►
From YouTube: Kubernetes Sig Docs 20180306
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 06 March 2018.
https://github.com/kubernetes/kubernetes.github.io
A
A
A
All
right
so
move
on
for
now
to
talking
about
migrating
kubernetes
io
migrating
the
website
from
Jekyll
to
Hugo,
so
I
don't
see
Jared
yet
either
so
Joe.
If
you,
if
at
any
point
you
want
to
chime
in
and
Jennifer,
if
you
want
to
chime
in
about
anything
from
our
discussion
this
past
week,
please
feel
free,
but
basically
we
met
with
a
dude
named
Buren
Eric
Patterson,
who
at
this
point
has
contributed
80%
of
Hugo's
codebase
to
it.
He
answered
a
lot
of
our
technical
questions.
A
We
did
kind
of
a
deep
dive
with
him
about
some
of
the
specific
concerns
and
interests
we
had
with
you
go
and
while
I
don't
think
that
we're
a
hundred
percent
on
all
of
those
questions,
I
think
we
are
satisfied
enough
to
be
able
to
confidently
move
forward
and
say
that
yes,
migrating
to
Jekyll
is
something
that
we
want
to
do.
There
are
compelling
reasons
for
it
that
are
in
our
interests
as
a
sig
that
are
in
the
service
of
making
contributors
lives
easier
and
in
will
make
it
easier,
I
think
to
make
developers
happier.
A
We
need
to
continue
setting
the
scope
of
what
does
it,
what
actually
needs
to
happen
in
a
migration,
making
sure
that
we
scope
out
the
process
properly
and
completely
as
completely
as
we
can
before.
We
actually
begin
based
on
our
preliminary
estimates,
and
this
is
nothing
more
than
just
kind
of
gabbing
at
each
other
and
going
what
do
you
think?
Well,
what
do
you
think
we're
guessing
that
the
actual
migration
process
will
take
between
three
and
five
weeks,
the
contractor
we're
going
to
contract
with
Baron
Patterson
to
do
the
work?
A
He
estimates
that
it's
going
to
take
him
three
weeks.
I
remain
somewhat
skeptical
that
enough
of
our
sort
of
situational
use
cases
could
be
done
in
only
three
weeks.
So
my
my
guess
is
four,
and
it's
always
possible
that
you
know
some
some
edge
case
arises
and
it
could
take
five.
So,
relatively
speaking,
it's
a
relatively
rapid
migration.
A
Our
goal
is
that
there
shouldn't
really
be
any
disruptions
to
the
contribution
pipeline.
There
are
going
to
be
some
moments
where
we
have
to
say:
hey,
stop,
stop
merging
long
enough
for
us
to
do
some
cut
over
pieces,
but
we'll
have
more
specifics
about
that
as
as
we
get
them
so
I
guess
the
the
take
away
from
it
is
that,
yes,
we
are
going
to
migrate
the
kubernetes
website
from
Jekyll
to
Hugo.
A
We
estimated
it
will
take
between
three
and
five
weeks.
We
don't
have
a
specific
time
yet
for
when
that
will
begin,
we
do
need
to
peg
it
around
to
be
sensitive
to
the
kubernetes
release
cycle.
In
whatever
time
that
we
we
pick
for
migration
and
for
the
details
of
forthcoming
now
is
it
Jennifer
Andrew
Joe
did
I
get
everything.
Is
there
anything
that
we
need
to
add
I?
Think.
B
You
covered
most
of
it.
I
did
in
the
thread
that
we
had
going.
I
had
responded
with
a
here's.
Some
suggested
checklists
for
that
migration.
I
haven't
heard
anything
back
on
that
process
and
I
haven't
done
anything
to
make
it
more
public
that
be
a
worthy
thing
to
do
at
this
point,
is
it
now
stable
enough
that
we
think
yeah
we're
gonna
head
this
direction,
that
we
should
do
that
I.
A
C
E
D
A
So
the
two
driving
concerns
well,
the
two
biggest
selling
points
are
that
Hugo
builds
much
more
quickly
with
large
sites.
It
gets
Jekyll
when
you
get
over
about
a
hundred
hundred
fifty
paisa
deckle
starts
to
slow
down
significantly
where
Hugo
is
kind
of
renowned
for
having
near
instantaneous,
builds
Wow,
so
the
increased
or
the
the
much
speedier
build
time
for
for
Hugo
was
one
of
the
big
selling
points.
The
other
big
selling
point
is
built
in
internationalization.
A
A
A
A
We
will
be
moving
to
will
be
moving
away
from
liquid,
conditionals
and
sort
of
away
from
a
ruby
stack
in
general,
so
we're
moving
away
from
liquid
conditionals
and
towards
what
are
the
short
code
so
more
towards
golang
templates.
So
there
will
be
some
differences
in
in
how
we
structure
documents,
but
from
what
I
understand
the
that
process,
that
conversion
should
be
fairly
straightforward
and
not
a
whole
lot
of
in-depth
knowledge
for
contributors
is
required,
though
that
may
be.
That
may
be
an
oversimplification
depending
on
what
it
actually
looks
like.
D
B
Follow
on
is
the
internationalization
is
one
of
those
areas
that
I
didn't
feel.
It
was
definitely
a
selling
point,
but
I
didn't
know
that
we
had
a
particularly
concrete
mechanical
plan
of
how
we
were
going
to
enable
that
yet
and
when
I
asked
for
more
details
from
Bjorn,
he
got
well.
He
was
a
little
short
with
the
response
there,
but
he
sent
me
towards
the
go
hugo
site
where
I've
asked
the
question
and
I
did
get
some
interesting
responses
there
in
terms
of
strategies
and
patterns
to
use.
A
C
The
other
thing
that
we
haven't
really
talked
about
much
explicitly
with
a
move
to
Hugo
on
is
the
fact
that
it's
written
and
go,
and
it
I
know
that
for
myself,
I've
been
having
to
work
with
go
templates
for
some
chef.
Tio's
projects,
I'm
not
a
lot,
but
a
little
and
and
I
was
one
of
the
ones
who
said
well,
hey
wait
a
minute,
let's
stop
and
backup,
and
looks
look
at
additional
requirements
when
the
proposal
was
first
made.
C
No
Beth
one
thanks
I
mean
this
seems
obvious,
but
nobody
has
really
come
out
and
said
it
quite
explicitly.
I
think,
except
maybe
a
little
bit
in
the
meaning
that
we
had
with
the
Orang.
So
I
thought
I
would
just
put
it
out
there.
It
felt
a
little
like
not
exactly
the
elephants
in
the
living
room,
maybe
the
the
sleek
cuddly
kitten
like
tiger.
You
know
the
thing
everybody
likes,
but
it
feels
like
they
don't
need
to
say.
E
A
Ok,
so
moving
on,
let's
talk
about
rotations
for
the
PRQ
I
missed
the
dawn
of
the
PRQ,
but
Andrew
I
would
like
to
say
from
my
heart
to
yours:
I
love
you
so
I
am
really
glad
that
we
have
a
queue
for
PR
anglers.
This
is
amazing,
so
Joe
I'm
gonna
ask
you
to
talk
a
little
bit
about
what
you
learned
as
a
PR
Wrangler
in
your
first
week.
B
Okie-Dokie
see
a
number
of
people
that
contribute
fixes
to
our
docks
are
little
one-off,
typos
at
least
eighty
percent
would
be
my
rough
estimate
of
mostly
contributions
or
in
that
form,
that's
offset
a
little
bit
more
recently
by
the
110
contributions,
which
are
kind
of
coming
in
Fast
and
Furious
a
lot
of
people
when
they're
interacting
with
the
PRS
they're
trying
to
say
at
milestones.
You
can't
also
to
try
to
do
lgt,
M's
and
cants,
because
they're
not
a
kubernetes,
org
member
or
things
along
those
lines.
B
So
a
lot
of
the
standard,
boilerplate
messaging
sort
of
fails
for
them,
which
means
you
know
somebody
else
is
coming
along
behind
it:
kind
of
sweep
and
clean
it
up
afterwards.
Bread's
run
into
that
same
thing
where
he
has.
You
know
manual,
merge
capabilities,
but
not
approved
at
this
point.
So,
while
he's
doing
it
this
week,
I'm
sure
there
was
some
some
complications
there
that
made
things
a
little
bit
more
quirky
beyond
that
it
was
fairly
straightforward.
The
feedback
process
was
was
obvious.
It
was
not
inconsiderate.
B
It
was
nothing
terribly
efficient
in
terms
of
time
management.
It
took
probably
about
an
hour
to
two
hours
a
day
to
do
the
sequence
that
I
did
so
I'm.
There's,
probably
something
I
could
have
done
to
make
that
a
little
bit
more
efficient,
but
that
was
the
gist
of
it.
I'm
trying
to
go
through
all
of
the
PRQ
the
whole
set
every
day
for
a
week.
A
D
D
D
You
know
the
thing
I
see
is
sometimes
I'll
look
and
I'm
going
through
and
then
there's
somebody
already
assigned
so
now,
I
think
of
myself.
Okay
did
the
person
just
haphazardly,
assign
it
or
the
person
got
assigned
and
I
suggest
ignore
that
one,
and
just
you
know,
put
my
name
and
worry
about
the
ones
where
somebody
didn't
already
assign
and
if
the
person
the
sign
is
not
one
of
the
folks
on
this
call,
I
get
a
little
nervous
right.
D
The
people
that
are
typically
on
this
call,
so
that
was
the
one
question
I
had
is
you
know,
was
that
just
somebody,
oh,
they
were
suggested
to
assign
this
person,
so
they
did
it,
but
maybe
that
person
isn't
as
engaged
as
the
rest
of
this
group
here
and
I
should
still
come
behind
it
and
and
put
my
name
on
it
as
well.
That
was
kind
of
my
kind
of
one
question
to
worry
about.
That's.
B
D
It's
been
not
too
hard,
I
think
to
find
technical,
reviewers
right,
I
saw
one
come
in.
That
was
poor,
DNS
and
I.
Think
you
know
the
only
issue
I
had
a
dealing
with
that,
but
but
now
that
I
can
just
say:
hey
try
do
an
LG,
TM
and
then
I
think
we
just
had
one
come
in
on
sort
of
hi
available
would
be
at
CD.
It
looks
like
a
really
minor
change,
I'm
trying
to
look
at
that
and
I'm.
D
Like
I
kind
of
know,
this
I
think
it's
fine,
a
little
change,
but
in
the
back
of
me,
I
kind
of
have
this
magnet
feeling
that
no,
let
me
go
get
a
@cd
to
do
the
quick
tech
review,
even
though
I'm
like
99%
sure
it
was
a
very
simple
change
to
not
use
a
duplicate
name
for
the
multiple
Etsy
D
servers,
so
sometimes
I
kind
of
struggle
with
that.
Where
I
feel
like
oh
I,
this
well
enough,
I
could
just
just
say
it's
fine,
and
then
part
of
me
is
like
oh
get.
C
And
reading
reading
Jose
duck
I
mean
I,
haven't
you
know,
I
haven't
been
in
the
the
Q
rotation,
but
just
reviewing
pr's
in
the
past.
I've.
Definitely
gotten
stuck
and
reading.
Jose
just
really
helps
helped
me
think
about
reviewing
going
forward
and
getting
unstuck
and
in
sort
of
when
in
doubt,
I'll
slap.
A
tech
review
were
on
there
and
Joe's
got
good
guidelines
about
you.
C
So
so
review
was
kind
of
built
in
there
and
because
I'm
tracking,
based
on
because
I'm
tracking
those
PRS
based
on
the
features,
it's
it's
pretty
easy
to
do
that,
where
I'm
not
sure
but
in
general
I
think
I
think
that
Joe
had
really
good
advice
and
going
forward.
I
will
not
let
myself
get
hung
up
on
a
PR,
because
I
think
I
ought
to
be
able
to
techyv
you
and
I'm
not
sure.
Essentially,
that
just
holds
things
up
well.
A
A
A
Finalists
just
firstly,
I
must
I'm
sorry,
it
is
in
the
meeting
agenda
and
it
will
look
okay,
the
spot
in
the
meeting
agenda
from
henceforth,
because
it's
really
awesome
to
have
it
there
and.
A
A
Can
you
can
you
update
that
and
he
for
updates,
to
highlight
certified
kubernetes
solutions
and
I
said
sure.
I
can
do
that
and
I
started
looking
at
that
whole
set
up
site
and
I
realized
that
it's
symptomatic
of
what
I
see
as
a
larger
as
a
larger
problem
in
our
documentation
set,
which
is
when
we
are
hosting
provider,
specific
instructions
and
documentation.
A
Oftentimes
that
information
goes
stale
and
there
is
not
anyone,
there's
no
clear
or
chain
of
ownership.
It's
not
clear
who
owns,
who
owns
provider,
specific
documentation
or
who's
in
charge
of
maintaining
it
and
keeping
it
fresh
the?
Consequently,
we
end
up
fielding
PRS
for
other
other
OS
other
platforms
that
are
not
that
we
have
like
some
instructions
on
our
website,
but
my
thought
the
thought
I
have
is
whether
it
would
be
better
to
point
to
community's
own
instructions.
I
Yeah,
yes
I'm
the
this
is.
This
is
Chris
I'm
the
CEO,
since
one
of
the
stig
OpenStack
leads-
and
this
is
a
discussion
we
had
last
week
at
one
of
our
big
events
as
we're
moving.
You
know
to
a
more
full
OpenStack
upstream
external
provider.
That
was,
you
know
coming
to
this
Friday.
Might
one
of
my
goals
was
to
actually
get
the
OpenStack
documentation
updated,
but
but
instead
that
goal
shifted
to
helping
you
know
kind
of
everybody
likes.
You
know
you
know,
develop
a
path
for
everyone
to
externalise
their
documents.
I
That
would
be
great
because,
because
we
realize
that
you
know
we
identify
a
ton
of
gaps
and
it's
hard
to
keep
on
top
of
it
and
right
now
we're
spending
more
time
removing
bad
Doc's.
Just
so
people
don't
have
a
bad
experience.
You
know,
as
we
sort
out
how
you
know,
things
are
changing
in
our
ecosystem.
So
you
know
as
a
sig
leader,
you
know
external
provider
sig
leader,
I
totally,
would
be
on
board
with
this
possible.
A
J
J
Else,
yeah.
Okay,
thank
you.
Steve
go
ahead,
so
this
question
has
come
up
several
times
over
the
last
year
and
a
half
and
we've
gone
back
and
forth
on
and
at
first
we
were
thinking.
We
can't
possibly
maintain
this
material.
Let's
just
have
other
other
people
host
the
material
on
their
own
sites
and
we'll
link
to
it.
And
then
you
know
over
the
last
year,
or
so
we've
kind
of
drifted
more
towards
thinking,
there's
a
value
in
having
the
one-stop
shop
so
that
we
will.
J
We
will
host
it
on
our
site,
but
we
won't
be
responsible
for
its
accuracy
or
main
so
that
that's
just
kind
of
how
the
thing
has
drifted
over
the
last
last
year
or
so
and
I'm
perfectly
open
to
the
idea
of
removing
it
from
our
site
and
just
having
it.
You
haven't
posted
on
other
sites,
I
think.
If
we
do
keep
it,
we
probably
want
to
make
sure
each
one
of
those
topics
in
the
front
matter
has
a
reviewer
list.
J
H
So
I
was
recently
thinking
of
something
around
like
the
getting
started
guides.
It
would
be
really
cool
if
the
documents
that
we
had
had
some
sort
of
mechanism
to
signify
stableness,
so
maybe
every
six
months
or
so
a
doc
that
was
specified
for
a
set
of
reviewers
and
be
tagged.
A
like
tags
like
this
doc
might
be
stale
and,
and
that
would
trigger
like
someone
from
that
shig.
If
it
was
an
external
cloud
provider
or
something
going
in
and
maybe
updating
the
doc
is.
A
A
H
Definitely
in
some
instances
is
it's
more
worthwhile
to
have
that
that
external
community
hosted
documentation,
but
I
feel
like
our
getting
started.
Guides
should
have
some
way
for
people
who
don't
want
to
who
want
to
have
some
sort
of
agnostic.
Kubernetes
solution
be
able
to
actually
get
started
so.
A
I
would
do
the
only
counter
that
I
would
put
to
that
is
cube.
Atm
and
the
idea
that
cube
ATM
is
the
preferred
configuration
tool
meant
with
cube
ATM
things
just
work
and
that
that,
like
keeping
a
whole
bunch
of
platform-specific
or
OS
specific
getting
started
guides
may
be.
One
of
the
exact
solutions
were
trying
to
figure
out
and
fix
by
preferring
external
hosting.
A
H
Yeah
sort
of
I
guess,
but
you
know
it's
different
strokes
for
different
sites
yeah,
so
someone
mentioned
the
the
core
OS
documentation
and
now
that
documentation,
OTO
redirects
to
the
tectonics
activity,
which
not
everyone
wants
to
touch
so
like
one
of
my
tasks
at
some
point,
is
going
to
be
forth,
lifting
all
the
the
four
OS
specific
to
the
site.
That's
still
wanted,
but
but
yeah
I
mean
like
not
everyone's
gonna
want
to
go
with.
You
cube
a
do.
It
is
a
good
starting
point,
but
you
know
maybe
not
the
right
solution,
driver
sure.
A
And
I
think
there's
a
larger
conversation
there
to
be
had
about
what
makes
sense
to
what
makes
sense
to
host
rather
than
well.
We
are
hosting
these
things
now,
so,
let's
find
a
way
to
shoehorn
and
make
the
things
that
we
do
have
work.
I
think
it's
time
we
start
thinking
about
the
larger
question
of
what
is
the
best
thing
to
do
with
this
material
and
other
material
like
it
in
general?
What's
like
not
just
what
do
we
do
with
what
we
have,
but
what
can
we
do
to
maximize
developer
happiness
or
user
happiness?
A
I
I
mean
it
might
be
worth
bringing
up
at
the
cloud
provider
working
group
meeting
too,
because
because
you're
gonna
have
representatives
from
you
know
from
a
lot
of
the
interest
you
know
like,
especially
for
the
for
the
public
clouds.
You
know,
they're
they're
gonna
have
representation
and
they're
gonna
have
opinions
in
there
and
they're
gonna
be
involved
with
a
lot
of
the
stig's.
Like
you
know,
it's
hard.
Turning
on
how,
like
you
know,
sig
AWS
or
you
take
GCE,
or
so
these
would
feel
about
it.
I
You
know
it
feels
like
if
you're
not
going
to
external
eyes
it
you
know,
having
ownership
and
responsibility
within
the
kubernetes
community
makes
makes
a
lot
of
sense,
because
I
know
that
an
experience
that
we
had
was
the
documentation
was
brought
in.
You
know
from
just
from
a
third-party
company
and
they
just
not
maintaining
it.
I
You
know,
and
then
you
know,
then
we
kind
of
had
to
come
in
and
clean
it
up,
and
you
know,
but
you
know
about
having,
but
having
a
single
location
I
mean
in
my
mind,
having
a
single
location,
even
if
it
redirects.
We
say
these
are
the
docks
that
we
know
have
some
level
of
standard,
for
you
know,
maintenance
and
installation.
You
know
I
think
it's
important
and
I
think
it's
valuable
to
these
are
community
I.
A
Think
that's
fair
yeah
I,
just
I'm
thinking
about
when,
when
that
third
party
documentation
goes
stale
or
goes
out
of
date,
and
and
that
information
is
hosted
on
on
kubernetes
diet,
I
Oh,
whose
stocks
suck
you
know
it
ends
up
being
kubernetes
as
well.
Kubernetes
docks
suck.
So
that's
that's
what
kind
of
the
problem
that
we
need
to
solve
for
here
and
it
seems
it
seems
to
me
like
linking.
A
H
Get
a
tracking
issue
open
for
what
to
do
with
the
docks.
What
to
do
with
the
getting
started
docks
we
maybe
start
in
cig
cloud
provider
like
I
forgot
to
mention
that,
but
I
mean
that'd,
be
a
good
start
to
see
like
what
are
the
best
docks
to
use
for
each
provider
right.
Consider
linking
those
and
I
know
there's
supposed
to
be
like
a
chart
of
supported
providers
and
what's
the
level
of
support,
I'm,
not
sure
if
that's
been
updated
in
a
while.
So
maybe
this
could
be
like
a
dual
effort
there.
So.
I
A
B
Little
bit
of
a
counterpoint
just
to
flesh
out
some
of
the
conversation
there
I've
seen
PRS
recently
that
do
both
add
more
to
the
docks
in
line
and
move
things
out,
including
one
for
VMware
cloud
provider.
Pieces
are
shifting
out.
It
looks
like
in
the
process
they
dumped
half
the
data,
so
there's
there's
some
weirdness
happening
there
and
then
there's
also
been
an
issue
open
fairly
recently
that
really
caught
my
attention.
B
In
fact,
I
think
you
actually
replied
to
it
where
someone
was
very
frustrated
about
jumping
around
that
is
necessary
because
of
the
way
that
we
linked
out
in
some
of
our
other
documentation.
So
it's
something
we
might
want
to
keep
at
least
an
eye
on
and
try
to
make
the
flow
as
smooth
as
possible,
so
that
they'll
have
to
hit
six
pages
to
get
one
thing
done:
okay
and.
D
So
if
there
is
enough
value
to
have
that
content
in-house
per
se,
we're
gonna
have
to
think
about
some
mechanism,
a
process
that
says
here's
how
we're
able
to
make
sure
this
content
is
good
enough
and
if
content
goes
in
bit
rots
that
there
is
a
way
to
aggressively
say:
listen,
you've
not
kept
this
up
to
date
and
if
you
don't
get
it
up
to
date
and
and
in
a
week
or
so,
we
will
pull
it
out
of
tree.
That
was
what
we
used
to
do,
and
the
old
world
that
Chris
and
I
are
in.
D
When
we
have
third-party
providers
for
drivers
that
either
you
met
those
minimum
standards
and
you
weren't
making
us
look
bad
or
we
threatened
to
pull
you
out,
and
then
you
weren't
even
there.
So
we
should
think
about
that.
You
know
if
we're
going
to
keep
it
in
entry,
that
we
become
a
little
bit
forceful
and
people
do
pay
attention
to
that
I
when
they
were
threatening
to
pull
our
drivers.
Out
of
my
the
IBM
products
out
of
the
tree
and
OpenStack
I
I
gave
the
guy
my
cell
phone
and
she
say
you
have
any
trouble.
D
A
When
I
put
this
on
the
agenda,
I
was
thinking
more
about
the
conversation
at
the
high
level.
Do
we
is
this
something
that
we
want
to
get
specific
about
which
which
which
of
these
to
which
of
the
paths
open
to
us?
We
want
to
go
down,
and
it
sounds
like
so
I
guess
by
by
show
of
hands
if
you
are
in
favor
of
linking
to
rather
than
hosting
other
provider
platform
OS
stocks
on
kubernetes
on
the
kubernetes
website.
If
you
are
in
favor
of
instead
pointing
to
external
sources
for
that
documentation,
raise
your
hand
I'm.
D
E
E
But
then
we
can,
you
know,
pull
it
in
similar
to
maybe
not
similar,
but
like
conceptually
the
same
as
like
how
we're
thinking
about
pulling
in
Docs
from
other
organized
repos
like
so
you
know
we're
like
a
single
source
for
people
to
like
you,
these
Doc's,
but
not
necessarily
like
be
on
the
canonical.
You
know
source
where
they're
they're
kept
up
to
date.
You
know
so
I'm
just.
A
E
Well,
I
mean
I,
think
yeah,
so
that
if
you
still
applies
but
like
I
mean
we
can
put
a
disclaimer
that
says
like
this
is
sourced
from
that
location,
I
mean
and
then
then
I
think
also
there.
There
would
need
to
be
some
sort
of
notification
mechanism.
So
so,
if
a
user
has
a
problem
with
the
docs
like
and
they
you
know
they
can
hit
like
hey,
you
know,
I
have
an
issue
and
it
goes
to
the
owner
of
doc.
Not
to
us.
You
know.
A
E
A
Mean
that's
I
sure
that
I
share
that
same
and
desire
I
think
we
all
do
of
wanting
to
make
our
users
happy
and
making
sure
that
they
have
access
to
the
information
that
they
need
in
a
way
that's
going
to
make
the
most
sense
for
them,
but
I
also
look
at
the
like
the
infrastructure
required
to
import
and
maintain
and
indicate
freshness
for
external
content,
and
that's
another
moon.
That's
another
tooling
piece
that
you
would
own
and
I
think
about.
Well,
what
is
the
return?
A
I
I
would
I
would
tend
to
agree
with
something
like
that
that
you
know
that
it's
it's
you
know
if
you're,
if
you
make
a
commitment
to
to
have
a
provider
or
an
installation
that
that
you
know
I,
think
that
you're
making
a
commitment
to
to
keep
the
documentation
up
to
date.
On
that
and
to
me
it
seems
like
hosting
it
on
the
Nakuru
Nettie's
website
is
something
of
an
endorsement
of
that
and
it's
and
it
it.
It
hides
the
the
ownership
and
the
responsibility
for
that
ownership.
I
A
Thank
you.
That's
helpful.
No,
it's
just
about
a
really
good
discussion.
I
feel
like,
like
all
of
the
points
that
folks
are
raising,
are
really
salient
and
are
really
important
to
consider
so
yeah.
Thank
you.
So
what
I?
What
I'm
hearing
in
terms
of
next
steps?
Is
they
get
an
idea
of
the
impact
of
what
we're
talking
about,
because
right
now
there's
a
feeling
of
a
big
impact,
but
not
necessarily
a
quantifiable
like
a
quantifiable
amount
of
documentation
that
externalizing
would
actually
affect
so
the
the
I
will
take
an.
E
C
Will
one
and
then
I'll
add
another
one
either
in
the
Select
channel
or
next
week?
Folks
have
talked
about
not
wanting
to
fragment
the
user
experience
by
pointing
them
off
the
kubernetes
website,
but
if
someone
is
using
OpenStack
and
running
communities
on
OpenStack
they're
already
in
the
OpenStack
Doc's,
they
need
they're.
Gonna
need
to
navigate
between
the
two
I
think
that
it's
important,
if
we're
looking
at
you
know,
I
mean
their
various
getting
started
solutions
right,
but
insofar
as
any
of
them
represents
alert
on
another
ecosystem
that
the
user
is
dealing
with.
C
We
might
want
to
consider
that
as
an
issue,
it's
not
just
a
matter
of
navigating
away
from
the
kubernetes
dock
site.
In
order
to
get
getting
started
information,
then
you
navigate
back
and
I
know
that
there
are
other
unfortunate
you
X's,
you
know
into
that.
But
that's
just
my
little
comment
about
it:
cool.
A
Thank
you,
no
I
agree,
it's
a
you
know
the
the
I
think
it's
a
reasonable.
A
reasonable
thing
to
expect
that
users
who
are
working
with
different
operating
systems
will
have
multiple
docks.
That's
up
at
the
same
time
and
I
think
the
way
it
may
be.
We
may
make
users
happier
I
mean
well,
we
definitely
make
them
unhappy
if
they're
looking
at
different
dock
sets
and
going.
One
of
these
is
wrong.
One
of
these
is
out
of
date,
so.
A
A
H
J
H
E
H
E
H
And
kind
of
making
it
this
kind
of
goes
back
a
little
to
the
sorry
trans
everywhere.
This
kind
of
goes
back
a
little
bit
to
like
how
do
we
organize
Docs
that
maybe
from
external
providers
right
so
so
in
my
mind,
there
is
elimination
between
the
docs
provided
from
a
provider
that
she
flies
operating
system
there.
Docs
and
lab
provides
because
they're
all
all-in-one
solutions
that
are
are
either
operating
system
or
five
provider,
agnostic,
so
I
think
a
lot
of
those
things
and
II
mapped
out.
H
Maybe
we
should
we
shouldn't
drop
a
proposal
for
our
surname
issue
for
it
and
get
the
conversation
going
around
that,
whether
it's
to
looping
all
of
within
the
sig,
leaves
from
cloud
providers
and
get
that
conversation
going
I
definitely
still
want
to
bring
or
less
dachshund
I
think
it's
I
think
I
just
think.
It's
a
larger
conversation.
H
So
one
thing
that
I
noticed
from
some
of
the
docs
is
that
it
includes
a
little
mapping
of
the
types
of
distributions
that
you
can
use,
whether
or
not
they're
there
community
meeting
or
our
commercial,
and
we
should
get
a
table
together
of
some
suggestions.
Some
suggested
dishes
that
are
our
CKA,
take
a
route
and
go
from
there.
I,
don't
know
what
I
think.
A
That's
the
goal
with
the
so
for
that
top
level
page
and
set
up
for
picking
the
right
solution
as
being
able
to
point
to
to
certified
kubernetes
solutions
and
and
listing
those
there's
a
URL
where
those
are
already
listed.
I
don't
know
if
it
makes
sense
to
add
another
one
specifically
to
documentation
or
simply
to
to
reference
that
the
existing
link.
But
that's
that's
something
that
we
can
talk
about.
Yeah.
A
Okay,
great
just
a
reminder
to
everyone.
Our
first
annual
sigdoc
summit
is
coming
up
on
May
9th
in
Portland,
and
it's
the
day
after
right,
the
docks
and
it's
in
the
central
office
in
the
Pearl
District.
So
hopefully
you
could
just
whatever
whatever
stay,
Arrangements
you're
making
for
right
the
docks.
If
you
were
attending
that,
whatever
stay
Arrangements
you're,
making
just
carry
them
over
for
an
additional
night
central
is
really
easily
walkable
or.
A
A
No,
we
would
love
to
have
you.
Is
it
a
one
day
or
two
day?
It
is
a
one
day.
We
will
it's
not
even
a
full
day.
We
will
likely,
we
will
likely
work
and
discuss
from
9:00
to
3:00
and
then
go
out
and
do
something
fun
as
a
community
in
the
afternoon
and
evening,
as
I
know,
by
by
midday,
Wednesday
I'm
going
to
be
hashed
and
I
was
about
to
go
out
and
relax
and
get
to
get
to
know
everyone
better.
D
I
just
I
was
thinking
about
this.
We've
got
so
many
cool
things
going
on.
We
had
Joe
putting
up
a
review
document.
I
was
refreshing
myself
with
the
style
guidelines
which
are
really
really
cool,
and
so
I
do
a
lot
on
social
media.
I
do
a
lot
of
tweeting
and
you
know
I
didn't
do
a
good
job
of
getting
everybody's
a
Twitter
handle.
So
we
can
sort
of
amplify
our
message.
I
think
one
of
the
things
you'll
start
seeing
me
tweeting
out
is
I'm
going
to
start
doing.
Some
style
guide
tips.
D
So
you
know
the
the
way
the
style
guide
has
written
each
little
section
kind
of
kind
of
would
map
to
like
one
single
tweet.
You
know
so
you
know,
for
example,
you
know
using
active
voice
or
what-have-you,
so
you
might
see
something
that's
going
out
so
before
I
started
doing
that
and
then
of
course,
I
would
link
to
the
style
guide
so
just
just
kind
of
a
great
way
to
get
more
visibility
for
the
doc
work
going
on
and
I
just
thought.
D
Oh
well,
let
me
make
sure
I
get
everybody's
Twitter
handle,
so
you
know
I
always
do
hashtag,
kubernetes
I
didn't
know
if
there
any
other
hashtags
to
add.
But
if
I
had
the
right
group
of
folks
who
you
know
whoever's
tweeting,
we
could
amplify
the
message
and
then
especially
as
we
go
from
Chapel
to
Hugo,
we'll
probably
want
to
kind
of
kind
of
get
that
word
out
too.
Just
trying
to
use
these
new
modern
fancy
tools.
A
Would
be
kind
of
cool
I
have
to
be
honest,
I'm
not
much
of
a
tweeter,
so
I
mean
Andrew
I
do
do
you,
love
Twitter,.
A
It's
not
that
we're
opposed
to
it.
It's
just
I
have
to
be
really
honest.
I
wouldn't
do
anything
with
it.
Unless
Brad
was
like
house
I'd
be
like
ok,
I'll
post
it,
so
it
would
not
be
well-maintained
under
my
watch.
Mm-Hmm
all.
D
D
A
D
A
great
idea,
okay,
like
I,
said
I,
usually
do
okay,
because
you
know
I'm
going
to
an
okay
following
I
got
like
1750,
so
I
you
know
follower
so
usually
get
a
decent
worn
out
if
I
get
some
retweets
and
stuff
to
the
community.
I'm
sorry
I'll
just
keep
doing
what
I'm
thinkin.
If
we,
an
official,
walk
and
figure
out,
manage
that
Brad.
C
I
just
followed
you
my
I'm
on
Twitter,
a
fair
amount,
and
mostly
mostly
professionally,
with
regard
to
write
the
docs
and
the
intersection
of
these
two
would
be
a
really
cool
thing
to
promote
I.
Think
so.
I'm
there
with
you
I
will
I've
been
Twitter
fitful.
Lately
I'll
try
to
get
better
on
top
of
it,
at
least
with
hashtags,
okay,.
A
C
C
So
it's
it's
not
a
huge
big
deal,
but
one
thing
that
happened
this
morning
as
a
couple
came
in
without
the
Wanda
ten
mill
stone,
so
I
learned
two
more
things
about
query
and
github
and
but
I'm
you
know
if
if
people
could
just
keep
an
eye
out
for
1.10
things
as
we
reach
merging
PR
deadlines,
this
Friday
I'd
appreciate
it.
I
am
trying
to
stay
on
top
of
that
multiple
times
a
day
at
this
point,
but
we've
got
a
few
outstanding
issues.