►
From YouTube: Kubernetes Community Meeting 20161027
Description
We have PUBLIC and RECORDED weekly video meetings every Thursday at 10am US Pacific Time.
https://docs.google.com/document/d/1VQDIAB0OqiSjIHI8AWMvSdceWhnz56jNpZrLs6o7NJY
Demo Stackpoint.io; Sig Storage; K8s components to client-go; StatefulSet rename PR; Reference Docs; Releases; Kubecon.
A
So
hello
today
is
when
october
27th
couple
days
before
halloween,
and
this
is
the
weekly
cooper
Nettie's
community
meeting,
I'm
joe
betta
and
I
will
be
playing
the
part
of
sarah
today
since
she's
out
and
we
have
a
pretty
full
agenda.
It
looked
like
it
was
going
to
be
light,
but
then
some
other
folks
put
themselves
on
the
agenda.
So
that's
exciting
I
need
a
note
taker.
Do
we
have
any
volunteers?
A
A
Or
just
add
you
again,
this
dock
is
getting
really
long.
After
the
meeting,
I'm
gonna
break
out
some
of
the
old
stuff.
It's
like
a
hundred
pages
now,
which
is
kind
of
crazy.
Alright,
so
let's
go
ahead
and
get
started.
First
on
the
agenda
is
matt
from
stack
point
man.
Are
you
around
I'm
here,
yeah
Oh,
awesome
there
you
are
okay.
Great
matt
is
a
fellow
Seattle
light
when
I
moved.
He
took
my
boxes.
A
C
So
I
will
share
my
screen
here
so,
like
Joe
said
where
I'm
at
from
second
cloud,
where
a
product
and
services
company
focused
on
goober,
Nettie's
and
microscopes
to
lane,
we
also
actually
host
the
meetups
in
Seattle
New,
York,
City,
San,
Francisco
and
Berlin.
So
if
you
guys
are
interested
in
that
feel
free
to
reach
out
so
I
have
the
app
shared
out.
C
Can
everybody
see
my
my
screen
all
right,
I'll
take
that
is
yes
on,
so
what
we've
done
is
we've
taken
an
approach
to
managing
coronaries
clusters
and
playing
for
nice
clusters
with
a
exceptional
usability
and
exceptional
design
on
we
wanted
to
put
we,
we
wanted
to
follow
a
digital
ocean
like
experience,
so
that
users,
don't
really
have
to
think
about
too
many
things
to
get
something
up
and
running.
So
we've
enabled
a
few
stories
in
here
right
now,
so
the
first
story
that
we
have
is
easily
being
able
to
build
a
cluster.
C
We
have
it's
a
three
step
process.
Just
choose
your
cluster.
We
allow
you
to
drop
in
your
ssh
key.
If
you
have
one,
if
you
don't
have
one,
we
generate
one
for
you.
That
is,
you
know,
download
one
time
you
submit
your
go
to
the
next
step
in
this.
In
the
next
step.
We
allow
you
to
compose
the
infrastructure
right
now,
we're
supporting
one
da
word
up
for
we
by
default,
install
the
urban
Eddie's
dashboard
and
we
default
to
core
less
stable
as
well.
We've
also
done
it.
C
We've
also
done
a
native
implementation
of
a
GF
proxy
as
an
ingress
controller,
and
we
ship
that,
with
every
cluster,
that's
built
through
our
our
tooling
on
reactively,
how
we
actively
have
that
as
a
pull
request
in
the
project
right
now
on.
We
also
allow
you
to
add
solutions.
So
you
know,
we've
been
focused
mostly
on
solutions
are
a
little
bit
more
complex
to
do
things
like
fabric
eight
day
as
a
tectonic
on.
C
So
we
only
have
one
on
a
scalar
and
that
supports
all
the
public
cloud
providers
that
we
support
on
when
we
add
a
new
public
cloud
provider,
auto
scale
with
auto
scaling
for
that
node,
auto
scaling
for
that
public
alarm
providers
automatically
supported
through
the
tooling.
So
there's
no
need
for
us
to
update
or
add
create
another
autoscaler
strictly
for
that
a
cloud
provider
on
we
do
support
features
such
as
trusted
platform
module
for
bare
metal
providers,
so
that
we
can
tell
a
distributed
trusted
computing
story
right
now.
C
C
We
also
give
you
a
management
kind
of
like
a
list
view
of
the
clusters.
We
show
you
the
provider,
we
show
you
the
communities
version
arm.
We
pre
create
the
coop
config
file
for
you,
so
you
simply
just
have
to
download
that
and
you
can
begin
using
a
good
cuddle
Spanish,
your
cluster
hold
on
removing
this
dialog
out
of
the
way.
If
you
want
you
can
dive
into
a
cluster
and
actually
start
to
see
details.
C
This
is
a
cluster
that
we
actually
so
today
we
also
partnered
with
fabric
a
team
over
at
Red
Hat
to
deliver
a
production
story
for
fabric
eight
on
amazon
web
services.
What
that
means
is
we
we
actually
create
all
the
volumes
we
set
everything
up.
We
set
up
a
cheap
proxy
and
we,
you
know
within
15
minutes.
You
have
a
developer
platform
that
you
could
begin
to
use
and
deliver
whatever
you,
whatever
app
you
want
to
deliver
into
that
platform
on.
C
C
What's
going
on
in
the
build
process,
what's
going
on
in
the
solution,
install
process
and
a
variety
of
things,
let
me
try
to
find
another
cluster
here
hold
on,
so
we
also
allow
you
to
navigate
on
this
left
side
to
look
around
and
do
things
on
like
here.
I've
got
an
error
in
the
event
log,
and
this
was
something
where
I
was
doing
a.
This
is
another
tool
that
we
just
added,
which
is
the
ability
to
onboard
using
docker
compose.
C
So
what
we've
done
is
we
made
it
where
you
just
drop
into
another
three-step
workflow,
where
you're
able
to
simply
drag
and
drop
your
daugher
compose
file
into
the
interface
name?
Well,
I'll
just
go
with
my
neck
green.
My
green
cake
solution
is
my
name
converted
into
pernetti's,
and
so
we
allow
you
to
do
and
then
we're
going
to.
Let
you
so
we
catch
an
error.
We
allow
you
to
go
in.
You
know
just
the
compose
file
before
you
convert
it,
and
then
you
just
simply
say:
let
me
review
the
manifest.
C
We
automatically
convert
it
into
a
manifest
and
then
allow
you
to
simply
jump
into
a
cluster
build
flow.
And
then,
when
you're
going
through
here,
you're
able
to
go
here
and
begin
to
look
and
find
your
solution
and
choose
it
and
then
add
it
to
your
cluster
and
build
and
basically
initiate
the
build.
C
So
that
is
pretty
much
the
app
actually
at
the
moment
we
also
have
an
API
that
allows
you
to
manage
multiple
clusters
through
us,
so
our
own
application
is
sitting
on
top
of
a
RESTful
API.
That's
all
syndicated
on.
We
allow
you
to
generate
a
API
token.
If
you
want
one
and
then
be
able
to
interact
with
us
and
your
question,
so
there
yeah
any
questions.
C
We
we
work
with
Sebastian
and
talk
to
him
about
that,
and
so
we
leverage
that
behind
the
scenes
to
do
things
on
this
will
be
harder,
and
so
we
automatically
update
update
that
whenever
there's
an
update
to
it
so
that
and
then
we're
also
starting
to
collect
metrics
on
things
are
problematic
with
it
and
we're
going
to
feed
that
back
to
stashed
at
some
point
on,
we
also
the
other
thing
we
do.
We
also
publish
our
build
numbers
publicly
on
our
app
site.
C
We've
built
over
a
look,
so
we
have
over
1100
clusters
that
have
been
built
through
this
application.
Thus
far,
we
have
over
1,000
organizations
that
are
using
this
tool
right
now
to
build,
deploying
to
whatever
whatever
those
organizations
do
with
kerber
Nettie's
on,
and
so
that's
helping
us
drive
a
lot
of
feedback
into
the
application.
We
also
monitor
all
the
logs
and,
as
we
discover
weird
things
with
providers,
we
patched
that
in
our
code
base
on
so
that
we
can
give
a
consistent,
consistent,
build
experience
to
everyone
on.
We
have
a
pretty
high
build
right.
C
C
Install
failure
on
down
the
road
we
do
have
plans
to
support
helm,
but
we
feel
helm
is
a
much
lower
hanging
fruit
and
we
wanted
to
focus
on
things
like
fabricate,
so
we
can
actually,
we
actually
have
customers
who
are
trying
to
do
like
developer
platform,
Peppa
stories
in
so
we
needed
to
do
that
type
of
stuff.
Before
we
did,
you
know
smaller
things,
hey
man,
how
do
you
handle
upgrades
on
clusters?
Yeah?
C
That's
a
problem
still,
so
we're
we're
following
the
life
cycle
right
now
to
see
what
the
directions
going
to
be,
but
we
don't-
we
haven't
approached
it
yet.
We
have
taken
a
look
at
the
core
OS
self-hosted
model,
and
that
is
by
the
way
that
we
may
be
thinking
of
going.
But
as
yeah
we
haven't
made
decision,
hey.
E
F
Yeah,
I
think
that's
what
matt
was
referred
to
from
the
sig
right,
that
the
costa
life
cycle
sick
should
come
up
with
that
tonight.
I
guess
the
other
question
is:
if
you
can't
do
upgrades,
do
you
have
a
sort
of
easy
mechanism
to
sort
of
clone
a
cluster
in
reversion
like
take
a
single
solution
and
deploy
those
in
a
new
cluster
at
you
know
145,
and
that
comes
out
yeah.
C
That's
actually
on
the
road
map
on,
but
we
don't
have
that
in
the
app
at
the
moment,
we're
highly
it.
So
we
ran
on
funded
company
by
the
way
and
so
we're
highly
area
of
on
the
app.
So
we
release
features
every
two
to
three
weeks
on,
and
so
we
do
have
gone
to
remember.
We
don't
have
that
actually
in
the
product
at
the
moment
and
you'll
notice,
I'm
actually
demoing
from
staging
the
main
reason
we
do,
that
is
to
not
include
our
build
numbers
yeah.
C
A
E
H
That's
checked
in
to
include
these
end-to-end
tests,
I
think
a
lot
of
time
and
open
source
and
really
any
software
project.
If
we're
all
honest
here,
testing
and
documentation
become
the
last
thing
that
everyone
wants
to
do
so.
I
think
storage
faces
some
unique
challenges
in
that
plugins
that
we
write
necessarily
aren't
and
cannot
be
directly
put
into
the
CI,
so
kind
of
a
dummy
is
in
its
place
to
provide
testing
and
we
don't
have
a
way
to
automate.
You
know
the
durability
of
those.
H
So
with
the
end
and
tests
we're
hoping
til,
they
simulate
what
we
can
to
find
any
bugs
make
it
more
stable
push
it
out
to
our
customers,
make
it
more
secure.
So
we
have
I
think
we're
well
ahead
of
the
mark,
where
we
thought
we
would
be
working
on
these
tests
and
we've
had
really
great
interactions
within
the
community
to
help
contribute
to
the
effort
sod.
Did
you
want
to
add
anything
to
that?
H
I
I
I
think
again,
these
are
just
really
good
four
steps,
and
so
you
know
we
are
trying
to
start
an
ongoing
sort
of
effort
within
the
storage,
sig,
community
and
I.
Think
the
structure
is
there
now
for
new
people
to
come
in.
We
just
had
someone
join
like
a
week
ago
and
because
of
the
work
that
darn
inside
has
put
together,
he
looked
at
a
spreadsheet.
I
He
said:
oh
cool,
here's
work
that
no
one's
doing
and
then
started
coding,
and
so
you
know
it's
not
just
writing
these
first,
it's
actually
having
a
vision
for
what
we
want
to
do
going
forward
and
on
top
of
that,
making
things
readily
available
for
people
to
contribute,
which
is
really
impressive,
that
they
did
it
so
quickly.
Thanks.
A
A
Sounds
great
I
think
you
know
it
might
be
and
I'm
sure
you
guys
have
are
aware
of
this,
but
there
there
is
provisions
as
part
of
the
testing
sig
for
federated
end
and
casting
and
reporting
of
this
stuff.
So
you
know
it
might
be
worthwhile
thinking
about
like
how
can
we
get?
You
know,
vendors
who
have
pretty
esoteric
equipment
to
actually
dedicate
some
of
that
equipment
for
continuous
testing
I.
B
H
B
A
A
J
Hello,
everyone,
so
we
are
going
to
migrate
all
the
components,
including
the
our
controllers,
the
schedulers
and
cube
literally
using
can
go.
If
you
can
go
it's
it's
a
separate
repository
now
and
that's
what
that's
our
blaster
go.
Client,
the
major
reason
we
do
we
are
doing
this
is
because
first
we
want
to
dogfooding
Clank.
J
Oh,
we
want
to
use
and
we
want
to
let
you
could
not
his
core
components
to
use
the
same
library
we
provide
to
our
give
to
other
developers
and
does,
and
also
it's
more
conceptually
correct,
because
client
go
doesn't
do
any
conversion
devoting
and
other
AP
AP
I
machinery.
All
those
things
will
be
done
by
server
and
components
like
controllers
and
schedulers.
Acuity
should
be
done,
clients,
so
that's
also
conceptually
correct,
but
this
change
will
be
huge.
It
will
touch
almost
every
file
in
the
our
repository.
J
J
A
How
invasive
is
that
going
to
be?
Is
that
going
to
be
mostly
rewriting
the
animals
and
a
few
names
here
and
there,
or
is
it
an
interpretive
major
API
change,
I'm?
Sorry,
how
difficult
is
it
going
to
be
to
do
that
conversion
for
people
to
update
their
their
in-flight
PRS?
Is
it?
Is
it
just
updating
some
imports
and
a
few
names
here
and
there
it.
J
A
K
D
I
D
And
we
use
that
machinery
for
cube
config.
We
use
it
for
the
scheduler
config,
basically
we're
trying
to
use
that
machinery
for
anywhere
where
we
serialize
it
in
east
of
any
kind
in
the
system.
So
we
just
have
one
approach
to
dealing
with
versioning
and
all
that
you
know
the
one
place
in
Cuba
it
will
not.
Change
is
the
static
pots,
because
those
actually
need
to
run
the
defaulting,
for
example,
on
the
key
blitz
for
the
API
server.
But
again
those
are
sort
of
all
non
client
scenarios.
A
B
I
On
polynesia
sure
that
we,
you
know,
I'm,
really
excited
that
the
storage
stuff
is
going
out
and
people
are
taking
a
look
at
like
this
testing
sort
of
system,
but
I
also
want
to
make
sure
that
we
don't
drown
it
with
cool
intention.
You
know
it
is
making
progress,
and
so
you
know
realize
it's
not
polished
for
external
and
everyone
quite
to
jump
in
it's
something.
Now
that
we're
using
is
a
way
for
everyone
to
coordinate
and
make
progress
internally
in
the
sig.
H
A
A
E
Think
the
results
already
been
announced,
I'm
just
discussing
that
is
we're
beyond
the
point
of
any
discussion
about
that
as
if,
in
my
view,
I'm
just
giving
people
heads
up
that
we
have
engineers
several
under
is
actively
working
on
the
rename
PRS
to
be
able
to
go
in
for
the
code
freeze.
And
then
it's
going
to
be
split
across
several.
So
the
docs
will
happen
after
the
code
freeze.
E
But
people
are
eagerly
actively
working
towards
getting
that
all
done
for
1.5
and
I
just
wanted
to
give
everyone
a
heads
up
if
they're
that
they
might
want
to
start
thinking
about
renaming
their
config
and
such
to
prepare
themselves
for
1.5.
If
they're
eager
for
that
and
I
can
answer
questions
not
about
the
name
but
about
the
mechanics
of
the
rename.
That
is
happening.
Didn't.
E
A
L
But
I'll
just
give
you
a
quick
walkthrough
of
what
I'm
working
on
and
the
goals
the
goals
are
to
provide
an
experience
more
like
the
stripe
documentation
where
you
can
provide
more
discoverability
and
examples.
Next
to
the
reference
for
the
fields
and
I've
broken
up
the
documentation
up
into
a
couple
of
categories
of
resources,
so
workloads
discovery
and
config.
And
as
you
dive
into
these,
you
can
see
that
the
different
operations
are
called
out
and
you
can
as
you
scroll
through
it.
It
will
update
the
left-hand
nap.
A
A
Right
so
there's
a
bunch
of
comments
in
the
end,
a
thing
here,
a
great
work,
nice
I,
think
everybody's
very
bullish
on
it.
I
think
you
know
it's
gonna
be
one
of
those
things
where
you
know
well,
you
know
I
have
some
feedback.
I
can
send
that
to
you
separately,
but
it's
awesome.
I
think
it's
a
it's.
An
awesome.
Awesome
start
here
now.
Is
this
being
generated
from
scraping
types
code
directly
or
is
it
being
generated
the
from
the
schema
files,
the.
A
L
A
Me
feedback
all
right.
Yes,
please
send
feedback
to
Phillip,
it
looks
great
and
then
I,
you
know
eventually
I
assume
this
will
get
hosted
on
on
goober
net
ETO
and
become
the
official
documentation.
Is
that
the
goal?
Yes,
that's
the
goal,
all
right,
very,
very
cool,
all
right,
then
next
step
is
we're
going
to
be
talking
about
releases,
and
so
the
the
first
point
there
is
is
Jess
on
on
1.4,
just
Arirang
yeah.
A
G
A
I
If
this
is
just
sort
of
a
for
the
whole
group
right
now
in
the
storage
team,
we're
trying
to
figure
out
how
far
back
to
backport
some
fixes,
we
found
some
fixes
for
some
volumes
and
some
other
issues
that
revolve
around
I
think
some
AWS
support
and
some
other
things
involving.
If
the
couplet
we
starts.
We
know
we're
going
to
put
them
alone,
for
their
all,
of
course,
set
up
for
15
or
there
enough
13
usage
out
there
that
it's
worth
us
back,
porting
them
two
one,
three
two
one
three
also
apologies
just
curious.
G
A
I
I
was
kind
of
erring
on
the
side
of
going
to
130
just
sort
of
curious.
What
the
communities,
both
feelings
and
policy
is,
we
can't
put
them
in
12
because
it
involves
a
rewrite
that
happened
in
13
and
thank
you
very
much
for
a
great
answer
and
Aaron
yep.
Oh
it's
air
and,
of
course,
thanks
guys,
of
course,
yeah
yeah.
A
F
I,
don't
I,
don't
think,
there's
consensus
on
exactly
how
far
back
to
patches
you
get
the
the
CVE
they
got
patched
in
143
and
they
get
back
port
back,
boarded
213
and
didn't
get
back
forward
into
12
until
someone
from
the
community
reached
out
to
Jordan
I
getting
myself
directly
so
that
that
was
not
sort
of
in
proactively
backported.
All
the
way
to
12.
A
All
right
great,
so
Aaron
is
there
a
do.
You
know
if
that
support
policy
is
documented,
someplace
I
was
asking
he's.
Ok,
he's
gonna,
take
a
look
so
so
we'll
see
if
that's
actually
written
down
anymore
or
if
it's
just
word
of
mouth.
Hopefully
it's
right
back,
ok,
so
1.5
a
sodding
and
kill
1.5
reminder
that
both
please
is
less
than
a
week
and
a
half
away.
You're
your
mic
is
super
annoying
I,
don't
know.
What's
going
on
there.
N
A
N
A
Right,
great,
okay,
so
so
Eric's
asking
just
sort
of
a
point
of
order
here.
Why
is
sods
name,
six
storage
and
my
name
is:
is
six
scale
it's
because
we
only
have
a
certain
number
of
official
zoom
accounts
to
go
around
and
it's
really
easy
to
leave
yourself
logged
in
when
you
join
this
meeting
so
and
then,
because
I'm
recording
I
need
to
have
a
real
account
to
in
case
anybody
had
a
question
on
that:
okay,
all
right
and
then
Aaron
was
able
to
dig
up
the
the
supporting
and
releases
and
components
q
stuff.
A
A
A
So
the
first
thing
there
is
that
juke
con
is
sold
out
and
if
these
things
are
always
kind
of
tough
to
figure
out
how
big
a
venue
to
actually
buy
to
actually
schedule
and
how
many
people
are
going
to
come
so
actually
selling
out
it's
exciting.
It
makes
you
feel,
like
you're
part
of
an
exclusive
club,
but
it's
not
necessarily
the
best
thing
for
the
community,
and
so,
if
you
were
or
somebody
else
that
you
know
wanted
to
come
but
weren't
able
to
because
it
got
sold
out.
A
Please
have
them
register
for
the
wait
list,
because
that
actually
ends
up
being
a
strong
signal
in
terms
of
how
big
a
venue
to
book
for
next
year
and
but
it
is
exciting
that
that
it
sold
out
and
there's
a
bunch
of
exciting
stuff
happening
around
cube
con
the
monday
before
their
whole
bunch
of
cigs
are
hosting
face-to-face
meetings
and
I
know.
There's
been
a
bunch
of
discussions
around
that.
So
talk
to
your
local
sig
about
what
their
plans
are.
A
This
stuff
has
been
a
little
bit
ad-hoc,
so
I,
don't
think,
there's
any
super
official
schedule
around
that.
However,
the
day
after
cube
con
is
the
dev
summit,
which
is
pretty
official,
and
so
there
is
a
schedule
now
posted
there
and
Cameron's
been
doing
a
lot
of
work
to
pull
that
together.
I
mean:
are
you
online?
Do
you
want
to
give
a
quick
summary
yeah.
O
I'm
online,
I
basically
I'm
here
just
to
talk
about
this.
In
case
people
have
questions,
but
yes,
we
posted
the
schedule.
It's
in
the
github
communities,
community
/
community
link.
We've
got
three
separate
tracks.
It's
gonna
be
12,
sort
of
kind
of
open
round
table
type
discussions.
There
are
facilitators,
but
not
presenters.
For
those
and
all
the
topics
came
from
the
community,
everybody
voted
in
a
like
a
conversating
pole
and
and
that's
directly
what
we
chose.
O
A
B
B
B
M
B
B
Think
that's
the
idea.
Okay!
Consider
it
like
I
mean
my
view
on
it.
Is
there
going
to
be
a
lot
of
people
wandering
around
not
even
knowing
what
the
cigs
are,
for
example,
and
so
the
idea
of
having
the
tape
having
the
cigs
sit?
You
know
grabbin
man
tables
so
that
people
can
wander
by
and,
like
oh
I
didn't
know,
there
was
a
cig
for
this
and
chat
with
the
people
that
are
involved
in
the
community
seems
like
a
good
outreach
thing
to
do
for
the
community
folks
attending
the
conference.
B
A
It
sounds
like
a
good
idea,
so
we
should
get
some
details
on
that
and
maybe
send
an
update
on
to,
or
once
you
figure
out
who
can
can
you
make
sure
that
an
update
gets
sent
out
and
Coburn
any
step
so
that
people
know
just
sign
up
for
that
and
maybe
we
can
make
that
get
some
more
details
and
make
that
happen
alright.
So
that
is
the
end
of
the
agenda.
We,
we
ended
up
going
through
it
pretty
quickly
up
wit,
there's
another
question
in
the
chat
here.
A
A
So
this
is
a
I
mean
what
you
know.
What
are
we
looking
for
reasons
bottlenecks
causing
that,
and
you
know
how
do
we?
How
do
we
actually
get
folks?
Looking
at
those
PRS
I
I
mean
this
has
been
an
ongoing
thing
that
we've
been
talking
about
for
quite
a
while
in
terms
of
what
is
the
review
capacity?
How
do
we
assign
stuff
and
incentivize
people
to
do
reviews
and
essentially
start
growing
the
community
to
be
able
to
handle
with
the
influx
of
new
people,
contributing
I?
A
D
I
have
a
couple
of
comments
in
terms
of
the
longer
term
solution.
Yes,
we're
actively
working
on
what
we
call
the
owners
efforts
of
actually
looking
at
growing
the
reviewer
pool,
so
we're
doing
a
lot
of
mining
of
github
data
at
a
future
meeting.
May
we
can
get
Garrett
to
come
on
and
talk
about
that,
but
if
you
are
interested
in
helping
with
the
effort
for
figuring
out
how
to
grow
the
community,
the
contributor
experience
working
group
is
focused
on
this
exact
topic.
D
D
D
So
some
yars
that
were
lgtm,
backed
up
in
the
queue
in
terms
of
things
that
are
not
getting
reviewed,
yeah
I,
don't
know
what
we
can
do
about
that
and
basically,
if
you
didn't,
have
a
partner
already
lined
up
ahead
of
time
to
review
your
PRS
and
in
general,
you
can't
assume
this
stuff
is
going
to
go
in.
If
everybody
the
only
writes
features-
and
nobody
is
lined
up
to
review-
then
yeah
you're
kind,
I.
A
D
D
A
P
A
Alright
well,
are
there
any
other
comments
or
notices
or
items
to
bring
up
before
we
can
adjourn
15
minutes
early
and
we
return
15
minutes
to
258
people
who
are
online
right
now,
nope,
you
know
calculate
that's
a
lot
of
man-hours,
all
right
or
person,
hours,
I
apologize
all
right!
Thank
you!
Everybody!