►
From YouTube: Kubernetes SIG CLI 20180926
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
Okay,
good
morning
afternoon,
wherever
you
are
just
a
reminder
that
this
meeting
will
be
reported
and
so
anything
you
say
will
be
available
publicly
online,
be
mindful
of
that.
So
welcome
to
another
sixty
line
meeting
we
have
a
few
announcements
for
today.
Master
is
now
open
for
113
development
and
yeah.
Basically,
just
PRS
are
not
able
to
marks.
Once
again
we
have
a
test
grade,
update
from
BAM
hoping
he's
joined.
A
So,
let's
see,
if
he's
in
the
chat,
looks
like
he's
not
with
us
yet
so
I
guess
we
will
go
ahead
and
wait
to
see
if
and
joins
us
later
and
did
a
test
coverage.
Update.
I
did
take
a
look
at
the
test
grade
earlier
and
there
was
not
much
activity
so
as
of
this
morning,
at
least
we've
had
no
no
failures
were
occurring,
flakes
or
anything
like
that.
A
So
it
should
be
good
I
remind
her
that
the
test
grid
is
essentially
phone
call,
so
we
do
need
volunteers
to
sign
up
and
basically
you
know
share
their
time
to
watch
the
flakes
for
us
and
help
resolve
issues
with
the
tests.
So
if
you
essentially
click
on
the
Google
Doc
write
and
write
a
little
test
grade,
item
you'll
be
able
to
sign
up
and
volunteer
for
two
weeks,
and
we
encourage
you
to
do
so
if
you
have
not
already
in
the
past.
So
with
this
said,
we
can
move
on
to
our
topics.
For
today.
B
Awesome
hi
everyone
again.
My
name
is
Paris
Pittman
and
I
work
at
Google.
I
am
a
co-chair
of
the
special
interest
group
for
contributor,
experienced
quick
plug
that
we
have
a
survey
out
right
now
and
we
would
love
for
you
to
take
it.
I
will
paste
the
link
into
your
meeting
agenda
today
as
well,
but
this
will
shape
our
priority
list
for
the
upcoming
months
as
well
as
next
year,
and
you
all
are
contributors
to
cover
Nettie.
B
So
you
all
should
care
thanks
for
everyone's
work,
for
for
1:12
and
during
this
retrospective,
please
feel
free
to
write
your
thoughts
down
in
the
retro
doc,
especially
if
it's
an
improvement
or
a
delta.
Please
put
your
name
on
next
to
it,
so
that
we
can
call
on
you,
so
we
can
have
some
robust
conversation
if
you
would
like
to
say
something
anonymously,
ping
me
and
chat
or
slack,
or
something
and
I'll
find
it,
and
we
can
talk
about
it
from
you
know
without
a
name
attached.
That's
totally
fine!
On
that
same.
B
In
that
same
regard,
please
make
sure
that
when
you
are
talking
about
things
that
did
not
go
that
great,
that
it's
a
respectful
solution-oriented
nugget
of
communication
and
that
it's
not
blaming
an
individual
so
to
speak,
both
blaming
the
process
or
tooling
or
something
like
that
can
have
some
concrete
feedback.
Also,
please
meet
when
you're,
not
speaking.
It
sounds
like
everybody's
already,
muted
right
now,
which
is
great,
and
since
we
only
have
a
very
short
time
period
for
this
retro,
we
will
need
to
respect
the
time
boxes
as
well.
B
Okay
and
I'm
not
going
to
fit
I
did
not
follow
CLI
this.
This
release,
so
I'm
also
kind
of
excited
selfishly
to
hear
what
you
did
accomplish
and
what
did
go
into
the
release.
So
this
is
also
an
education
session
for
me
and
I
do
apologize
in
advance,
I,
don't
know
all
of
you
by
first
name.
So
I
may
ask
ahead
of
time
if
you
could
introduce
yourself
alright,
so
first
things,
first
follow-up
from
the
last
retro
for
111.
Did
you
do
what
you
said
you
would
do?
C
Think
we
didn't
do
a
planning
sheet
per
se,
although
there
is
so
whenever
we
are
doing
a
planning
like
we
will
be
doing
following
these.
This
retro
we
have
the
main
goals
for
the
sick
to
work
on
and
we
can
easily
identify
whether
the
goals
were
were
met,
but
I
think
yeah
that
the
tracking
sheet
is
a
good
point
that
I
completely
forgot
about
and
I'm
going
to
create
one
up
repair
at
this
meeting.
D
Just
a
quick
point,
so
I
just
noticed
that
in
the
coop
cuddle,
repo
and
github
there's
a
projects
tab
and
it
it
provides
some
really
decent
tools
to
keep
track
of
our
current
status,
especially
related
to
two
issues,
not
sure.
If
that
would
be
a
a
tool,
would
want
to
use
instead
of
just
like
a
planning
sheet.
Maybe
maybe
that
might
be
a
little
bit
more.
B
Also,
just
put
something
in
chat
to
contributor
experiences,
using
a
project
board
on
just
the
kubernetes
org,
which
lets
us
take
issues
in
from
other
repos,
and
things
like
that
which
has
helped
us
a
ton.
So
I
don't
know.
If
that's
that's
something
you
can
use
as
well
and
you
all
can
create
a
board
against
against
that.
Alright,
so
we've
already
gone
over
our
time
and
I
just
butted
in
and
it
took
some
of
that
so
I
apologize.
Does
anybody
have
any
last
thoughts
on
what
you
said
you
were
going
to
do
for
111.
B
A
C
B
A
We
didn't
that
from
which
prior
or
we
know
that
the
talent,
primarily
that
the
things
we
did
was
improvements
around
server-side
printing
for
keeping
forget.
We
had
deliverables
like
plugins.
We
essentially
had
I
believe
encrypt
testing
Shawn
started.
You
know
quickly
way
towards
eliminating
dependencies
that
the
control
has
on
the
the
core
repo,
which
is
a
big
ole.
We
have,
for
you,
know
the
near
future.
So
I'd
say
you
know.
In
summary,
it
was
basically
you
know,
since
the
infrastructure,
I
would
say,
interest
rates
of
cleanup
and
features
like
plugins.
That.
C
I'm
gonna
name
the
same
item
that
I
named
before,
which
is
the
dirt
of
fact,
ranks
and
cleaning
the
kikyo
bits
so
that
it
is
easier
and
we
are
able
to
faster
extract
the
cube,
CPL
repo
I.
Think
we,
if
we
put
enough
energy
more
of
us
and
spend
a
little
bit
more
time,
we
can
get
again
that
one
better.
A
D
Just
as
a
quick
plug
at
the
top
of
the
sheet
I
put
five
separate
under
current
status
and
put
five
separate
links
to
the
three
of
them
to
current
issues,
one
of
them
to
recently
merged
PRS
and
then
one
of
them
to
the
test
grid.
I
just
thought:
I'd.
Try
that
out
see
what
you
guys
think
to
address
this.
This
same
visibility
issue
to
keep
us
focused
and
to
have
a
very
easy
place
to
to
find
out
what
the
current
status
of
what
we're
working
on
right
now.
A
D
B
All
right,
sorry
about
that
I
was
typing,
just
wanted
to
dig
in
a
little
bit
to
the
refactoring
I
think.
Was
it
maisha,
except
that
especially
the
need
to
put
more
energy
and
time
into
this?
Was
there
other
items
or
did
problems
come
up?
That
would
take
away
time
and
energy
from
from
contributors
and
focusing
on
that
or
just
what
it
just
wasn't
a
priority.
A
Okay,
mainly,
we
do
I
guess
we
need
a
methodology
in
place
for
it,
so
a
more
well-defined
approach,
but
yes,
I
think
that
I
don't
think
there
was
anything
particularly
blocking
that
was
preventing
anyone
from
from
you
know,
exercising
their
their
contributions
inside.
Something
I've
noticed
lately
that
I
have
liked
is
social.
Knightley
has
been
even
there
going
through
issues
in
the
acute
control,
repo
and
essentially
tagging.
The
appropriate
person
you
know,
would
be
the
be
like
the
main.
A
E
A
E
A
Who
doesn't
agree
that
you
know
email
filters
definitely
do
not
help
when
you
get
tagged
on
a
bunch
of
issues,
so
I
guess
if
there
is
an
alternative
way
to
ping
people,
maybe
slack
we
could
have
like
I,
don't
know
a
way
to
to
maybe
have
the
boss
on
github
no
duping.
People
is
like
something
that
needs
immediate
attention,
or
things
like
that.
But
I
think
that
help
we.
E
A
Think
it
sounds
good,
and
particularly
because
so
at
least
but
not
yet,
for
example,
things
I've
done
in
the
past
is,
if
I
run
into
an
issue,
for
example
that
I
I
need
him
to
approve
fer
or
it's
an
issue
relevant
to
his
his
area
of
expertise.
Then
I,
I
I'd
put
it
on
his
radar,
so
to
speak
by
directly
linking
it
to
a
moans
black,
and
he
knows
that
it
fits
us
like
it.
It
should
probably
receive
a
needed
immediate
attention,
so
I
like
that
approach,.
D
E
Think
the
question
just
more
specifically
I,
think
a
question
I
had
was
like.
Thank
you.
First
of
all,
thank
you
also
for
doing
that.
I
think
we've
already
said
it's
appreciated,
but
it's
definitely
great
to
have
you
doing
that
and
just
like,
would
it
be
a
huge
burden
to
also
ask
you
to
when
you
paying
someone
on
the
issue
to
maybe
just
throw
it
in
slack,
it
seems
like
something
we
want
someone
to
look
out
for
sure.
Yeah.
D
Sure
yeah
I
think
that
you
know,
especially
if
it's
one
of
our
P
1
or
P
zero
items
that
that's
totally
appropriate.
You
know
I
think
if,
if
we
think
it's
maybe
a
p2
or
p3
issue,
you
know,
maybe
you
know
we
wouldn't
need
to
you
know
to
go
to
that,
go
to
those
lengths,
but
definitely
I,
think
on
the
p1
and
and
p0
issues
that
that's
more
than
appropriate.
It's.
B
To
feel
free
to
come
to
contributor
experience
to
because
we're
always
testing
and
doing
stuff
that
we
think
might
improve
from
an
automation
perspective,
your
experience
and
how
you,
how
you
assign
people
and
and
stuff
like
that
and
I
know
Christoph
and
others
in
in
the
crew,
would
be
super
stoked
to
talk
to
you
about
how
we
can
do
that.
So
we
can
take
that
conversation
offline,
all
right,
and
then
we
are
wrapping
this
up.
What
can
we
do
differently
for
1:13
and
beyond.
A
Yes,
I
think
a
few
of
the
items
recovered
just
now
regarding
you
know:
pinging
singing
people
on
slack
for
key
ones
and
zeroes,
for
example,
to
make
sure
immediate
attention
to
something
is
it's
something
I
think
would
be
useful,
perhaps
using
the
projects
tab
on
the
github
repo,
you
know
to
add
disability
to
our
current
features,
beckoning
to
deliver
and
who's
building,
and
all
that
so
yeah
I
think.
We
think
that
all
of
the
videos
we've
discussed
in
the
past
couple
moons
are
things
that'll.
A
B
E
Just
say
one
thing
to
consider
in
the
future
is
when
we
do
eventually
move
to
control
out
of
the
core
repo.
We
can
begin
to
decouple
it
from
the
main
release,
and
so
the
retrospectives
might
look
different
and
we
might
like
planning
on
a
three
month
cycle.
Maybe
look
different
in
me.
Maybe
that's
the
wrong
scope.
Maybe
it
should
be
six
months
or
maybe
it
should
be
one
month.
A
So
with
that,
we
are
ready
to
move
into
the
next
topic
and
thank
you
again,
Paris
for
leading
the
retrospective.
So
our
next
topic
in
our
agenda
today
is
113
planning.
We
lay
down
a
few
topics
that
you
know,
we
think
are
our
core
to
the
113
discussion.
However,
if
you
see
a
topic,
if
you
do
not
see
a
topic
that
you
feel
like
it's
important
to
the
113
timeline,
please
feel
free
to
and
it's
time
so
starting
out.
We
have
servers
referencing
improvements.
A
We
have
two
issues
linked
to
this,
mainly
around
improving
watches
for
queue,
control,
gifts,
sorting,
improvements
for
cubicle,
yet
as
well
as
promoting
server
by
printing
to
GA
and
deprecating.
The
old
printing
paths,
so
as
really
having
moved
to
server-side
printing
has
meant
that
we
essentially,
we
still
have
the
code,
for
you
know
the
old
printers
it's
you
know
within
the
cube
control
tree
and
it
would
be
nice
to
be
able
to
completely
remove
any
dependencies
only
because
if
there
are
any,
if
not
completely
reflected
from
the
control
tree
and
another.
F
A
We
wish
to
have
for
113
is
promoting
plugins
to
beta,
so
right
now,
plugins
are
still
in
alpha.
Despite
the
major
rewrite.
We
hope
that
by
moving
them
to
beta
it
will
give
users
more
confidence
on
the
system
and
essentially
promote
a
sense
of
reliability
with
the
current
API,
so
to
speak
so
that
authors
are
more
inclined
to
write
something
for
the
long
term,
rather
than
with
the
expectation
that
their
plugins
will
break
in
the
next
release.
A
A
These
are
mainly
differences
that
you
essentially
have
to
control
dependent
on
most
of
every
second
sort
of
server
side,
helper
API
helpers
and
things
like
that,
and
we
wish
to
even
remove
those
helpers
to
a
separate
repo
that
is
shared
by
the
query
launch
of
control
for
move
them
into
a
queue
control
tree
if
they're
humans.
Thank
you
control.
So
at
this
time,
if
anybody
would
like
to
particularly
go
into
detail
about
any
of
these
items,
you
have
the
floor.
D
D
F
C
Regards
to
gathers
to
get
commands
and
we
agree
that
we
are
going
to
copy
over
the
current
printers
as
is,
and
since
we,
if
we
are
lucky,
we
will
deprecated
the
printers,
which
basically
means
both
freeze
them,
assays
inside
of
the
CTO
subtree,
and
they
will
be
as
is
over
there,
so
that
will
break
the
dependency
on
the
printers.
Secondly,
it's
a
describer
and
I
need
to
input
clayton
about
his
idea
of
describer.
C
Is
we
need
to
externalize
the
describer,
because
they
are
very
tired
to
keep
CTL
itself?
I'm
not
convinced
too
much
that
they
should
move
to
the
server
side
as
with
against
so,
and
there
are
some
technical
difficulties
around
what
we
should
be
getting
for
to
describe
its.
I
think
that
at
the
current
state
it
will
probably
be
part
of
the
regular
cube
CTO,
so
we
will
just
externalize
him
and
move
over
today.
C
G
G
Haven't
made
any
progress
on
describe
on
the
server
side
and
the
last
couple
releases,
while
we
were
focused
on
get
so
I,
think
that
that's
a
pretty
pragmatic
approach
like
there's
a
there's,
a
bunch
of
there's
some
arguments
originally
around
one
format
might
look
like.
So
it's
definitely
something
that's
going
to
take
a
couple
of
releases
to
sort
out
and
I
wouldn't
want
that
to
block
you
know
the
the
short-term
stuff
we
could
definitely
than
113
and
there's
some
ideas,
but
there
it's
a
very
high-level,
vague
idea,
rather
than
something
concrete.
C
F
C
Side,
so
it's
a
lot
of
the
issues
by
default,
because
we
are
always
working
in
the
context
of
the
user,
whereas
doing
that
server
side
would
allow
additional
complexity
to
be
implemented
to
fulfill
the
same
requirements
so
that
that's
that
was
the.
That
was
a
discussion
that
we
had
with
with
who
on,
and
that
would
be
probably
the
simplest
approach,
and
you
know
yeah.
H
G
The
risk
of
a
security
exposure
that
way
is
a
lot
higher
and
described
like
we
originally
were
like
Oh
describers
are
going
to
be
really
complex
and
we're
gonna
go.
Do
this
cool
stuff
but
I
the
current
set
of
describers
are
doing
a
fairly
minimal
set
of
special
case
logic,
and
you
know
I
totally
exactly
Percy.
Doesn't
you
know
my
I
have
struggle
to
imagine
what
we
would
do
for
CR
DS?
That
would
make
them
useful,
and
so,
if
we
can't
come
up
with
something,
that's
gonna
make
see
our
DS
useful.
G
You
know
the
effort
to
come
up
with
a
generic
solution
on
the
server
side
is
kind
of
bound
up
by
that.
So
maybe
we
want
to
focus
by
saying
how
do
we
make
a
good
CRD
describer
and
what
are
like
the
minimum
things
that
get
us
a
better
user
experience
than
what
we
have
today
and
is
it
worth
it?
And
maybe
we
just
I
mean
I'm
some
advice.
If
we
can't
come
up
with
a
good
describers
thing,
that's
gonna
be
better
than
what
we
have
leaving
the
external
types
and
cute
control
for
another.
E
E
G
E
C
And
then,
additionally,
at
the
same
time,
there
is
the
plugin
where
we
haven't
sold
the
idea
of
being
able
to
override
the
built-in
commands.
If
we
tackle
the
problem
or
we
allow
people
to
extend
the
built-in
commands
such
as
that,
you
will
be
able,
through
a
plugin
mechanism,
to
describe
your
crv.
That
will
also
solve
the
problem
for
the
CRD,
and
this
is
a
an
issue
that
was
that
was
raised
several
times.
People
have
been
asking
about
that
one
we
will
have
to
solve
the
problem
sooner
or
later.
A
A
C
E
E
C
Where
we
would
be
saying
here
are
some
sample
commands
that
you
can
try
out,
but
we
are
we're
just
saying
that
they
will
change,
they
might
change
significantly.
These
are
all
alpha
you
can
play
with
it.
You
can,
you
know,
provide
feedback,
but
we're
not
taking
responsibility
for
you
to
run
them,
or
rather
people
would
prefer
to
go
through
a
alpha,
plugins,
promote
them
somehow
and
then,
eventually,
when
we
are
happy
with,
then
we
will
be
including
them
in
in
a
core
tube
studio.
I,
don't
know
how
powers
feel
about
it.
C
F
E
A
So
we
thought
being
said:
I
guess
if
there
are
no
more
specific
topics
of
discussion
around
you
know
splitting
up
to
control.
Does
anybody
else
have
any
other
topics
for
one
thirteen
planning
or
anybody?
Does
anybody
have
anything
else
to
focus
on
the
30
or
so
like
watches
or
sorting
improvements,
I
think
everything's,
pretty
self-explanatory
and
I
feel
like
we've
covered
describers
to
a
good
extent,
to
begin.
D
So,
just
as
a
quick
note,
we
we
have
umbrella
issues
in
our
repo
for
the
coop
Codel
independence
that
were
using
to
to
also
help
track.
The
the
PRS
I
was
wondering
if
there
were
umbrella
issues
or
if
somebody
wanted
to
create
umbrella
issues
for
these
other
efforts
for
the
so
the
plugins
I
guess
is
already
that's.
That's
already
been
put
in,
but.
D
F
A
A
C
The
way
I
went
ahead
this
morning,
I
went
ahead
this
morning
and
close
to
of
the
umbrella
issues
for
the
cube,
CTL
extraction
and
linked
to
just
one,
because
I
figure
out
that
it
would
be
easier
to
track
one
issue
rather
than
three
separate
that
are
basically
covering
the
same
topic.
I'm
not
sure.
Have
you
seen
my
comment
on
it
asking
for
you
to
provide
either
the
link
to
the
Google
Doc
or
maybe
provide
a
list
of
the
items
that
are
left
so
that
it
will
be
easier
for
us
to
track
the
progress?
E
D
A
D
D
As
we
know,
the
the
coop
con
is
the
11th
through
the
13th
of
December
in
Seattle.
If
you
haven't
signed
up,
please
do
what
one
of
the
things
that
I
was
hoping
to
bring
up
to
this.
This
crowd
was:
do
we
want
to
have
a
cig,
CLI,
there's
going
to
be
a
contributor
summit
the
day
before,
but
I
was
hoping
that
we
could
get
the
six
CLI
folks
together
in
order
to
to
dig
into
a
more
look
at
this,
the
strategy
and
the
vision
and
the
long-term
goals
that
we
have
a
lot
of.
D
What
we're
talking
about
in
these
in
these
meetings
is,
is
tactics
and
and-
and
it
feels
like-
maybe
we
don't
have
quite
the
you
know.
What
does
coop
cuddle
look
like
in
two
to
three
years,
and
maybe,
if
there's
a
critical
mass
of
us
at
Seattle,
that
we
can
have
a
meeting
and
try
to
at
least
come
to
some
agreement
on
some
of
the
long
term
goals
for
for
coop
cuddle
in
version
six
Eli
I'm,
not
sure?
Well,
what
does
everybody
think
about
that?
How
many
of
us
are
gonna
prank,
cuddle,
pucon,.
C
I'm
going
to
who
wants
coming,
you
said
you
will
be:
there
am
ng
I'm,
hoping
Phil
will
show
up
as
well,
because
he
promised
to
show
me
the
awesome,
create
dynamic,
create
I'm,
hoping
some
other
either.
Both
newcomers
and
additional
folks
will
be
able
to
show
up
as
well
so
I
think
the
the
necessary
majority
that
you've
mentioned
should
be
available
there
and
I'm
more
than
happy
to
have
that
discussion
that
you
just
mentioned.
C
D
I'll
talk
with
Paris
to
see
if
we
could
get
a
room
and
yeah
and
I'll
float
some
ideas
on
the
on
the
dates
I
plan
on
showing
up
on
a
Sunday
that
the
day
before
the
contributor
the
night
before
the
contributor
can't
get
together,
and
so
you
know
maybe
we
could
we
could
do
it
even
before
the
the
the
official
coop
con
starts.
Well,
we'll
figure
out
those
details
later
I
just
wanted
to
get
a
feel
for
what
you
guys
think
about.