►
From YouTube: 2021-11-16 CNCF TAG Observability 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
B
B
So
I
guess
we'll
get
started
so
welcome
everyone
to
the
latest
meaning
of
tag
observability.
This
is
a
cncf
event.
The
cncf
code
of
conduct
applies,
please,
if
you,
if
you
feel
moved
to
there's
a
link
in
the
chat
here
for
our
document
feel
free
to
sign
in
and
if
anyone
has
agenda
items
to
add
feel
free
to
do
so.
B
B
Is
jennifer
on
the
call?
No
is
anyone
here
from
that
has
context
on
the
chaos
engineering,
working
group
charter.
That
looks
like
the
first
thing
on
our
agenda
for
today.
C
Yeah
should
be
now.
Can
you
hear
me
yeah?
Thank
you,
okay,
so
hello,
everyone
yeah,
so
I
put
it
because,
as
I
got
this
context
from
slack
our
slack
channel
and
essentially
cause
engineering,
as
you
might
remember,
it
was
part
of
observability
tag.
We
used
to
have
cars,
engineering
projects
like
litmus
house
monkey
and
anyway,
like
it,
was
kind
of
weird,
because
there
were
there
was
no
better
place
for
for
them
and
now
there
is
yeah
idea
of
having
another
charter
from
that.
C
So
it
will
help
to
focus
conversations,
so
that's
pretty
pretty
nice
and
they
ask
us
to
get
the
feedback
to
them.
So
there
is
a
report.
There
is
a
proposal
so
feel
free
to
you
know,
come
and
suggest
you
know
other
plus
one
if
you
agree
with
that,
but
I
think
it's
a
great
idea
so
yeah
that's
what
I
wanted
to
share
really.
B
Okay,
well,
if
that's
it,
as
people
are
coming
in
either
folks
that
haven't
been
here
before,
that
want
to
say
hi.
A
I'll
introduce
myself
real
quick,
my
name
is
eric
tyse.
I
work
for
wipro
technologies.
I
lead
the
open
source
program
office,
consulting
and
coe
around
open
source
technologies.
A
We
are
members
of
cncf
as
well
as
linux
foundation
and
many
other
foundations,
and
I'm
here
to
you,
know,
assess
some
of
the
new
projects
and
start
looking
at
places
where
my
teams
can
start
to
contribute
more.
B
So
from
my
side
at
least
it's
been,
I
don't
know
it
seems
to
be
feast
or
famine.
Now
we
we
came
through
kubecon
today.
Actually
the
tlc
meeting,
which
was
an
hour
ago
or
would
have
been,
was
actually
canceled,
so
the
toc
could
catch
up
on
back
business
and
administrative.
B
I
have
a
backlog
about
two
pages,
long
of
github
things
to
do
various
pr's
and-
and
things
like
that,
so
I
know
a
lolita
just
chimed
in
that
she's
unable
to
come
today,
as
is
richie,
so
is
there
anyone
is
there
anything
folks
would
like
to
chat
about.
I
could
give
a
couple
updates
on
on
some
of
that
backlog,
but
I
don't
want
to
monopolize
time
I'd
rather
talk
last,
if
there's
there's
really
anything
that
other
folks
want
to
cover
or.
D
I
mean
it's
on
here,
it's
the
last
bullet,
but
I
was
kind
of
curious
with
the.
I
was
looking
at
the
agenda
from
a
couple
weeks
ago
and
I
saw
the
observed
kubernetes
project.
It
doesn't
have
a
ton
in
it,
yet
I
guess,
and
so
it
doesn't
yeah.
I
was
kind
of
curious
what
the
what
the
motivation
was
there
and
like
what
the
goal
is.
Yeah.
B
I
can
jump
to
that
then,
if
you
like,
if
there's
nothing
else,
my
entire
family,
all
four
kids
and
my
partner
and
I
were
whacked
with
like
a
cold
somebody
brought
it
home
and
the
whole
house
got
it.
So
I'm
about
a
week
behind
I'd
hope
to
roll
out,
at
least
in
broad
strokes
what
it
is,
but
I
started
kind
of
thinking
about
how
could
we
the
tag?
B
Have
you
know
a
project
that
folks
could
contribute
to
around
the
observation
of
you
know
kubernetes
and
its
constituent
workloads,
including
itself,
and
so
that
github
work,
observe
dashcades
is
just
a
shell
home
to
to
to
prototype
some
ideas
around
that
I
did
manage
to
secure
the
I
o
and
dev
domains,
and
so
in
the
in
that,
in
that
context
there
are
two
other
kind
of
interested
parties.
B
I
think
one
is
the
get
ops
work
group
and
back
before
kubecon
there
was
discussion
I
had
raised,
you
know
with
scott
and
a
couple
others
there
that
it
would
make
a
lot
of
sense
to
have
some
sort
of
open
standard.
Like
you
know,
we
have
open
metrics
for
metrics,
we
have
open
telemetry,
and
then
we
have
these.
B
You
know
open
protocols
that
have
enabled
an
ecosystem
to
thrive
where
vendors
can
take
different,
slants
and
different
different
takes
at
you
know
letting
those
open
standards
provide
a
marketplace
right
where
everyone
can
can
grow
and-
and
and
it's
been
wildly
successful
in
this
emerging
space
of
get
ops
tooling,
you
know
many
of
the
observability
vendors
as
well
as
others
you
know
could
take
different
to
could
take
different
shots
at
how
to
communicate
the
list
of
changes
that
are
happening
to
to
production,
environments
or
otherwise.
B
But
there
is
no
real
open
standard
around
deployments
coming
from
something
like
flux
or
argo
or
jenkins
or
whatever
get
ops
stuff.
So
the
gitops
working
group
is
sort
of
interested
there.
I
think,
and
that
there's
a
place
where
you
know
across
tags,
that
the
get
ops
working
group
is
part
of
the
app
deploy
tag.
So
you
know
that
would
be
a
cross
tag
thing
that
could
be
done.
B
That's
one
element
in
my
backlog
and
then
the
other,
and
this
is
that
that
might
be
relevant
for
the
observed
kh
general
organization,
not
a
particular
repo
within
there.
The
other
is
the
sig
instrumentation
upstream
in
kubernetes.
You
know.
If,
if
we
start,
you
know
building
out
tools
in
in
the
open
here
and
it
and
it
you
know,
that's
a
natural
choice
that
more
instrumentation
is
needed
upstream,
general
generally.
B
So
so
the
idea
for
observed
k8s
was
you
know,
there's
a
there's,
a
working
dock
that
I
need
to
drop
some
stuff
into,
and
I'm
kind
of
curious
if
others
are
interested
in
this
too.
But
you
know
what
sorts
of
tools
would
make
sense
in
that
in
that
context,
and
so
for
the
last
couple
of
weeks
or
a
month,
I've
been
kicking
around
a
prototype
and
I'll
have
some
some
some
code
up
and
some
rebates
and
stuff
and
some
pictures
in
the
next
week.
B
But
the
the
general
idea
is
something
that
I'm
not
sure
should
be
called
either
cube
graph
or
time
graph,
but
basically
a
I'm
prototyping
with
neo4j
because
of
the
the
broad
set
of
libraries
that
you
can
use
with
it,
but
imagine
something
that
listens
for
all
crud
operations
to
all
krm
objects,
so
all
creates,
updates
and
deletes
for
all
kubernetes.
B
You
know
pods
and
everything
else
with
a
dynamic
you
could.
You
could
use
dynamic
informers.
You
know
so
combined
with
a
mechanism
that
listens
for
new
kinds
of
things
like
crds
being
installed
with
an
operator
being
installed
to
a
cluster
and
then
listen
to
those
events
too.
And
if
you
have
these
stream
of
you
know,
crud
events,
you
can
put
them
into
a
graph
and
you
know
start
modeling
some
of
the
relationships.
B
That
is
the
the
system
of
things
you
know,
crds
that
refer
to
other
c,
crds,
etc,
and
if
you
make
that
graph
be
able
to
keep
track
of
all
of
those
states,
you
know
so
you
have,
and
this
is
where
pictures
will
help,
but
you
build
the
versioning
into
the
graph.
So,
every
time
an
update
comes
in,
you
know
you
use
copy
on
right
semantics.
You
just
you
make
a
new.
The
new
state
of
that
krm
resource
becomes
the
node
in
that
graph
and
then
associated
with
that.
B
E
B
No,
I
was
saying
that
in
this
broad
observed,
k8's
organization
that
I'm
proposing
that
we
the
tag
kind
of
foster-
this
is
one
example
of
the
kind
of
thing
one
could
implement
under
that
umbrella,
but
you
know
to
have
to
have
both
the
visibility
and
contribution.
I
don't
want
to
go
anywhere
in
any
more
detail
or
to
say
that
what
I
just
explained
with
the
graph
thing
you
know
is
part
and
parcel.
That's
just
an
example
right.
E
So
the
challenge
that
I
see
is
a
attack
doesn't
own
code
and
b,
so
we
would
need
to
to
do
it
through
a
you
know,
some
project
or
or
sig
or
whatever
in
in
code.
It
is
but
a
tag
doesn't
own
code
by
definition
right
and
b.
It
sounds
to
me
a
little
bit
like
we
are
talking
about
concrete
implementations
in
various
details
without
really
thinking
what
do
users
out
there
need
what
what
are
the
expected,
ux
and
and
whatnot
like.
E
Let's
maybe
start
there,
let's
start
with
the
requirements
with
what
people
really
need,
because
I
think
most
of
us,
more
or
less
representing
vendors
are
so
far
ahead
in
the
bleeding
edge
with
you
know
all
kinds
of
fancy
things
where
the
majority
of
users
out
there-
I
I
can
only
imagine
just
like
wow.
E
B
Think
you're,
I
I
think,
you're
correct
again
the
the
the
specific
graph
example.
Any
of
us
could
conjure
up.
You
know
a
half
a
dozen
others
at
will.
B
What
I
meant
to
say
is
that
you
know
similar
to
how
the
the
app
deploy
tag
spawned
the
get
ops
working
group,
the
get
ops
working
group
met
and
worked
for
almost
a
year
and
put
in
place
policy
and
governance
for
what
has
now
become
or
what
is
quickly
becoming
open,
get
ops,
an
actual
organization
that
does
own
code,
but
you
know
with
a
bunch
of
work,
done
ahead
of
time
around
how
that's
structured
and
talking
with
with
some
of
the
members
of
that
working
group
at
kubecon
and-
and
previously
I
think
many
you
know
if
we
wanted
to
do
something
similar
here
like
have
an
open,
observability
organization
right.
B
That,
similarly,
is
not
the
tag
but
as
an
open
organization,
but
the
tag
can
launch
in
the
correct
way
with
all
the
governance,
then
the
things
like
I
had
been
describing,
you
know,
there's
a
fertile
ground
for
that.
Where
open
core
nuggets
you
know
can
be
can
be
implemented.
So
I
perhaps
put
the
car
in
front
of
the
horse
there
in
terms
of
explaining
things
right,
but
that
was
kind
of
the
notion
of
the
organization
so
that
it
could
foster
these
kinds
of
things
either
open
standards.
Or
would
it
interesting.
E
Problems
would
it
make
sense
to
invite
someone
from
the
githubs
working
group
over
to
quickly?
E
If
someone
volunteers,
I
don't
know,
you
seem
to
have
better
connections
to
summarize
what
they
are,
what
they
have
been
doing,
how
they
went
about
stuff
and
and
really
get
an
insight
there,
because
I,
like
I,
used
to
do
githubs
a
year
ago
and
up
until
then
with
refworks
and
whatnot,
I
I'm
a
little
bit
out
of
the
loop.
I
don't
know
what
the
working
group
is
currently
doing,
but
it
sounds
from
what
you
said
that
it's
like
went
super
well
and
they
have
great
deliverables
and
so
on.
E
B
A
great
candidate
as
he's
kind
of
spearheaded
a
lot
of
the
policy
and
and
governance
and-
and
I
think
you
know
there
are
both
collaboration
potentials
around.
You
know
the
two-
the
two
tags
working
together
to
help
foster
open
standards
as
they
emerge,
but
also
to
model
you
know
after
them
and
I'm
I
will
extend
an
invitation
to
him.
B
I'm
pretty
sure
scott
would
come
and
give
an
overview
of
their
experience
in
launching
open
get-ops
from
the
working
group
that
was
spawned
by
the
tag
right
and
and
also
kind
of
talked
with
some
others
on
the
tlc
as
well
as
chris,
and
they
they
concurred
that
you
know
chris's
specific
advice.
You
know
our
cto
for
presidential
was
in
fact
to
model
something
like
observe,
k8s
or
whatever.
B
It
might
be
called
after
what
the
get
ops
working
group
has
done,
so
there's
been
slow,
rolling
and
again,
apologies
is
not
documents.
First,.
E
I
guess
the
main
challenge
is
that
the
majority
of
us
have
not
been
like
yourself
involved
in
all
those
things,
so
we
don't
really
share
the
the
same
understanding
and
insights
that
you
already
have.
You
seem
to
already,
you
know,
invested
quite
a
lot
and
I'm
just
trying
to
understand.
What's
you
know
what
what
is
it
that
you're
suggesting
I
like
the
overall
idea,
but
I
I
really
love
to
work
backwards
from
you
know
what
what
users
actually
want
and
need,
rather
than
you
know,
we
are
all
like.
I'm
pretty
sure.
E
If
you
ask
bartek
hey
come
up
with
a
solution,
he
has
pretty
much
all
the
components,
I'm
pretty
sure
he
has
all
of
the
things
together
and
and
can
can
say
this
is
the
best
way
to
do
it,
but
you
know,
let's,
let's
maybe
start
with
what
is
actually
needed.
What
where
are
people
currently
and
what
do
they
need.
C
I
think
if
I
would
would
like
to
add
that
you
know
I
totally
feel
matt's.
You
know
intention
to
really
like
hands-on
and
just
you
know,
deploy
something
and
finally
show
how
easy
it
is
for
people
right.
So
I
feel
like
there
is
opportunity
you
know
within
our
group
to
really
yeah
just
just
have
fun
demos
going
on,
and
I
totally
see
that
we
don't
want
to
maybe
replace
the
each
project's
work.
C
We
don't
want
to
put
them
in
maybe
wrong
use
case
that
doesn't,
I
don't
know
like
there
are
so
many
things
that
that
we
have
to
be
careful
with,
and
ideally
we
have
demo
of
everything
to
have
like
an
equal,
fair.
You
know
overview,
but
there
is
a
risk
of
course,
but
in
the
same
time
there
is
also
like
this.
This,
like
value
incredible
value
of
this
kind
of
tutorial
effort.
C
So
yeah,
that's
a
good
question,
my
my
thinking,
my
main
kind
of
idea
I
have
right
now
is
that
why
not
collaborating
with
each
of
those
projects
and
somehow
encourage
them
to
to
do
those
tutorials
in
a
consistent
way?
Maybe
that's
the
solution.
E
D
A
E
D
E
E
What
I
understood
from
that
was
more
like:
let's
create
this,
this
kind
of
like
open
standard
or
whatever
great,
but
what
what
you
are
now
saying,
which
I
love
is
essentially
imagine
you
take
the
the
use
cases
that
we
have
in
the
white
paper
and
we
pick
one
random
one
and
show
how
using
cncf
components
and
where
there
is
not
one
other
open
source
to
actually
implement
that
right.
That
could
be,
I
don't
know,
demo
dot,
but
whatever
the
domain
was
that
that
met
already
had.
But
we
can
actually
say
like
look.
E
You
know
here
is
one
instance.
This
is
not
the
only
way
how
to
go
it,
but
this
is
one
instance
where
you
can
actually
click
around
and
see
how
it
actually
implements
this
use
case
right,
but
we
need
to
own
that
we
can't
defer
or
ask
others
like
hey.
You
know,
cortex,
hey
thanos,
would
you
mind
putting
that
together
for
us.
C
Yeah,
I
definitely
used
wrong,
encourage,
wouldn't
solve
the
problem,
but
I
would
still
work
with
them,
because
if
we
create
demo-
and
we
already
in
tunnels-
have
catacola
tutorials,
so
we
already
have
some
of
it.
It's
kind
of
waste
of
time
right,
because
we
kind
of
spread
the
focus
as
well.
So
it
has
to
be
collaboration,
absolutely
good
right.
E
No,
no
I'm
just
saying
I
I
I
did
not
say
or
did
not
mean
to
suggest
that
we
should
ignore
what
what
what
the
projects
are
doing
and
start
from
scratch,
but
also
we
can't
just
turn
it
around
say.
Like
hey.
We
would
like
to
see
this
demo
by
next
friday.
Go
and-
and
you
know
here,
deploy
it
there,
we
need
to
own
it,
and
then
we
can.
You
know
in
certain
cases
like
thanos
has
an
awesome
community,
which
is
you
know
to
a
great
extent
you're
like
you.
E
You
did
that
arctic,
so
there
you
can
benefit
from
many
people
who
already
have
done
something
you
have
to
uncut
kappa
koda,
et
cetera.
We
can
build
on
something
others
might
not
have
that.
You
know
big
of
community
that
that
actually
say,
like
oh
yeah,
sure
our
demos
over
there
jaeger
has
a
wonderful.
I
I
I
know
they
have
this
wonderful
hot
rod
thing
right.
There
are
many
many
projects
that
already
have
something
we
can
just
build
on
that
or
invite
people
to
help,
but
we
need
to
own
that
we.
B
This
gets
to
kind
of,
I
guess,
you're
doing
a
better
job
of
explaining
this
idea
than
I
am
for
sure.
So
so
thank
you
working
backwards.
In
other
words,
you
know,
I,
I
think
one
of
the
things
that
kept
rattling
around
in
my
cage
and
has
been
for
last
year
is
a
case
study
compendium
right,
like
not
the
tag
saying
this
project
is
that
project,
but
people
who
use
the
tools
having
a
place
where
they
can
say.
This
is
how
I
run
thanos.
This
is
how
I
run
cortex.
B
Here's
what
it
looks
like
so
you
know
having
you
know
it's
just
like
like
an
open
observer,
not
open,
observability,
that's
actually
a
thing,
but
you
know
just
having
some
place.
You
know
again
that
can't
really
be
the
tag
it
should.
It
needs
to
be
something
that
we
could
facilitate
the
launch
of
that
has
its
own
legs,
but
having
a
space
to
to
to
have
that
neutral
place
to
just
say:
hey,
here's,
here's
what's
working
for
me!
Here's
what's
not,
and
and
have
that
be
cloneable.
B
You
know
what
I
mean
so
that
if
somebody
were
new
to
all
of
this
open
source
stuff
and
new
to
all
of
these
tools,
they
could
they
could
have
a
place
to
to
see
what
worked
for
other
people
not
not
necessarily
do
anything,
be
right,
the
right
way
so,
for
example,
that
graph
thing
I
was
yammering
on
about
probably
a
little
early.
It
is
me
working
backwards
for
wanting
to
literally
use
a
hololens,
2
or
some
ar
glasses
to
visualize
networks
like
how
would
I
do
that?
B
What
data
do
I
need
so
working
backwards
from
there?
Maybe
a
data
model
like
this
would
look
cool.
Where
would
I
go
find
people
that
might
want
to
work
on
that
with
me,
some
sort
of
umbrella
organization
that
facilitated
that
finding
of
each
other,
you
know
finding
of
contributors
versus
just
you
know
a
needle
in
the
github
haystack,
so
to
speak.
B
D
I'm
kind
of
curious,
so
I
mean
you
mentioned
the
or
michael
mentioned
the
like
hot
rod
for
jaeger
example,
kind
of
thing
it
sounds
like
for
any
of
this.
We
would
need.
I
mean
you're
talking
about
like
the
tooling
that
you
would
use
with
a
sort
of
abstract
project,
but
it
sounds
like
for
whatever
tooling
you
want
to
like
show
off,
or
you
know
show
examples
of
that.
D
You'd
have
to
talk
about
like
what
the
sort
of
requirements
for
that
kind
of
project
are
like
what
you
know,
as
I
assume
you
know,
is
a
project
that
should
be
able
to
sort
of
like
exemplify
the
like
use
cases
of
like
kubernetes
in
some
kind
of
way
and
like
all
that
kind
of
stuff,
and
so
I'm
curious
like
is
there
any,
you
know,
would
you
just
use
existing?
You
know
example:
project
like
hot
rod
or
like?
Is
that
something
that
you'd
be
willing
to?
D
Like
talk
about
here,
of
like
coming
up
with
a
new
sort
of
you,
know
fake
project
that
can
be
used
as
sort
of
like
a
base
to
show
all
the
rest
of
this
stuff.
With.
E
Given
that
we
already
have
written
the
the
white
paper,
I
think
it
would
make
sense
and
also
in
terms
of
a
consistent
story,
to
pick
one
of
like
starting
with
one
of
those
scenarios
saying
like.
Okay,
let's
you
know
pick
one
that
we
actually
show
end
to
end,
but
again
it
would
be
like
you
know.
E
If
I
look
around
here
on
that
call
who
would
actually
volunteer
to
you
know
sit
down
meet,
I
don't
know
once
a
week
for
for
an
hour
or
whatever
and
actually
work
on
that
thing
like
okay,
what
other
requirements
identify
components
integrate
that
we
need.
You
know
there
are
a
number
of
things
we
we
need
to.
You
know
actually
figure
out
how
to
do
it.
How
to
go
about.
It
might
be
that
cncf
has
resources,
certain
vendors
might
be
able
to
throw
in
resources,
but
you
know
someone
needs
to
actually
sign
up
for
that.
E
I'm
more
than
happy
to
you
know
put
some
some
of
my
cycles
in
there,
but
I
I
can't
do
that
alone
right,
so
we
would.
If
we
want
to
go
down
that
route,
we
would
need
to
at
least
say
like.
Okay,
how
do
we
go
about
that?
Do
we
start
this?
Whatever
we
call
it
working
group
or
whatever,
but
we
actually
think
about
how
to
implement
it.
What
do
we
use
as
a
starting
point?
E
A
Yeah,
I
mean
just
just
to
throw
in
one
more
name.
It
would
be
what
I
find
always
very
good
to
use,
is
hipster
shop
or
how
it's
now
called
this
google
cloud
demo
that
could
be
easily
used
for
a
lot
of
things.
E
It's
actually
the
the
sock
shop,
we've
worked
sock
shop
and
I
think
he's
not
on
the
call
today,
but
has
forked
that
and
has
open
telemetry
open
telemetrized
it.
So
it
does
now.
There
is
an
issue
there,
but
the
I
think
we
didn't
so.
A
I
think
I
mean
real
quick,
just
one
more
thing,
and
then
we
see
that
there
is
some
need,
I
think,
in
the
community
as
a
whole
for
some
generic
application.
That
is
well
maintained,
because
that
was
also
the
problem
with
shop
because
it
kind
of
was
out
of
date
at
some
point
so
yeah.
Just
throwing
that
in
not
saying
that
I'm
volunteering
for
that.
But.
B
Yeah,
if
I
could
respond
to
something
that
was
said
earlier,
that
I
think
is
a
really
a
a
really
good
point.
I
think
ryan
you're
you're
talking
about
this.
You
know
when,
when
I
say
like
the
governance
and
the
policy
piece,
which
is
actually
quite
a
bit
of
work,
that
we
could
leverage
some
of
the
stuff
from
the
get
ops
working
group,
you
know
consider
consider
wanting
to
do
a
demo
of
some
open
source
projects
and
have
that
be
contributed
to
this
right.
B
There
should
be,
there
should
likely
be
a
workflow
where
the
people
that
are
interested
in
that
particular
thing
reach
out
to
those
projects
right
because
so,
if
you
look
at
the
toc
call
a
couple
weeks
ago,
they
were.
We
were
talking
about
the
issue
of
the
end
user
technology,
radars
being
taken,
as
you
know
how
you
know.
If
one
just
looks
at
those
surveys,
it
might
say
that
the
cncf
says
this
right
versus
the
cncf
end
user
community.
Similarly,
were
we
to
do
this
and
just
say:
okay?
B
Well,
if
somebody
wants
to
to
put
a
demo
of
how
they're,
using
this
particular
open
source
project,
to
solve
a
real
problem,
cool
go
ahead
and
do
it
right,
it
might
seem
that
that
demo
speaks
for
the
project
right.
B
So
we
should
have
a
workflow
that
goes
that
has
the
person
go
or
the
or
the
team
or
whoever's
doing
this
go
to
that
project
and
make
sure
that
project
you
know,
give
that
project
input
and
and
and
an
ability
to
to
work
with
that,
and
in
that
way
you
know,
I
think,
consistent
with
our
mission.
You
know
we
can.
B
We
can
provide
a
ladder
for
community
members
to
be
involved
with
projects
right
and
we
can
educate
users
but
with
unbiased,
accurate
information
accurate,
because
the
project
itself
was
given
an
opportunity
to
contribute
to
this,
and
that
project
may
not
even
have
even
known
about
this.
This
tag
at
all
or
the
or
these
people
right.
So
some
of
that
governance
work
it's
a
little
bit
meta,
but
I
think
it's
really
important,
I
think,
to
nail-
and
I
think,
michael
you
were
getting
at
that
that
point
as
well.
B
So
I
hope
that
does
that,
does
that
make?
Does
that
make
sense.
E
Yeah
I
mean
to
to
me
the
question
is:
what's
the
next
step
right
I
mean,
I
don't
know
if
we
have
anything
else
on
on
that
and
for
today,
but
if,
if
we
want
to
do
that
kind
of
demo
bet
or
whatever
we
call
that,
then
I
would
be
interested
like
right
here
on
the
call
like
who
would
come
to
such
a
working
group
meeting
and
and
potentially
invest
like
I
I,
as
I
said,
I
certainly
can
invest
something,
but
I
I
don't
like
yeah.
So
we
have
three.
B
I
I'm
interested
I'm
also
kind
of
there's,
there's
also
a
cncf
lab
by
the
way
where
we
can
get
some
hardware
to
like
run
things
and
always
host
things
as
well.
So
there
are
resources
that
you
can
hard
hardware.
B
E
Virtual
hat,
oh
you
mean
the
cloud
so
should
we
should
we
start
a
little
like
indie
in
the
google
docs
there
like
yeah.
B
There's
actually
I'm
scrolling
down
to
get
it.
There
is
a
doc
that
you
can
just
put
your
name
into.
I
haven't
touched
it
since
it's
a
it's
an
almost
empty
doc,
but
I'm
looking
for
it
now.
Yes,
I'm
I'm
pulling
it
out
from
a
previous
meeting.
It
was
basically
we
we
had
some
of
the
stuff
in
flight
and
then
kubecon
happened
I'll,
find
it
in
a
short
minute
here.
F
So
I
definitely
agree
that,
having
these
examples
based
on
the
white
paper
use
cases
is
very
cool
and
very
important,
but
I
think
also
something
matt
was
saying
is
definitely
worth
capturing
as
well.
Now,
maybe
that's
not.
This
isn't
quite
the
right
place
the
whole
idea
of
capturing
what
users
have
done
in
terms
of
their
blueprints
architectures,
whatever
how
they're
using
these
things,
I
guess
the
one
thing
in
my
head
is
if
we
foster
something
like
that.
F
B
So
in
in
terms
of
the
cncf
and
approval
I'll
be
really
careful
that
so
so
this
tag
and
really
in
general,
the
cncf
you
know-
has
an
ethos
of
not
really
being
a
king
maker
like
we're
here,
to
foster
an
ecosystem
and
to
facilitate
projects,
thriving
plural
without
sort
of
picking.
So
so
you
know
we're
not
an
arbiter
or
a
decider
or
an
approver
of
anything.
B
We
need
to
be
careful
to
not
have
that
perception
so
so
that
that
point
is
well
put,
and
it's
worth
repeating
every
meeting
where
this
kind
of
thing
is
talked
about,
but
that's
on
point.
Yes,.
E
In
the
in
the
doc
notes,
thank
thank
you.
So
we
met.
Are
you
essentially
suggesting
that
whoever
is
interested
in
puts
their
name
there
and
then
like
who,
in
terms
of
you,
know
arranging
a
meeting
or
whatever?
Who
who
drives
that.
B
I'll
drive
it
yeah,
just
if
you're
interested
throw
your
name
in
the
dock.
That
was
another
recommendation
from
from
scott
rigby
over
at
the
get
outspoken
group
was
to
kind
of
slow
roll
things.
You
know
put
up
a
doc,
those
that
are
interested
will
emerge
and,
and
we
can
iterate
and-
and
we
can,
you
know,
I
think,
that
the
guidance
that
I've
received
from
everyone
I've
talked
to
is
to
to
kind
of
go
slow
and
take
time
to
do.
B
You
know
the
governance
and
things
before
you
know,
rushing
into
you
know
doing
just
so
that
just
so
that
it's
really
clear
from
the
get-go
that
that
things
are
in
water
and
have
had
rounds
of
review
and
things
like
that
not
only
with
ourselves,
but
you
know
with,
for
example,
folks
that
have
done
it
before
or
with
the
tlc.
B
If
it
involves
anything
that
starts
to
look
like
a
like
a
cncf
seal
of
approval
or
anything
anything
like
that,
in
fact,
to
do
anything
remotely
perceived
as
that
would
actually
have
the
impact
of,
I
think
you
know,
fragmenting
and
splintering.
B
B
Yeah,
I
can
share
my
screen
if
you
like,
but
yes,
it's
quite
specifically,
we've
got
that
in
there
on
purpose.
So
you
know
so
this
is
this.
Is
our
charter
document
right,
and
so
you
know
our
mission?
Has
these
main
main
points
you
know
any
or
all
of
which
we
can?
We
can?
B
We
can
engage
it
and
I
think
what
we've
been
talking
about
covers
a
few
of
these
and
then
you
know
we
have
some
example
areas
in
and
out
of
scope,
but
but
yes,
so,
basically
in
terms
of
governance,
this
is
the
cncf's
operating
model
for
for
tags
and
in
general
like
what
what
what
they
do
there
we
go,
what
they
do
and
how
they
work
and
and
tags
specifically
in
their
operating
model.
Have
you
kind
of
hear
working
groups.
B
It's
in
here
somewhere,
but
but
the
long
infrared
atari
is
a
working
group,
is
meant
to
have
defined
outputs,
they're
actually
approved
by
the
toc
and
the
tag
together
so
like
the
tag
proposes
them
tlc,
approves
them
and
then
they're
formed,
and
they
have
a
chair
or
chairpersons,
ideally,
two
at
least
for
continuity
and
and
all
of
that
and
then
they're
time
time.
B
Boxed
and
you
know
they
they
report
back
and
and
they
define
their
own
goals
and
things
like
that
and
have
their
own
meetings.
But
there
is
a
there
is
a
formalized
process
for
that
here.
It
is.
B
We
can
follow
up
later,
but
they
must
have
factored
it
out
to
the
working
groups,
but
in
any
event,
I
don't
know
if
that
answers
I'll,
find
a
specific
link
and
drop
it
in
the
dock
to
where,
where
where
working
groups
are
described,.
E
Ken
has
an
interesting
question
specifically
around.
Like
I
don't
like
the
question
is,
you
know,
should
we
first
kind
of
finalize
the
white
paper
and
then
move
on
to
the
demos
like
at
least
to
me.
The
white
paper
is
not
something
that's
written
in
stone
like
oh,
you
know,
we
did
it,
it's
a
pdf,
you
can
print
it
out
and
that's
it,
but
it
we
managed
the
first
version
out
there,
but
it's
a
living
document.
It's
in
git,
it's
in
github
and-
and
you
know
you
you,
you
know
we
we
can
extend.
A
E
Can
change
things
we
have
the
the
chinese
mandarin
translation
there.
It's
like
it's
a
living
document
that
hopefully
extends
and
refines
as
we
learn
from
from
users
and
and
the
community
at
large.
F
B
F
F
He
just
made
the
comment
before
he
left
that
he
was
that
finalizing.
The
whiteboard
white
paper
is
on
the
shoulders
of
the
group,
so
I
don't
know
what
exactly
he
means
by
that,
but.
E
We
haven't,
we
haven't
officially
published
it
yet
right,
matt,
we
we
haven't,
like
you
know,
the
toc
hasn't
like
or
like
whoever
officially
publishes
somewhere
on
on
a
blog
post.
Saying
like
here
is
the
first
iteration
of
the
white
paper
right,
but
we
haven't
done
that
yet
maybe
that's
the
context.
Maybe.
F
But
yeah,
if
it's
purely
like
a
process
thing
of
finalizing
it
and
getting
it
published
and
things
like
that,
then
that's
totally
fine.
I
just
the
way
he
wrote
that
I
wasn't
sure
if
there
was
like
some
pieces
of
work
that
still
needed
to
be
done
on
the
white
paper
and
would
therefore
therefore
impact
the
demos
we're
thinking
about.
D
It
still
seems
like
there's
like
a
good
place
to
start.
You
know
even
the
like
situations
that
the
the
white
paper
talks
about.
I
feel
like
it's
never
easy
to
create
a
sandbox
project
that
can,
like
you
know,
satisfy
all
those
situations
that
you
can
then
exemplify
the
solution,
and
so
I
don't
know
I
feel
like
there's,
probably
plenty
of
stuff
that
can
be
done
in
parallel
with
the
even
you
know,
yeah
like
finalized
or
not
finalized
or
whatever.
D
I
bet
a
lot
of
that
stuff,
probably
won't
change
or
change
too
drastically
of
the
you
know,
situations
where
certain
technologies
become
useful,
and
so
we
might
still
be
able
to
kind
of
work
on
something
related
to
that.
While
you
know,
while
the
finishing
touches
get
put
on
the
white
paper,
yeah.
B
Yeah
so
so
working
working
groups
are
are
responding,
are
actually
meant
to
be
a
mechanism
to
enable
sort
of
the
horizontal
scale
of
the
tag
like
they're
meant
to
be
asynchronous
and
in
parallel
we
don't
have
to
serialize
one
thing
for
another
and
in
fact
we
would
we
would
as
we
grow.
Actually,
I
think
you
know
most
tags
have
multiple
tech
leads.
For
example,
you
know
two
three,
four
with
you
know
any
number
of
working
groups
in
in
in
flight
at
any
point
in
time,
so
yeah.
B
So
I
don't,
I
don't
think,
there's
a
need
to
serialize
or
do
one.
I
think
what
bartek
might
have
been
referring
to
was
how
he's
prioritizing
some
of
his
time
on
the
white
paper.
So
we
had
it's
a
long
story,
but
you
know
a
whole
lot
of
work
has
been
done
there
and
and
the
folks
driving
that
are
nearing
completion.
B
I
believe
some
of
the
final
steps
for
the
white
paper
are
review
from
the
toc.
I
know
liz
had
asked
for
that.
The
server
I
think
back
in
september
or
august
in
one
of
the
tlc
meetings.
So
but
again
I
think
bartek
is
is
driving
that
now
or
I
think
arthur
had
to
go
away
for
a
little
while
from
from
the
project.
B
E
That
makes
sense
cool
so
like
we
have
only
a
few
minutes
left
and
I
kind
of
like
don't
want
to
risk
to
you
know
let
that
now
hang
there
for
two
weeks.
Nothing
happens
so
shall
we
stay
like
we
meet
at
some
point
in
time.
The
people
who
are
interested
in
in
a
potential
demo-
and
we
can
also
just
you
know-
use
the
dislike
channel
or
whatever,
but
to
get
something
off
the
ground
to
to
kind
of
like
brainstorm.
E
You
know
how
we
go
about
it
right.
B
Yeah,
I
can
send
out
a
doodle
and,
and
we'll
just
say,
if
you're
interested
put
your
name
in
that
doc
and
we'll
have
a
doodle
open
until
what's
like
thursday
of
this
week,
maybe
or
and
we'll
find
the
time
early
next
week,
that
works
for
everybody
who
who
might
be
interested
yeah
more
of
the
merrier
and
then
yeah.
And
then
that
can
go
in
in
parallel.
E
A
E
Then
you
will
show
up,
and
you
know
we
can
decide
how
to
go
about
it,
and
I
mean
this
is
kind
of
like
I
think
the
first
step
right,
it's
the
kind
of
like
internal
and
informal
like
is
there
enough
potential
for
working
groups?
Are
there
enough
people?
Is
there
enough
horsepower
right
here?
If
not
right,.
B
E
B
And
we
are
free
to
make
our
own
roles
as
well,
so
you
know,
for
example,
we
said
for
a
while,
like
if
anybody
wants
to
help
curate
content
or
do
interviews
for
a
youtube
channel
or
do
branding
work,
or
you
know,
make
cool
video
and
you
know,
there's
all
manner
of
creative
things
that
are
not
engineering
per
se
and
those
can
be
run
by
working
groups.
B
Gibbs
cullen
from
from
chronosphere
is
working
on
and
has
been
for
a
while
and
and
has,
I
think,
is
approaching
critical
mass
and
is
working
on
a
write-up
on
personas.
That
was
another
sort
of
work
stream.
So.
A
E
A
E
Yeah
it
has
this
this
nature.
We
did
that
once
internally,
for
for
like
engineering
and
pms,
and
an
empathy
customer
and
empathy
workshop,
saying
like
okay,
you're
developing
that
service.
Now
you
get
to
use
it
and
build
x
and
that's
fun
right
if
you're
the
person
who
who
actually
builds
something
and
then
you
have
to
actually
use
it
to
build
something,
that's
something
completely
different
and
it's
it's
very
nice.
It's
great
learning,
so
I
agree.
Yeah.
B
A
E
A
For
for
one
of
those
remaining
like
tasks
to
do
in
there,
I
think
it's
like
they're
looking
at
the
roi
of
observability.
That
feels
like
a
dense
enough
topic
that
it
almost
my
gut
says,
put
it
in
a
separate
dock
and
link
to
it.
But
is
that
kind
of
against
the
norms
of
the
white
paper.
E
I
don't
think
that
there
are
norms
per
se.
We
can
do
whatever
we
like,
but
think
of
it
from
a
consumption
point
of
view
right
if
you
are
totally
someone
who
googles
and
figures
out.
Oh
there
is
a
tag,
cncf
tag,
observability
white
paper,
and
you
know
you
probably
want
to
have
it
like
self-contained
right,
like
a
pdf
that
has
everything
in
there.
E
B
Okay,
well
have
a
great
day.
Everybody
54321
class
call
thanks
for
joining
and
welcome
to
the
folks
that
are
new
and
see
you
in
a
couple
weeks.