►
From YouTube: Kubernetes Sig Docs 20181120
Description
Meeting notes: https://docs.google.com/document/d/1Ds87eRiNZeXwRBEbFr6Z7ukjbTow5RQcNZLaSvWWQsE/
The Kubernetes special interest group for documentation (SIG Docs) meets weekly to discuss improving Kubernetes documentation. This video is the meeting for 20 November 2018.
https://github.com/kubernetes/website
A
C
C
D
It's
not
a
new
title,
really
new
rankings.
Superb.
D
E
Sure,
hey
folks,
just
as
a
reminder:
I'm
the
docs
lead
for
the
113
release.
Apologies
for
the
making
the
last
few
meetings,
I've
been
traveling,
so
general
updates.
So
yesterday
we
had
deadlines
for
PRS
to
be
ready
for
review.
We
had
a
bit
of
a
crunch
yesterday
to
try
and
get
everything
ready
for
review,
and
now
we
have
all
of
them.
Already,
apart
from
three,
we've
got
a
pretty
tight
timeline,
so
we
need
everything
to
be
ready
by
November.
27Th
is
the
target
and
the
release
is
gonna,
be
on
December
3rd.
E
So
with
the
holidays
and
everything
I
think
the
review
timeline
is
pretty
tight,
so
I
feel
a
spreadsheet
where
I'm
tracking
all
of
the
issues
there's
a
link
in
the
minutes
there
will
be
really
really
helpful
is
if
anybody
is
able
to
take
a
look
at
issues
that
they
you
know
would
like
to
be
approvers
on
or
would
would
be
keen
to
take
a
look
at
and
kind
of
add
themselves
if
possible.
That
would
be
helpful
to
make
sure
that
we
get
everything
reviewed
as
quickly
as
possible
over
the
next
seven
days.
A
E
So
those
two
are
currently
like
they're
empty
commits,
because
just
so
these
are
two
of
the
PRS
which
we
haven't.
There
hasn't
been
any
work
on
yet,
unfortunately,
so
we're
reaching
out
to
the
PR
owner
to
get
that
done
and
then
hopefully,
I
think
once
the
commit
comes
in,
the
reviewers
should
be
assigned
automatically.
E
D
So
so
we
had
a
bit
of
a
situation
committing
dachshund,
because
we
didn't
get
well
for
a
variety
of
reasons.
They
weren't
having
difficulty
reviewing,
especially
the
one
complicated
PR
that
has
includes
four
generated
files.
That's
one
that
we
have
a.
We
have
a
process
solution
too,
but
we've
come
up
with
it
only
in
the
last
release,
or
so
so
it's
still
a
bit
of
a
hiccup
and
that
one
I
think
should
be
back
on
track.
D
E
Ok,
yeah
I
think
we've
got
lots
of
people,
lots
of
awesome
people
working
on
that,
so
I'm
I'm
sure
it's
gonna
be
fine
and
thank
you
very
much
for
kind
of
offering
to
work
over
the
holidays
to
to
get
that
done.
I
know
it's
an
awkward
timing
for
sure
I'm.
E
D
So
I
I
wanted
to
call
out
I
wanted
to
thank
chiming
for
taking
on
PR
wrangling
this
week
because
he
switched
because
of
Thanksgiving.
So
since
I
wasn't
originally
selling
tea
this
week,
I'm
extra
appreciative
and
then
I
think
that
the
rest
of
the
agenda
items
that
are
currently
in
the
dock
are
mine,
except
for
the
one
against
Andrew,
who
doesn't
seem
to
be
here
so
without
further
ado.
D
I
guess
maybe,
if
I
share
it
will
help
for
folks
to
understand.
What's
going
on
here,
yeah
so
many
times,
I
believe.
D
Okay,
so
let's
go
take
a
look
at
the
pull
requests
in
question
and.
D
Things
are
breaking
links
in
metal
docks,
but
the
problem
is
these
are
generated,
and
this
particular
poll
request
does
what
we
keep
telling
people
not
to
do,
which
is
manually
fix
on
generated
and
chiming
did
a
nice
job
of
digging
in
and
offering
several
different
options.
So
we
can
fix
the
upstream
codebase.
We
can
fix
the
code
generator
or
week,
because
the
cube
shuttle
docks
are
generated
with
a
different
script
kind
of
like
the
one
that
generates
the
API
rift
ox
and
that's
why
we
get
these
docks.
F
D
Let's
take
a
look,
the
only
thing
I
want
to
do
is:
go
back.
I
raised
a
couple
of
other
issues,
just
in
case
we
get
into
the
weeds
with
this
one
which
I
will
try
not
to
but
I
think
this
one
probably
trumps,
the
other
two
in
terms
of
being
worth
spending
some
time
on
the
agenda,
I
will
make
that
judgment
call
and
we
can
defer
the
other
two
if
we
need
to
so.
Let's
I'm,
just
gonna
have
to
navigate
I.
Don't
have
this
stuff.
D
Not
the
world's
loveliest
navigation,
but
this
is
what
the
current
generated
docks
for
kids
going
to
look
like,
and
just
so
you
have
for
comparison.
Oh
no
cube!
Atm
is
not
a
good
option.
Let's
just
take
a
look
at
the
API
server.
Oh,
this
is
what
the
API
server
ref
docks.
Look
like
I've
got
this
right
done.
B
But
hang
on
a
second
Jennifer
did
this
is
the
this
is
for
launching
the
binary,
and
so
we
should
compare
that
to
the
ref
page
for
cube
control
for
launching
the
cube
control
binary,
but
that's
a
different
different
thing
from
the
ref
docks
for
the
cube
control
commands
I'm.
Trying
to
think
of
what
might
be
a
comparison.
I
mean
we
might
look
at
the
Federation
API.
They
use
a
standard
generator
for
fur
for
an
API.
That's
more
like
what
we
could
yeah
I.
D
B
B
E
B
D
D
This
is
the
those
are
the
pages,
or
that
is
the
file
on
that
this
PR
is
against.
Alright,
that's
why
we
can't
look
at
because
it's
good-
let's
not.
D
B
D
B
D
D
The
only
thing
I
called
out
was
that
historically,
we've
had
difficulty
getting
changes
from
Doc's
book
approved
and
merged
in
the
upstream
codebase,
but
I
think
there
I'm
willing
to
take
on
the
task
of
finding
a
person
the
way
person
to
bunk
in
person,
quite
possibly
not
until
after
the
release.
D
D
D
In
that
case,
it
looks
as
though
we
have
time
for
a
couple
more
items
and
still
get
done
early,
so
we
keep
getting
pull
requests
against
old,
blog,
post
and
and
most
of
the
time
I.
Don't
know
what
the
rest
of
you
do.
I
merge
them,
they
look
harmless.
They
don't
seem
like
the
greatest
use
of
anybody's
time,
but
it's
much
faster
just
to
merge
them,
then
send
them
back
saying.
Please
don't
do
this
anymore,
also,
not
very
friendly
to
do
that,
but
I.
D
Last
week
when
I
was
wrangling,
I
came
across
this
poll.
Request
that
didn't
have
there
wasn't
a
good
solution
right.
The
link
is
now
broken,
because
some
interface
no
longer
exists,
basically
the
whole,
so
the
architecture
has
changed.
D
Somebody
merged
it
I
was
concerned
because
it
didn't
in
fact
point
to
anything
useful
because
there
wasn't
anything,
you
know
if
I
merged
it.
Oh
this
is
this.
Is
it
sorry?
I
was
confusing
it
with
another
one
and
that's
what
I
get
for
using
memory.
D
Do
we
want
us
to
say
something
about
that?
You
know
I
mean
broken
and
you
can
tell
what's
going
on
right,
somebody's
decided
to
run
a
link
checker
and
then
they
you
know
in
this
case
anyway.
The
link
checker
cache
is
a
broken
link
from
ancient
ago,
and
so
somebody
says:
okay,
here's
a
broken
link,
I'll
fix
the
broken
link
and
not
really
paying
attention
to
the
fact
that
it's
not
a
useful
link
anymore
still,
Oh
with
spell
checkers.
D
I
don't
or
do
you
have
any
other
ideas
about
how
to
deal
with
the
situation
because
it
seems
as
though
they
come
in
small
waves,
especially
the
fixed
type
of
ones.
If
we
don't
have
anything
in
the
contributor
guide,
that's
probably
the
place
to
start
and
I,
don't
know
where
else
we
would
go
or
what
else
we
might
do
to
try
to
wyrex
contributor
energy
a
bit
more
productively.
Shall
we
say
I'm.
A
Playing
up
the
PR,
how
old
is
the
homeless,
the
blog
post
like
at
a
certain
point?
Do
we
just
want
to
say,
like
a
blog
post
over
a
year
olds,
we
don't
work
on
some
arbitrary
time
that
we,
let
me
no
longer
support
it-
is
we're
also
not
writing
blog
posts
either.
So
it
seems
weird
that
we're
maintaining
it
I'm.
D
Particular
one
dates
to
2016
yeah.
A
It
seems
I
think
it
I
mean
it'd,
be
nice
to
take
some
of
those
they
get.
What
I
think
there's
like
there's
a
couple
like
points
that
we
should
take
a
look
at
one
is
at
what
point
do
we
stop
maintaining
it
and
then
there's
another
point
where
we
might
want
to
consider
dropping
it
from
search
results
that
was
a
bit
more
dramatic
and
probably
going
to
erase
a
few
more
eyebrows
but
I.
Think
after
like
I
think
after
a
year
we
shouldn't
be
reviewing
any
of
these
dr's.
D
I
agree
with
your
conclusion,
but
the
question
is,
then:
how
do
we
enforce
that?
Do
we
on
do
we
do
it
hey
by
putting
specific
on
the
contributor
guidelines,
do
we
do
it
more
actively
by
refusing
to
you
know,
by
closing
unmerged
pull
requests
that
come
in
like
this,
you
know:
do
we
do
we
kind
of
waffle?
It
depending
you
know,
like
I
kind
of
hate,
if
somebody
actually
fixed
a
broken
thing,
even
if
it's
not
a
useful
thing
in
their
first
I
kind
of
hate
close
emerge
that
maybe
we
should.
A
Think
you
raise
a
good
point:
there's
a
there's,
a
compassionate
way
to
do
this,
where
we
we
definitely
should
document
whenever
we
decide
in
the
contribution
guideline
for
sure,
I
also
think
we
should
I.
You
know
I
think
if
somebody
submits
a
fix
on
their
own
and
that's
outside
this
group
I
think
it's
totally
fine
to
accept
this
I
think
within
this
group,
and
we
definitely
have
way
bigger
priorities
to
be
working
on.
So
none
of
us
should
be
working
on
it.
A
So
I
think
that
might
be
the
way
to
do
it
is
so
we
documented
in
a
contributor
guidelines.
We
we
we
tell
folks
here
in
this
meeting.
Please
don't
work
on
any
of
this
and
I,
don't
think
anybody's
working
on
it
now
I
think
we
have
a
pretty
good
idea
of
our
priorities
and
then
outside
of
this
group,
if
people
submit
it,
we
could
just
say
like
hey
thanks
for
submitting
those,
we
normally
don't
support.
We
normally
don't
update
these.
A
D
A
D
Think
folks,
on
this
call
tend
to
I
mean
I
tend
to
see
I
can't
at
least
tend
to
see
these
PRS
come
in
from
first-time
contributors
we
had
a
rash
of
typo
fixes.
Last
fall,
I
think
it
was
probably
a
year
ago
now
from
some
fairly
new
contributors
in
Asia.
That
all
seems
to
settle
down.
It's
not
happening
anymore.
Oh,
you
know
what
is
getting
out
some
places,
but
you
know
we
can't
so
we're
not
to
everybody.
So
I'm,
like
yeah.
D
A
D
Past,
when
you
that's
good,
okay
awesome
does
anybody
else
want
to
volunteer
to?
We
are
against
the
contributor
guys.
D
Good
item
Jared
I
saw
yours
and
let's,
let's
take
that
one
up
first
then,
because
I'm
tired
of.
A
Yeah
I
think
this
set
up
landing
page
has
a
lot
of
those
which
it
has
a
lot
of
confusion.
So
so
I
guess
like
manage
solutions
would
be
yeah
cos
like
we
have
local
machine
solutions,
hosted
solutions,
a
turnkey
doesn't
make
any
sense,
mm-hmm
and
then
custom
solutions
yeah.
These
are
all
very
confusing.
A
H
Basically,
it
goes
from
easy
to
hard
configuration
there.
The
only
thing
was
that
there
were
some
people
who
were
pretty
passionate
about
keeping
turnkey,
as
it
was
important
for
their
marketing
and
I.
Don't
know
whether
or
not
like
our
organization
should
be
affected
by
external
marketing
practices.
A
E
D
D
Someone
who
contributes
from
a
company
that
cares
very
much
should
be
at
the
marketing
impact
of
its
upstream
work.
I
couldn't
agree
with
you
more
Geralyn,
I
I
think
that
this
is.
This
is
like
straining
at
usability.
We,
as
a
group
see
Doc's,
are
about
making
the
documentation
useful
for
folks
who
want
to
learn
or
manage
with
kubernetes
we're
not
about
helping
third
party
marketing
solutions.
D
A
I
think
we
should
strive
for
the
most
useful
content
for
users
and
the
most
accurate
and
honest
concept
as
content,
so
I,
don't
yeah
I.
Think
this
I,
like
that.
You
I,
like
that
you
dropped
rank
II
like
that.
This
makes
sense,
because
I
think
people
would
want
to
do.
People
want
the
easiest
solution.
First,
more
customizable
solutions
at
the
bottom.
G
A
A
Yeah
I'll,
but
Andrew
make
the
call
a
lot
but
I
think
I.
I
think
this
looks
good
to
me,
so
maybe
I
was
just
should
I
just
send
I
think
we
just
list
here.
That
is
this
good
with
everybody
here
it
seems
like
it
is
I.
B
I,
like
it
I
think
we
might
stumble
a
bit
with
the
word
hosted
because
I
know
gke
refers
to
itself
as
a
hosted.
We
have
a
host
of
master,
but
we
can.
We
can
pound
out
the
details
as
we
go.
I.
I
Think
this
work
also
has
some
relevance
to
the
cloud
providers.
Effort
that's
happening
as
part
of
the
the
other
sig
and
Andrew
has
been
working
with
some
internal
folks
here
at
Google
to
think
about
kubernetes
on
GCE
as
a
hosted
solution,
and
part
of
that
is
thinking
about
the
onboarding
pads
into
those
hosted
solutions.
More.
That's
part
of
the
context
for
this
as
well.
I.
D
Wouldn't
mind
being
part
of
a
conversation
about
the
overall
sort
of
rework
of
that
set
up
page
because
I
to
help
wrangle
a
lot
of
the
cue
medium
ducks
which
lived
there
appropriately
I
think,
but
in
a
different
relationship
than
these
categories
and
I.
Don't
know
that
information
architecture
is
the
solution
to
that
problem.
But
I
think
it
would
be
useful
to
consider
all
of
the
content
that
we
currently
have
under
setup
together
as
a
way
of
both
framing
these
revised
categories
optimally
and
making
sure
that
the
other
content
on
that
page
is
appropriately
placed.
D
Some
of
it
might
not
even
necessarily
continue
to
live.
There,
there's
really
a
reference
topic
about
certificates,
which
I
put
there
for
one
1101,
12
I,
forget
which,
because
it's
really
closely
related
to
cube
ATM
and
we
couldn't
figure
out
where
else
to
put
it,
but
it
might
be
that
there's
a
better
solution.
If
we
look
at
all
of
the
content
together
so
and
there
will
be
another
search,
there's
another
search
topic
coming
in
at
40
115.
Also,
all
this
stuff
together,
I
think
it's
timely,
yeah.
A
C
B
I
So
I
think
we
need
Andrew
for
that
context.
He's
been
chatting
with
Jago
McCleod
about
this
particular
topic
and
the
set
up
there.
There
have
been
a
lot
of
different
conversations
about
the
setup
page.
The
Cody
was
doing
in
his
noogler
onboarding
up
until
now,
with
the
kubernetes
on
GCE
content
and
the
paths
into
that
content,
and
so
andrew
has
better
context
than
I.
Do
I
can't
comment
beyond
that.
I
Believe
that
the
I
believe
that's
the
case
and
I
think
that
the
provider
sig
is
led
by
Jago
or
somebody
on
jegos
team
and
that
their
main
interest
is,
is
the
kubernetes
on
GCE
being
an
exemplar
for
other
hosted
solutions
in
terms
of
the
depth
and
quality
of
content?
And
so
part
of
that
is
the
is.
How
do
you
find
that
content
and
fixing
that,
and
then
part
of
it
is
the
actual
hosted
solution.
H
The
sick
cloud
provider
right
now
is
basically,
there
are
seven
companies
and
they
are
just
going
to
provide
like
the
baseline
documentation
and
what
we're
working
on
is
like
or
like
as
google
we're
just
gonna
provide
one
piece
of
that,
but
each
company
is
gonna,
be
providing
the
same
types
of
documentation
as
an
example
for
everyone
else.
Who
wants
to
be
listed
on
that
page
yeah.
A
I
think
right
now
we
list
a
whole
bunch
of
setup
like
additional
technical
setup
information
that
should
probably
live
on
other
company
sites
instead
of
ours
and
I,
know
I
see
you
nodding,
see
if
it
may
Steve
I
think
I'm
just
something
that
we
have
wanted
to
push
off
our
site
for
a
while,
because
it's
just
hard
to
maintain
consistency
and
accuracy
of
content
that
doesn't
belong
down.
So
we
should,
which
is
the
I,
think
edges
spearheading.
B
A
A
G
D
Hearing
no
doesn't
disagreements
let's
and
we've
got
only
a
few
minutes
left.
So
let's
get
started
talking
about
Kip
Kahn
Seattle
and
we
can
defer
the
redirects
question
to
next
week
or
possibly
even
take
that
one
too
slack
I
will
start
a
conversation
there.
Maybe
next
week
also
Jared
did
you
have
thoughts
about
coupon,
no.
A
I,
just
I
just
got
back
from
vacation
I'm,
currently
in
like
in
Breezy
vacation
notes,
though
so
I
I
just
know
I'm
going
to
Seattle
in
a
couple
weeks
and
it's
a
couple
things
I
know:
we've
we've
talked
about
all
meeting
up
and
having
a
discussion
like
a
cig,
Docs
discussion
and
I.
Think
Zac
has
already
planned
that
and
has
that
well
in
hand.
Yes,
I
think
there's
I
think
everybody's
gotten
details
about,
but
if
you
haven't
I
think
that
we'll
send
out
a
couple
reminders,
additional
information,
yeah.
D
A
A
If
not
then
I'd
like
to
talk
about
some
specific
efforts
that
we
might
put
forward
or
things
you
might
want
to
do
so
I,
it
could
just
be
a
an
effort
to
pay
down
a
bunch
of
stats
that
we
have
with
off
the
shoes
or,
if
there's
something
more
specific,
that
we'd
like
to
do.
My
biggest
had
to
do
that
so
I'm
seeking
information,
and
if
that
information
doesn't
exist,
I'd
like
to
make
some
wet.
D
Snow,
timely,
timely
inquiry,
Jared
I'm
what
what
Zach
and
Andrew
and
I
have
got
to
thus
far
Zach's
main
focus
at
this
point
is
the
docks
summit
he's
not
as
interested
in
a
dock
sprint.
I
agree
with
you
and
I've
made
this
clear
and
I'm
to
Zach
that
we
should
go
prepared
to
run.
You
know
something
in
a
way
of
a
dock
sprint
and-
and
that's
as
far
as
I
got
I
said
that
I
was
willing
to
wrangle.
Whoever
showed
up
wanting
to
do
it
and
I
was
prepared
to
wing
it
on
the
spot.
D
We've
done
this
before
was
not
unsuccess
anyway,
but
if
you
would
like
to
help
do
more
planning
for
such
an
effort.
That
would
be
amazing
and
awesome,
and
wonderful
and
I
would
be
happy
to
help
I
didn't
feel
like
I
had
the
bandwidth
to
take
on.
You
know
planning
a
dock
sprint
ahead
of
time,
but
I
was
willing
to
run
the
wrangling
in
improv
as
a
Tory
fashioned,
so
fearing
that
you
know,
he'd
have
the
energy
to
pull
something
together
with
other
people
in
the
room.
D
A
I'm
more
worried
about
coordinating
it
against
the
rest
of
the
event
and
because
we
did
our
in
the
contributor
summit
last
time.
I
know
a
lot
of
contributors
at
a
Copenhagen.
We
had
the
dock
sprint
at
the
same
time
as
a
contributor
summit,
which,
frankly
kind
of
fact,
because
it
was
a
lot
of
people
competing
people
had
to
make
a
decision
of
like.
A
Are
they
going
to
stay
at
the
contributor
summit
and
get
to
know
what's
going
on
and
all
the
other
cigarettes
and
get
to
meet
all
the
other
contributors,
or
are
they
going
to
come
and
participate
in
the
contributing
to
Docs?
So
I
think
it
might
be
nicer
to
do
the
if
we
do
a
tox
print
to
do
it?
Maybe
during
a
like
a
slot
in
the
actual
conference
instead,
but
I,
you
know
want
to
sort
coordinate
that
with
you
know,
with
yeah.
You.
A
A
D
A
I
think,
even
if
it's
just
like
a
group
of
us
here
want
to
get
together
and
mic
work
on
Doc's
I
think
that
would
actually
be
really
useful
because
we
rarely
are
in
the
same
status
at
the
same
time.
It'd
be
nice
just
to
meet
up
work
through
a
couple.
Prs
maybe
like
might
be
a
good
thing
to
do
before
or
after
the
place
of
DOC
planning
summit.
A
Maybe
before
so,
we
kind
of
get
a
feel
of
some
of
the
problems
with
dots
and
you
can
have
discussed
them
and
hash
out.
What's
important
and
what's
not
but
I,
don't
know
I'm
open
to
ideas,
so
yeah
I
agree
with
you.
Jennifer
I've
done
this
a
bunch,
so
I'm
not
I'm,
not
worried
about
try
something
the
night
before
it's
just
notifying
people
about
it's
happening
and
having
the
space
inside
and
having
it
in
the
schedule
and.
D
D
Depending
on
what
your
calendar
says,
we're
either
two
minutes
over
or
we've
got
eight
minutes
left
I'm
going
to
call
it
for
being
two
minutes
over
because
I
don't
think
we
want
to
take
up
another
item,
this
late
in
the
game.
Does
anybody
have
anything
else
they
want
to
mention
briefly
before
we
call
the
meeting.