►
From YouTube: zoom 0Kubernetes Sig Docs 20180403
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 03 April 2018.
https://github.com/kubernetes/kubernetes.github.io
A
C
A
B
D
C
D
A
So
just
some
quick
updates,
we
are
meeting
the
need
originally
scheduled
to
meet
last
week
with
the
contractor
who's
migrating
was
migrating.
The
website
from
Jekyll
to
Hugo
that
was
postponed.
One
week
meeting
is
now
tomorrow,
so
we
don't
have
any
updates
on
the
Hyuga
migration
yet,
but
that
meeting
is
has
been
rescheduled
for
tomorrow.
So
they'll
have
more
information
about
the
Hyuga
migration
next
week,.
A
A
Rotations
for
the
pull
request,
queue
just
a
reminder
if
you
have
not
done
a
PR,
regular
rotation
yet
and
you're
interested
in
doing
so
go
ahead
and
sign
up
and
the
link
is
the
rotation
is
in
the
weekly
agenda
and
if
you
want
to
shadow
someone
before
you
do
a
rotation
on
your
own.
That's
great
just
indicate
when
you
sign
up
that
you
want
to
be
a
shadow.
A
E
E
E
E
You
know,
you
know,
you
know
other
things
so
that
you
know
so
so
that
you
know
and
and
and
what
support
would
you
need
from
us
in
order
to
help
you
you
know,
do
that
generation
and
do
that
work
so
that
it's
not
you
know
so
that
maintaining
these
isn't
entirely
on
your
shoulders
and
rather
is
on
the
shoulders
of
people
who
are
producing
the
documentation
and
so
I.
Don't
know
if
I
expect
any
answers
right
now,
but
I
wanted
to
make
sure
that
we
started
that
conversation
and
we
were
communicating
back
and
forth.
Thank.
A
It
was
the
the
solution
that
I
had
been
looking
at
would
be
to
contact
everyone,
individual
contributors
who
had
contributed
all
of
these
Doc's
over
time
and
give
them
a
certain
window
to
move
the
documentation
off
of
the
site,
but,
as
long
I
think,
I
think
the
solution,
you're
proposing
of
hosting
platform
or
provider,
specific
documentation
and
individual
repositories
and
automating
content
delivery
is
is
worth
exploring,
as
our
list
is
really
long.
Today's
Andrew,
here
and
drew
is,
are
like,
as
our
automation,
evangelist.
A
For
cloud
providers,
conversation
is
going
to
be
to
have
a
conversation,
an
offline
conversation
with
the
working
group
and
folks
from
from
Doc's
or
interest
folks
from
documentation
who
are
interested
so
I
will
take
an
action
to
set
up
a
meeting
that
is,
would
you
is
that
agreeable?
Would
you
like
to
meet
offline
and
have
a
more
specific
and
more
confined
discussion?
Yeah?
Absolutely!
Ok!
So
just
a
really
quick!
We
if
you
are
interested
in
joining
a
conversation
about
isolating
an
automating
cloud
provider
documentation.
E
Yeah
I
saw
that
the
so
for
from
our
point
of
view,
the
user,
the
end
the
end
users
is
the
most
important
audience
that
we're
trying
to
reach
with
these
different
providers.
You
know
being
able
to
create
a
consistent
experience
for
them
so
that
when
they
arrive-
and
they
want
to
run
kubernetes
on
that,
you
know,
as
your
open
stack,
that
they
have
that
they're
successful
at
that
like
guaranteeing
their
success
is
the
most
important
thing.
E
So
that's
the
this
will
be
looking
at
to
start
off
with,
but
I
also
think
it's
valuable
to
have
documentation
in
place
that
allows
developers
to
also
show
up
successful
and
to
enter
contribute,
because,
because
a
lot
of
times
these
consumers
wind
up
turning
into
developers
as
they
uncover
needs,
but
but
to
start
off
with
just
using
it
using
it
correctly.
As
our
audience.
A
C
Rbac,
yellow
files
or
also
networking
files,
as
these
are
the
two
cases
actually
were
run
into
issues
with
this
with
networking
before
and
with
our
days
seated
last
week,
and
because
there's
actually
some
trouble
to
debug
this-
and
it's
not
always
obvious-
and
my
question
would
be
is:
is
it
just
wrong
that
there
is
destroyed
API
group
name
and
the
reference
documentation?
What
doesn't
need
to
be
like
what
should
it
be
the
wrong
one?
F
So
I
have
the
same
question.
I
was
hoping
to
get
the
answer
to
that
from
you
and
I
was
thinking.
The
answer
is
we
probably
should
be
displaying
the
long
name
that
the
code
that
does
that
generation
was
developed
by
Phil
wit
rock.
He
chimed
in
a
little
bit
on
our
slack
channel
just
to
say
that
he
did
it
that
way,
because
there
was
some
that
the
group
names
were
were
weird
at
the
beginning
is
how
he
phrased
it.
But
I
don't
have
any
more
detail
as
to
why
he
made
that
decision.
F
F
G
F
One
of
the
things
I
notice
in
our
you
know
now
that
the
the
the
generation
code
has
been
revised
is
that
the
left
pane
gets
him.
The
things
get
cut
off,
they
get
truncated
and
so
I
was
wondering
if
we
could
keep
the
short
names
in
the
the
left
nav,
but
have
the
long
names
displayed
in
the
main
body
of
the
content.
What
do
people
think
of
that.
C
D
F
C
F
H
C
A
A
A
So
in
case
you
don't
know,
cookie
licking
is,
if
you
see
a
cookie
that
you
want,
you
lick
the
cookie,
put
it
back
down
and
say
you
can't
have
that
cookie,
I
licked
it
it's
mine
and
it's
a
way
of
talking
about
claiming
a
poll
or
claiming
an
issue
and
commenting
on
an
issue
either
formally
or
informally
and
saying
hey
I
commented
on
that
issue.
That's
mine!
You
can't
work
on
that
or
discouraging
other
people
from
working
on
it
and
what's
happening.
A
Is
that
there's
a
lot
of
cookie
licking
happening
with
issues
that
are
tagged
good
first
issue
and
it's
inhibiting
folks
from
working
on
it?
So
I
would
like
to
formally
propose
that
we
have
a
no
cookie
licking
policy.
That
look
like
formally.
Is
that
if
someone
comments
on
an
issue,
if
there
is
not
a
if
there
is
not
an
active
poll
request
on
that
issue
in
less
than
a
week,
it's
fair
game
for
everyone
or
a
week
max
you
have.
A
If
you
comment
on
an
issue,
you
have
one
week
maximum
to
get
a
poll
request
open
and
actively
worked
on.
Otherwise
that
issue
is
fair
game.
Good.
First,
issues
are
unsurprisingly
popular
ones
and
I
want
to
make
sure
that
they
that
they
keep
keep
available
and
keep
in
circulation
rather
than
going
say.
So
is
anyone
super
pro
cookie
licking?
Is
this
controversial
in
any
way.
I
I
By
I
accident,
two
different
folks
had
picked
the
same
glossary
item,
and
so
okay,
thanks
Joe,
that's
awesome,
and
so
you
know
what
happened.
There
wasn't
the
worst
thing
in
the
world.
I
felt
bad
I
had
to
tell
one
person:
hey
I'm,
really
sorry,
but
the
other
is.
The
southern
port
is
really
pretty
much
good
to
go,
and
yours
had
some
problems
and
I
had
to
make
a
decision.
They
both
were
trying
to
to
spot
for
entry.
It
wasn't
the
worst
of
the
world
right.
I
mean
it
wasn't
the
worst
thing
the
world.
G
D
I
I
think
doc.
Splints
are
kind
of
a
corner
case,
so
I
I
think
forgot
experience.
We
would
likely
tag
a
bunch
of
issues
and
say
these
are
for
a
doc
sprint.
We
don't
claim
them
and
then,
within
the
doc
sprint
I
think
it's
useful
to
have
like
a
separate
tracking
mechanism
like
a
spreadsheet
or
somehow
I
thought.
G
D
I
D
A
J
A
L
A
L
H
A
A
H
L
A
L
A
Is
both
confusing
odeon?
Yes,
yeah
sure,
let's,
let's
take
that
with
a
get
a
PR
open
within
three
days.
Works
in
progress
can
be
works
in
progress.
If
someone
else,
if
someone
is
also
interested
on
working
something
they
can
contact
the
person
with
an
open,
PR
and
work
it
out
between
is
that
agreeable
is
that
is
that
sufficiently?
Is
that
sufficiently
specific
to
address
the
majority
of
use
cases.
A
A
We
had
a
request
from
Dan
Kahn,
the
the
CEO
for
the
CN
CF.
Apparently,
the
kubernetes
website
is
having
long
load
times
in
China,
because
the
kubernetes,
the
fonts
for
the
kubernetes
website
are
being
served
from
his
google
font
server
and
the
Great
Firewall
tends
to
slow
things
down
significantly.
So
we
have
a
request
to
find
an
alternate
means
of
serving
fonts
for
the
kubernetes
website.
A
A
L
A
H
A
H
L
You're
gonna
run
into
educators,
where
you're,
adding
links
that
you're,
like
your
PR,
is
creating
links
that
you're
then
linking
to
in
the
same
CR.
So
the
link
checker
I'll
fail
because
well
maybe
it
won't
fail,
because
the
link
will
be
in
the
PR
but
like
this
is
a
chicken-and-egg
problem.
I've
seen
in
the
past,
where
you're,
adding
new
endpoints
and
then
linking
to
them
can.
L
A
J
A
J
L
I've
used
link,
link,
live
in
the
past.
It's
a
little
bit
slow,
it's
probably
too
slow
for
CI
CD.
It's
super,
thorough,
I.
Think
one
of
the
things
that
you
have
to
do
is
decide.
You
have
to
balance
out
thoroughness
with
time
constraints
because
it
actually
takes
a
long
time
to
validate
weblinks,
and
you
also
sometimes
have
issues
with
looking
like
you're
doing
denial
of
service.
So
sometimes
you
have
to
add
artificial
delays
and
timeout.
So
you
don't
get
blocked
by
somebody
like
it's
everything
right.
A
A
You
a
sure,
I
understand
the
the
reluctance
joke
to
put
it
on
a
CID,
CI
CD
side,
but
I
mean
ultimately
like
what
are
we
gay
checking
against
is
right
now,
there's
not
really
any
gay
checking
that's
happening
and
I
feel
like
we're.
Seeing
enough
broken
links
that
having
having
a
link
checker
serve
as
a
gate.
Test
for
for
PRS
would
be
really
valuable.
F
I'm
on
the
fence
about
this
I'm
on
the
fence
between
doing
it
that
way
and
having
the
link
test
run
automatically
once
a
day
and
generating
a
report
for
us
and
I
guess
if,
if
it
turns
out
that
it
adds
a
lot
of
time
to
being
able
to
you
know
when
you,
let's
see
how
does
this
work,
you
submit
a
PR
and
you
and
it
kicks
off.
You
know
the
Travis
builds
in
the
net
laughs
I.
F
L
Another
thing
to
think
about
is:
if
you
have
a
link
checker
running
on
every
PR,
it
should
only
be
checking
new
links,
they
get
added.
But
if
this
thing
hi
I'm
your
PR
may
be
breaking
a
link
that
your
PR
doesn't
realize
is
breaking
so
you
probably
mean
the
car
PR
checks
for
new
things,
but
you
probably
also
need
a
periodic
overall
website
check.
L
Checking
like
crawling
the
whole
website
for
links
takes
a
really
long
time
and
that's
where
you
start
having
problems
with
another
website
being
down
or
blocking
link
checkers
or
you
looking
like
denial
of
service.
The
other
thing
you
can
choose
to
only
validate
internally,
not
externally
I,
don't
know
it
sort
of
depends
on
the
pattern.
I
mean
I'm
too
new
to
the
doc.
That
opinion.
A
No
that's
validating
external
links.
Often
my
experience
with
that
is
that
you
get
all
kinds
of
HTTP
errors
depending
on
what's
going
on
with
a
provider,
and
you
can
get
a
failing,
GAE
check
for
a
an
unrelated
external
reason,
and
if
we,
if
we
use
a
if
we
use
a
link
checker
as
a
gate,
check
function,
I
don't
want
to
slow
it.
I
don't
want
to
slow
down
anything
happening
because
it's
like
this.
The
site
is
HTTP
500,
for
example.
F
F
A
A
Let's
revisit
this.
Let's
revisit
this
again
next
week
and
and
review
review
initial
findings
from
what
I'm
hearing
it
sounds
like
folks
are
thumbs-up
in
general
for
a
CIC
d-link
checker
that
validates
links
introduced
or
removed
by
a
PR
and
I'm
am
I
correctly
hearing
that
yes,
excellent,
okay,
all
right,
let's
do
research
and
review
in
a
week.
A
J
When
you're
documenting
something
and
if
you're
referencing
to
a
text
or
a
topic
in
a
different
page,
you
rather
use
this
section,
exact
section
name
than
using
a
custom
mean
to
reference
it.
So
if
there
is
a
broken
link
and
if
you're
planning
to
fix
it,
you
wouldn't
know
what
it
was
initially
referring
to
right.
So
unless
you
have
clear
guidelines
about
what
was
the
section
or
the
topic
that
it
was
referring
to.
J
Usually,
you
know
if
the
topic,
if
they
were
referring
to
a
cube,
CTL
overview
section
and
referring
to
that
as
just
cube
CTL
in
their
text,
the
referencing
text
was
just
you
just
cube
CTL.
That
would
end
if
you
would,
if
it
was
broken
because
cube
CTL
overview
was
renamed
to
just
say
overview,
or
you
know
that
didn't
work
correctly
and
if
the
link
broke.
So
unless
we
have
some
guidelines
about
what
should
be
the
referencing
text
that
you
need
to
add
when
you're
referencing
and
what
should
be
the
guideline,
when
you're
renaming
I.
A
J
Yes,
yeah
right,
that's
what
I
was
referring
to,
because
some
places
we
are
referring
them.
As
you
know,
cube
C.
We
are
referring
it
as
dash
board
and
some
places.
We
are
referring
it
as
web
UI.
There
is
no
consistency,
so
you
would
know
you
would
not
know
I
mean
at
least
for
newcomers.
Like
me,
I
found
it
a
little
difficult
to
actually
understand
what
was
the
topic.
It
was
referring
to
and
also.
J
A
All
right,
moving
on
from
CI
CD
improvements,
just
an
update
to
folks.
We
have
got
this
morning
a
proposal
to
request
from
a
Korean
translation
team
who
wants
to
begin
working
on
the
Korean
translation.
So
yeah
we're
growing,
and
this
is
makes
me
really
glad
that
we're
thinking
about
Hugo
and
easier
internationalization
yeah
can.
A
A
D
H
H
A
big
dawgs
maintainer
because
you
know
a
lot
of
documentation
we
have
is
for
the
general,
like
community
membership
and
that
sort
of
thing.
So
we
would
like
to
sort
of
add
that
to
the
docs
contributor,
augmentation
and
clean
that
up
and
also
I,
think
see
we
reorganize
the
contributor
section
in
our
site
to
make
it
easier
to
for
people
to
find
like
kind
of
like,
depending
on
where
they
are
in
the
process.
You
know
find
the
proper
instructions
for
that.
D
What's
in
the
what
extent
lien
box
when
it
comes
to
contributing
appearances,
there's
kind
of
a
we
smashed,
a
bunch
of
content
like
Jekyll,
includes
a
hands
like
really
advanced
topics
in
that
section
that
are
completely
irrelevant
to
a
new
user
and
you
contributor,
and
then
there's
really
clear
guidance
around
how
to
file
with
the
are,
and
let's
do
so,
I
think
it's
important
to
separate
those
two
out
and
set
it
I.
Think
there's
this
shoe
part
to
it.
Jared.
F
D
D
So
if
you
open
up
that
link
and
and
no
fault
of
anybody
who
created
the
stage
but
I
would
expect
this
page
to
be,
how
do
I
join
this
exact
group?
How
do
I
understand
what
to
do
in
listening
doc
screw?
How
do
I
become
apart?
It
was
the
exact
screw
and
none
of
that
contest
actually
in
this
document,
so
one
this
content
needs
to
get
written
or
should
be
renamed
and
then
a
separate
doc
needs
to
get
creative.
J
D
It
really
shouldn't
be
here.
It
might
need
to
be
in
another
bucket.
That's
married
on
a
level
and
with
the
custom.
Jekyll
include
snippet,
which
is
going
to
be
irrelevant
in
a
couple
weeks
when
we
migrate
and
then
localizing
kubernetes
documentation,
also
a
pretty
highly
specialized
topic,
so
we're
not
giving
good
guidance
to
new
contributors
who
want
to
contribute
to
our
Casa.
This
is
something
that
Misty's
pointed
out
and
Thomas
both
pointed
out
as
their
new
users,
trying
to
test
out
the
content
and
trying
to
contribute
and
they're
really
struggling
to
get
involved.
A
D
Just
clear
the
participation
section
that
I
linked
to
it's
useful
information,
because
it's
like
who
has
write
access
to
what
that
information
is
definitely
useful.
That's
definitely
more
of
an
advanced
topic.
How
to
participate
should
be
a
separate
document.
That's
more
about
like
how
to
get
involved.
You
have
to
leave
the
site.
F
So
my
memory
of
how
to
how
to
participate
topic
is
that
it's
mostly
about
how
to
review
peels,
and
so
we
could
maybe
combine
that
with
the
one
on
reviewing
issues.
You
know
we
could
have
a
topic
called
something
like
reviewing
PRS
and
issues
and
then
a
different
one
called
you
know
something
about
being
in
cig
docks.
Have
you
know
how
do
you?
How
do
you
join
and
where
the
meeting
is
that
kind
of
thing,
yeah.
D
A
Yes,
I
think
that
and
I
think
that
falls
into
two
contribution:
buckets
as
well,
I
think
so,
like
misty
and
Tom
Metropolitan
can't
assign
you
anything,
but
this
would
be
a
really
awesome
opportunity
to
write
up
that
journey
because
you
are
seeing
it
with
fresh
eyes,
specifically
be
how
do
how
to
join
the
Cates
org
and
how
to
become
a
sig
Docs
maintainer.
That's
like
joining
the
Cates
org
is
pretty
researchable
the
sig
Docs
maintainer.
You
would
probably
have
to
like
interview
folks
and
a
really
good
opportunity.
A
I
think
he
would
be
the
appropriate
people
to
like
take
that
on,
but
then
the
reorganizing,
the
content
of
that
entire
section,
I
think,
is
I
think
he
would
become
very
frustrated
very
quickly
and
I
reference.
My
own
long
existing
work
in
progress,
because
I
tried
to
tackle
that
when
I
was
new
and
that's
why
the
HBR
is
still
WIP
Lodi's
months
later,
so
I
am
happy
to
close
my
PR.
You
know
in
favor
of
someone
else
who
wants
to
tackle
this
work
because
I
will
be
honest.
I
cannot
do
it.
A
I
cannot
with
coop
Connie.
You
write
the
docs
and
the
the
summit
coming
up
at
the
end
of
the
month.
I
can't
take
on
anything
new
right
now.
So
if
someone
wants
to
take
on
reorge
amazing,
the
the
contributor
section,
this
would
be
a
golden
opportunity
to
do
it
because
I
agree
with
Jared.
This
is
it's
really
important
to
have
this
the
very
high
quality
and
be
as
explanatory
and
as
helpful
as
possible
and
as
soon
as
possible.
A
L
Seems
like
the
right
way
to
work
on
this
is
to
get
a
couple
of
newbies
to
collaborate
with
somebody
who's
totally,
not
a
newbie
who
really
understands
the
process,
because
if
you
have
newbies
trying
to
write
with
the
processes
that
it
it's
like,
we
don't
know
what
the
process
is.
That's
the
whole
point
so
like
they
would
be
like
a
lot
of
research
or
it
would
be
a
having
people.
We
could
just
ask,
and
then
we
can
kind
of
write
down
what
they
say
like
I.
A
G
F
A
N
A
A
Tom
is
it
okay,
if
I
I
don't
want
to
fumble
over
Tom,
Benton
or
Tyson
in
notes?
Can
I
just
call
you
tell
Tom
in
the
notes?
Yes,
that's
that's
why
that
nut
moniker
exists.
Thank
you
all
right.
We
have
a
couple
of
minutes
left.
Is
there
anything
that
folks
want
to
bring
up
and
anything
to
discuss
before
we
before
we
call
it?
A
K
I
was
just
going
to
say
enough
to
get
ahead
of
ourselves
on
the
contributing
stuff,
but
love
to
have
like
a
link
to
this
eventually
on
the
new
contributor
site,
as
well
as
the
single
point
of
truth,
contributor
guide,
that's
housed
in
the
community
repo
as
well.
It
just
lets
other
folks
know
that
there's
a
different
way
to
contribute.
You
thought
that
would
be
cool.
K
That's
all
I
wanted
to
say
doctor
that's
going
to
be
a
good
place
for
people
to
get
started
in
Gavin,
and
we
and,
as
we
have
been
doing
more
and
more
contribute
to
kubernetes
talks
at
conferences.
I
think
one
of
the
areas
that
we've
been
highlighting
is
back
as
a
best
single
point
of
entry
into
the
project,
so
yeah
cool.