►
From YouTube: Kubernetes Community Meeting 20170202
Description
We have PUBLIC and RECORDED weekly video meetings every Thursday at 10am US Pacific Time.
https://docs.google.com/document/d/1VQDIAB0OqiSjIHI8AWMvSdceWhnz56jNpZrLs6o7NJY
Release watch - 1.6, 1.5, 1.4; SIG Service Catalog; SIG Big Data; Demo: Kops; Governance discusssions
B
Makes
me
want
to
make
try
to
make
some
joke
about
self
hosted
Cooper
nettings
that
I
can't
quite
right
back
pull
it
together.
If.
B
Morning,
please,
alright!
Well,
it's
a
good
thing:
I'm,
not
a
stand-up
comic
I
have
a
hard
time,
but
I'm
making
living
so
today
is
fab.
Second,
this
is
the
key
to
benetti's
committee
meeting.
We
are
recording
few
things
on
the
agenda
today.
Looking
here
at
the
notes,
I
did
one
of
the
one
of
the
advantages
Sarah
gets
by
asking
me
to
host
today
as
I.
Just
have
someone
on
my
team
to
please
be
the
note-taker,
so
we
don't
have
to
be
I
volunteered
in
so
thanks
about
thanks,
Eleanor
girl.
B
A
damn
can
cover
releases
we're
going
to
probably
get
a
sub
bullet
talk
there
by
igor
release,
update
by
assad
fat,
I'm
not
sure.
Hopefully
I
was
takin
a
nice
and
then
some
big
update
and
then
really
be
good.
If
we
could
have
some
time
at
the
end
talk
about
governance,
that's
an
important
topic.
I'm
gonna
make
sure
than
me
that
I
knew
things
along
such
a
way
that
we
have
some
on
talk
about
at
the
end
and
with
that,
I
will
pick
it
off
and
hand
it
to
Dan
go
down.
C
C
Cool
as
I
guess,
if
everyone's
okay
with
staying
with
this
is
sort
of
confirmed,
that
would
work
for
us
cool
all
right,
there's
confirm
so
yeah,
so
we
put
out
upper
one
they're
playing
well
with
that.
So
while
we
are
comfortable
secure
releases,
there
is
a
couple
things
that
we
should.
I
keep
my
supporter
to
make
it
so
that.
E
C
More
likely
will
be
able
to
make
it.
So
this
is
the
son
of
sort
of
status
checks
that
these
are
things
that
we
totally
have
confident
that
we'll
be
able
to
do
right.
But
if
they're
not
done
correctly,
they
gave
me
to
a
song
but
having
just
listen
release
orders
across
teacher
short,
so
the
first
thing
is
actually
feeling
easily
seen
we're
still
remember
short
of
what
we
said
that
we
have.
So.
F
C
The
second
one
is
to
respect
the
country
so
having
people
try
to
get
features
inaccurate,
we
have
closed
the
French
and
our
shortages
were
crowned,
Bobby
really
explosive
sound,
and
if
we
believe
that
I'm
here
able
to
do
sometimes
but
having
people
add
features,
Africa
trees,
expand
took
a
bunch
of
the
place
where
we
don't
actually
know
yet.
So
that
is
definitely
is
like.
So
this
is
a
project.
I've
been
working
on
a
little
bit
trying
to
get
each
sleep
in
the
owner
style
of
scientists
in
30
APR
in
there
as
I.
C
How
long
as
earlier
I
think
of
it.
This
we
have
to.
Basically
it
would
allow
test
its
high
ownership
of
the
sacred
place
and
I'm,
suggesting
that
we
have
your
targets.
Resync,
where
they
need
to
do
is
eliminate
two
thirds
of
those
Spears.
We
don't
have
the
hopefully
she's
trying
to
run
it
later
stay
and
help
this
numbers.
We
can
actually
say
what
those
numbers
are
proceeding,
but
the
idea
is
like
we
did
with
the
expectation
that
two-thirds
of
PT
replace
a
little
bit
by
the
time,
Oh
trees.
C
C
Okay,
cool:
it
is
this
major
elements
that
it's
like
a
cherry
cracking
and
if
there's
any
other
major
things
that
we
should
talk
about,
that
sort
of
add
some
risks.
We
released
the
first
one.
E
C
Three
being
enabled
by
default
for
a
large
number
of
users
that
are
upgrading
their
buses
just
will
require
a
migration,
incredibly
different
area,
that
we
need
more
documentation
that
forearm
she
ground.
So
that's
just
something
that
we
should
keep
an
eye
on
the
next
one
is
PRI
enable
that
people
are.
You
chose
this.
C
Requires
every
starty
grade,
so
is
that
operationally
to
let
go
of
subsidies
to
users,
and
it
also
in
terms
of
code
paths,
a
lot
of
that
alongside
yet
to
be
three
change?
Basically,
every
single
code
baptism
by
this
release
in
a
way
that
has
necessary
enforcement,
is
attested
by
that
necessarily
app
in
a
actual
release
and
the
last
part
is
pad
level
secrets
which
changes
which
will
also
require
in
their
grain
in
distress
and
changes.
C
C
Yet
then,
you
can
circle
back
to
that
accessible,
alright,
but
the
biggest
thing
that
I
sort
of
told
I
previously
services.
It's
suggestive,
only
seem
to
ensure
that
the
leases
go
out
on
time
and
durable,
high-quality,
and
this
is
a
tuesday
that
we
should
be
thinking
about.
So
that's
my
general
understanding
how
I'm
doing
this,
but
I'm
totally
welcome
to
do
interpretations
with
once
before
formal.
B
C
Hoping
that
as
if
we
can
right
now,
circle
circle
back,
maybe
at
the
end
this
discussion
or
we
can
do
it
out
of
the
one
and
two
tvs
that
if
you
determine
the
first,
is
the
release
date,
the
team
actually
gained
filled,
and
the
second
is
setting
that
target
for
seeds
and
the
most
hey
with
we
can
put
off
it.
We
can
wait
until
the
actually
as
numbers
with
how
many
stages
your
game
aside
perfectly
perfectly
right,
but
instead.
H
I
D
So
I
have
a
brief
update.
We
have
42
the
features
on
getting
106
release
and
just
because
a
great
products
lated,
it's
really
skilled.
The
stabilization
really
smooth
as
most
of
the
features
socket
and
stable
beta
stage,
and
also
it's
a
good
time
for
the
future
owners
to
start
working
on
them
documentation.
You
haven't
here
and
like
component
documentation
if
you
have
existing
documentation
and
preparing
to
release
those
forgeries
muscle.
If
all
for
me.
B
J
Think
the
floor
is
yours,
cool,
so
1.53
is
scheduled
to
go
out
on
friday
of
next
week.
Februari
10
we're
hoping
to
get
any
cherry
picks
in
by
wednesday.
If
you
have
any
cherry
picks
for
the
1.5
branch,
go
ahead
and
assign
them
to
me
and
I'll
make
sure
that
they
get
merged
in
time.
That's
it
for
1.5
I,
don't
know
it's
jess
is
here,
come
on
that.
K
B
Oops
volume
there
we
go
all
right,
I!
Think,
let's
see
here.
Let's
do,
let's
do
the.
Let's
do
this
thing
update,
let's
circle
back
on
a
couple
of
the
commentaries
around
the
release,
release
steam
and
so
forth.
Just
to
give
us
give
folks
a
little
bit
more
time
to
think
about
to
get
up
the
courage
to
volunteer
and
then
we'll
do
the
demo
I
fit
up
age
fall.
You
go
hello,
everybody
so
yeah
the
sig
I'm,
providing
enough
update
for
a
cig,
Service
Catalog
laughs.
B
Last
week
week
before
last,
we
had
a
really
great
base
to
finish
our
second
face
to
face
for
the
sig
in
Boulder
Colorado,
we're
working
on
some
remaining
design
issues
for
our
MVP
milestone
and
what
the
TLDR
of
the
MVP
milestone
scope
is
is
that
we
have
a
separate
API
server
with
the
Service
Catalog
API
and
integration
with
the
open
service
broker.
Api,
for
those
of
you
that
aren't
aware
open
service
broker,
API
is
the
is
the
like
next
generation
of
what
started
out
originally
as
class
boundaries.
B
Service
Worker,
API
and
the
what
a
service
worker
is
is
something
that
allows
you
to
provision
new
instances
of
services
and
binded
em
in
your
applications,
so
binding
being
I
want
to
consume
credentials,
coordinates
and
config
items
related
to
to
using
this
particular
service.
As
an
example
say
it's
like
a
Postgres
database,
so
we
we
have
the
API
server
working
now,
which
I
want
to
give
a
shout
out
to
Morgan
Bauer
and
Morgan
Bauer
at
IBM
and
Jeff
peeler,
add
red
hat.
Who
did
a
lot
of
hurdles
in
the
dark?
B
B
B
We're
working
on
adapting
the
controller
from
pprs
to
the
actual
API
server,
okay,
Oh
litter
question
you
mentioned
Cloud
Foundry
is
this:
is
there
some
attempt
to
around
cloud
foundry
portability,
or
is
it
just
kind
of
taking
the
sub
concept
and
so
yeah?
That's
a
really
good
question.
Bob
I
think
that
I
would
say
that,
like
an
end
goal
would
be
an
end.
Goal
would
be
to
have
service
workers
that
are
compatible
both
with
Cooper
Nettie's
and
with
Claude
boundary.
B
I
know
that
there
are
excuse
me
there
are
potentially
other
platforms
that
might
want
to
use
it,
but
I
would
say
that
that
probably
having
multiple
consuming
platforms
is
like
a
pretty
significant
cultural
shift
for
brooker
implementers,
so
I
can't
make
any
guarantees
or
necessarily
point
to
any
service
workers.
That
would
work
that
way
out
of
the
box,
but
that's
definitely
a
goal.
I
would
say
for
folks:
in
the
open
service
broker,
he
goes
jump
yeah.
H
H
Some
of
the
binding
information
between
the
service
coordinates
credentials
and
additional
client
configuration
and
that's
a
desirable
improvements.
Also,
the
parameters
that
need
to
be
provided.
The
various
service
broker
calls
are
not
discoverable
and
that's
a
big
inhibitor
as
well,
so
that
you
can
use
evolve.
Hopefully,
cloud
battery
and
other
orchestrators
will
adopt
extensions
and
that's
being
driven
to
the
open
service
broker
ever
yeah.
B
In
fact,
those
two
gaps
that
Brian
is
just
pointed
out
right
now
are
both
things
that
that
we
have
proposals
in
progress
for
on
the
open
service
broker,
API
working
group,
so
we're
hoping
that
we
can
close
those
gaps
pretty
rapidly
because
they're
they're
important
for
us
in
Coober
Nettie's,
at
least
like
the
the
binding
credential
differentiation
of
like
what
is
the
what
is
the
type
of
a
particular
piece
of
data,
but
more
so
like
in
general.
The
parameters
thing
is
generally
desirable.
That
answer
your
question.
Bob
I'll.
H
That
was,
it
was
great
thinking
and
I.
Just
also
want
to
say
that
this
is
going
to
be
a
huge
boon
to
kinetic
users
to
microservice
developers
and
others
at
the
big
gap
that
has
been
missing
board
has
one
at
this
for
a
long
time
actually,
and
it
was
never
develop
their
its
kind
of
facilitate
portability
from
death,
fraud,
environments
and
all
kind
of
business.
G
So
a
lot
of
people
there
you
decide
our
first
big,
leading
this
race.
Basically,
it
successfully
I
reckon
listen.
Lee
is
currently
focusing
on
the
public
park.
In
addition,
our
source
of
starts
and
go
into
it
alpha,
and
we
also
have
some
talking
like
a
changer,
slow
blink
and
support
for
others.
It
the
assessment
go
on,
so
I
would
like
to
say
anyone
who
has
an
interest
in
fig.
They
are
together
flat
house
and
community
on
arguing
on
been
phase
of
coming
up.
The.
C
A
So
the
13th
of
February's
in
Monday
I
think
it's.
This
is
a
date
that
we
are
raising,
alarms
that
having
a
phone
week
after
we
raise
alarms
is
probably
not
a
bad
thing.
We
could
also
move
it
back
to
the
thursday
before
that
and
be
that
puts
us
at
a
community
meeting
again
and
we
can
start.
At
least
you
know
rallying
your
own
video.
If
I'm
of
the
community
meeting
is
it
I
think.
B
C
C
N
N
The
intention
is
I
mean
your
system
features
that
hopefully
are
relevant
to
everyone,
whether
using
cops
or
AWS
or
not.
Even
if
you
are
misguided
enough
to
have
some
other
preference,
which
is
acceptable,
then
hope
you
found
those
will
be
relevant,
so
I'm
going
to
start
quickly
by
running
a
command
the
basically
jumped
to
the
conclusion,
because
it
uses
one
of
our
new
I'm
one
of
the
new
way
shaded,
where
services,
which
takes
a
while
to
spin
up
so
I'm,
going
to
just
start
that
going.
Why?
Then,
okay?
N
So
we
are
creating
a
cluster
with
all
sorts
of
fancy
features
which
were
then
going
to
describe,
and
it
is
going
to
preview
a
ridiculous
number
of
changes
that
are
going
to
happen
because
it
is
a
very
complicated
cluster.
So
the
AWS
objects.
We
are
creating
this
to
me
even
to
page
through
so
and
I'm
just
going
to
get
this
going
with,
which
I'm
going
to
say
go
ahead
and
do
that
all
right.
N
So,
while
that
runs,
let's
talk
about
what
that
what
that
is
did
so
we
have
a
lot
of
new
features,
but
the
the
sort
of
Q
really
big
headline
features
our
support
for
overlay
networks
and
the
private
apologies
they're
critical
features
that
are
very
important
for
everyone.
So
overlay
networks,
we
have
44
overlay
networks
from
we've,
calico
co
p.m.
networking,
and
we
have
to
bring
your
own
option
for
NEC
and
I
plugin
that
you
want.
We
have
support
for
flannel
and
romana
there
that
are
coming
in
and
brought
graphs.
N
It
turns
out
dead
on
AWS,
it's
particularly
important
because
the
default
networking
mode
that
started
this
once
the
default
networking
mode,
this
shifter
benetti's
prior
to
CN,
I
was
DPC
rallying
and
it
has
a
limit
on.
They
do
us
about
fifty
or
hundred
knows.
If
you
raise
your
limit
and
when
we
want
to
get
to
larger
numbers
of
nodes,
then
networking
overlays,
like
Weaver
calacatta
bar
what
you
need.
F
N
If
you're
not
running
on
AWS,
it's
great,
because
some
of
them
can
do
encryption,
for
example,
which
can
be
handy
if
you
want,
if
you
have
to
do
encryption
on
the
wire,
it
also
enables
our
other
big
headline
feature,
which
is
private.
Apologies,
I,
think,
actually
that
we're
lucky
enough
to
have
the
author
private
apology.
The
private
apology
feature
on
the
call
today,
Chris
Nova,
so
I
mean
if
you
can
speak
and
jump
in
and
tell
us
a
little
bit
more
about
it.
Okay,.
E
So
private
apologies,
basically
let
you
run
all
the
nodes
and
masters
in
the
Cooper
dennys
cluster
without
a
public
IP
address.
So
it's
just
a
name
for
the
network,
orchestration
that
we're
using
to
secure
our
notes
and
masters
to
a
nice
resources.
The
default
way
has
traditionally
been
to
run
everything
with
a
public
IP,
and
then
you
use
cloud
firewalls
and
on
machine
firewalls
to
secure
that.
But
if
you're
a
particularly
security
sensitive,
we
have
people
working
with
HIPAA
PCI
compliant
clusters.
E
It
makes
a
little
bit
more
sense
that
have
that
lock
down
into
a
private
IP
subnet
in
networking
space.
It's
also
meaningful
for
our
users,
who
have
any
sort
of
constraints
with
running
a
hybrid
cloud.
It's
important
to
call
out
those
two
big
challenges
with
private
topologies
there's.
One
of
them
is
how
the
clients
can
reach
the
crew
benetti's
API
and
how,
of
course,
the
instances
can
reach
out
to
the
Internet
the
Cooper
Nettie's.
E
A
TI
is
simpler
because
we
can
just
put
a
wsdl,
be
load
balancer
and
pregnant
cluster
and
I
will
poke
a
hole
into
the
network
and
field
around
traffic.
On
the
notes.
Talking
to
the
public
internet,
however,
we
had
to
take
advantage
and
use
products
called
eew,
SMS
gateway.
It
only
allows
redress
and
users
not
so
all
of
the
traffic
comes
from
a
single
IP
of
the
internet.
That's
pretty
much
the
long
and
short
of
it.
I
think
I
touched
on
everything.
Xd
deafen,
yes,.
N
N
B
N
N
N
So
actually
so
that's
everything
that's
nicely
onto
the
Sebastian.
Well,
I,
guess
I
guess.
So,
if
you're
not
running
on
the
AWS,
we
took
it
up
private
eye,
peas
and
like
what
we
don't
end
with
this
or
not
only
to
us,
you
could
run
an
instance
with
the
public
IP
to
act
as
max
gateway
and
probably
in
the
next
version
of
costs.
We
want
to
support
GCE,
which
often
have
you
know
out
of
the
box,
not
a
gateway.
So
we'll
probably
do
something
like
that.
N
Another
option
would
be
to
run
a
proxy
like
squid,
which
could
be
really
nice
if
you
needed
more
control
or
auditing
of
you
know
what
exactly
your
notes
can
access,
rather
than
just
blanket
bumi
and
yes
or
no
access,
and
then
to
do
all
that
we
added
support,
for
instance
this
in
different
roles.
So
if
I
don't
skip
I
G
orange
ID,
you
can
see
that
num
s3
is
really
slow
today
on
the
line,
but
so
you
computer.
We
have
the
option
of
interesting
in
different
roles,
so
we
have
master
roles
which
are
the
masters.
N
We
have
nodes
and
had
I,
not
typo'd,
my
continuation
character,
we'd
have
a
third
row
or
another
another
instance
group
for
bastions
and
evacuated
tremendous
a
server
with
the
public
IP.
The
USS
ation
do
and
then
you
message
from
there
into
any
of
your
instances.
It's
optional
soon,
as
you
can
see,
so
it
can
break
anything,
and
it
is
useful
for
diagnostics
and
I.
L
G
N
Yeah
I
think
it
generally.
We
get
good
money
from
ingress,
but
we
can.
We
can
go
to
that
separately.
So,
while
we're
waiting
I'm
one
of
the
things
I'm
really
happy
about
with
this
release
is
for
the
e
de
testing
that
we've
done,
like
I
feel
like
if
something
isn't
running
under
ed
is
probably
broken,
and
then
my
hypocrisy
shows
through
because
we
are
not
see
de
testing
like
a
J
or
weave
or
vegan
pirates
apologies,
but
zach
has
done
a
neck.
N
So
thanks
to
that
might
also
like
to
give
a
big
shout
out.
Everyone
contributed
to
the
custom
point
for
release,
in
particular
like
Chris,
no
Chris
love,
Eric
hall
and
some
and
honestly
too
many
people
to
lift
give
a
large
and
growing
number
of
contributors,
and
we
also
are
very
welcoming
we
think
to
new
contributors
so
come
join
us.
What
we're
going
to
be
working
on
in
16
is
authentication
and
our
batch
I
am
functionality.
N
Maybe
some
support
for
some
other
clouds
and
they've
been
creating
an
HTTP
service
to
wrap
comps
and
we're
gonna
be
working
with
tube
ADM
to
define,
hopefully,
a
common
way
to
install
and
manage
add-ons.
So
if
you're
interested
in
any
of
those
hope,
you'll
join
us.
If
you
want
to
there's
a
cost
channel,
we
have
office
hours
tomorrow
at
ten
a.m.
pacific
think,
but
it's
on
the
head
and
not
the
channel,
and
we're
going
to
talk
about
some
documentation
and
refactoring
and
maybe
even
a
new
website
and
we're
very
big
on
github
issues.
N
We
have
a
lot
of
issues
and
we
think
that's
a
good
thing
there
we
are
so
you
can
see
that
we
spun
up
here
3mm.
So
we
did
an
HJ
cluster
with
three
masters
master
count
three,
so
we
have
three
master
node
1,
2
3.
We
did
two
nodes
and
if
we
can
cuddle
subscribe,
no
boom,
you
can
see
that
they're.
The
address
is,
and
we
do
not
have
those
are
all
internal
addresses.
We
do
not
have
a
public
IP.
N
The
only
way
to
reach
this
is
through
the
non-existent
bastion
or
through
the
EOB,
which
is
in
front
of
the
API
server,
and
it's
it
is
easy
to
add.
If
you
want,
if,
if
we
want
to
fix
that,
when
we
just
do
cops,
create
IgE
Bastian
will
ask
in
would
actually
go
and
create
the
bastion.
So
we're
not
we're
not
stuck.
N
We
don't
have
the
nuke
anything
and
upsets
you
like
to
edit
your
Custer's
and
updated
easily,
so
that
was
some
of
the
things
we're
doing
invaded
too
and
hopefully
not
just
about
addressing
cost
about
everyone's
can.
If
you
are
interested
in
more
secure
clusters
elsewhere
and
look
what
we
check
out
with
getting
caught.
I
N
B
B
So
there
is
a
governance,
envie
de
Bruyne
grant
document
which,
as
the
notes
here
say,
because
the
see
how
we
often
think
operate
on
the
project
today,
there's
been
a
patient,
some
very
good
debate
on
the
higher
level
project
governance
I
made
a
proposal
about
three
branches.
Someone
has
pointed
out
that
what
this
really
is
kind
of
a
call
for
Constitution,
so
I
think
this
is
a
really
critical
project.
Critical
items
for
the
project
looks
as
if
there's
a
call
for
a
special
meeting,
governess
special
meeting
I'm,
not
sure
what
the
latest
isn't.
O
O
So
like
and
the
one
thing
I
realized
I
didn't
put
on
there
that
maybe
I
should
put
on
there
is
like
preference
for
the
next
few
weeks.
Right,
like
I'd,
have
days
and
hours,
but
I
didn't
have
somebody's
going
on
vacation
or
whatever
I.
Don't
have
that
so
I'm
going
to
try
I'll,
add
that
in
and
we
can
get
together
and
have
that
discussion,
because
I
think
that
we're
just
not
going
to
accomplish
it
in
the
context
of
community
meeting,
because
we'd
have
to
dedicate
an
entire
community
meeting
to
it.
Basically
or
maybe.
O
B
E
O
L
I
O
H
H
Discussion
community
talking
past
each
other
and
trying
to
solve
different
problems,
yeah
context
of
the
existing
proposals,
another
just
a
clarification
again
on
the
survey
when
you
plan
to
start
the
meeting.
The
open
source
leadership
summit
in
tahoe,
pristine,
CF
and
some
other
folks
in
the
community
is
in
two
weeks,
and
I
actually
have
yet
another
thing
that
we
next
week,
which
yeah.
O
L
O
I
didn't
actually
I
looked
at
what
it
looked
like:
I
didn't,
actually
try
and
click
on
today
it
looks
like
it
might
work,
but
it
open
all
right.
That's
what
I
get
for
doing
it
early
this
morning
before
the
meeting
and
then
I
think
actually
something
that
occurs
to
me.
That
I,
don't
think
we've
ever
actually
done
is
like
a
community
happiness
survey.
All
right.
You
know
they
know,
but
I
mean
I.
O
Think
it's
true
like
we
do
this
like
we
did
these
at
Google,
and
we
do
this
or
similar
thing
Microsoft
and
like
we
do
these,
like.
How
happy
are
you?
What
are
the
issues?
A
lot
of
people
of
all
kind
of
service
and
I
think
that
that's
actually,
probably
to
speaking
to
a
Brian,
was
saying
I
think
that's
actually
probably
a
really
good
way
to
get
this
data
and
find
common
theme,
something.
H
O
L
A
Surprises
point,
and
really
all
of
this
discussion
I,
would
love
to
say
that
we
need
to
hear
for
more
people
more
broadly
in
the
community,
because
we
are
getting
all
the
usual
voices
and
you
don't
have
to
be
a
community
or
sorry.
A
project
founder
and
or
community
lead
voice.
In
order
to
give
us
suggestions
on
this
and
add
to
the
conversation
and
get
up
on
all
of
this
or
even
attend
the
governance
meeting,
because
we
really
want
to
figure
out
what
works
for
everybody.
A
O
H
O
So
I
will
draft
that
I'll
put
it
on
to
kerbin
Eddie's
a
dev
announced
I.
Guess
there
could
be
an
indecent
out.
I,
don't
actually
feel
like
Cooper
native
dev
announced
has
gotten
much
love,
so
I
mean
I've,
been
kind
of
cross.
Posting
and
I'll
continue
to
cross-post
and
we
can
iterate
from
there
like
I'll
put
it
out
there
as
a
candidate,
not
as
something
people
should
fill
out,
but
as
something
people
should
comment
about
the
question.
O
B
B
Great
all
right
with
that
I
think
there's
one
more.
I
wanted
to
bring
up
and
I
guess
I
would
take.
I
would
take
some
input
as
to
whether
we
think
this
should
be
deferred
to
that
meeting
or
something
that
we
want
to
try
to
move
ahead
of
that,
which
is
a
I
proposal
to
lift
the
PM
group
to
a
safe,
I'm,
not
sure
that's
actually
controversial
in
any
quarters
at
this
point.
So.
A
Let
me
give
you
an
example
of
why
I
made
its
differences
may
occur,
burn
Eddie's
death,
which
is
not
a
cig.
We
have
to
burn
Eddie's
p.m.
which
is
not
mistake.
When
I
was
establishing
special
interest
groups
inmates
to
burn
a
DCM
initially,
not
one,
the
idea
in
my
head
was
to
have
special
interest
groups
online
with
technology,
whereas
p.m.
was
across
the
whole
of
the
project.
So
something
like
elders
are
lead
would
also,
then
be
a
comfortable
of
the
project,
and
so
they
would
not
be
a
cig
himself.
So
that
was
my
initial.
B
B
One
I
think
by
the
way,
Sarah
would
be
great
if
you
put
some
commentary
in
on
the
PR
around
the
promotion
on
this
hey,
if
you
put
but
I,
think
the
expectation
that
there's
regular
meeting
the
meeting
to
recorded
you
there's
a
bunch
of
work.
That's
been
done
around
defining
expected
behavior
from
things,
and
there
really
isn't
that
for
the
work
for
the
working
group
and
working
working.
H
P
A
H
Not
obvious
that
it's
valuable
to
distinguish
it,
and
even
for
things
like
all
the
renewals
that
is
expanded,
so
much
at
this
point
that
we
might
want
to
consider
making
the
release
function
as
they
take
itself
into
the
biggest
self
I.
Think
it's
useful
to
not
have
excessively
different
accessibly
many
different
models
of
the
ways
of
doing
things.
That's
why,
when
writing
up
the
governance?
H
H
Oh
and
one
other
point
about
this,
I
would
like
all
areas
of
the
project
as
much
as
possible
to
be
covered
by
a
cig
is
possible,
or
at
least
some
other
well
defined
role,
and
particularly
our
like
code
to
be
owned
by
cigs.
We
do
have
a
few
things
right
now,
which
don't
have
clear
ownership
in
the
project
and
that's
a
problem
for
accountability
and
for
what
problem,
making
sure
that
someone
can
still
own
code.
When
the
original
author
moves
on,
and
things
like
that,
so
I'll.
A
O
I
would
actually
sit
for
internationalization
I
think
we
should
probably
treat
that
separately
as
a
separate
sig.
Just
because
it'll
be
a
distraction
to
most
of
the
CLI
work
to
talk
about
it
there,
it's
really
pretty
orthogonal
and
then
there's
also
a
lot
of
this
process,
which
is
basically
just
like.
How
do
we
use
de
nada?
How
do
we
use
poet
like
that?
Actually
I'm,
all
just
learning
that
so
it
makes
sense
to
have
to
be
separate,
yeah.
H
Another
comment
on
big
versus
working
group
I
think
it's
fine
for
cigs
to
come
and
go.
We
had
big
data
come
up
and
then
go
for
a
while.
Now
it's
resurrected,
I,
don't
think
we
need
a
separate
concepts
of
working
group
just
because
we
think
you
may
be
bound
in
duration,
like
contributor
experience.
Originally
it
was
hypothesized
that
might
be
bounded
duration,
I
think
whenever
in
and.
O
I
would
say
if
anyone's
interested
in
internationalization,
please
please,
please
contact
me
or
you
know,
put
something
on
the
list
or
something
because
you
know
I,
basically,
organically
discovered
people
who
had
some
degree
of
expertise,
fabiano
and
then
from
Warsaw
actually
had
some
experience
and-
and
so
that
was
helpful,
it
would
be
great
if
other
people
who
are
interested
also
we're
willing
to
kind
of
talk
about
their
experience,
because
I've
actually
never
done
this
before.
So
this
is
the
new
learning
experience
for
me.
O
F
I'm
something
about
cigs,
especially
ones
that
are
cross-cutting,
like
sagat's,
because
we're
running
into
this
now
with
all
the
various
operations
focused
cigs.
It's
getting
hard
to
track
all
the
things
that
are
being
duplicated
against
each
Sega
and
whatnot
and
I
recently
ran
a
retro
for
cig
app
and
it
really
drew
out.
It
may
be
one
thing
that
would
help
us.
F
So,
for
example,
if
we
have
one
person
that
attended
as
note-taker,
retro
door,
etc
for
all
the
obstacles,
it'd
be
really
easy
for
those
for
the
sig
people
for
the
sig
facilitators
list,
there's
like
four
or
five
of
them
across
all
the
cigs
to
get
in
a
room
and
have
sort
of
scrum
of
scrums
like
communication
about
what
is
happening
on
all
the
figs
and
give
us
a
view
of
it.
That
is
currently
really
impossible
to
get.
F
It
would
also
help
because,
as
cig
leaders,
you
want
to
be
able
to
participate
in
retros
and
whatnot
and
having
some
facilitation
there
would
be
really
great
to
to
free
the
fig
donors
up
to
also
give
feedback.
So
when
I
was
doing
a
cig,
apps
retro,
the
the
sig
apps
leaders
were
able
to
participate,
and
it
was
a
huge
benefit
for
them
to
be
able
to
do
that
in
the
neutral
way.
F
E
F
Like
getting
meeting
notes
or
presentations,
your
retros
and
all
that
stuff,
so
that's
just
my
thing
and
I'm
willing
to
write
up
a
proposal.
Sort
of
to
this
extent,
if
that's
useful,
would
but
I
really
feel
strongly
that
we're
worth
a
scale.
Now
with
so
many
things
that
we
need
some
sort
of
crosscut
view
to
be
able
to
see.
What's
going
on
so.
H
I
think
that's
a
great
idea.
I
want
to
also
say
that
the
effort
to
split
the
difference,
github
notification
in
two
different
channels,
for
example,
differentiating
bugs
from
design
proposals,
which
is
a
standard
strategy
used
by
pretty
much
every
team
up
and
on
at
every
company,
is
an
effort
to
try
to
enable
the
kind
of
visibility
across
the
things
they're.
A
So
I
would
actually
ask
Jason:
is
you
and
Caleb
can
get
together
and
see
if
this
is,
if
there's
a
way
to
integrate
your
idea
with
what
the
PM
group
is
asking
for
representatives
from
the
cigs,
because
Caleb
had
a
good
idea
that
that
might
be
a
way
to
handle
this
as
well?
So
maybe
you
two
can
pull
together
some
sort
of
proposal
or
find
out
that
that's
not
it
at
all
and
keep
on
yeah.
H
All
right,
I
think
I
actually
had
one
more
item.
I
dropped
on
the
agenda,
maybe
after
the
you
know
was
drafted
or
some
doubtless,
which
is.
I
want
to
make
people
aware
that
we're
making
progress
on
the
on
the
definition
of
the
core
and
the
communities
layers,
how
to
manage
extensions
on
cadet
ease
and
can
conformance
testing
and
things
like
that,
all
right,
so
working
doc.
H
A
B
Very
good
I'll
say
I'll
say
at
least
us
on
this
topic,
which
is
some.
The
topic
of
court
has
come
up
just
casually
in
conversation
between
me
and
my
various
folks
around
it
comes
up,
and
I
really
have
heard
almost
nobody
anything
other
than
in
favor
of
it.
So
I
think,
there's
a
lot
of
community
a
lot
of
community
support
for
the
concept
that'd
be
great
to
get
that
feedback.
A
There's
a
note
from
elan
in
here
saying:
are
there
any
plans
for
dead
or
community
sig
meeting
certain?
Could
con
carne
de
caunes
to
work
what
we
had
to
seattle?
So
the
answer
to
that
is:
there's
not
going
to
be
a
dev
summit
or
anything
that
looks
like
what
happens
the
day
after
could
calm
in
seattle.
I
have
I'm
happy
to
help
facilitate
people
who
want
to
have
big
meetings
there
there
any
six
at
once,
if
have
meetings
there,
but
I
was
not
blending
anything
formal
and
project
cross-cutting.
A
The
next
leadership
summit,
so
we're
an
alternate
dev
summit
and
leadership
summit.
The
leadership
summit
is
being
we're
working
on
getting
space
for
that
right
now
and
the
two
options
for
places
that
it
would
be-
and
maybe
this
is
a
quick
straw
poll
in
chat
the
two
options
right
now
that
we're
looking
at
our
Boston
around
the
OpenStack
summit
or
San
Francisco
jury
or
OFS
or
after
Korra
Lessa
before
after
process.
A
So
if
anybody
anybody
wants
to
vote
on
those
I
would
love
to
know,
hear
your
opinions,
whatever
my
numbers
suggest
that
everybody
would
have
to
travel
to
boston
or
nearly
everybody
would
have
to
travel
to
boston
as
fewer
people
would
have
to
travel
if
we
put
it
in
san
francisco,
so
my
gut
is
leaning
toward
san
francisco
right
now.
I.
B
Don't
let
me
put
in,
let
me
put
in
a
clip
I'll,
say
a
considered
opinion
on
us
and
I
think
I
have
a
little
bit
of
my
GF
board
hat
on
with
this
opinion,
which
is
this.
The
open
sky
project
and
community
has
been
super
welcoming
Reuben
Eddie.
Did
you
go
to
the
communities
conference
you're
going
to
see
all
kinds
of
discussion
sessions,
presentations
and
by
contrast,
at
a
doctor
conference
at
least
last
time,
around
c'mon
Yakir.
G
E
B
That's
great,
I
just
think
we
should
make
sure
we
have
a
the
guiding
principle
should
be.
We
should
have
it
at
a
place,
that's
friendly
to
the
community,
in
which
case
I
withdraw
my
preference
for
specifically
for
openstack
or
sm.
Let's
go
to
great,
ok,
so
Austin
Rob
Rob
mentioned
something
here:
economy,
not
cons,
obviously
been
friendly,
so
I
just
think
we
need
to
have
a
venue
at
a
place,
that's
open
and
welcoming
welcoming
to
the
contributed
community
overall.
So.
A
All
con
is
the
same
week:
Emma's
open
sex
on
it,
which
makes
it
really
complicated
for
people
to
buy
locate
and
a
lot
of
people
are
going
to
be
those
go,
eat.
Openstack
summit
ofs
is
the
weekend
between
somebody
correct
me.
If
I'm
wrong,
mange
une
right,
sorry
to
leave
it
the
week
to
cover
so
so
it's
Memorial
Day
weekend,
while
down
and
so
I,
don't
remember
off
the
top
of
my
head
with
the
dates
are
for
Austin
and
open
economic
slavery,
I.