►
From YouTube: Charts Chat 20171128
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
B
B
B
A
I'll
double-check,
okay:
well,
we
can
follow
up
afterwards.
The
other
thing
that
I
wanted
to
talk
about
was
owners
files
and
the
the
two
reasons
I
wanted
to
talk
about
owners
files
is
one
they're,
probably
about
ready
to
be
able
to
turn
on
prowl
doing
owner
stuff
for
us,
they've
been
testing
it
in
a
few
places,
and
some
projects
have
already
taken
it
on
and
so
I
was
actually
liking
to
shoot
for
doing
the
migration
the
week
after
coop
gun
to
enable
owners
files
on
the
charts
repo.
A
But
that
means
that
we
have
to
generate
a
whole
lot
of
owners
files
here,
which
is
a
whole
lot
of
bumped
charts
versions
and
helm,
ignore
files
altered
and
all
that
stuff
right,
and
so
what
I
was
looking
to
say
is:
should
we
do
this
in
one
PR
that
affects
everything
or
is
there
another
way?
We
should
do
it
to
add
owners
files
everywhere,
I.
E
Think
it
would
be
easy
to
try
to
get
folks
on
ax,
or
at
least
a
couple
of
us
on
our
long
call
to
just
work
through
it
and
do
them
one
at
a
time
and
have
them
run
through
the
test
and
just
kind
of
have
a
pipeline
of
all
of
them
running
rather
than
blast
the
99
PRS
or
whatever
it's
going
to
be,
or
the
99
changes
that
have
the
CI
test
it
all
together.
Okay,.
A
E
Is
the
time
just
just
rock
through
it
and
and
fix
the
linting
too
and
I
mean
if
we,
if
we
do
it,
you
know
with
attention
to
detail
on
each
one,
then
we
get
added
benefits
to
at
least
maybe
decide.
Do
we
even
made
me
want
to
start
deprecating
these
things
and
and
not
move
them
and
actually
just
move
them
to
a
new
folder?
E
You
know
the
the
long-lost
chart
folder
or
something
like
that,
because
the
the
big
PR
is
gonna,
be
I,
think
difficult
to
validate
whether
or
not
the
little
changes
have
broken
things
or
not.
I
mean
I,
guess
technically
we
can
make
sure
it's
just
tell
them
ignores,
and
you
know
the
chart
file.
Maybe
that
would
be.
A
Getting
them
yeah
and
I'm
okay
either
way
it
might
take
us.
You
know
a
couple
hours
in
a
week
to
go
through
all
of
these,
because
there's
a
bunch
of
them
get
a
few
of
us
together
and
I'm
totally
on
board
with
doing
that,
I
just
I'm.
Looking
for
what
path
should
we
do
and
who
will
be
my
cohort
in
and
getting
this
going.
E
I
can
I
can
definitely
do
at
least
at
least
an
hour
of
just
focused
back
and
forth.
Pra
PRA
PRA
video
I
think
we
can
get
a
good
chunk
of
them
in
an
hour
and
probably
take
at
least
two,
but
we
probably
need
a
planning
session
before
that
to
have
a
cadence
for
how
to
do
it
and
stuff
like
that.
So
I'm
happy
to
it
is.
A
F
A
E
A
E
A
B
A
E
A
B
B
Discussed
in
the
email
thread
how
this
was
gonna
work,
but
I
just
wanted
to
make
sure
that
that
at
least
I
remember
exactly
the
workflow
for
it.
So
the
idea
is
that
this
would
allow
maintain,
maintain
as
a
transfer,
not
necessarily
communities,
org
members
to
run
commands
like
/lg
TM
and
then
yeah,
okay
and
then
would
that
then
automatically
merge
office
sometime
after
test
pass,
and
you
know
is
it?
Do
we
have
a
merge
robot
now
that
I
can
go
and
do
that
you
know:
will
we
have
a
minute
robot
or
does
someone
else
actually.
A
It's
the
kubernetes,
merge
robot
that'll,
do
it
it's
part
of
the
test,
infrastructure
called
prowl
and
prowl.
Well,
we'll
look
for
things
like
looks
good
to
me
and
approve
and
be
able
to
act
appropriate
on
those
and
kick
off
the
merge.
It's
entirely
a
piece
of
prowl
functionality.
Now
I
used
to
be
in
something
called,
merge
or
sorry,
munch
github,
and
they
moved
it
over
to
prowl,
and
so
we
should
be
able
to
do
that.
A
It'll,
look
at
things
like
owners,
files
and
collaborators,
and
so,
if
you're,
not
a
member
of
the
kubernetes
project,
as
many
people
aren't,
we
can
add
them
as
a
read-only
collaborator
on
this
and
then
that
would
put
them
on
the
collaborators
list.
So
we
could
take
people
who
aren't
members
and
just
say
you
know:
we
trust
them
enough.
We
can
make
them
emergently
or
a
read-only
collaborator
on
this
repo
and
then
it'll
work
and
it'll
give
them
the
permission.
E
A
Yeah
and
what
what
the
bot
does
is,
it
actually
says
is
this
person
trusted
and
they're
not
actually
just
going
to
look
at
the
owners
file
to
see
it?
They
actually
want
to
make
sure
they're
trusted
by
looking
at
the
organizational
settings
and
all
the
bots
does
is
it
says,
is
this
person
a
collaborator,
doesn't
look
at
their
level
of
permission,
which
means
read-only
will
work.
A
Collaborators
or
a
member
of
the
org
for
kubernetes
org,
and
if
we
want
to
get
a
bunch
of
people
as
members
of
the
org,
there
is
a
process.
We
can
go
fire
off.
Here's
a
list
of
people,
but
in
the
basics
we
can
just
also
just
make
them
a
collaborator
on
the
repo
and
a
read-only
capacity,
and
that's
fine
that'll
work
too.
A
B
A
Okay,
so
yeah
I'll
start
the
doc
and
start
just
listing
out
some
of
these
things
and
if
I
gather
information
I
can
link
off
to
it.
No
there's
can
do
the
same
thing
if
you
all
want
to
jump
on
board.
Early
I
probably
have
more
excitement
to
do
that
than
some
of
you
right
now,
but
I
really
want
to
get
this
switched
over,
because
I
think
that'll
alleviate
pressure
on
us.
Oh,
the
other
thing
we
got
to
do
is
make
the
tests
required
right
now,
they're
optional
and
you
can
merge
without
it.
A
E
E
Yeah
warning
or
something
that
we're
gonna
want
test
for
the
charts,
that's
something
that
we
have
had
in
our
back
pocket
as
far
as
a
quality
measure,
you
know,
but
not
all
charts
and
probably
only
a
handful
actually
implement
a
home
test,
something
that
we
can
use
to
accelerate
our
kind
of
review
process.
Yeah.
A
Yeah
well
the
two
things
that
I
had
on
my
next
two
dues:
where
one
were
tests
and
charts,
but
I
actually
wanted
to
start
linting
the
basically
using
helm,
template
to
generate
the
output
of
charts
statically
and
then
run
them
through
a
linter
and
do
it
for
the
N
and
n
minus
1
versions
just
to
make
sure
it
works
for
both
and
that
they
validate
them.
Both
I
think
that
would
be
a
good
static
test.
We
could
do.
C
A
F
E
E
A
So
maybe
there's
a
little
confusion,
and
this
is
something
we
might
want
to
chat
about.
Stable
and
incubator.
Aren't
the
state
of
the
charts
they're
the
features
they
implement
and
so
stable
doesn't
in
implement
alpha
features,
it's
it's
just
stable
features
and
then
incubator
can
implement.
Things
like
alpha
features
and
be
a
little
bit
more
of
a
playground
there
and
I
think
that's
the
intent
than
the
quality
of
the
year
charts,
which
I
think
is
confusing,
but
I.
A
E
We
need
a
little
bit
of
time
just
to
get
things.
I
think
we
need
to
do
an
audit,
something
like
like
Kenzi,
said:
they're
right,
like
everything
see
where
things
lie.
I
think
we'll
see
that
there's
a
probably
a
third
of
the
charts
that
aren't
at
the
same
level
as
all
the
rest
of
them
from
an
involvement,
maintainer
ship.
E
C
Was
thinking
about
statement
versus
incubator
is
wrong,
didn't
be
like
just
about
making
if
something
goes
into
it,
incubate
everyone
it
like
in
a
certain
time
frame
to
be
promoted
to
stable
or
something
or
that
would
rule
out
other
actually.
So
the
question
is:
do
they
really
need
incubate?
We
wanted
at
all.
Do
we
need
a
place
where
people
can
talk
your
half-baked
stuff,
so
I
mean
why
don't
we
just
require
create
a
chart
that
meets
table,
and
then
you
already
train
a
PR
for
it.
E
Yes,
I
think
I
think
it's
its
transition
to
to
more
or
less
a
practice
ground
for
four
charts.
If
people
submit
stuff
to
incubator
I,
don't
stop
them
from
submitting
it
there
anymore.
If
that's
where
they
want
it
to
be.
That's
fine!
Don't
thing
that
stable
gets
you
today
over
incubators
that
gets
into
the
vein
repository
for
help
right
so
I,
don't
know
that!
That's
you
know
a
great
feature
or
not.
B
Originally
so
I
think
it
really
liked
it.
The
incubator
was
kind
of
you
thought
it'd
be
a
good
place
for
people
to
collaborate
on
things
well,
actually,
initially
like
there
was
a
lot
of
alpha
stuff
that
people
wanted
to
use,
and
you
know
stay
full
set,
for
example,
or
pet
set
back
then
right,
so
we
wanna
the
place
for
that
originally,
but
also
we
kind
of
saw
it
as
a
place
where
people
could
collaborate
and
work
on
things
and
slowly
burn
em
out,
but
we
haven't
really
seen
that
happen.
B
B
B
D
E
D
E
Effort
we're
gonna,
take
ours,
many
many
people,
hours
of
time
to
do
one
by
one
review
of
things.
This
should
be
one
of
those
flex
life.
Has
it
been
dated
in
next
time?
You
know
like
let's,
let's
do
a
little
report
card
kind
of
thing
where
we
can
say
this
one
needs
to
go
into
a
give
it
or
not,
and
make
that
very
clear,
because,
maybe
someone's
you
know,
okay,
why
they
might
get
moved?
You
know
I
moved
out
of
a
stable
repo.
E
B
Know
we
have
the
deprecated
field
in
helm,
but
as
far
as
I
know,
it
still
doesn't
really
actually
tell
the
user
that
it's
dedicated,
mostly
used
in
things
like
monocular,
which
then
you
know
we
can
all
sort
of
that's
something
that
we
can
change
in
home
as
well
as
those
can
actually
treat.
Definitely.
E
F
We
want
to
while
we're
doing
this,
since
we
were
doing
a
kind
of
a
comprehensive
assessment.
At
this
point
we
want
to
go
back
and
look
at
the
old
classic
charts
and
look
for
things
like.
Oh
there
was
a
classic
chart
for
this,
but
now
there's
actually
a
stable,
updated
chart.
Do
we
want
to
keep
the
classic
chart
around
anymore?
F
F
E
E
C
E
A
E
E
F
B
E
E
So
that's
a
mechanism
I
think
we
could
add
more
to
that
if
needed,
I
think
I
would
want
to
figure
out
is
if
prowl
can
do
post
submit
stuff
for
us
I,
don't
think
it
does
any
post
submit
today,
and
then
we
would
just
bake
that
into
too
proud
to
push
to
the
right
repositories.
For
for
the
chart.
Did
you
have
a
particular
one
that
you
think
I
should
get
delegated.
B
E
I
I
have
the
delegation
doc
I'm
gonna
link
it
in
there
and
if
people
have,
this
is
no
brilla
proposal
from
awhile
ago,
and
if
people
have
any
comments,
they
can
have
them
there.
But
basically
the
idea
is
to
have
some
sort
of
meta
repo
information
where
people
can
add.
You
know
some
entries
into
there
that
say
which
are
the
up-to-date
tags
for
their
or
references
for
their
git
repo
where
the
chart
lives,
and
then
we
will
take
and
extract
that
chart
when
it
through
our
CI
and
then
polish
it
to
our
charts
RIBA.
G
G
We
got
there
because
they
are
all
stateful
in
nature,
so
I
would
be
doing
that
by
this
week
and
the
other
one
is
about
area
of
interest,
because
one
other
thing
I
felt
as
a
matron
of
coming
and
trying
to
help
out
the
chart
repose
that
the
number
of
PRS
that
are
waiting
to
be
reviewed.
That's
like
hundred
fifty
two
hundred
the
moment
you
see
that
number.
It's
like
that's
really
demotivating
so,
but
there
are.
G
There
could
be
like
a
few
charts
which
are
really
easy
for
me
to
start
reviewing
and
move
on,
but
there
could
be
a
lot
of
charge
which
I
cannot
actually
start
looking
at
right
now,
for
instance,
like
our
back,
though
I
I
may
not
know
anything
about
our
back.
So
if
that
is
the
first
chair,
I'm
looking
at
then
I
might
not
start
the
VM
here.
G
Also,
in
order
to
motivate
the
chart,
maintenance,
so
I
thought
that,
could
we
introduce
the
concept
of
a
like
area
of
interest
or
the
area
in
which
the
chart
is
actually
being
proposed
or
requires
review
in
so
so
a
PR
creator
could
actually
copy
or
add
that
labels
say,
for
instance,
he
could
say
that
area
is
like
persistence,
volume
or
areas
like
security
for
areas
like
abs,
which
might
include
affinity
and
all
the
code
controllers.
Then
those
can
be
selectively
assigned
to
those
people
who
have
demonstrated
interest
in
that
area.
G
E
The
only
concern
I
have
I
think
that's
a
good
idea.
The
only
concern
I
have
is
having
people
who
send
the
PRS
at
the
labels.
I,
don't
know
that
that's
gonna
work.
If
we
can
find
a
way
to
automate
that,
then
you
know
I
think
we
could
get
the
right
labels
on
there,
but
in
general
I.
Don't
think
that
they
would
be
willing
to
add
that
extra
effort
or
label
it
properly.
For
that
moment,
the
other
side
is
is
we
can
always
have
a.
E
You
know
to
your
first
point
of
like
coming
in
and
just
wanting
to
review
and
not
knowing
what
to
do.
We
could
have
maybe
a
link
in
our
repeat
that
goes
to
a
filter.
That's
like
extra
small
PRS,
for
example.
Those
are
pairs,
are
labeled
by
size
right
now,
so
we
can
say,
like
here's
like
an
extra
small
PR,
if
you
could
just
look
at
and
here's
what
you're
gonna
do
and
have
like
a
little
bit
of
a
kind
of
contributing
from
a
reviewer
standpoint.
A
Also,
and
maybe
here's
a
suggestion
so
reinard
started
this
great
document
called
review
guidelines
where
we
can
document
here's
the
stuff
if
you're
gonna
go
review
it
right,
and
so
some
of
these
areas
where
people
aren't
comfortable
with
like
our
back
right,
that's
a
complicated
thing.
If
you
know
how
to
review
our
back
or
some
of
this
stuff,
can
you
do
a
PR
against
this
file
to
add
details
on
it
to
help?
A
E
A
H
A
H
Okay,
just
that
I
do
like
the
idea
of
having
areas
to
look
at
I
also
agree
that
it's
not
a
good
idea
whatsoever
to
ask
any
users
to
label
those
themselves.
I
think
that's
that's
a
that's!
Just
like
Michael
was
saying:
that's,
that's
not
gonna!
That's
not
gonna
work,
but
but,
however,
those
get
categorized,
whether
they're,
actually
we
label
them
or
whether
we
just
have
a
checklist
kind
of
like
what
Matt
was
saying
within
our
review
guidelines.
H
I
think
there
are
some
areas
that
we
want
to
almost
have
a
areas
of
common
concerns
that
we
want
to
make
sure
that
we
we
check
off.
We
do
have
a
label
for
UX
review.
I've
noticed
that's
almost
a
that's,
become
almost
a
useless
label
again.
I.
Don't
want
a
solution
eyes
right
now
about
how
it's
done,
but
I
think
that
yeah,
our
back
is
one
area.
I
think
someone
else
mentions
persisted
persistence.
H
You
know
like
whether
we
want
to
have
you
know:
stateful
set
integration
across
our
charts.
I.
Think
one
of
the
issues
is
that
we
don't
have
a
ton
of
consistency
in
all
of
our
charts
yet,
and
so
it's
really
hard
for
people
to
know
even
what
what
needs
to
be
included
in
a
new
chart
or
what
needs
to
be
included
in
an
existing
chart
to
get
it
up
to
speed.
You
know,
I
think,
that's
something
that
we.
It
would
be
good
for
us
to
determine
us.
H
G
I
mean
trying
to
coming
in,
there
is
a
small
section
in
the
actual
kubernetes
core
repo
which
says
that
how
to
get
your
peers
reviewed
faster.
So
we
could
have
a
little
section
similar
to
that
and
say
that
if
you
have
this
in
this
requirement
and
probably
request
for
these
levels,
so
I
totally
understand
that,
and
it's
not
I
mean
for
a
season.
The
chart
contribute
of
this.
G
This
wouldn't
be
a
problem
you
would
know
which
area
his
chart
needs
to
be
reviewed
for,
but
for
somebody
who's
new,
he
wouldn't
really
know
it
so
I
mean
you
could
look
at
whether
we
can
take
advantage
of
the
straw
to
allow
people
to
add
labels
themselves
or
whoever
is
planning
to
screen
this
cha
I
mean
the
PRS
at
first
we
could
add
the
labels
out
ourselves.
I
think
that
is
how
it
is
happening
in
the
brain
and
community
as
well.
So.
A
I'm
gonna,
you
know
just
to
throw
in
here.
We
don't
have
where
kubernetes
kubernetes
has
a
lot
of
people
to
go.
Do
reviews
it's
very
easy
to
break
down
areas.
We've
only
got
a
handful
of
people
who
are
doing
reviews
right
now,
and
some
of
them
aren't
necessarily
strong
in
all
of
the
different
areas
and
comfortable
and
reviewing
them,
and
so
what
we
probably
need.
First,
a
set
of
guidelines.
A
We
got
a:
how
do
we
bring
up
reviewers
and
help
them
to
be
able
to
do
that
because
I
mean
you
could
have
our
back
if
nobody's
comfortable
doing
our
back
or
the
one
person
who
is
goes
on
vacation,
then
all
our
back
stuff
stops.
We
really
need
to
look
at.
How
do
we
help
more
people
be
able
to
review
that?
And
maybe
how
do
you
automate
as
much
of
it
as
you
can
and
and
because
there
just
aren't
that
many
of
us
right
now
and.
C
What
we
do
have
now
is
kind
of
a
set
of
best
practices.
We
have
one
for
a
bank,
we
have
one
for
persistence,
I
guess
we
should
probably
document
this
somewhere
so
and
this
is
already
used
in
many
in
many
charts.
Now
our
bag
style,
like
Mike
Lee,
invented
it
I,
think
and
and
how
persistence
and
storage
classes
are
configured.
C
A
D
We've
we've
done
a
pretty
good
job
of
creating
these
best
best
practices
or
these
patterns,
and
then
you
know
hammering
people
over
the
head
with
them,
but
we've
not
done
a
very
good
job
at
all
of
actually
documenting
them,
so
that
yeah
I
think
that
would
go
a
long
way
towards
not
having
to
you
know
hammer
people
would
quit
them
over.
You
know
again:
Michael.
H
D
Think
they
definitely
I
think
those
definitely
belong
in
in
in
the
home,
create
command
alternately,
I
mean
they
shouldn't
throw
it.
You
know
they
should
be
going
into
the
I'm.
Sorry
on.
What's
the
name
of
the
honey
yeah
but
yeah
the
common
chart,
and
you
know
and
and
and
then
you
know,
we
should
be
really
and
then
we
can
start
hammering
people
to
start
using
bats.
You
know
but
yeah.
It's
significant
air
hum
create
and
documentation.
H
Yeah,
where
is
the
canonical
source?
That's
that
was
actually
that's.
Probably
my
first
week
looking
into
this
I
was
like
that
was
what
I
was
like
just
super
focused
on
and
where
I
got
to
is
that
we
don't
know
yet,
but
we're
working
toward
it
and
then
that's
why
I
focus
my
effort
on
chart.
You
tell
because
it
seemed
like
well,
it's
baked
into
helm,
so
so.
H
A
So
if
we
can
get
some
of
this
stuff
documented
and
hot
it'll
enable
a
whole
lot
more
people
to
actually
review
stuff
which
will
speed
things
up
and
then
there
is
making
it
into
something
like
chart
Utah.
So
in
the
future,
some
better
case
and
I
think
that's
a
good
idea
too,
to
make
more
of
it
over
to
chart
Utah
I,
like
that,
we.
F
H
F
B
F
Because
I
feel,
like
we've,
got
at
least
I've
seen
things
on
some
of
this
other
stuff
like
ingress
and
are
back
and
I
I've.
Seen
in
the
notes
back
on
August
31st
network
policies
and
charts
was
an
agenda
item,
but
I
don't
know
that
anything's
actually
been
standardized
there.
So
that's
a
discussion.
We
should
probably
have
at
some
point
yeah.
D
F
F
That's
fine,
but
like
do
something
that
gives
the
lowest
common
denominator
and
the
most
common
case
that
it
will
actually
be
accessible
because
I've
seen,
especially
with
a
lot
of
the
older
charts
where
the
Arabic
stuff
isn't
in
there,
people
are
trying
to
use
them
now
in
our
back
enabled
clusters
and
they
don't
work,
whereas
they
used
to
work
before,
because
our
back
is
blocking
subjects.
So.
A
So
let's
maybe
turn
some
of
this
actionable
because
it's
it's
I
like
talking
about
this
stuff,
but
what
we
really
need
is
something
documented
and
then
once
it's
documented,
that's
something!
Maybe
we
can
even
look
at
writing
tools
against
the
charts
repo
to
see
who's
missing
it
we
can
do.
We
can
do
all
kinds
of
neat
stuff
that
and
so,
let's,
in
the
charts
meeting
minutes
an
agenda,
we've
started
to
collect.
Who
is
up
for
documenting
which
things,
and
so
let
me
drop
that
back
in
here
for
anybody
who
missed
it.
H
C
F
C
A
But
it
should
also
help
anybody.
Who's
got
a
review.
No
okay,
I'm
gonna.
Do
our
back!
I'm
gonna
go
read
over
here:
here's!
What
I
need
to
know
about
our
back,
so
I
can
get
better
at
it,
because
we've
got
different
areas
of
expertise,
and
that
means
some
charts
get
reviewed
more
than
others.
Some
are
easy
to
get
in
some
people
go.
You
know
what
I
don't
know
how
to
review
this
and
I.
Don't
know
where
to
learn
how
to
review
this,
and
so.
E
A
If
we
can
help
people
learn
how
to
do
this
that'll
help
with
velocity,
especially
as
we
go
to
owners
files
and
lots
of
people
not
knowing
how
to
review
these
things.
We
need
to
be
able
to
teach
all
of
these
different
people
are
gonna,
be
owners
how
to
review
this
stuff
and
do
it
well
and
what
that
means,
and
they
need
some
place.
A
They
can
go
to
say,
here's
how
I
should
do
pvcs
here's,
how
I
should
do
our
back
when
I'm
reviewing
my
own
charts
and
the
changes
to
that
and
what
it
means,
because
if
we
don't
get
it
as
a
group,
when
we
go
to
owners,
files
and
you're
gonna
have
a
whole
lot
more
people
doing
it.
A
whole
lot
of
them
aren't
gonna
know
how
to
do
this
stuff
either,
and
we
need
to
help
them
be
successful,
agreed.
H
F
B
Pvcs
are
pretty
I
think
at
this
point
like
fairly
consistent
across
the
destro
ingress
I,
think
you
know
someone
who
I
think
WordPress
Drupal.
Those
are
the
chart,
studies
that
I
know
of
that
have
pretty
good
ingress
model,
I'll
back
I'm,
not
sure,
actually
and
I'm
sure
about
network
policy
over
that.
A
B
B
G
A
Awesome
so
yeah,
and
you
know
it
may
I,
make
a
suggestion
that
for
the
first
pass
we
can
put
it
all
into
just
one
document
and
then,
when
we
see
what's
there,
we
can
see
how
we
can
break
it
up.
I
love
having
a
better
organizational
structure,
but
I
would
rather
have
it
all
documented,
where
we
can
find
it
and
then
figure
out
how
to
better
organize
it
afterwards.
If
that's,
okay,
just
PR.
A
A
H
H
Gaps
can
just
weigh
in
because
most
of
those
people
can't
be
there
in
person
to
help
contribute
to
that
certification.
The
application,
developer
certification
and
if
it
didn't
and
I,
also
mentioned
your
distinction
earlier
Matt
about
the
difference
between
operators
and
and
an
application
developers
when
Michelle
was
on
the
call-
and
she
was
said
yes,
that's
great.
We
don't
really
necessarily
all
have
a
common
language
for
that
yet,
but
we
would
love
to
at
least
somehow
bring
that
distinction
into
it
anyway.
H
I
think,
would
be
a
missed
opportunity
if
we
didn't
at
least
try
to
have
a
conversation
about
what
we
might
want
to
make
sure
is
addressed
on
that
certification,
exam
and
so
I.
Don't
expect
to
do
that
hearing
this
call,
but
I
just
wanted
to
note
it
mention
it
so
that
you
all
know
if
you
didn't
already
hi.
G
Squad
Scott
I
wanted
to
actually
talk
about
this
last
time
when
you
brought
up
it
was
yesterday,
but
I
couldn't
because
I
had
bad
network
connection.
The
thing
is
that
I
am
part
of
that
committee,
which,
which
is
meeting
this
kook
on
I,
think
two
days
before
Conan,
who
is
a
con.
G
But
I
don't
think
I
would
be
able
to
contribute
to
that,
because
I
think
it
only
matters
to
people
who
are
actually
there
and
they're
not
doing
any
video
conference,
but
anything
that
sort.
The
other
thing
is
that
for
people
who
couldn't
actually
help
them
physically
right
now,
we
could
also
help
them
in
terms
of
standardizing
the
I
mean
standardizing
the
marks
for
each
questions
and
trying
to
take
in
the
alpha
test
and
stuff
like
that.
G
So
I
think
there
was
a
there
was.
The
initial
mailing
list
was
sent
out
to
those
people
who
actually
participated
in
cka,
so
I
was
part
of
the
CK
mailing
list,
so
I
got
invited
to
that
call.
So
I
could
forward
that
mailing
list
to
you.
So
basically,
whoever
is
interested.
You
I
have
to
sign
an
NDA
non-disclosure
agreement,
saying
that
you
know
you
wouldn't
release
any
of
the
questions
or
anything
things
like
that,
and
then
you
would
be
part
of
the
committee.
G
H
Iii
have
that
I
sign
the
NDA
and
I'm
gonna
be
joining
those.
What
my
goal
was
just
to
try
to
connect
with
other,
because
I
can't
extend
that
invitation
to
anyone
I'm
not
on
cience
I'm,
not
part
of
C&C
effort,
Linux
condition
or
any
that
I
can't
do
that.
But
I
would
like
to
just
communicate
ahead
of
time
that
that's
what
that
was
what
my
goal
was
just
to
make
sure
that
there
was
some
important
concerns
or
interests
that
people
thought
should
at
least
be
addressed,
whether
they're
large
scale
conceptual.
H
A
H
So
that
I
could
so
that
I
and/or
dilip
it
sounds
like
you're,
they
might
pronounce
your
name
right,
yeah
yeah,
if
you're
going
to
be
there
as
well,
that
we
could
at
least
just
make
sure
that
some
of
those
things
are
brought
up
in
the
discussion
and
covered
there
are
that's
that
is.
That
is
the
extent
of
it.
From
my
point
of
view,
oh.
G
Sorry
about
that,
so
what
my
actual
objection
to
that
was
that
I
mean
this
should
be
something
that
should
be
built
from
the
feedback
from
the
absolute
community,
because
we
are
the
people
who
are
trying
to
standardize
or
tell
how
the
applications
to
be
deployed
and
what
are
all
the
ways
to
deploy
applications.
But
it
appears
that
it
is
being
done
from
those
set
of
people
who
were
not
actually
actively
involved
in
the
gaps.
So
that
was
my
first
objection.
G
G
So
if
there
is
a
way
that
we
could
bring
it
up
to
our
notice
and
then
we
could
have
another
set
of
India
rolled
out
and
we
could
all
participate
and
then
look
at
the
questions,
then
that
would
be
really
helpful
because
I
I
see
a
lot
of
people
taking
that
certification,
Dan
CK
in
my
company,
so
I
would
love
to
participate
in
that
and
then
see
what
should
they
know.
If
they
have
certified
in
terms
of
application
development
in
two
minutes
so.
A
So
so,
let's
turn
this
practical
use.
We
got
about
10
minutes
left.
Is
there
anything
else?
Besides
this
that
we
wanted
to
discuss
today?
I'm
gonna
take
silences
now
so
Oh,
a
couple
of
things.
One
cig,
apps
and
people
in
my
community
have
gone
to
Dan,
both
cig
apps
and
the
app
dev
working
group.
People
there
have
told
Dan
you
know
doing
it
this
way
and
trying
to
hurry
up
and
get
it
done,
and
not
necessarily
looping
in
a
bunch
of
the
apps
people
is
probably
gonna
lead
to
something
that
doesn't
hit
home
at
first.
A
In
fact,
the
cluster
operator
exam
has
run
into
the
same
problem
where
there
were
huge
outcry
over
it
in
a
very
negative
fashion
when
it
first
came
out
because
of
the
content
of
it
and
that
same
fears
here
for
the
app
dev
certification
effect.
We
don't
even
have
the
difference
between
an
app
developer
and
an
app
operator
like
a
developer
and
operator
is
the
same
role
and
in
some
companies
it
is
and
lots
of
others
it
isn't,
and
so
all
of
this
has
actually
been
brought
to
dann.
A
In
this
way
of
doing
it
and
they're
driving
ahead
with
it
anyway,
and
so
this
is
kind
of
out
of
our
control
and
the
best
we
can
do
is
look
at
influencing
it
as
positive
as
we
can.
We've
brought
it
up
in
cig,
apps
to
say
hey.
If
anybody
has
the
time
and
is
capable
of
going,
please
sign
up
we'll
make
the
connection
dan
came
and
presented
at
cig
apps
in
order
to
kind
of
kick-start
some
of
that
going
as
well.
We
can't
change
whether
or
not
they
do
it.
H
A
A
How
do
you
use
them
and
when
do
you
use
them,
but
also
because
the
that
it's
not
just
the
test
there's
a
training
that
they
want
to
go
along
with
the
test
where
they
teach
people
how
to
do
things
and
so
I'm,
looking
at
what's
really
gonna
be
useful
for
people
and
then
how
do
you
test
it,
and
this
is
about
a
popper
ation
if
I
understand
it?
Basically,
the
idea
is
we'll
hand
you
something
like
an
ojs
app
and
now
you
figure
out
how
to
operate
in
kubernetes.
That's
the
way,
they're
approaching
it.
Not.
A
How
do
you
build
the
app,
but
now
how
do
you
just
run
any
app
in
kubernetes,
and
so
it
is
that
app
operator
and
so
I
would
say
you
know
how
do
you
use,
which
is
the
different
kind
of
for
workload,
API
things
in
order
to
do
it,
but
then
look
at
how
do
you
set
the
application
up
to
use
our
back?
Our
back
is
incredibly
painful
and
I
suck
at
it,
and
a
lot
of
people
do,
and
so,
if
there's
gonna,
be
training
focused
around
the
exam.
A
If
we're
able
to
really
help,
you
know
the
training
around
our
back
and
I
would
say
maybe
Gress,
and
maybe
even
network
policy
like
something
about
those
in
there
partially,
because
it
gets
you
into
the
training
and
for
anybody
who's
gonna
go
take
this
and
I'll
probably
be
included
in
that
it'll
force
me
to
really
make
sure
I
know
what
I'm
doing
I
think
those
are
useful
things.
Some
of
the
things
that
we
talked
about
today
is
kind
of
that
guidelines.
Here's
the
areas
we
need
review
in
the
charts.
A
H
I,
just
twinkle
fingers
up
that
support.
That
message,
mainly
you
know,
like
I,
believe
that
if,
if
some
of
us
didn't
talk
about
this
ahead
of
time,
you
know
I'm
not
convinced
that
that
helm
or
charts
would
even
be
discussed
or
or
included
in
that
you
know
what
I
mean
and
I
think
that
given
given
that
helm
is
a
is
the
package
manager
for
kubernetes,
or
at
least
billed
as
such.
It
really
should
be
part
of
that
yeah.
A
H
A
So
how
do
you
tie
it
into
some
of
these
I'm,
not
sure
how
far
we
get
into
it,
because
then
it
starts
to
get
opinionated
and
deployment
processes.
But
how
could
you
go
about
using
some
of
these
core
standard
tools
in
the
CI
tool
chain
or
just
in
general?
What
are
the
base
objects?
I
mean
it
comes
down
to
the
basics.
What
are
the
base
objects?
How
and
when
do
you
use
them?
A
And
then,
after
that,
what's
the
next
level
tooling
and
how
and
when
would
use
those
and
some
this
gets
hard
with
it-
the
app
operator
stuff,
because
not
everybody
uses
something
like
helm.
Lots
of
people
do,
but
not
everybody
does
it's
not
completely
pervasive
or
the
kubernetes
core
objects
are,
but.
H
That's
a
question
see
and
that's
part
of
the
question
that
I
think
is
I.
Don't
that's
it.
A
foundational
distinction
between
an
app
operator
and
app
developer
is
not
really
fleshed
out
if
it
is
fleshed
out
like
based
on
what
a
conversation
was.
Maybe
you
and
others
have
done
work
on
fleshing
it
out
more,
but
I
only
saw
one
one
line
mentioning
mentioning
is
and
I
don't
think
everybody
even
knows
what
the
difference
between
those
are.
In
fact
I.
H
A
I
might
go
write,
something
like
this
up,
because
I
mean
I've
been
around
conversations
on
the
distinction
and
I've
worked
at
companies
with
the
distinction
for
a
very
long
time,
but
I
don't
hear
a
lot
of
it
in
practical
terms,
put
out
there
or
and
I,
don't
really
read
a
lot
of
it
and
writing
yet
DevOps.
That's
a
good
thing.
You
know
its
traditional
been
you've,
got
ops
and
then
you've
got
developers,
and
then
you
kind
of
got
DevOps
in
the
middle
and
people
on
both
sides
and
sres
right,
because
that's
are
easier
than
once.
A
You
do
it
and
go
and
DevOps
doing
what
it
would
be,
and
so
it
kind
of
depends.
It's
a
role
thing
right,
I
mean
some
companies
have
their
application
developers
also
operate
it.
Some
people
have
embedded
people
on
the
team
who
could
help
with
the
operations,
but
it's
shared,
and
it's
that
kind
of
same
thing
here.
But
it's
you
know
not
always
just
the
person
who
writes
the
up.
A
The
people
who
write
WordPress
don't
necessarily
operate
it
right
and
the
people
who
write
your
add-on
functionalities
for
your
UI
may
not
be
the
same
people
at
your
company
who
are
going
and
operating
that
same
WordPress
site
you
might
have
to
you
know.
One
person
does
all
the
fancy
UI
stuff
and
then
somebody
else
goes
ahead
and
operates
it
and
that's
that's
not
an
uncommon
distinction
and
I
say
that
cuz
WordPress
is
used
in
most
of
the
Fortune
100
companies,
at
least
that's
my
case.
Okay,
yeah.
G
My
opinion
is
that
the
app
developer
is
somebody
who's
writing
application
when
he
doesn't
really
care
where
the
application
is
going
to
be
deployed
to
be
held
to
or
to
Panetta's
our
PCOS.
If
that
is
the
role
that
the
app
developer
is
taking,
then
why
should
he
be?
He
be
even
bothered
about
kubernetes
or
getting
certified
in
kubernetes.
G
So
when
we
mean
that
kubernetes
application
developer
or
in
this
context
means
that
somebody
who
could
deploy
application
and
communities
so
which
means
that
would
want
to
operate
the
application
so
I,
don't
think
that,
for
me,
it
is
very
difficult
to
understand
what
what
is
the
role
of
an
application
developer
inside
Google.
It
could
be
dwelling
without
as
well
yeah.
A
In
fact,
one
of
the
concerns
that
was
part
of
to
me
was
you
know
there
isn't
something
special
about
writing
an
application
to
run
in
kubernetes.
You
don't
have
to
do
any
kind
of
magic.
You
can
it
kind
of
config
files.
I
can
use
environment
variables.
There's
there's
no
special
way
to
make
it
run
in
kubernetes.
A
You
can
make
all
kinds
of
stuff
run
in
there
and
by
calling
it
the
kubernetes
app
developer
certification,
you
might
give
people
the
impression
that,
in
order
to
write
an
application
for
kubernetes,
you
have
to
do
something
special,
because
app
developers
write
the
application
they're,
not
necessarily
the
rule
that
operates
it,
and
that
could
cause
confusion,
and
so
this
was
conveyed
to
me
by
somebody
and
I
hadn't
thought
of
it.
Well,.
H
What
about
people
who
are
writing,
kubernetes,
plugins
and
and
operators?
You
know
like
ours
that
considered
apps.
Is
that
part
of
what
is
it
within
the
scope
of
that
exam?
You
know
and
I
I
I,
wonder
about
that
and
I
wonder
if
there's
just
too
much
of
a
gulf
again
I'm
not
trying
to
to
just
raise
a
paradoxical
questions:
I'm
just
really
practically
speaking
like
there.
Yes,
like
you,
said
like
we
can't
stop
them
from
start
doing.
H
The
certification
I
personally
wasn't
involved
those
earlier
conversations
that
you
have
with
Dan,
so
I
didn't
even
attempt
to
try
to
stop
them.
It
just
seemed
to
me
like
there
was
a
lot
of
gray
area
and
a
lot
of
on
I
lacked
a
lot
of
lack
of
clarity.
You
know
and
I'm
glad
that
you
invited
him
in
last
minute
to
to
to
discuss
with
the
Vista
gaps
group,
because
it
gives
all
of
us
the
opportunity
to
have
these
conversations.
I
guess
I
wanted
to
try
to
do
that
beforehand.
If
we
can
I've.
F
Started
to
think
of
it
in
basically
three
domains.
There
are
applications
that
run
in
kubernetes.
There
are
applications
that
leverage
kubernetes
and
then
there
are
things
that
extend
kubernetes
so
like
the
Prometheus
operator
would
extend
kubernetes
because
it
creates
a
resource
type,
something
that
had
like
the
elasticsearch
has
a
plugin
that
uses
the
kubernetes
api
to
discover
other
masters,
though
there
again
that's
more
than
just
running
an
application,
but
then
something
that
just
like
you
know,
you're
running
a
web
server
inside
kubernetes,
like
you
said
it
doesn't
need
any
special
capabilities
or
knowledge.
F
A
And
and
so
we're
getting
a
little
depth
in
technical
here,
I,
don't
know
that
the
CNC
F
folks
are
going
this
depth
in
thinking
about
the
technicalities
of
kubernetes
the
way
we
are
I
think
they're
looking
at
it
as
something
pretty
simple.
We
had
a
cluster
operators
certification
and,
quite
frankly,
there
are
a
whole
lot
more
people
who
are
going
to
do.
A
Apps
and
kubernetes
then
run
the
clusters,
and
so
it's
a
really
big
market,
and
if
we
call
it
app,
devs,
that's
a
huge
market,
but
what
Dan
told
me
that
we're
really
looking
at
is
grab
something
like
an
ojs
app
and
we
hand
it
to
you
and
you
figure
out
how
to
operate
it
in
the
cluster
and
so
that's
kind
of
where
they're
looking
at,
and
this
is
marketing.
This
is
sales.
This
is
have
a
very
wide
certification
for
another
app
or
you
know
group.
A
That
is
not
this
and
if
you
say
app,
operators,
that's
even
smaller
than
the
app
developers
they
want
to
go
broad
with.
It
is
the
impression
that
I'm
picking
up
and
so
I
think
a
lot
of
the
naming
and
terminology
around
it
isn't
to
get
technical
and
how
we
can
do
things
in
kubernetes.
It's
really
to
go
wide
and
get
a
whole
lot
of
people
certified,
because
that's
kind
of
something
that
the
Linux
Foundation
does
is
certifications
all
over
the
place.
I.
H
A
You
mean
probably
not
they
don't
even
think
about
it.
That
way,
they
say
it
at
the
devs.
Did
you
know
so?
The
nodejs
project
is
under
the
Linux
Foundation,
and
did
you
know
that
there's
a
nodejs
application,
developer,
certification,
I,
don't
know
anybody
who's
got
it,
but
there
is
one
because
they
do
this
I
think
this
is
something
that
the
Linux
Foundation
does.
They
do
lots
and
lots
and
lots
of
certifications
and
app
developers
are
far
and
wide
groups,
and
so,
if
they
are
targeting
lots
of
people
who
do
this,
ok
there's
no
special
way.
A
So
what
can
we
focus
on
them?
Doing
and
so
I
would
target
is
what's
pragmatic
for
the
wide
variety
of
people
they
probably
won't
rename
it
because
that
shrinks
the
target
market.
This
is
not
a
technical
thing
as
far
as
being
how
do
you
be
the
most
technically
proficient?
This
is
sales
and
marketing
around
you
know,
certifications
and
training
courses
right.
H
G
Like
to
the
Department
of
the
exam
is
totally
interesting,
I
mean
they
actually
give
you
a
real
cluster
and
they
ask
you
to
debug
and
then
fix
it
so
and
so
be
the
given
time.
If
you
could
solve
like
a
couple
of
them,
then
it
would
be
awesome
if
you
could
solve
things
like
that
in
the
real
world
in
just
three
hour,
be
awesome.
So
some
it
is
interesting,
but
a
lot
of
question.
G
H
No
I
was
just
you
know.
Basically
what
I'm
thinking
is
just
ultimately
what
is
this
exam,
at
least
at
a
high
level,
even
as
you
said,
Matt,
if
the
goal
is
largely
sales,
but
also
just
just
to
gain
a
wider
adoption,
that's
still
a
decent
goal,
you
know
I
mean
it
may,
but
you
know.
So
how
is
it?
How
should
it
be
distinguished
and
essentially
I'm
just
wondering?
Is
there
any
value
from
the
Civic
Lisa
Gups
community
weighting
in
you
know
what?
If
so,
what
value
can
we
bring?
H
A
And
and
I
would
just
start
by
the
very
basics-
the
kinds
of
things
we
talked
about
earlier
today
needing
to
get
documented
in
charts.
But
how
are
those
things
for
air
kubernetes
objects
and
how
can
people
know
how
to
use
it,
because
that'll
then
tailor
the
trainings
to
pass
the
exam,
and
then
that
makes
the
trainings
useful
to
a
wide
slot.
You
know
wide
group
of
people
and
its
operation,
but
it
actually
makes
it
useful
for
people
to
know
how
to
operate
their
apps
and
I.
Think
that
would
be
useful
as
a
first
go
and.
G
One
more
small
thing:
I
mean
I,
totally
agree
with
what
Matt
said
about
creating
I
mean
creating
a
positive
influence.
So
to
add
to
that
one,
the
scenario
that
we
should
look
at
is
that
whoever
is
getting
certified
as
kubernetes
certified
application
developer.
They
should
at
least
know
when,
should
they
use
what
sort
of
workload
so
I
mean?
What
when
should
they
use?
Stateful
said:
when,
should
they
use
deployment?
When
should
they
use
jobs?
That
should
be
very
clear,
so
somebody
who's
coming
out
of
that
certification.