►
From YouTube: Steering Committee Meeting 20180103
Description
Recording of the Steering Committee meeting. See also:
https://github.com/kubernetes/steering
Chat transcript:
00:01:26 Brandon Philips: Notes, FWIW http://bit.ly/k8s-steering-wd
00:03:23 bburns: are other people hearing things
00:03:25 bburns: ?
00:03:46 SIG Beard: I gather you’re not hearing me say stuff
00:03:49 Brandon Philips: We can all hear each other bburns
00:04:48 Joe Beda: I can't hear anything. Is taht on my end?
00:04:54 bburns: YES
00:04:57 Joe Beda: gah
00:04:59 SIG Beard: check your zoom settings?
00:05:02 bburns: oops, didn't mean caps
00:05:07 bburns: sry
00:05:18 bburns: yes
00:05:21 bburns: can hear you
00:06:52 SIG Beard: gfi
00:09:08 Brandon Philips: Notes: http://bit.ly/k8s-steering-wd
00:16:39 bburns: I just checked, kubernetes-dashboard doesn't actually exist
00:18:38 Clayton: BE SAFE
00:18:43 Clayton: NO RUNNING WITH SCISSORS
00:22:36 Timothy St. Clair: I’d love to spend some time walking through Brendans doc.
00:24:28 SIG Beard: https://docs.google.com/document/d/1yauW9zMtWgXN8xh4q6144B2xBTPuIOLGc0L6aQPMj1I/edit - doc in question
00:34:28 Joe Beda: s/associated repos/collab repos/?
00:38:40 Timothy St. Clair: Part of adding the CLA bot would add the repo to that list.
00:44:20 Timothy St. Clair: It’s also the “descheduler” fwiw, I lost the vote.
00:47:47 Joe Beda: What should happen to graduated projects already? Not all of those would meet any new bar from sig-arch...
00:56:40 Joe Beda: https://docs.google.com/forms/d/1h41mwLlcQwQDP6c4VDZzjiJepaQ1irHjP4mXYQFQAIo/edit
A
B
B
A
A
A
D
G
D
C
Issuing
your
parrot
in
was
I,
guess
I
could
add
a
commit
to
your
branch
to
add
what
the
I
guess.
The
amendment
to
the
Charter
process
was
going
to
be,
which
is
why
I
think
I
counted
on
that
doc
to
say.
Maybe
what
language
you
had
their
previously
should
go
in
the
initial
charter
daf,
but
otherwise
I
was
trying
to
look
at
what
other
communities
do
for
amendments.
I
think
we
talked
about
lazy
consensus,
I.
Think.
C
C
G
C
About
adding
that
more
weeks
was
sufficient
holiday
coverage
and
then
honestly,
if
we
don't
have
one
steering
committee
meeting
between
the
time
and
amendments
posted
and
anyone
saying,
maybe
we
should
have
a
vote
like
then.
Obviously
we
don't
have
lazy
consensus,
though
I
honestly,
don't
think
we'll
be
amending
this
Charter
too
frequently
so
yeah.
A
A
A
D
D
Good
intent
with
the
community
I've
been
doing
a
lot
of
chopping
wood
carrying
water
stuff,
where
I
like
disabled
the
wiki
in
the
community,
repo
and
I,
revoke
direct
direct
write
access
to
the
community
repo
I'm
doing
the
same
thing
for
the
wiki
and
kubernetes
repo,
and
starting
to
ruffle
feathers
by
suggesting
that
we've
denied
access
to
the
kubernetes,
secure
nineties
repo.
This.
D
There's
an
open
issue
in
community
which
I'll
at
least
found
that
before
you
call
on
me
to
document
the
list
of
or
semi-dome
so
I
sort
of
found
that,
after
the
fact
but
I've
been
using
that
and
then
this
issue
here,
which
I
just
closed
the
meeting
notes.
That's
great
I've
been
using
my
effort
to
had
the
same
code
of
conduct
file
everywhere
as
sort
of
a
tracer
:
to
see
like.
D
Is
it
every
single
repository
in
the
kubernetes
project
covered
by
this,
and
it
made
me
think
that
for
administrative
purposes
it
would
be
a
lot
easier
or
it
might
be
a
good
signifier.
If
the
steering
committee,
as
a
team
was
publicly
listed
as
owners
of
each
of
these
organizations
within
github
that
are
part
of
the
kubernetes
project.
So
it's
really
clear
who
you
can
contact
and
that
it
is
in
fact
part
of
the
kubernetes
project.
D
This
would
also
kind
of
having
admin
access
to
all
the
repos
for
the
reasons
I
suggested
in
that
mailing
list
thread
to
steering.
It
also
helped
me
a
lot
with
pushing
some
of
this
administrative
Australia
stuff
through
assuming
I've
built
good
faith
within
the
community
for
all
this
stuff
I
plan
on
acting
the
same
way
to
push
through
your
ownership
of
all
of
the
repos
by
all
of
the
six.
H
So
I
think
creating
some
lists
somewhere
of
here's
all
the
organizations
that
are
part
of
the
kubernetes
project,
making
sure
them
some
consistency
that,
like
hey
for
governance
of
this
organization,
all
repos
here
in
go,
look
at
XYZ
I.
Think
that
sounds
super
reasonable
anybody's
going
to
have
you
know
ownership
across
the
project.
Having
a
you
know,
a
set
of
folks
from
the
steering
committee.
You
know
the
owners
on
every
single
organization
seems
totally
reasonable.
F
D
F
F
D
Sounds
like
something
we
maybe
ought
to
scope
out
as
a
chunk
of
work.
We
could
subcontract
out
to
somebody
from
the
CN
CF,
using
this
mythical
pool
of
money
that
they
have.
It's
something
I
think
could
be
doable
via
prowl
within
testing
from
just
like
point
it
at
an
organization
and
then
have
it.
Matically
managed
teams
within
that
organizations
based
on
some
central
repo
that
defines
all
the
teams
and
usernames
blah
blah
blah
exactly.
F
F
D
A
D
One
like
define
the
list
of
organizations
that
are
part
of
the
kubernetes
project,
there's
an
issue
that
I've
linked
from
the
meeting
notes.
That's
in
the
community
repo
I've
tried
to
link
the
ones,
I
know
about
the
ones
that
don't
not
sure
about,
and
then
all
the
ones
that
you
may
have
preemptively
registered
and
I'm.
Assuming
this
is
something
I
pretty
much
just
can't
answer
until
right
back
from
vacation,
but
sure
what
that
list
is
I.
Don't
keep
code
of
conduct
PR
to
every
single
repo.
D
F
Does
it
make
sense
so,
first
of
all,
the
list
of
orgs
I
think
right
now
is
very
small.
There's
a
bunch
that
we've
preemptively
squatted
on,
but
unless
we're
using
them
for
something,
let's
not
consider
them
to
be
part
of
the
project
like
I,
make
full
commitment
that
the
20,
or
so
that
I'm
squatting
on
are
free
for
for
us
to
use.
F
G
F
To
know
it's
mine,
green
graveyard
should
go
away.
The
only
thing
that's
stopping
it
from
going
away,
is
Brian
being
back
to
give
me
permission
to
finish,
moving
all
that
stuff
over
other
than
that.
Okay,
some
of
those
feel
weird
to
me,
but
fine
I.
What
I
can
do
in
the
very
short
term
is
at
least
for
the
couple
of
those
that
I
have
or
that
men
access
to
I
can
create,
or
any
one
of
us,
I
guess
who
has
or
get
men
can
create.
A
group
it
has.
F
D
F
D
D
Otherwise,
we're
just
and
I'm
I'm,
not
just
I,
don't
want
us
to
necessarily
get
blocked
behind
that
tooling
I'm,
just
trying
to
document
the
State
of
the
State
and
get
enough
minimal
stuff
together.
So
I'm
happy
to
like
help
scope
out
the
proposal
for
the
tooling
and
how
to
do
it
going
forward
I'm
just
trying
to
figure
out
what
are
the
guidelines
to
follow
right
now.
So.
F
I
think
I
mean
I.
Think
the
tooling
is
the
right
Avenue
in
the
end,
in
the
short
term,
I'm
looking
at
it
right
now,
you're
right
there
isn't
a
way
to
make
a
group
have
org
owner
access,
so
I'm,
just
gonna
click
on
everybody
in
the
order
who
doesn't
already
have
owner
access
and
make
them
all
org
owners.
Anybody
disagree.
D
H
But
I
mean
but
backing
up
a
little
bit
so
like
when
I
started.
Looking
this
one
of
the
things
that
got
me,
a
little
wound
up
was
like
I,
don't
know
Crowell
that
well,
it
was
clear
that
it
might
should
be
perhaps
part
of
prowl
and
I'm,
like
oh,
my
god
like
getting
like
crowd
to
the
point
where
I
can
develop
on
it
and
start
adding
it
and
like
playing
with
it,
was
kind
of
intimidating.
H
F
We're
all
or
go
nurse
for
kubernetes
I
think
we
should
talk
to
Dan
and
see
if
we
can
use
his
infinite
money
to
hire
contract.
To
do
this
work,
I,
don't
think
it's
particularly
complicated
work
I
think
we
can
spec
it
reasonably
well.
As
this
group,
we
can
produce
a
specification
that
they
can
then
code
to
yeah.
D
H
D
F
D
F
H
Yeah,
alright,
that
seems
like
a
separable
issue.
Well,
that's
one
of
the
other
things
is
that,
as
we
start
looking
at
the
synchronisation
of
like
we
want
to
have,
we
want
to
have
yeah
mo
files
that
drive
the
team
definitions.
Do
we
want
to
have
those
then
actually
remove
people
who
are
not
part
of
those
yeah
mo
files?
So
have
those
be
the
only
things
that
actually
get
driven
by
those?
Yes,
yeah,
no
we're
terrifying,
but
I
think
the
answer
is
yeah
and
I
really
have
to
rationalize
on
it.
Cuz
like
we
have
295
teams.
F
D
Yes,
yeah
I
was
gonna,
say
I.
I
also
think
that
Brendan's
doc,
probably
is
the
more
substantive
thing.
I
was
really
just
asking
for
give
me
permissions
to
go.
Do
stuff
I'll
write
up
some
proposals
of
next
steps
that
I
will
take
with
those
and
make
sure
that
I'm
doing
the
right
things
right
time
and
place
all.
F
H
E
G
The
goals
are
basically
to
so
I
think
that
one
of
the
reasons
we
ran
into
trouble
was
that
we
were
kind
of
trying
to
define
an
all-or-nothing
process
and
I
I
tried
to
kind
of
identify
three
different
sort
of
sub
groupings
of
process
that
we
could
come
up
with
in
a
sort
of
concentric
circles,
of
more
more
seriousness
or
more
alignment.
And
hopefully
then
we
can
find
the
right
for
each
project.
We
can
find
the
right
blend
that
fits
for
them.
So
I
don't
know
any
substantive
discussions,
though
III
don't
have
the
doc.
F
G
The
other
concern
that
was
raised
is
effectively
like
if
and
when
this
thing
becomes
something
we
want
to
bring
in
it's
much
much
harder
to
do
it
if
it's
not
if
it
hasn't
been
created
under
the
CMC
la
to
start
out
with
so
it's
kind
of
like
a
safe,
harbor,
I
guess
is
the
best
way
to
come
up
with
it.
It
really
is
to
sort
of
just
get.
G
H
G
D
I
was
gonna,
say
from
an
administrative
perspective,
it's
a
nightmare
to
support
arbitrary
organizations
right
now.
The
current
bot.
You
have
to
specifically
ask
somebody
as
a
human
to
add
each
individual
org
for
our
own
automation
that
we
have
control
over.
It's
still
going
to
be
an
entry
for
each
organization,
slash
repo
and
then
you'll
have
to
make
sure
that
the
bot
has
the
appropriate
admin
access
to
you
set
repo
to
do
the
things
it
needs
to
do
you.
G
Don't
necessarily
have
to
do
it
that
way,
I
mean
I
mean
obviously,
I
talked
about
discovering
changes,
but
if
the
bot
instead
looks
at
repos
that
it
is
an
owner
of
and
then
goes
and
or
it
is,
you
know,
it's
a
collaborating
member
of
and
uses
that
as
the
authoritative
list,
then
that
ensures
both
permissions
and
lists
in
the
one
fell
swoop,
and
then
people
actually
effectively
only
have
to
add
the
bots
themselves
without
any
human
input.
So
this
issue.
D
G
We
can
table
the
full
discussion,
I
think
right
now,
because
I
feel
like
this
is
in
the
technical
details
of
the
weeds.
Is
the
proposal
I
mean
so
like
like
if
we
can
highly
confident
that
we
can
make
it
work
for
associated
repos?
If
we
believe
that
it
is
the
right
structure,
and
so,
let's
not
fight
about
the
technical
bits,
let's
see
if
the
structure
is
right.
So
sorry.
D
G
Repo,
well,
is
it
as
if
we
decide
we
want
it
to
be
I.
Think
that's.
The
whole
point
is
it's
up
to
the
project
to
decide
if
it
wants
to
be
associated
its
opting
it's
an
OP
associated
is
an
opt-in
step.
Basically
right,
cuz
I
think
one
of
the
things
one
of
the
one
of
the
rocks
that
we
fell
on
Gates
or
one
of
the
big
reasons
we've
stalled
for
so
long
is
because
we've
been
like
well.
What
is
the
endorsement
like?
What
does
it
mean
to
be
a
kubernetes,
endorsed,
project
and
I?
G
Think
that's
all,
because
we
were
talking
about
kubernetes,
adopting
repos
as
opposed
to
repos.
Saying,
hey
I
want
to
be
part
of
this
thing,
and
then
there
is
no
implied
endorsement.
We
don't
have
to
worry
about
that
whole
thing
for
that
level
of
repo,
and
yet
we
get
a
lot
of
the
regularization
that
it,
which
is
what
we
want,
which
is
which
helps
enable
collaboration
right.
G
F
G
E
G
F
I
decide
I
want
my
org
to
be
associated.
I
put
the
CLA
on
and
I
collected
a
whole
bunch
of
contributions
to
the
CLA
and
then
I
turn
to
CLA
bought
off,
have
I
actually
revoked
the
legal
right
for,
since
you
have
to
distribute
those
changes,
I
don't
think
so
right,
because
the
CLA
says
that
you've
W
their
copyright
exhibit
and
I'm.
Saying
that
putting
this
in
the
end,
users
control
is
suspect.
G
Well,
but
I
mean
I,
don't
they
still
think
it
matters
right,
like
I
mean
really
like?
Do
you
I
mean
either
it's
so
unimportant
that
who
the
hell
cares,
whether
they
try
and
do
something
like
this
or
it's
important,
and
someone
will
notice
it
will.
You
know,
pursue
legal
alternatives
or
whatever
right
like.
F
G
F
Across
our
unit,
I
think
we
need
to
figure
out
the
DCO
thing.
Apparently
there
is
some
concern
from
Google's
lawyers
and
maybe
others
that
DCO
isn't
satisfying
enough.
I've
asked
Google
our
legal
team
for
clarity
on
what
their
position
is
on
that
I'm,
not
sure
that
they
get
to
dictate
what
C
and
C
F
for
what
kubernetes
does
at
this
point,
but
I
think
it's
interesting
and
if
we
switch
to
DC
o,
then
we
wouldn't
need
to
deal
with
the
CLA
bought
at
all
right
sure,
and
then
that's.
D
Yeah
so,
like
I,
didn't
mean
to
distract
things
with
the
scale
I
think
so.
My
hang-up
is
just
that,
even
though
you're
using
the
word
associated
to
me,
I
still
think
it
implies
that
we
are
taking
this
repo
and
we're
now
putting
it
under
the
kubernetes
project
umbrella.
Thus,
the
coda
that
becomes
our
current.
D
G
An
imagine
imagine
that
it's
best
practices
right
if
you
write
a
best
practices
talk,
and
somebody
then
goes
as
follows.
Your
best
practices
dog
are
you
adopting
their
project,
no
they're
following
their
best
practices,
we're
enabling
people
to
follow
our
best
practices
in
ways
that
make
it
easier
for
them
to
come
further
in
and
it
enables
it
our
community
to
contribute
they're
more
easily.
My
understanding.
D
D
Thing
like
sig
labels
and
the
consistence
that
our
priorities
everywhere,
/lg
TM
/
approve
for
average
automation
and
all
that
stuff,
where
you
kind
of,
would
consistently
expect
every
repo
that's
poor
to
our
project
or
managed
by
our
project
to
operate
sort
of
the
same
way.
But
my
understanding
was
associated.
We
know
certain
people
just
experiment
and
they
don't
have
time
for
all
that
heavy
waiting
wrap,
but
they
still
need
vendor
neutral
places,
with
legal
protection
to
experiment
and
play
around
so
I'm.
Trying
to
understand.
G
And
know
I
think
it's
more
than
that
right.
It's
it's
like.
It
makes
the
collaboration
easier
so,
for
example,
I
think
that
CLA
or
whatever
we
use
for
copyright
and
code
of
conduct
I
think
those
two
things
are
required
because
that's
what
it
means
to
be
compliant
like
that.
You
know
we
imagine
I
mean
imagine
that
someone
created
some
horribly
toxic
non
code
of
conduct
compliant
Rico,
we're
not
gonna
bring
that
into
the
kerbin
Nettie's
org
right
so
like
what
I
want.
G
If
we
really
want
to
invite
about
this
for
forever,
if
we
really
want
to,
but
there's
a
pressure
gradient
here,
a
very
real
projects
that
want
an
answer
to
this
and
we
have
had
exactly
zero
progress
in
making
us.
They
work
right
and
I.
Think
this
is
the
closest
thing
we
could
come
to
to
something
that
we
can
actually
start
thinking
about,
implementing
and
would
release
the
pressure
wave
right
unless
we
want
to
do
onesie
twosie,
like
with
Dawn's
thing
and
say
like
okay.
Well,
that's
a
no-brainer!
Let's
go
right,
which
I
mean
I!
G
G
The
well
I
mean
I,
don't
care
that
when
we,
the
other
I,
think
that
the
value
of
it
is
actually
the
one
catch-all.
Lord
is
actually
a
stronger
endorsement,
but
I
don't
super
care
if
we
want
to
say
associated
as
playground,
but
but
I
think
the
trouble
is
then
we
have
to
establish
the
logistics
of.
How
do
you
join
and
what's
the
committee
that
decides
how
you
join
in,
what
meeting
do
you
have
to
go
to
Bob?
You
got
people
over
the
wall
right.
G
G
That
the
administration
necessary
to
add
yourself
to
the
CLA
bot
would
be
to
add
the
CLA
bot
as
an
owner
of
your
of
your
organization
or
of
your
project,
which
is
something
that
you
can
do
using
your
github
permissions
with
zero
input
from
anybody
else
in
the
future.
No
right
now,
right
now,
I
mean
assuming
that
the
bot
could
do
it
like
I
need
to
get
a
profession.
You
could
do
it
right
now.
I
do.
E
Think
there
does
need
to
be
some
level
of
enforcement
and
auditing
right
like
there
needs
to
be
like,
as
Tim
mentioned,
you
know
if
it's
like
my
malware,
whatever
there
needs
to
be
some
way
for
us
to
check
and
vet
to
ensure
that
you
know
anything
associated
with
this
larger
project.
Isn't
you
know,
at
least
on
the
relative
up-and-up
right
yeah.
G
G
Anybody,
if
you
think
about
this
in
terms
of
like
the
terms
of
service
for
our
logo,
like
people,
could
be
taking
our
logo
and
like
putting
all
kinds
of
crap
on
it,
printing
stickers
and
distributing
them
at
conferences
and
like
yeah
I,
will
find
out
about
it.
If
it's
big
enough
and
we'll
go,
try
and
knock
it
down
if
it
is,
but
nothing
that
we
do
actually
stops.
That
kind
of
stuff.
Okay,
I.
F
G
F
H
H
Another
thing
that
we
could
do
is
we
could
have
something
as
simple
as
like
somewhere
on
kubernetes
io,
a
list
of
collab
repos
with
pointers
to
these
things
and
you,
if
you're
a
member
of
the
community,
you
can
sponsor
it
by
just
opening
a
PR,
and
if
we
find
that
somebody
is
like
you
know,
you
know
spewing
hates
and
not
conforming
to
the
code
of
conduct.
Then
we
remove
them
from
that
page
and
they're.
H
D
I
was
in
faith,
I
am
in
favor
of
Brendan.
Need
scoping,
found
this
proposal
to
just
kind
of
solve
the
discrete
problem
and
I
hear
there
is
an
urgent
need.
I
guess:
I
thought
the
urgent
need
was
more
for
repose
that
we're
going
to
be
created
in
kubernetes
kubernetes,
but
we're
going
to
empower
them
to
follow
the
core
repo
standard.
That
seems
a
little
commercial
to
me
and
the
only
other
massive
open
question
I
mean
is
whether
or
not
each
sake
should
get
its
own
organization
or
not.
I.
Think.
G
E
Iii
I
really
liked
the
idea
of
having
simulator
repos
even
having
multiple
targets
within
a
repo,
because
right
now
that
means
that's
the
sink.
Ownership
for
graduation
is
well
defined,
because,
right
now
we
have
many
repos
who
owns
what?
Where
and
how
does
it
get
graduated?
At
least
you
can
go
to
a
single
Clearing
House
for
understanding
what
is
owned
there
and
how
you
graduate
that
if
it.
F
Work
but
I
point
on
the
email
like
I'm,
worried
that
sticking
things
in
sig
orgs
produces
weird
artificial
boundaries
that
wouldn't
be
needed
for
collaborative.
Like
hey,
am
I
I'm?
Not
a
member
of
st.
fubar
am
I
allowed
to
send
patches
to
sig
fubar,
so
like
I?
Just
it
feels
weird,
nothing
and
sing
specific
work.
I.
Think
it's
like
a
question.
G
F
G
That
makes
sense,
no
I
mean
that's
what
I
said
in.
Maybe
we
agree
now
that
you
went
through
it,
but,
like
I,
think
that
we
should
make
sure
that
you
can't
graduate
things
generally
can't
graduate
things
to
GA
without
moving
them
into
the
core
org
right
which
mean
that
if
you
had
developed
a
cute
proxy
v2
or
whatever
insignificant
you
couldn't
make
it
part
of
a
shipping
release
until
you
or
a
GA,
you
can
GA
and
as
part
of
a
release
until
you
and
I
think
that
will
hopefully
accomplish
what
you're
talking
about
I'm.
H
G
Effectively
because
I
felt
like
that
was
the
biggest
log
jam
because
and
the
details
of
what
each
endorsement
level
means
I'm,
totally
open
to
refining
right,
like
I,
think
that
it's
the
levels
of
endorsement,
that
is
the
biggest
logjam
that
prevented
us
from
making
progress
here
and
was
the
biggest
thing
that's
separated,
like
hey
three
people
from
the
community
want
to
collaborate.
Projects
from
hey
sig
scale
needs
a
bunch
of
tests,
testing
tools
versus
like
hey.
This
is
something
that's
actually
like
API
machinery
that
we
know
is
essential
and
part
of
our
main
project.
G
G
H
G
G
C
G
I
mean
I,
guess
I
would
I
would
think
that
there
would
be
a
transition
anyway,
like
you
wouldn't
start
out
with
a
vanity
URL,
but
my
suspicion
is
by
the
time
you
get
to
be
a
GA
product.
Maybe
you
do
want
to
have
a
vanity
URL
I,
don't
know,
I
mean
I,
don't
think
that
we
can
introduce.
Maybe
we
can
disassociate
vanity,
URLs
and
github
organizations,
so
you
could
always
be
Cades.
Org,
slash
reschedule
or
even
if
you
were
in
the
six
scheduling,
see
note
that
it
could
be
an
issue
all
right.
G
D
Headed,
thank
you
so
I
just
want
to
straw
man.
What
let's
say
we
decide.
You
know
what
this
more
or
less
looks
good.
What
implementation
would
look
like.
So
let
me
just
for
simplicity,
say
associated
repos,
let's
just
say:
that's
everything.
Inside
of
kubernetes
incubator,
let's
rename
kubernetes
incubator,
kubernetes
playground.
That's
our
list
of
associated
repos,
create
Cooper
native
organizations
for
all
the
snakes.
D
Next
up,
as
we
say
sakes
who
want
to
take
ownership
of
things,
please
movies,
repost
out
of
play,
dubs
and/or
organization,
the
remaining
pressure,
point,
height
or
pain,
point
I've,
heard
of
our
repos,
such
as
coops
grey,
which
felt
like
they
have
gone
through
the
incubation
requirements
to
go
GA.
What
would
we
tell
them
like
I
think
tell
them
to
go.
They
move
to
stick.
I
would
tell
everybody
they
have
to
move
this
thing.
D
G
Right
and
I
think
that
that
was
kind
of
also
the
pressure
try
out
the
forcing
function
which
is
I
really
want
to
force.
I
don't
want
people
to
join
or
to
want
to
join
the
Corps
or
just
as
an
endorsement
I
think
it
has
to
like
you.
Something
ever
needs
to
move
to
the
core
right.
I
would
agree.
I
would
agree
with
that
right.
I,
agree:
I
was
just
trying
to
run
through
this,
but
I
think
you're,
right,
no
you're
right
to
work
through
labor
with
the
people.
G
Who
will
ask
questions
and
I
think
the
question
I
think
the
question
I
wouldn't
even
create
cute
playground.
I
would
say:
incubator
stays,
as
is
because
we
don't
want
to
screw
people
up,
incubator
stays,
as
is
if
you
thought
that
you
were
ready
for
graduation.
Your
job
now
is
to
find
a
cig
to
own
you
and
you'll
move
into
a
cig
org.
If
you
want
to
start
a
new
project,
go
start
an
Associated
project
or
find
a
cig
to
own
you,
your
choice
of
bureaucracy
versus
freedom,
and
then
we're
done.
E
G
Mean
for
incubator
yeah,
we
could
do
that.
I
mean
I,
suppose
so,
I
don't
see
a
lot
of
harm
and
leaving
things
in
there
as
long
as
it's
understood
and
we
put
a
label
on
it,
that's
a
it
says.
Like
hey
we're
done,
you
know
like
we're,
not
adding
anything
new.
There's
no
progress
happening
here.
Things
like
I
guess.
D
G
D
Complaint
there
I
forget,
if
it
was
maybe
sig
note
was
talking
about,
maybe
was
crying
or
something
where
how
can
you
differentiate
the
sort
of
level
of
maturity
or
stability
or
readiness?
So
let's
say
it's
something
that
the
sink
has
worked
on
for
a
year,
but
it's
something
that's
never
really
going
to
get
to
court.
If
this
takes
like
don't
worry,
this
is
totally
stable
and
cool.
How
would
a
user
be
able
to
differentiate
between
that
and
a
bunch
of
other
experiments
happening
within
the
same
thing
right
solving
all
the
world's
problems,
but.
D
G
C
I
think
the
example
Erin.
There
was
not
a
signet
example.
I
think
it
was
when
I
offered
around
Service
Catalog
itself
and
that,
like
there
is
a
sig
service
catalog
and
the
only
output
of
that
sig
is
a
debater
project.
Today,
so
I
guess
and
the
concern
there
was
the
incubator
label
made.
It
feels
yeah
to
some
people
for
some
reason,
I
think
in
this
scenario,
I'm
trying
to
work
through
what
that
sig
would
look
like,
though,
like
so
we
are.
C
We
have
a
sig
service
catalog
that
only
owns
an
incubator
repo,
so
it's
kind
of
a
weird
chicken
egg
thing
where,
like
other
things,
wouldn't
have
produce
two
sig
necessarily
to
incubate.
But
let's
say
a
new
sake
is
formed.
They
get
a
sig
or
right
away
and
then,
in
this
case
catalog
would
have
gone
air.
You.
E
A
E
G
Think
it
depends
I
think
it
really
depends.
I
guess:
I
think
that
there
are
gonna
be
some
tools
that
go
GA
out
of
their
SIG's
space,
Service
Catalog.
Maybe
it's
specific
enough
and
not
core
enough
and
an
add-on
enough
that
it's
a
cystic
that
AG
is
out
of
there,
but
I
think
you
have
to
get
permission
for
that
exception,
that
the
expectation
should
be
a
graduate
decor
and
if
you
need
all
right,
Joe
swings
yeah.
H
This
is
the
thing
that,
like
like
I,
think
saying
that
you
have
to
graduate
to
core
to
go.
Ga,
that's
going
to
create
a
lot
of
pressure
on
stuff,
I!
Think
if,
instead,
we
say,
if
you
want
to
be
GA
and
be
part
of
the
release
bundle,
you
must
be
part
of
core,
but
then
there,
but
having
six
to
find
their
own
release,
bundles
that
actually
ship
out
okay
and
make
a
ton
of
sense
right.
That's.
H
Is
this
question
of
like,
and
you
know,
I
look
at
something
like
helm
right,
it's
like
helm
is
currently
a
graduated
project.
It
has
its
own
release
time.
Ehrlich
Ops
is
in
the
same
case.
It
has
its
own
release.
Timeline
has
its
own
release
bundle.
It
has
its
own
mechanisms
right.
It's
a
completely
sort
of
parallel
project
like
should
that
be
part
of
the
the
Cooper
Neysa
gaps
for
for-4
helm,
for
example,
and
have
it
be
GA
and
actually
be
helm
out
of
there
right
I
mean
that
doesn't
seem
insane
to.
G
Me,
no,
that
seems
really
fun
too
I.
Think
that,
and
maybe
we
make
a
distinction
between
like
Sig
GA
versus
kubernetes,
GA
and
kubernetes
GA
means
sega
architecture
has
to
say
it's
GA,
even
like
just
sort
of
a
criteria.
It
doesn't
mean
that
you
move.
It
just
means
that
someone
else
don't
go.
Look
at
what
you
said
was
GA
and
we
have
some
degree
consistency
across
the
project.
Well,.
H
G
G
Are
we
in
enough
agreement
as
I
watched,
Joe
sort
of
coffin
stuff
DX,
and
are
we
in
enough
agreement
around
this
to
say
like
hey,
let's
and
Isaac
Irina
have
agreement
read
this
to
ship.
It
externally
movies,
sort
of
shifted
externally
anyway,
but
like
and
get
the
comments
and,
and
today
we're
gonna
go
discuss
this
and
narrow
it
down.
I.
E
E
Like
a
fact
like
what
like
what
we
just
did
just
look
just
exactly
the
question-and-answer
session
that
we
just
did
I
think
with
you,
know,
Aaron's
outlining
of
different
user
stories,
and
what
that
actually
means
would
be
super
helpful
to
alleviate
what
will
essentially
be
a
bunch
of
people
are
going
to
take
umbrage
with
some
of
the
ideas
here.
That's
gonna
happen,
but
at
least
we
have
the
user
stories
thought
out
enough
that
you
know
if
there's
a
path
for
a
or
B
or
C.
D
H
D
G
The
GA
court,
a
little
thing
is
good,
but
I
feel
like
yo
L
stuff
is
I,
mean
I,
just
don't
care
and
I.
Don't
think
anybody
else
does
either
I.
Think
if
you,
if
you
associated
then
clearly
you
don't
care.
If
it's
sick,
then
clearly
it
gets
to
decide
how
serious
or
not
they
are.
If
it's
cooler
than
architecture
should
the
side.
A
Okay,
so
it
sounds
like
we
need
another
rev
on
the
docks.
I
recorded
some
of
the
open
questions
from
this
discussion
in
the
notes.
Brendan,
alright,
guys
I'll
riff
I'll
rip
tonight
tomorrow,
something
like
that
and
get
a
drop
by
the
end
of
the
week.
Yeah
as
I
didn't
really
speak
much,
but
this
person
he
helped
the
kid
off
Thank,
You,
Bader
I,
think
this
proposal
makes
a
lot
of
sense
insults
the
issues
I
want
to
create
a
repo.
How
do
I
do
that
and
have
some
path
to
kubernetes.
H
So
so
Phil
and
Michelle
and
I
I
met
and
to
talk
about
and
I
can
pull
up
the
notes,
but
to
talk
about
sort
of
what
is
the
interfaces
between
SIG's
and
sort
of
the
rest
of
the
project
and
what
are
the
rights
and
responsibilities
and
expectations
out
of
that
and
I?
Think
one
of
the
the
core
things
there
is
a
recognition
that
there
is
no
one-size-fits-all
solution
here
and
that
there
are
different
modes
that
different
SIG's
operate
in
and
so
before
we
make
any
sort
of
decision
there.
H
H
H
H
A
A
A
H
H
B
That
sounds
right,
I
think
maybe
the
other
things
were
they
from
a
higher
level.
We
want
to
provide
a
baseline
template
of
organizational
structure
that
states
if
they
don't
have
submission
structure
right
now.
A
couple
of
things
we're
targeting
were
like
how
decisions
are
made
within
the
sake
both
like
technical
decisions
and
process-related
decisions
and
like
conflicts
are
resolved,
how
decisions
are
escalated,
that
sort
of
thing
and
so
from
the
survey,
hopefully
we'll
figure
out
what
people
are
doing
today.
If
anyone
has
something
like
that
works,
we
can
maybe
cannibalize
that
okay.