►
From YouTube: Kubernetes Community Meeting 20180111
Description
We have PUBLIC and RECORDED weekly video meetings every Thursday at 10am US Pacific Time.
(Sound missing for first 30 seconds, but kicks in right after, sorry about that!)
Notes: https://docs.google.com/document/d/1VQDIAB0OqiSjIHI8AWMvSdceWhnz56jNpZrLs6o7NJY
A
A
A
You,
okay
I,
will
try
my
best
to
speak
as
louder
as
I
can,
but
thankfully,
as
a
moderator,
I,
don't
have
to
speak
as
much
as
I
should
be
doing
so.
Anyway,
we
have
a
packed
agenda.
We
have
two
demos
and
a
quick
view
updates
from
the
Saints
as
well.
So
let's
make
it
real
quick.
Do
you
want
to
start
with?
A
B
I'm,
really
pleased
to
announce
that
we
have
fully
staffed
the
110
release
team
and
there
are
still
some
shadow
spots
open.
So
if
you
have
an
interest
in
learning
more
about
the
release
process
or
you're
passionate
about
committee
and
don't
know
where
to
pitch
in
definitely
consider
talking
to
myself
or
one
of
the
other
release
team
members
listed
as
the
person
you
want
to
shadow-
and
you
can
definitely
definitely
connect
with
them
and
learn
a
lot
more
about
the
release
process.
B
By
doing
it
and
watching
it
so
I
highly
encourage
that
there
is
some
new
documentation
out.
We
have
a
roll
guide
for
the
communications
coordinator
position.
This
was
formerly
known
as
marketing,
but
thanks
to
some
great
help
from
CN
CF
marketing
and
also
Swarna,
who
helped
review
these
documentation.
We
have
a
document
up
that
helps.
You
understand
what
the
role
entails.
B
I
also
have
a
PR
listed
in
the
the
notes
for
the
release
lead
documentation,
which
is
essentially
a
way
of
looking
at
all
of
the
things
that
need
to
be
done
during
the
lifecycle
of
a
release
in
a
relatively
high
amount
of
detail.
So,
if
you're
ever
curious
about
what
the
what
the
life
of
a
release
looks
like
that,
PR
has
a
pretty
good
idea
of
what
that
is.
B
C
Here,
yes,
so
the
next
major
milestone
in
our
world
wanted
generally
cycle
is
the
future
freezing
it's
coming
in
two
weeks
in
June
in
less
than
two
weeks.
So
if
you,
if
you
are
targeting
a
new
feature,
if
you
are
planning
a
new
feature,
one
to
ten,
please
made
it
as
usual
to
the
features
ripple.
Today
we
have
only
a
few,
he
suppose
so
I
can
expect
it
it's
much
lower
number
than
usual.
D
C
B
Alright,
thanks
for
that
on
the
features
I,
just
a
quick
touch
base.
It's
about
the
other
releases
that
are
happening.
One
eight
seven
is
gonna
be
due
early
next
week
and
one
ninety
I
believe
is
going
to
come
out
today.
We
can
check
with
Medi
and
find
out,
if
that's
the
case
or
not,
but
I
think
that's
it
for
releases
back
just
wanna.
A
Thank
you
very
much
and
for
anyone
that
is
here
that
has
not
been
part
of
the
release
team,
yet
I
will
vouch
for
the
release
lead
here,
at
least
for
1:10.
Jase
has
been
a
phenomenal
release,
lead
and
working
with
him,
or
even
learning
from
him
was
pretty
enlightening
for
me.
So
if
you
haven't
done
that
yet
I
would
highly
recommend
being
part
of
the
release
team,
either
as
a
shadow
or
as
an
active
participant
and
I'm.
Glad
that
the
audio
is
fixed.
E
E
D
E
The
graph
of
the
week
this
week
is
the
graph
that
shows
all
the
pull
requests
that
currently
have
the
means.
Rebase
label
applied
over
time.
You
can
see
that
we
currently
have
285
Col
requests
that
have
this
label
applied,
which
means
that
we
can't
actually
merge
them
because
they
would
conflict
with
what's
in
head.
We
walk
you
through
how
we
got
here
so
first
off.
If
you
have
any
having
any
questions
about
what
those
various
remote
groups
are
within
the
graph,
let
me
get
back
real
fast.
E
You
can
sort
of
see
how
I'm
showing
that
the
Green
Line
here
is
the
kubernetes
repo
group,
but
the
red
line
is
the
clustered
life
cycle
group
I
scroll
down
here,
I'm,
seeing
this
as
a
percentage
of
all
pull
requests
that
are
open
across
all
of
the
kubernetes
organizations
that
were
tracking
these
repo
groups
are
defined
in
a
sequel
file.
So,
if
I
were
to
look
for
the
word
closed,
their
life
cycle
in
here
keep
searching
there,
we
go
it's
all
of
these
repos
that
might
have
the
needs.
Freebase
label
applied.
E
These
are
things
that
that
we,
as
contributors
or
approvers
or
viewers,
cannot
do
for
you.
The
poll
requests
author.
So
if
I'm
just
gonna
take
a
look
at
this
hey,
why
not?
Second,
try
using
a
manatee,
GCR
name
here
if
I
scroll
down
I
can
see
that
all
of
these
files
have
a
merge
conflict
right
now
and
so
I
wouldn't
be
able
to
merge
this
pull
request,
so
I
would
want
to
go
talk
to
you,
Tim
Hawk
and
can't
read
Asus
pull
requests
so
that
we
could
merge
it.
E
The
code,
that's
responsible
for
applying
this
label
is
a
muncher.
It
is
here
winters
run
as
part
of
this
thing
called
lunch.
I
have
talked
about
here
in
the
past,
which
sweeps
through
github
things
and
lunges
them.
We
are
gradually
in
the
process
of
moving
wanders
out
of
lunch
github
as
we
look
to
deprecate
lunch
github
here
are
all
the
monitors
on
the
left-hand
side,
any
issues
that
track
what
we're
doing
to
replace
them.
On
the
right-hand
side,
you
can
see
that
some
things
like
cherry-picking
we're
trying
to
look
at
re-implementing
entirely,
not
necessarily
launchers.
E
The
reason
we
are
doing
this
is
because
we
have
to
have
a
separate
munch
github
instance
for
each
repository
that
we
lunch,
whereas
we
can
have
a
single
instance
of
prowl
operating
for
all
repos
in
the
organization.
We
recently
extracted
them
under
that
applies.
The
means
read
a
label
into
a
prowl
plugin,
but
I
do
not
believe
we
have
yet
roll
it
out
to
all
repositories
yet,
and
you
can
look
for
an
email
on
kubernetes
that
when
we
do
this.
A
B
Thank
you
for
the
lovely
handoff,
as
always
yeah,
so
sig
a
sure
we
are
doing
some
I
think
some
interesting
things.
This
go-around
110,
one
of
the
things
that
we're
trying
to
do
is
move
all
of
our
upstream
planning
from
the
features
we
want
to
deliver
for
Microsoft
into
the
sig.
So
essentially
the
sig
is
acting
as
the
main
coordinator
of
that
work.
B
I've
even
gone
so
far
as
to
not
leverage
internal
tracking
tools
and
Microsoft,
so
that
it
forces
people
to
keep
updates
and
PRS
and
then
issues
and
keep
that
all
super
visible,
and
we
want
to
also
continue
our
work
in
110
on
the
cloud
provider
breakout
and
we'll
be
working
with
the
great
people
that
we
met
the
people.
We
got
some
great
means
with
face
to
face
during
a
coop
con,
so
it
looks
like
there's
some
great,
create
teamwork
and
happening
around
that
and
collaboration,
and
we'll
hear
more
about
that
moving
forward.
B
We
did
our
110
as
you're
planning
yesterday
and
there's
a
recording
of
that
available
on
YouTube
and
we're
going
to
experiment
with
doing
the
features
process
a
little
differently.
This
time
around.
We're
gonna
maintain
the
standard
features
repo
process,
but
alongside
I'm
also
going
to
be
creating
tracking
issues
in
kubernetes
kubernetes.
B
So
the
the
idea
is
simply
that
we
want
to
try
and
find
ways
better
to
tie
proposals
back
to
issues
to
pull
requests
and
create
a
better
chance
custody
for
for
the
work
that
we're
delivering
110
and,
as
always,
the
cig,
as
your
channel
on
slack,
is
a
great
way
to
reach
any
of
us.
Yes,
that's
pretty
much
it
any
questions.
A
E
There's
a
working
group
for
this
that
is
across
six
I
do
not
believe
that
we
decided
on
using
a
single
state
to
drive
all
provider
work,
but
it's
since
that
WG
cloud
provider
and
I
think
y'all
go
I.
I
can
point
I'll
post
a
link
in
chat,
but
there
are
regular
meetings.
I
have
the
last
sometimes
I
saw
from
those
folks.
A
G
Signal
that
we
still
have
that
that
we
have
the
firm
the
signal
that
we
found
off
the
resource
management
world
rule
and
there
is
going
to
continue
for
at
least
for
half
a
year.
So
there
are
a
lot
of
discussing
around
the
signal
and
we
also
think
we
are
still
in
the
process.
We
talked
about
our
2011
release
and
roadmap
in
the
Tuesday
signal
meeting
and
still
in
the
process.
Feminist,
Act
and
I
can
share
talk
about.
G
G
Yeah,
it
is
a
sure
to
the
people
next
step.
Okay,
anyway,
I
can
I
can
talk
so
so
the
next
thing
you
know
we
know
Dean.
We
can
grass
our
activities
into
the
cavity
for
the
area,
so
no
the
management,
application,
workload,
management
and
security,
your
sauce
management
and
also
the
monetary
logging
de
Baca
p90.
So
there's
the
fire
area
so
each
area
every
quarter.
We
have
some
working
item.
We
prioritize
and
ranking
and
decided
what
it
is.
G
It
should
be
good
for
you,
so
this
year,
we've
somehow
that's
like
yeah,
okay,
so
for
the
this
year
the
one
of
the
big
deal
first,
a
half
year,
one
of
the
big
I
think
it
is
a
lot
of
the
one
time.
Last
year
we
spent
among
the
different
alternative
moving
that,
and
so
they
see
here,
those
red
times
is
I
at
least
the
front
file,
and
also
continuity
boasts.
They
are
go
into
the
production
writing
stage,
so
so
for
them
for
the
application
and
the
working
of
the
management.
G
So
so
so
first,
the
thing
it
is:
we
want
to
be
fine,
because
we
want
you
when
we
start
that's
a
continental
interface.
We
told
the
community
and
also
the
user.
We
want
the
user
feel
free
to
mix
of
the
different
paths
and
the
instant
worry
about
we're.
Gonna
run
time
to
serve
that
purpose
so,
and
the
Khan
Academy
actually
based
on
that
production
of
requirement
and
you
choose
the
proper
run,
including
for
their
use
cases.
G
So
one
of
the
things
to
make
that
when
top
things
will
make
those
wanting
to
run
time
fee
the
productivity.
So
the
top
thing
is
we
need
to
define
criteria
what
it
is,
the
minimum
requirement
for
those
to
have
those
runs
hard.
So
what
kind
of
the
test?
What
is
confirmed
as
a
full
partner
on
time
and
what
is
the
feature
required
and
what
it
is
other
ones
the
feature.
So
we
are
working
on
to
define
those
policy
and
also
requirements,
and
and
also
working
on
the
suggest,
opposed
the
rollout
plan
for
the
community.
G
So
that's
the
that's,
the
one
of
the
item
and
another
two
I
think
it
is
so
we
try
to
part
nice.
It
lies
afters,
krile
and
also
the
CRI
continent
lead
or
the
CI
continuity.
Well,
I've
heard
it
eros
is
we
merge
murder,
desire,
continuity
and
also
company
together?
So
then,
yeah
I
become
too
late
here.
Work
services
provided
by
the
continent
in
future,
so
that's
kind
of
one
performance
booster
for
the
CI
continuity.
So
another
thing
is
we
plant
you
in
the
q1
graduate
to
gradual
to
back
up.
G
So
when
we
talk
about,
we
want
to
use
of
no
worry
about
working
off
the
runtime
they
should
run
so,
but
to
use
they
already
have
the
experience
that
we
used
to
using
a
donker
content
to
do
the
testing
introspection.
So
we
want
to
have
the
give
them
similar
of
the
user
experience
so
so
because
that,
due
to
that
region,
we
also
found
either
prior
to
project.
So
so
we
plant
in
q1
and
the
bracket
I
have
to
type
the
project
to
the
bypass,
and
so
next
one
is.
We
also
talked
about
the
deprecated
regulated.
G
So
reckoning
is
right.
Now
it
is
how
to
hold
it
and
and
defined
it
into
the
Covenant
in
the
two
per
night,
and
if
so,
will
pine
ticket
I
ended
up
back
to
we
mass
here
we
are
shifted
to
regulate
history
project,
that's
a
separate
of
incubator
project,
so
we
in
queue.
When
we
plan
to
deprecate
ocular
this
project
and
in
q2
we
can
clean
our
code
base.
Alright.
So
then
there's
those
couple
icon
honestly,
I
enhancement
especially
in
an
interface.
G
Then
we
started
some
portioning
things
and
how
we're
going
to
in
the
future
to
do
the
backward
command
even
once
this
is
equal
to
the
production
so
and
on
no
the
management.
We
talk
about
open
source
remedy
system
and
another
thing
thing
and
also
enhance
of
existing
know
the
poverty
factor
and
there's
a
bunch
of
the
work
to
make
that
sir
experience
is
better
and
we
also
talk
about
major
and
include
noticed
and
for
latency.
G
First
thing:
we
need
to
make
sure
those,
often
that
is
visible
to
the
user
and
last
one
in
is
actually
online
ways
the
sig
I
drove
after
okay
are
we
want
to
have
the
node
each
we
tested
for
Windows
container
images
so
right
now
we
windows
support
if
sig
windows
claim
that
the
back
have
support,
but
our
Federation
testing
for
editors
no
test
at
all,
so
the
user
keep
asking
signaled
and
the
cygwin
dosis
or
what
it
is
mean
had
five
families.
So
we
think
about.
We
come
over
to
the
Sega
windows
Cascada.
G
We
want
at
least
the
standard,
honest
note,
ii/3
testers,
so
you
could.
We
should
profit
on
multi-part.
So
then
talk
to
the
security
security.
It
is
the
top
one
it
is.
We
are
going
to
knock
down
the
company's
equity
strategy.
What
we
should
you
align
align
with
the
tiger
container
effort,
and
so
we
want
to
define
working
off
the
policy
working
off
the
type
of
behavior
you
want
went
to
the
to
the
to
the
each
part
and
what
it
is
it
is
that
asset.
G
If
you
can
turn
there,
it
is
at
the
pot
level,
no
denial,
or
it
is
even
at
your
company
narrow
we
need
lockdown
are
our
decision
on
those
kind
of
things
and
another
one.
It
ends
the
calm.
We
should
put
promoters
that
come
to
the
backup
and
to
propose
the
rollout
plan
for
the
for
the
for
the
open
source.
So
now
we
have
the
bunch
of
the
resource
management
enhancement.
So
one
thing
it
is
the
triple
negative
action
enhancement
and
the
boss
offered
is
for
memory.
G
There
are
some
of
what
to
do
and
also
there's
the
couple
things
is.
Actually.
It
is
just
make
a
lot
of
feature
on
the
resource
management
in
town
last
year
to
the
beta
and
when
it
is
the
device
packing
and
there's
the
we
also
when
she's
had
it
is
how
we
are
going
to
do
the
registration
of
the
new
device
back
in
so
we
working
together
is
the
storage
came
on
the
60s
I
because
also
need
to
register
through
the
punic,
and
we
want
to
consistent
of
the
way
consistent.
G
We
can
I
think
you
register
those
things
to
the
turbinate
and
another
Renata
to
it
is
the
huge
cage
supported
to
the
beta
and
also
CPP
Rebecca,
so
the
last
one
we
want
to
started
to
talk
about
more
comprehensive
of
the
Army's
I
need
to
support
how
we
are
going
to
cut
you
off
the
number
of
the
size
number
of
the
open
fire
and
so
on.
It
is
a
Donald
purse
back,
it
is
the
per
pod
and
how
I've
been
living
in
those
kind
of
things.
G
We
started
with
design
those
things
and
in
front
of
monitoring
the
next
category.
Elastic
area
is
monitoring
not
being
a
native
a
community.
So
we've
been
talk
about
the
back
part
and
right
now
we
settle
down
all
the
first
emotion
of
the
API
to
the
98
year
and
then
I
am
the
really
expecting
we
have
the
first
of
it
means
I
offer
ironies
in
q1
for
the
debug
call
and
there's
the
other.
It's
several
of
the
enhancement
of
the
lobby.
G
C
Community
and
folk,
even
a
full
cloud
Native
community
bit
tomorrow,
is
the
final
day
when
you
can
submit
your
presentation.
Proposals
for
Cuba
include
native
corn.
That
will
happen
in
Copenhagen
at
the
beginning
of
May,
so
the
code
for
papers
form
will
close
at
the
midnight
on
January
12
and
also,
if
you
are
going
to
attend
this
event
as
a
as
a
generality,
Indian
artist,
speaker
or
the
pricing
deadline
is
also
finishes
tomorrow.
C
Last
we
could
jest
ask
the
question:
what
what
should
you
do
if
you
have
proposed
to
talk
and
have
here
and
you
purchase
the
ticket
risk
early
price?
If
you
talk,
will
be
selected,
you'll
receive
the
free
ticket
and
for
a
phone
feel
free
to
reach.
Us
back
feel
free
to
reach
the
ncf
back,
and
it
will
handle
this
question.
C
So
the
next
announcement
is
that
IBM
has
organized
an
index
conference
in
San
Francisco
in
around
the
months
in
late
February,
and
they
are
offering
the
free
space
for
face-to-face
meetings
the
space
way
and
it's
it
will
happen
at
the
Moscone
Center.
So
if
you're
seek
or
your
group
of
Cuban
artists
contributors
is
interested
in
having
a
space
for
face-to-face
meeting
in
San,
Franciscan
February
2004
to
reach
me
and
I'll
provide
Huey's
their
details
and
I've
see
that
Aaron
is
next.
Okay,.
E
Hi
everybody.
It's
me
again
bonus
graph
of
the
week
edition,
so
the
graph
we're
looking
at
is
pull
request,
merge
seven
day
moving
average
over
the
last
90
days,
I
have
inserted
by
repositories
that
have
merged
the
most
pull
requests
in
the
last
90
days,
and
I
only
have
a
couple:
repos
highlighted
basically
I'm
highlighting
the
repos
to
whom
I
have
turned
on
hide
and
merged
automation.
E
E
Here,
when
I
was
rolling
out
tide
before
the
community,
repo,
the
you
I
still
isn't
great,
but
it's
gotten
a
little
bit
better.
It's
basically
this
thing
that
makes
sure
that
all
pull
requests
that
have
these
labels
applied
and
do
not
have
these
labels
applied,
have
little
green
check
marks
next
to
their
commits,
and
then
it
will
merge
them
if
I
click,
this
github
search
link
here
I
will
see
that
right
now.
E
It's
looking
at
this
bullet
list
is
a
potential
pull
request
to
merge
if
I
scroll
down
to
try
and
find
out
why
it
hasn't
merged.
Oh
look:
it's
another
pull
request
for
Tim,
Hawk
and
I,
just
love!
Picking
on
you.
You
can
see
that
one
of
the
tests
has
failed
and
so
we're
not
merging
it.
Yet
I
have
sort
of
organically
been
going
through
and
picking
the
highest
traffic
repos
and
finally,
an
issue
with
the
mask.
E
If
they
would
like
me
to
turn
on
merge
automation,
here's
the
umbrella
issue
that
I
have
created
to
track
this
effort
across
all
repos.
Here's
an
example
of
me
doing
this
for
kubernetes
dashboard,
where
I
say
here's.
What
I'd
like
to
do
is
this
cool
I
get
some
feedback
from
folks
I
open
up
a
pull
request
to
make
sure
that
there's
an
odor
file?
If
there
isn't
one
or
if
there
is
one
I
taste
it?
E
E
Who
has
the
power
to
mainly
approve
all
requests
into
this
little
weird
iterating
on
the
discussion
here
they
said
in
their
case,
helm
does
not
want
any
of
the
siga
ops
leads
to
have
nerds
rights
in
this
repo,
so
I
have
removed
how
but
we're
still
kind
of
at
this
point,
where
I'm
trying
to
understand
as
a
community.
Would
you
folks
like
consistent,
merge,
automation
and
process
across
all
of
the
repos
that
are
in
the
kubernetes
organization,
or
am
I
going
about
this
all
wrong
here?
E
E
I'm
happy
to
continue
this
discussion
in
kubernetes,
dev
or
in
say,
contributor
experience
since
to
me
this
seems
like
a
consistent
contributor
experience,
question
or
sig
testing,
since
tied
is
kind
of
the
thing
that
sig
testing
has
interests
in
rolling
out,
so
that
we
don't
have
to
maintain
munge
github
or
whichever
SIG's
you
feel
appropriate.
You
know:
I,
went
and
introduced
this
automation
to
sig
Doc's,
who
informed
me
that
they
have
a
slightly
different
pull
request
workflow,
whereby
they
use
LG
TMS
from
both
Docs
and
Tegrity
people.
They
decided
they
could
conform.
E
The
workflow
to
saying,
like
the
improve
label,
the
more
substantive
yes,
this
is
good
to
merge
label
should
come
from
Docs
people
and
we're
making
sure
that
it
conforms
to
the
standards
and
norms
of
our
documentation
and
the
LG
TM
label
could
come
from
somebody
who
can
bet
the
technical
content
of
this.
So
I
just
thought.
I'd
start
the
discussion
here
and
take
it
to
wherever
we
feel
is
most
appropriate
comments.
E
H
I
That
is
a
putting
off
experience,
but
if
they
were
engaged
in
the
conversation,
it's
better
to
have
self-service
or
not.
How
do
we?
You
know
there's
a
thing
about
consistency,
but
I've
been
living
with
inconsistency
for
a
long
time,
and
we
have
to
be
careful
on
how
we
move
people's
cheese
in
a
way
that
doesn't
turn
them
off
or
alienate.
E
Agree
with
that,
so
for
what
it's
worth
everything
I'm
doing
is
100%
pull
request
proofing
right,
so
I'm
not
forcing
anything
on
anybody,
and
these
are
Cole
requests
that
other
people
are
free
to
make
for
themselves.
It's
true.
My
goal
has
been
to
make
sure
that
lunch
github
doesn't
exist
because
it's
an
operational
nightmare
and
so
I've
actually
replaced
lunch
github
for
every
repo
except
kubernetes.
E
My
attempt
at
rolling
tide
out
to
other
repos
was
to
build
consensus
that
this
is
a
way
enjoy
merging
our
code
that
enables
new
contributors
to
do
it
more
effectively.
We're
turning
it
on
I
feel
like
a
lizard,
I've,
never
hosted
time
looks
by
and
what
it
is.
It's
tend
to
be
more
documentation
for
that,
but
I'm
happy
to
stand
off
and
say
that
they're,
a
select
group
of
12
repos
that
can
merge
with
this
/lg
love
that
not
every
repo
happens
them
generally.
E
D
E
Doesn't
you
know
know
that
that
repo
exists
again
like
I,
could
totally
just
step
back
and
say
if
you
would
like
merge
automation
on
your
repo
here's
some
examples
of
how
I
found
this
in
the
past,
and
we
can
continue
to
hi
I'm
really
but
I.
Just
I'm
not
sure
how
what
I've
been
doing
has
been
like
changing
people's
labels
out
from
underneath
of
them
without.
E
E
So
Matt,
where
do
these
discussions
around
the
automation
and
in
contribution
happen,
happen
I?
Think,
you're.
Listening
to
one
of
those
conversations
right
now
and
I
suggested
to
other
forums,
contributor
experience
seems
to
be
the
natural
place
to
talk
about
how
we
improve
and
work
on
the
experience
of
contributors
to
this
project
and
safe
testing
kind
of
implements
some
of
the
automation
it
doesn't.
E
You
know
both
of
these
groups
seem
to
have
some
control
and
opinion
expressed
through
code
and
words
on
how
things
should
be
automated,
but
I
think
those
are
all
fantastic
forums
to
have
this
discussion.
I'm,
not
sure
if
I've
been
in
excluding
people
in
having
conversations
in
those
those
places
and.
I
I,
don't
know
that
anybody
has
been
trying
to
exclude
people
I,
think
it's
where
we've
naturally
gravitated
towards
these
things
happening
and
now
that
we're
starting
to
have
more
tools
and
we're
starting
to
spread
them
in
wider
use.
The
natural
conversation
is,
you
know,
is
it
self
service
opt-in?
Is
it
enforcement?
What
does
that
mean?
How
does
it
serve
people?
I
I
To
that
conversation
and
that's
one
example:
another
is
using
a
Munder
rather
than
the
merge
pull
request.
Button
are
actually
using
tide,
in
this
case,
not
the
muncher,
but
to
do
that,
that's
that's
a
process,
flow
change
and
how
you
do
it
to
use
automation
and
they
don't
know
the
benefits.
First,
not
right
so,
like
today,
I
was
sharing
with
the
helm.
Folks
that
you
know
you
can
have
tests
on
the
CIA
infrastructure
that
run
again
right
before
you
merge.
I
I
And
so
now
it
feels
like
it's
being
pushed
onto
people
out
of
their
control,
and
maybe
they
were
comfortable
with
other
things,
because
a
bunch
of
the
things
we
do
are
kubernetes,
isms
and
and
then,
if
you
don't
know
them,
they're,
not
like
a
broader
open
source
thing
that
you're
gonna
get
on
github.
And
so
where
do
you
learn
those
kubernetes
isms?
And
what
do
they
mean?
A
lot
of
people?
Don't
know
how
owners
files
works.
I've
had
to
explain
that
to
a
whole
bunch
of
people.
I
J
That's
a
fine
answer
and
I'm
concerned
that
you
even
perceive
an
in
crowd
out
crowd,
because
that
is
not
the
way
we
work.
We
work
in
distributed
distributed
groups
trying
to
make
efforts
better
for
the
whole
project
and
trying
to
push
the
decisions
closer
to
the
people
who
are
most
affected
and
Jorge
had
in
here
that
he
was
pretty
sure
that
the
label
change
has
been
committed
or
communicated
in
multiple
times
in
this
meeting.
J
I
guess
my
question
will
be:
where
would
you
expect
these
changes
to
be
communicated
more
broadly
because
we've
had
this
happen
in
the
past,
where
we've
had
emails
go
out
to
kubernetes
dev,
which
is
a
fairly
high
volume
list.
We've
had
emails,
go
out
to
specific
special
interest
groups,
we've
had
notices,
go
out
on
chat,
multiple
times
conversations
happen
in
this
community
meeting.
What
is
the
piece
that
you
wear
to
these
people
that
you're
talking
to
wish
to
see
these
notifications?
J
I
Know
I
don't
know
that
I
have
a
good
answer
to
it,
but
I've
gone
and
you
know
you
go
look
in
the
rooms
in
slack
right.
Some
of
the
things
like
kubernetes
user
have
28,000
people
in
it.
If
you
look
at
the
number
of
people
who
submit
PRS
and
issues,
there's
there's
so
many
people
who
are
involved
in
touch
cooper,
Nettie's
and
in
here
we've
got
86.
People
who
joined,
which
is
a
few
conference
rooms,
were
probably
just
over
a
hundred
of
the
thousands
of
people
who
are
involved.
I
J
I
J
I
I
E
E
E
Like
I,
said,
I'm
happy
to
continue
this
discussion
in
whatever
the
appropriate
forum
is
I,
don't
I,
don't
know
if
a
blog
post
is
a
good
to
weight
conversation,
but
you
know
you,
can
you
can
figure
something
out?
I
guess
I'm
still
feeling
stuck
is
like
how
should
I
proceed?
Should
I
should
I
just
stop.
Stop
the
efforts
I'm
doing
to
attempt
to
apply
consistency
or
repose
should
I
plan
on
having
a
more
substantive
discussion
about
this
with
the
proposal
document
or
something
in
a
you
know,
sake
Tim,
I
think.
K
There
needs
to
be
a
clarification,
and
it's
part
of
what
the
steering
committee
has
been
discussing
with
the
layers
of
what
is
in
core
and
not
of
course,
I
do
totally
agree
that
consistency
is
highly
beneficial
for
things
that
are
in
court,
anything
that
resides
out
of
cores
and
the
layers
outside
of
it.
Then
it's
you
know
it's
whatever
they
choose
to
do,
and
that
would
provide
the
mechanisms
for
opt-in
or,
if
I
those
details.
E
Just
I
want
to
pick
on
the
label
example
real
quickly.
The
labels
that
have
been
renamed
to
my
knowledge,
via
our
automation,
were
to
change.
The
word
help
wanted
to
help
space
wanted,
which
are
both
labels.
That
github
recommends
you
use
for
issues
that
require
new
contribution.
We
added
a
slash
help
and
the
bots
will
respond.
Are
you
sure
you
want
to
add
the
help
label
Help
Wanted
label
to
this?
I
Eric,
let
me
ask
you
a
quick
question
about
this.
If
somebody
is
working
on
a
sub
project
and
one
of
the
repos
and
the
experience
that
they've
crafted
around
their
particular
sub
project
is
really
happy
for
them,
it
may
be
a
slightly
different
bubble
from
the
other
ones,
because
each
one
is
a
little
bit
different.
Why
would
they
want
to
come
to
contributor
experience
to
contribute
to
an
experience
when
their
own
little
part
they're
happy
with
the
experience
on,
because.
L
We're
one
community
and
just
should
aim
for
consistency,
so
that
when
somebody
has
experience
contributing
to
one
part
of
kubernetes,
they
can
actually
translate
that
to
other
parts
like
we
don't
want
to
have
helm
and
then
the
rest
of
kubernetes.
You
know
having
different
contributor
guides
as
much
as
possible.
L
We
want
to
actually
have
one
one
community
across
this
stuff
and
I
think
this
is
part
of
the
calculus
that
we
need
to
make
in
terms
of
whether
they're
part
of
kubernetes
project
as
a
whole
or
whether
they
actually
are
probably
better
off
being
on
their
own
outside
of
cooper's
I.
Think
consistency
has
to
be
part
of
of
what
we
get
out
of
things
being
part
of
the
Trouper
Nettie's
project,
I
I,.
L
Really
think
it's
about
consistency
and
again,
if
those
projects
think
that
this
that
what
they
gain,
how
that
the
processes
for
the
consistency
across
the
kubernetes
project
outweigh
the
gains
that
they
get
about
being
part
of
kubernetes,
and
I
think
it
makes
sense
for
them.
You
know
they
can
be
perfectly
successful
outside
of
the
kubernetes
project
as
a
whole,
being
an
Associated
project
I,
you
know
there
are
rights
and
something
being
partisan,
sure.
I
So
if
I
understand
it
right,
though,
because
kubernetes
we
regularly,
you
talk
about
it
being
more
complicated
to
contribute
to.
This
is
something
people
regularly
say,
and
so,
if
the
sub
projects
have
found
a
way
to
be
simpler,
you're
saying
that
they
should
become
more
complicated
to
bring
about
consistency,
because
that
will
be
better
for
overall
contribution.
I,
don't.
I
J
He
is
saying
is
the
consistency
across
the
community
is
something
that
works
for
many
sub
projects,
not
all,
and
those
may
not
necessarily
need
to
remain
as
direct
sub
projects
of
kubernetes
if
they
don't
want
the
consistency
but
those
smaller,
less
complicated
sub
projects.
They
use
the
same
tooling
as
google
kubernetes
company
and
from
gateways
to
broader
and
deeper
contributions
across
the
community
across
multiple
sub
projects
and
across
and
into
kubernetes
kubernetes
I.
E
Will
say
for
what
it's
worth
cops
volunteered
themselves
to
use
time
to
merge,
stuff
and
mini
cube,
had
no
objections
when
I
suggested
that
they
use
tide
and
both
of
them,
because
for
automation
today,
all
the
repos
in
our
organization
are
consumers
of
some
of
our
automation.
You
can
slash
a
sign
or
slash
close
or
slash,
reopen
absolutely
any
issue
or
pull
requests
on
any
repo
in
kubernetes
or
kubernetes
incubator.
If
you
are
assignable
as
a
human
being,
which
means
you
are
either
a
collaborator
of
that
repository
or
the
organization.
E
Personally,
as
an
individual
contributor
who
only
like
two
weeks
ago
got
access
to
see
who
is
a
collaborator
of
a
repository,
it
was
kind
of
difficult
for
me
to
know
who
to
notify.
So
it
was
really
great
to
see
that
there's
this
pattern
of
teams
where
it's
like
helm,
admins
or
helm,
maintainer,
sir,
could
deploy
admins
cube,
deploy
maintainers,
so
I
could
at
least
like
ping
them
I'm,
just
speaking
as
somebody
who's
open
to
pull
requests
against
every
single
repository
and
had
varying
degrees
of
success
getting
them
merged.
D
Is
it
possible
to
have
sort
of
like
I
understand
the
concern
about
like
make
everyone
do
the
most
complicated
thing,
but
why
don't
we
make
everyone?
Do
the
most
simple
thing
and
then
have
every
other
project
or
related
sig
have
additional
guidelines
that
build
on
the
common
workflow
I
mean
I,
don't
know
how
different
of
the
workflow
people
want,
and
so
maybe
that's
a
slightly
different
conversation,
but
I
think
it's
a
little
bit
from
extreme
to
say.
D
But
if
we
make
everyone
do
the
most
complicated
thing,
then
it's
just
going
to
be
complicated
for
everyone,
because
I
think
for
most
projects
or
sakes
it
would
be
possible
to
just
go
with
the
basics
and
then
say
and
in
addition,
we
have
these
requirements
that
you
must
also
adhere
to.
But
that
allows
people
who
are
new,
like
myself
to
start
out
somewhere
that
is
simple
and
accessible
and
obvious
and
then
build
on
that.
E
That
that
sounds
good
to
me
and
I.
Just
I
was
kind
of
confused
that
the
whole
/lg
team,
/
approved
thing
is
really
for
in
confusing
people,
because
it's
it's
been
an
active
use
for
a
while,
so
I'm
feeling,
like
maybe
I,
should
stop
my
efforts
to
roll
that
out
until
the
steering
committee
is
put
forth.
The
proposal
that
mandates
exactly
what
conventions
all
projects
should
follow.
It
kind
of
bugs
me
a
little
bit
because
I
thought
I
was
making
things
better,
but
I'm
happy
to
wait.
E
I
Know
I
I,
don't
know
that
stopping
completely
is
a
good
thing.
Alright,
if
somebody
came
with
an
external
service
like
Travis
or
GoBot,
which
is
the
Box
people
are
using
on
github
a
lot
these
days,
they
put
a
certain
amount
of
documentation
and
presentation
together
and
sales.
So
people
understand
here's.
This
awesome
benefit
you
get
from
doing
that
right,
but
with
a
lot
of
our
stuff
we
don't
have
good
documentation.
We
don't
have
a
lot
of
clarity
and
how
you
use
it,
and
so
it
can.
You
know
like
when
you
do.
I
You
know
when
something
is
approved,
it
can
be
merged.
Where
does
it
say
that
it's
gonna
go
rerun
the
tests,
how
many
people
actually
catch
on
that
that
happens
to
know
that
it
reruns
the
test
right
beforehand
is
a
benefit?
How
do
we
you
know?
How
do
we
sell
this
so
that
people
know
what's
useful?
What
before
we
put
something
on
people?
What
kind
of
what
kind
of
presentation
do
we
have
to
have
for
them
to
be
users
and
consumers
and
and
get
support
and
understand
how
this
works
fully?
I
E
Let's
go
in
the
right
thing,
so
I
can
do
with
Sega
apps
related
repo
as
much
what
I've
done
with
website,
which
was
start
the
discussion
and
then
realized
that
things
are
a
little
more
complicated
there.
Leave
that
issue
open
for
us
to
hash
out
the
details,
maybe
show
up
to
say,
gaps
and
make
a
presentation
to
y'all
to
make
sure
we've
got
the
best
path
forward.
Again.
None
of
what
I've
proposed
thus
far
changes
how
you
do
things
it
just
lets.
You
use
additional
slash,
communes
and
I
can
try
out
other
repos.
E
Or
I
can
go,
be
a
marketing
guy
I
mean
that's,
that's
totally
cool
too
I
like
for
what
it's
worth
like
I.
Don't
it's
a
bunch
of
other
people
who
do
really
fantastic
work?
I,
just
think
it's
really
super
cool
and
useful.
So
I
just
sit
here
and
try
and
brag
about
it.
They
seem
to
prefer
to
be
heads
down
about
it.
E
A
No
but
I
think
this
is
about
a
discussion
for
sure.
The
next
announcement
is
just
me:
relaying
George's
announcement,
I
think
the
kubernetes
Alcazar's
is
this
Wednesday?
If
you
have
any
questions,
please
do
reach
out
to
him.
But,
more
importantly,
if
you
want
to
volunteer,
please
do
pink
George,
Vaughn
slack
but
I
do
appreciate
the
discussion
that
happened
from
the
various
groups
and
also
the
various
individuals.
It
helps
kind
of
reinforce
the
concept
of
community.
So
that's
definitely
helpful.
It's
not
just
one
item
dominating
the
entire
conversation.
This
is
an
important
discussion
to
have.