►
From YouTube: Kubernetes Community Meeting 20170209
Description
We have PUBLIC and RECORDED weekly video meetings every Thursday at 10am US Pacific Time.
https://docs.google.com/document/d/1VQDIAB0OqiSjIHI8AWMvSdceWhnz56jNpZrLs6o7NJY
Demo OpenStack-Helm; Releases 1.6, 1.5, 1.4; SIGs AWS, PM ; Incubation process update.
B
Job
with
the
name-
that's
not
typical,
see
here
if
I
can
start
sharing,
I
haven't
shared
under
before
shoving.
It's.
B
Okay,
cool
sorry
wanted
me
to
talk
about
the
about
the
open
sex
up
stuff
now
or
later
you
when
we
start
with
the
demo
car.
B
See
here
so
when
the
at
AT&T,
we
started
a
project.
Well,
it's
something
that
has
been
kind
of
went
to
doing,
I've
been
wanting
to
do
for
a
while
and
finally
landed
it
in
the
right
word,
so
change
positions
and
what
the
EIC
team
or
the
agency
integrated
cloud
teams.
Now
in
Barcelona,
we
collaborated
with
s80
and
also
the
cola-cola
communities
within
OpenStack
to
talk
about
how
we
would
do
Cola
to
burn
at
ease
at
the
time
and
essentially
just
a
way
to
deploy.
B
B
So
when
we're
looking
at
how,
when
we
had
to
kind
of
think
of
things
a
little
bit
differently,
so
the
first
thing
is
that
we
don't
really
use
home
Global's
and
there's
no
real
parent
chart
for
everything
we
want
each
of
the
charts
to
deploy
independently.
So
that
way,
an
operator
can
keep
art
their
own
cloud.
The
way
they
need
to.
They
may
not
need
a
certain
component,
and
we
want
to
allow
this
them
to
be
able
to
interchange
those.
B
So
the
second
thing
is
that
obviously
I'll
manages
all
the
installation
and
upgrades
in
this
case
I.
So
you
have
one
common
source
of
truth.
All
of
the
OpenStack
services
use
multiple
rap
because
and
source
resource
definitions
by
default.
This
is
in
our
philosophies,
guide,
which
I'll
show
in
a
minute.
This
is
all
in.
This
is
all
open
source.
It's
all
available
to
anyone.
The
charts
include
node,
selectors
pot
affinity
and
anti
affinity
for
labeling
with
scheduling
instructions.
Also,
the
duds
documented
a
little
bit
better.
B
Now
I'll
show
you
guys
where
to
get
to
that
and
then
face
cream
information
in
there.
The
big
thing
is
that
we
want
to
use
the
Cooper
nettings
natives.
We
don't
want
to
drop
that,
so
we
don't
want
to
have
a
sorceress
on
our
images
at
all.
That's
really
important
to
us.
We
want
to
have
a
source
of
truth,
that's
super
Nettie's
and
helm,
so
that
also
allows
us
to
be
image
agnostic,
so
we're
runtime
them
interface.
B
You
know
runtime
like
hyper
or
racket
that
should
actually
work
in
this
model
and
you
would
be
afforded
some
of
the
benefits
of
making
those
choices,
but
that
does
present
some
problems.
So
now
we've
got
to
have
some
amounts
of
giving
big
maps
for
configuration
in
correctly
that
were
using
SC
and
also
for
our
startup
scripts,
which
we're
placing
into
a
bin
directory.
B
So,
basically,
relaying
beeps
and
big
maps
phrases
over
right
over
where
we
want
starting
the
container,
the
way
that
we
want
and
that
we've
also
got
some
plans
for
tracking
away
from
users
so
that
users
can
be
defined.
Security
groups
set
camp
rules,
a
table,
selinux
policies,
etc.
So,
if
you've
got
a
bright
future
but
a
lot
of
work
ahead
of
us,
we
also
want
to
have
consistent
chart
layouts
for
naming
convention
in
ships
or
services.
B
For
instance,
when
you
look
at
the
project,
I'll
paste
a
link
later
you'll
see
the
config
hice
can
se
directory
and
it's
some
big
pipe
and
bin
directory.
So
what
we're
essentially
saying
is
you
don't
have
to
bring
these
as
EML
files
or
anything
you
can
bring
them
as
startup
scripts?
That
also
means
that
when,
when
you
changing
these
these
files
or
these
config
masks,
we
want
to
hash
them
so
the
annotation,
and
when
you
make
a
change
to
the
convicts
map
we
want
to,
we
want
to
start
a
rolling
updates.
B
So
that's
one
of
the
philosophies
as
well
for
us
in
building
the
stack
I'll
show
you
the
void
deployed
back
here
in
a
moment,
but
we
have
the
concept
of
a
seacoast,
so
we
have
lots
of
data
centers
and
quite
large.
So
what
we're
doing
is
starting
a
sea
toast
with
Cooper
netting
I
benetti's,
all
in
one
we're
labeling
it
according
to
how
we
would
use
it
internally.
But
what
we're
deploying
is
a
bunch
of
mash
chart.
So
that's
our
bare
metal
provisioning.
B
We
are
writing
some
tools
that
will
that
will
use
a
single
yeah
mo
file
to
define
the
hardware
and
we're
essentially
we're
essentially
launching
a
daemon
set
for
mass
rack
and
region
and
then
we're
launching
a
demon
set
for
Steph.
So
we
have
persistent
bottom
store
on
the
backend
for
things
like
database,
because
we've
broken
up
the
mass
application
into
several
components,
trying
to
get
it
closer
to
of
micro
source
appointment,
then
what
we
can
do
is
we
can
essentially
just
install
hardware
and
that
hardware
then
gets
booted
up
and
Cooper
Nettie's
deployment.
B
Now
it's
it
automatically
installs
the
post,
the
node
receives
the
cluster
token.
It
joins
the
Cuban
any
cluster
if
it
sees
a
label
depending
on
what
role
we
want
that
to
have,
and
then
and
then
the
appropriate
OpenStack
services
just
automatically
drilling
a
cluster.
We're
going
to
have
demos
on
this
video's
demos
that
are
a
little
bit
more
concrete,
so
stay
tuned
for
that
and
how
Allah
Austin
M
updates
to
the
group,
but
just
just
kind
of
go
over
everything.
B
B
B
B
B
B
We've
got
a
lot
of
interest
in
this
so
far,
so
I
think
that
folks
from
back-end
are
using
this,
so
they've
been
posting
some
things
out
on
Twitter
three
medicines
and
things
on
that.
But
this
is
the
project.
The
project
I've
go
ahead
and
paste
basis
into
the
chat
for
anybody
who
might
be
interested
for
you
open
sexist
and
is
there
any
questions.
A
B
Actually,
that's
working
pretty
well
Alan
I
kinda!
Let
you
speak
in
little
bit
too,
but
I
said:
that's,
that's
the
piece
that
we're
using
to
do
it
an
automatic
rolling
up
afraid
of
the
service
or
an
updated
the
disservice,
we're
basically
passing
passing
a
decent,
Big,
Macs
or
a
hash
function.
Now,
when
you
expand
on
that
a
little
bit
yeah.
E
I
would
I
would
I
would
say
the
one
challenge
that
we
have
right
now,
whether
it
is
it
worked,
but
it's
a
shotgun
approach,
so
anytime
there's
a
change
to
any
configuration
file.
Of
course
you
do
a
rolling
update
and
there
could
probably
be
scenarios
in
the
future.
We
have
collections
of
things
that
should
not
result
in
a
rolling
update.
We're
fine
to
have
those
just
organically
appear.
Whenever
there's
a
new
container,
so
I
mean
we
could
we
could
work
that
with
continued
with
annotations
by
making
like
I've,
maybe
a
third
config
mass
definition.
E
B
And
to
be
to
be
really
clear
to
we've
been
working
on
this
for
about
two
ass
month,
two
months
really,
so
it's
a
relatively
new
projects
and
really
only
a
few
people
working
of
it.
So
you
guys
have
heaven
information
and,
and
obviously
I
mean
we
would
love
to
have
some
has
and
support
and
have
some
questions
out
there.
So,
yes,
please
feel
free
to
you
know
to
reach
it
up
to
the
community
we're
trying
to
find
a
home
for
it
right
now.
So
we're
talking
to
everybody
we
can
to
see.
Where
is
the
most?
B
What
is
the
most
appropriate
place
at
the
end
of
the
day?
We
definitely
see
this
as
a
a
cloud
native
project,
although
it's
difficult
with
OpenStack
is
everything
is
so
intertwined,
but
we
are
approaching
things
from
a
very
cloud
native
perspectives,
even
though
that
firm
is
getting
abused,
fight,
they're
sleeping
so
that
that's
how
that
that's,
how
the
that's,
how
the
project
is
constructed
to.
F
F
F
Time
it
is
something
something
that
I
I've
been
working
on
this
project
as
well,
quite
a
bit
and
something
that
I
think
is
quite
quite
special
about
as
well
as
well
as
being
able
to
scale
up.
It
also
works
incredibly
well
at
a
small
scale
which
is,
and
something
I've
seen
that
very
few
other
OpenStack
deployments
were
quite
well
as
so
for
small
organizations
consuming
cinder
and
Marilla
and
things
it
makes
it
much
more
viable
proposition,
which
is
what
block
on
the
on
board
and
exchange.
Debater
yeah.
B
We're
actually
just
doing
development
in
mini
coop,
so
you'll
see
the
back,
go
in
two
different
directions:
they
go
in
a
getting
started
guide,
which
assumes
that
you
can
use
Seth
and
OBS,
because
container
izing
OBS
is
a
little
bit
wonky,
but
the
other.
The
other
guide
is
also
for
mini
cube,
because
that's
just
easier,
it's
an
easier
cost
of
entry
for
developers
who
want
to
just
get
started
and
commit
sharps
so
that
options
available
to
I.
B
So
I
would
like
it
if
all
of
you
could
look
out
for
an
email
from
Sarah
regarding
this
is
in
the
Cooper
net
e
cig,
open
faq
of
mailing
list
or
group.
There
is
an
open
source
day,
that's
being
kind
of
constructed.
If
you
will
for
the
open
sex
summit,
the
deadline
to
sign
up
is
februari
20th
idiot
futures
at
first
glance
that
this
is
to
get
the
communities
to
really
start
engaging
with
one
another.
B
G
C
A
Thank
you
for
coming
Elfie.
Are
you
online
I
am.
H
Welcome
you're
going
to
tell
us
about
1.6
release
right
I
am
Gannon
kay
Lazar
out
today,
assigned
I'm,
of
course
substitute
so
I
have
some
notes
from
them.
They
wanted
me
first
and
foremost
to
you
emphasize
that
we
really
need
to
fill
the
roles
on
Dan's
relief
team.
This
is
super
critical
to
the
release
states,
saying
what
it
is
currently,
if
you
have
the
capacity
to
throw
one
is
yours,
please
do
so
you
can
volunteer
by
adding
yourself
to
the
dot,
that's
linked
in
the
agenda.
H
On
some
other
things
of
note,
we
really
need
all
of
the
things
to
have
the
github
users
to
get
sub
teams
and
the
Google's
group
mailing
list,
so
issues
can
be
rabid
effectively.
If
you
have
any
problems,
doing
that,
please
reach
out
to
contraband
we'd
be
more
than
happy
to
help,
and
then
we
have
some
ete
tests
that
you
can
stick
owners
and
we're
looking
for
volunteers
on
those.
A
C
Were
some
drop-dead
dates
that
were
set
last
week
in
the
notes
as
well?
We
really
really
want
to
make
sure
we
have
be
the
release
to
you
decided
and
I
believe
the
date
was
I'm
not
going
to
remember
now.
This
is
a
note,
but
I'm
not
finding
it
029
there
we
go,
so
we
want
to
make
sure
we
have
these
done
when
29
by
the
way
is
today.
C
So
we
need
to
do
this
or
we
really
jeopardize
the
release
date
so
volunteers
and
for
those
of
you
who
don't
actually
volunteer,
you
will
get
no
chocolate
for
me
a
and
B.
You
will
also
no
longer
get
to
call
me
and
tell
me
how
terrible
this
project
is
in
ways
a
through
z.
So
you
need
to
step
up
and
chocolate
and
carry
water
with
the
rest
of
us
and.
A
A
I
D
Manager
was
a
lot
of
fun
is
actually
a
lot
nicer.
1.53
is
fine.
For
this
friday,
all
cherry-picks
should
have
been
approved
by
wednesday.
I
went
through
yesterday
and
improvements
to
them
on
the
submit
q
is
currently
blocked.
Phil
wittrock
is
the
build
cops
this
week,
but
I
believe
he's
out
and
jesse
is
filling
in
for
him
and
she's
looking
at
that.
Hopefully,
that
gets
unblock
soon
and
the
pending
PRS
get
merged.
If
not
I'll
just
go
in
and
start
to
manually,
more
the
1.5
PRS
and
then
hopefully
we
will
cut
1.53
tomorrow.
D
J
I'm
here,
yes,
the
same
as
far
as
when
we're
going
to
do
thoroughly.
First,
the
next
14
and
I
think
everything
should
be
in,
and
it's
not
all
sued
the
famous
merged.
If
this
doesn't
get
on
bus,
but
I
did
just
merge
something
that
will
hopefully
unblock
it.
But
if
you
have
any
ideas
as
to,
if
that
doesn't
work,
you
feel
free
to
ping
me
or
whatever,
on
black
or
on
SPR.
A
Okay,
anyone
out
there
is
just
all
the
sense
that
you
maybe
they
wanted
this
release
thing,
but
I'm
scared
about
being
on
the
critical
path
for
such
an
important
project.
I've
got
your
back.
Sobs
got
your
back.
If
you
couldn't
use
git
and
github
and
build
stuff
would
go
and
ask
questions
to
people
when
you
don't
know
what
you're
doing,
then
you
can
probably
do
the
release
role.
We
won't
be
mad
at
you.
If
you
like,
take
a
while
to
do.
A
K
K
So
that's
going
to
spin
across
updating
the
cloud
providers,
documentation
that
I
know
that
Joe
and
sto
have
a
use
case
for
work,
we're
going
to
start
exploring
some
qim
stuffs,
and
so
we
probably
have
some
exploratory
work
around
that
networking
in
AWS
was
one
of
our
goals
for
the
year,
so
I
personally
see
that
go
through
and
I
know,
Cora
lessons
and
stuff
on
reference
architecture,
as
well
so
trying
to
announce,
eliminate
those
and
get
past.
Confucian
update
also
we're
working
on
helping
fpo
with
the
new
plan
for
a
new
solution.
K
They
have
in
mind,
which
is
a
AWS
CloudFormation,
+,
q2
ADM,
so
we're
in
the
works
of
kind
of
figuring
out.
How
that's
going
to
play
out.
Also,
there
is
an
update
on
cops.
We
have
released
earlier
this
week.
It's
our
first
release
since
early
December,
we're
tracking
out
a
lot
of
new
features.
So
users
have
a
lot
more
control
over
network
topology
network
configuration
in
general
and
we
are
currently
trying
to
get
our
goals
documented
into
the
community
repository
as
well.
K
A
Thank
you
Chris.
You
can
always
put
a
question
on
the
chat
for
zoom
and
we'll
come
back
to
it.
Let's
move
on
Igor.
Are
you
here
to
talk
about
sig
p.m.
yes.
L
I'm
here
so
as
initial
discussed
last
weekend
before
object,
definitive
p.m.
group
p.m.
working
group,
it
is
promoted
to
the
six
stages,
the
actual
link
to
the
community
committee,
really
spaced,
a
change
in
the
meeting
halls
and
I
precepting
to
relish
it.
So
if
you
have
any
questions
are
getting
this
feel
free
to
update
the
melon
place.
Don't
sweat!
I
would
like
us
to
save.
My
personal
crystal
are
non
calc
ooh
quadra,
when
this
is
music
and
I
would
like
to
submit
feedback
on
the
Internet.
L
For
in
general,
this
tremendous
BMC
curve
will
cover
timorous
as
we
can't
waste
and
working
group.
So
it
our
what
what
you
go
get
so
they
won't
find
anything
you.
The
matter
is
meant
for
establish
me.
The
seeker
was
to
fold
in
the
guidelines
for
their
eye
for
the
60,
with
they're,
going
to
have
the
regular
agents
probably
feel
weekly
we
are
going
to
have.
We
are
going
to
follow
the
general
guidelines
with
the
team's
research
possibilities
with
delicious
store
journal.
L
Community
SPM
group
will
focus
on
them
on
the
errors
that
are
covered,
but
they
are
not
covered
by
anything,
especially
in
the
era
that
is
not
required.
Writing
some
code,
but
in
more
management
errors
in
Arras
that
in
the
communication
with
the
customers
with
the
community
to
fill,
if
so,
where
we
can
make
businesses
and
the
products
much
power.
L
A
C
M
What
we
wanted
to
to
to
maybe
check
out
is
that
keep
on
Europe.
You
want
to
do
some
face
to
face
it
again
and
if
he
again
we'll
have
some
kind
of
overview
sheet
where
we
avoid
collisions
with
other
fixes,
wanted
to
exercise
or
if
anyone
has
planned
something.
So
we
can
just
coordinate
the
six
and
face-to-face
meeting
I.
Don't.
C
L
L
M
A
O
A
O
Going
to
put
a
link
into
the
chat
I,
so
my
discussion
topic
is
that
in
six
service
catalog,
we
have
have
made
a
new
incubator
riso,
as
maybe
you
have
probably
at
least
heard
up
for
this
community
service
catalog,
and
we
found
drinking
that
experience,
that
the
incubator
process
is
really
helpful
from
a
standpoint
of
like
defining
when
you
should
make
an
incubator
resale
all
right.
Those
of
you
that
aren't
familiar
with
the
incubator
processual
like
two
main
rules
that
are
defined,
there's
the
repost
sponsor,
which
basically
is
one
of
the
owners
of
kuva
Nettie's.
O
That,
like
is
where
the
incubator
project
derive
their
authority
to
assist
from,
and
then
there's
the
repo
champion
role,
which
is
basically
someone
who's,
maintainer
urban
at
ease
and
are
that
person's
responsibility
is
to
educate
the
folks
working,
an
incubator
on.
What's
the
community
norms
and
practices
are
which
found
super
awesome?
O
However,
none
of
the
community
norms
and
practices
are
defined,
at
least
that
document,
and
so
I
opened
an
issue
with
a
call
to
action
to
help
close
some
of
those
gaps,
specifically
ones
that
we
get
during
the
process
of
getting
on
purpose
of
getting
me,
someone
needs
to
meet.
Maybe
alright
sounds
good.
Okay
during
the
process
of
our
you
know,
first
couple
months
we're
working
on
this
incubator
Rico,
so
do
check
out
the
issue.
Let
me
see,
let
me
jog
my
memory
and
see
what
does
guests
were
at
high
level.
A
O
Yeah
I
think
that's
right.
I
think
I
would
add
to
that
a
couple
things
so
some
context
that
I
should
have
provided
and
probably
when
I
started
talking,
is
that
the
books
that
work
in
service
catalog
things
generally
when
they
started
were
people
that
were
very
new
to
the
community.
O
So
some
of
the
some
of
the
gaps
that
I
discussed
in
that
document
steam,
like
a
second
nature,
to
me
after
working
on
this
project
for
more
than
two
years,
not
at
all
obvious,
if
you
have
not
been
doing
so,
some
of
the
things
that
we
ran
into
a
for
example,
were
things
like
what
should
the
review
process
for
pull
requests
be
how
which
robots?
Should
you
always
use
my?
O
What
what's
the
policy,
if
any,
towards
check
me
and
generated
code,
and
some
were
kind
of
fundamentals
about
Cuba
Nettie's
like
when
and
how
you
can
build
an
API
server?
How
do
you
write
a
controller
I
know
there's
at
least
a
couple
of
those
things
that
that
I
just
named
that
our
technical
classes
we're
doing
some
work
on
and
some
like.
There
was
a
controller
document
that
I
managed
not
to
find
so
our
that's
on
me,
but
do
take
a
look.
I.
O
Think
that
that
one
reason
that
it's
going
to
be
really
important
to
include
some
of
these
gaps
and
at
least
provide
guidance
for
that
hey,
there's
a
process,
you
should
use
it
or
you
should
do
whatever
you
want
and
it's
totally
cool
I.
One
of
the
reasons
that
we
want
to
do.
That
is,
as
probably
almost
everybody
on
this
meeting
as
to
where
the
decor
is
getting
really
really
big
and
we're
moving
towards
an
extension
pattern
where
to
extend
through
benetti's.
O
Instead
of
modifying
the
core,
you
build
a
goober
Nettie's
like
thing
which
exists
outside
the
court
has
its
own
API
server
that
gets
aggregated
together
and
present
like
single
endpoints.
That
just
gives
you
all
the
api's
all
the
discovery
of
those
api's
together
as
one
big
thing,
and
if,
if
you
think
about
like
play
the
movie
out
6-12
months
for
that
pattern,
it,
at
least
in
my
brain,
it
kind
of
indicates
that
we're
going
to
have
a
lot
of
new
folks
that
are
joining
the
development
spiders
community.
O
In
the
incubator
repos,
rather
than
in
the
true
benetti's
repost
cell,
so
as
part
of
making
that
process
of
getting
offensive
speed
on
things
of
really
easy
and
make
people
feel
like
they
want
to
work
on
it,
not
that
they're
working
on
it,
because
I
have
to
that
their
employer
told
them
they
have
to.
We
should
just
try
to
provide
this
kind
of
information
up
front
and
spit
specific
expectations.
We're
norms
should
be
followed
or
where
discretion
is.
You
know
of
the
elbow
eos.
A
Thank
you
Paul.
You
got
shot
a
bunch
of
good
ideas
there
and
hope
everyone
takes
a
look
at
your
dock
and
maybe
we
can
unwrap
that
into
some
to
some
actionable
and
I
think
we
should
have
a
discussion
about
this
again.
If
you
are
somebody
who
is
just
started,
doing
incubator
other
than
service
catalog
and
you
have.
A
C
So
there's
another
point
in
there
that
Paul
made,
which
is
the
core
definition,
is
underway,
and
this
is
Anna
notices
section
of
our
notes.
The
core
definition
is
underway.
If
you
are
part
of
the
Cooper
nebach
Cooper
Nettie's
dev
mailing
list,
you
can
take
a
look
at
the
document
and
we
want
lots
and
lots
of
feedback
on
it.
So
Jason
Lauren,
but
I
would
say
it's.
A
In
a
state
where
you
have
to
sort
of
be
brave
and
have
some
context
have
even
read
the
docs
get
shot,
another
thing
I
would
ask
for
is
if
you
have
built
a
distribution
or
a
hosted
or
a
managed
version
of
Cooper
netting
or
built
your
own
on
prem
singh
and
try
it
didn't
understand
what
parts
you
have
implemented.
What
parts
you
didn't
have
to
implement!
Please
talk
about
your
experience
and
what
questions
you
had
because
ultimately
I
don't
want
the
Accord
off
to
be
like
a
dense,
prescriptive
document.
C
C
P
Q
Go
for
it
all
right,
just
a
quick
pasa
if
you're,
not
monitoring
your
city
is
conspire
great
I,
highly
highly
recommend
that
you
start
for
doing
that.
Large-Scale
clusters
and
some
of
our
validation
is.
We
have
found
that
there
is
a
jump
of
between
four
or
five
times
the
number
of
I
us
man
writes.
Well,
we
caused
it.
You
know
what
the
problem
is
that
we're
going
to
have
to
have
a
discussion
on
how
to
want
to
fix
things.
C
Okay,
we
can
go
through
a
couple
of
quick
notices,
then
there's
a
discount
code
in
the
community
meeting
topics,
agenda
document.
So
the
thing
we
all
use
the
reference
for
this
and
there
is
a
discount
code
for
ten
percent
off
of
a
corporate
ticket.
If
you
will
need
to
still
purchase
tickets
going
to
coop
con
I
mentioned
core
definition
under
way.
C
Apparently,
stick
cliff
drops
is
moving
to
a
two-week
cadence
for
a
bit.
There
is
a
meeting
today
so
come
by
and
there's
a
full
agenda,
and
then,
lastly,
we
have
a
firms
date
for
the
leadership
summit,
which
is
Friday,
62
and
it'll,
be
in
the
sand
in
san
jose
at
the
samsung
campus.
This
is
the
same
week
as
core
OFS
in
San
Francisco,
so
people
that
are
already
traveling
to
core
OSS
that
are
invited
will
hopefully
be
able
to
just
extend
their
trip
by
a
day.
C
The
invites
to
that
will
be
going
out
to
the
special
interest
group
leads
and
then
top
company
contributor
technical
leads
leads
from
the
large
company
contributors
that
are
or
the
actions
of
a
large
company
contributors.
The
large
contributors
who
happen
to
be
companies,
so
those
will
be
going
to
save
the
date,
will
be
going
out
in
the
next
week
to
be
in
my
list
for
the
leadership
summit.
This
is
a
different
idea
than
the
dev
summit.
C
This
is
more
about
the
leadership
of
the
project
in
a
broad
sense,
including
things
like
governance
project,
a
technical
leadership
and
vision,
and
the
focus
is
trying
to
make
sure
we
collect
the
right
people
together
and
have
a
smaller
group
even
than
we
deserve
some.
We
will
be
repeating
the
dev
summit
at
coupon
in
Austin.
I
will
have
more
information
about
that,
as
that
gets
nailed
down
to
a
venue
as
well.