
telemetry
website,
they're,
going
to
put
on
a
banner
on
the
open,
telemetry
website
as
well
and
a
blog
post
and
then
we'll
throw
it
in
slack
and
the
slack
channel.
We've
got
a
couple
of
you
in
it.
A
Yeah
it's
not
officially
launched
yet,
but
I
I
figure
there
will
be
discussion
as
we
kind
of
like
grow
and
open
the
gates
of
awareness
versus
people
who
are
just
really
close
to
it.
Being
aware
of
it
right
now
and
thank
you
wren
for
volunteering
to
help
moderate
that
slack
channel
and
enforce
a
code
of
conduct.
F
Absolutely
and
I'm
I
have
not
run
a
lean
coffee,
but
I'm
happy
to
talk
about
how
to
sort
of
make
the
zoom
work
and
logistics.
I
run
a
bunch
of
zoom
events
outside
of
work,
as
well
as
occasionally
in
work.
F
F
A
I
don't
expect
it,
but
I
just
wanted
to
make
sure
that
we
had
consequences
published
sized
about
being
removed
to
the
group
if
they
did
violate
so
that
they
can't
be
like.
Oh,
I
didn't
know,
that's
really
harsh.
F
D
No,
except
for
there's,
probably
a
reason
why,
having
been
in
some
of
these
meetings,
I
don't
see
a
solid
level
of
accountability
in
the
sense
that
we
we
come.
We
go
right
like
volunteer
work,
you
know
it's
a
variable
quantity,
but
what
I
mean
by
accountability
is
like
you
know.
D
If
somebody
says
I
will
actually
actively
do
this
within
the
next
couple
weeks
that
that's
captured
as
like
what
the
what
what
some
of
my
other
volunteer
groups
do
is
like
an
essay,
a
suggested
action
which
is
a
nice
small,
intel
or
or
some
way
to
basically
put
your
name
on
a
thing
and
hold
yourself
accountable
to
that
thing.
D
Because
then,
if
you
end
up
having
you
know,
you
can
kind
of
track
it
week
over
week
and
if
somebody's
starting
to
grow
and
become
like
oh
gosh,
like
I've
got
30
things,
and
nobody
else
does
it
that
it
shows
as
balance
every
time
you
know.
So
I
have
a
tendency
to
take
one
more
thing
that
I
can
actually
do.
You
know
what
I'm
saying
so
it
helps
it
helps
folks
like
myself,
but
also
helps
the
balance
of
accountability.
D
So
I
wonder
if
there's
anything
that
we
do
to
just
make
sure
that
the
progress
is
rolling,
the
people
who
have
a
specific
action.
You
know
yeah.
D
D
Okay,
well
I'll
put
my
name
next
to
one
or
two
things
as
outcomes,
at
least,
if
you
don't
mind
in
this
doc
and
then
I'll
hold
myself
accountable
to
it.
How
about
that
absolutely.
D
It
was
not
only
helping
with
the
structured
interview
guide,
which
we
just
did
and
maybe
going
back
taking
about
a
half
an
hour
this
week
to
go
back
to
that
and
just
be
more
thoughtful
than
10
minutes
allows
me
to
about
some
of
the
questions
and
then
get
that
to
a
point
where
we
have
something
that
we
can
all
agree.
This
is
these
are
the
meaty
questions.
These
are
the
simple
questions.
D
Good,
okay,
close.
Let's
use
that
right.
I
can
help
to
drive
that
to
closure,
and
then
I
can
also
help
to
do
the
actual
interviews.
So
those
are
the
two
areas
I
know
I
can
help
turning
those
into
action
items.
D
A
B
Let's
see
yeah,
I
just
want
to
say
this:
is
this
is
kind
of
fun
getting
to
watch?
Watch
you
all
work
through
this
and
yeah.
I
guess
I
probably
won't
come
back
to
this
meeting
in
the
future
or
I
don't
know
it
doesn't
seem
like
it's
the
right
move
for
me,
but
yeah.
I
go
by
a
brunette
on
the
internet,
so
oh
you're.
D
You,
oh
really:
oh
wow,
okay,
okay,
okay,
very
cool
good
to
meet
you
in
person
and
yeah
like
this
is
more
of
the
ministry
of
getting
the
working
group
going.
Rachar.
D
A
Absolutely
we
we'll
reach
out
to
you
when
we
get
some
of
those
like
really
end
user,
focused
spaces
created
that
might
be
what
you're
looking
for
is
to
discover
best
practices
with
people
who've
actually
used
it
yeah.
A
Excellent
well
thanks
so
much
folks,
we'll
see
in
two
weeks,
bye.