►
Description
The OKD Working Group's purpose is to discuss, give guidance to, and enable collaboration on current development efforts for OKD, Kubernetes, and related CNCF projects. The OKD Working Group includes the discussion of shared community goals for OKD 4 and beyond. Additionally, the Working Group produces supporting materials and best practices for end-users and provides guidance and coordination for CNCF projects working within the SIG's scope.
https://okd.io
A
All
right,
folks
welcome
to
the
okd
working
group,
documentation,
subgroup
meeting
for
july
26th
of
the
year
2022
and
we've
got
a
relatively
light
agenda.
I'll
drop
the
link
in
the
chat
here
and
we'll
spend
maybe
30
seconds
taking
a
look
at
if
there's
anything
that
you'd
like
to
change,
move
around.
A
Please
do
put
your
name
in
the
attendees
section
just
so
that
we
know
that
you
were
here.
A
B
I'll
I'll
talk
about
it
when
I
get
to
the
getting
started
page,
but
I
I
want
to
talk
about
I've
got
a
to
do
about
doing
those
two
articles
to
guides,
but
when
I've
looked
at
the
guides,
my
question
is:
should
we
actually
get
rid
of
the
guys
all
together
or
rewrite
or
cre-create,
which
I
think
is
more
linked
to
getting
started
so
but
then.
A
Okay,
well,
we
will
get
to
that
when
we
get
to
it.
So
actually
I
can.
Let
me
add,
to
let
me
move
around
something.
So
I'll
put
this
under
technical
documentation.
A
A
That's
a
table
of
the
operators,
that's
pretty
snazzy,
and
so
my
quick
question
for
folks
here
is:
is
there
any
value
for
maybe
surfacing
that
page
by
wrapping
it
in
in
a
general
sort
of
hey?
This
is
how
it
works.
Type
five
paragraph
document
about
openshift,
slash,
okd
and
have
a
link
to
that
page.
That
says,
by
the
way,
here's
the
list
of
the
specific
operators
that
you
know
we
just
provided
an
overview
of
a
simple
diagram,
maybe
like
I
feel
like
there's
nothing
in
openshift.
B
I
was
that's
what
I
was
hoping
that
the
yeah
I
mean.
I
started
the
technical
section,
but
I
mean
I
think
the
overview
of
the
technical
section
should
be
exactly
that.
I
put
some.
I
threw
some
stuff
up
there,
which
were
random
thoughts
and
collections
that
I'd
gathered,
but
it's
not
really
proper
curated,
which
is
why
it's
got
that
under
construction
everywhere.
But
I
see
the
overview
page
should
be
that
sort
of
introduction.
C
I
I
I
keep
trying
to
find
a
raise
the
hand
button.
Every
time
I
come
back
to
blue
james,
it's
terrible.
I
decided
like
raising
my
hand
in
meetings
now
that
we
can
lift
and
shift
from
the
overview.
That's
in
the
docs.okd.io.
A
C
Yeah,
I'm
I'm
wondering
between
what's
in
just
not
that
I
don't
want
to
write
new
content,
because
you
know
hey,
we
all
want
to
write
more
stuff.
Is
that
either
on
the
openshift
landing
page
on
the
corporate
landing
page
for
openshift
there's
some
stuff,
but
actually
don't
quote
me
on
this
and
it's
being
recorded.
I
really
don't
like
that
content
that
much
it
doesn't
really
explain
what
we
do
from
you
know
from
a
basic
point
of
view.
C
D
A
Yeah,
that's
the
one
that
someone
earlier
this
morning
pointed
us.
B
I
I
think,
I
think,
the
challenge
the
challenge
is:
it's
the
getting
started
an
overview
for
a
user,
but
I
think
what
we're
talking
about
is
somebody
that
wants
to
start
getting
a
little
bit
more
technical,
a
bit
under
the
covers
and
work
out.
Oh,
if
I
want
to
change
that,
I
need
to
go
look
at
this
gate
repo,
so
I
mean
for
me
this
is
an
overview
for
somebody
who
wants
to
go
a
little
bit
deeper
than
just
being
a
user
of
openshift.
They
want
to
start
looking
at
the.
A
So
I
see
it
as
both
I
see
it
is
also
for
folks
that
want
to,
because
once
I
had
to
start
debugging
things
like
machine
config
operator,
and
you
know
all
of
once,
I
had
to
start
debugging
when
things
went
wrong.
Knowing
that
there's
an
operator
and
there's
pods
associated
with
that
operator,
really
helped
me
out
a
lot
and
in
fact
we
triage
a
lot
of
questions
in
the
chat
that
our
folks
don't
know
that,
there's
an
operator
that
handles
this
and
there's
a
pod.
You
know
one
that
comes
up.
A
Alert
manager
like
folks,
didn't
realize
that,
okay,
that
there's
the
the
logging
operator
and
then
there's
the
alert
manager,
pods
controlled,
you
know,
and
so
that
kind
of
stuff
is
sort
of
where
I'm
coming
from.
So,
in
addition
to
development,
also
sort
of
being
able
to
troubleshoot
at
a
at
a
level
more
than
just
hey,
it
doesn't
work.
Can
someone
tell
me
like
how
to
make
it
work.
B
B
Then
we've
got
the
operations
you
root,
which
is,
as
you
said,
it's
it
doesn't
work.
Or
why
doesn't
this
do
what
I
think
it
should
do
so
then
there's
that's
right
classes
operations,
but
then
I
think
we've
got
the
developer
where
they
want
to
go
that
one
step
deeper
and
maybe
start
looking
at
code
and
rebuilding
or
patching
things,
and
I
mean
am
I
wrong
there
or
would
people
think.
A
C
So-
and
I
just
put
the
link,
what
I
think
I
hear
you
saying
is
something
like
the
kubernetes,
what
is
kubernetes
page
and
then
at
the
bottom
of
it.
If
you
look
at
it,
there's
a
section,
that's
called
what's
next
and
it
takes
a
look
at
kubernetes
components
and
ready
to
get
started,
and
I
I
always
go
back
to
the
kubernetes
source.
Personally,
because
it's
it's,
you
know
they
do
a
really
good
job
and
they
have
a
huge
community.
C
But
I
I
think
that
structure
of
you
know
what
it
is
like
a
page.
That
is
what
is
openshift
from
or
what
is
okd
probably
is
better
way,
and
then,
in
paraphrase
from
the
kubernetes
structure
here
with
adding
what
is
the
flavor
of
openshift
stuff
in
there
and
then
the
components
breakdown
is,
I
think
the
thing
that's
the
interesting
bit,
because
that's
when
you
start
getting
into
the
operators
and
the
things
that
you
might
want
to
tweak
so
yeah
and
and
need
to
know
about
in
order
to
to
build
it
so.
A
And
I
could
see
links
at
the
bottom
that
are
like
develop
on
okd
or
administer
okd
or
or
whatever
something
like
that.
That
would
be
then
links
that
could
go
to
like
more
specific
documentation
for
each
of
those
spaces.
A
B
Yeah,
I
was
gonna
say
I
think
there
is
something
there
in
terms
of
yeah
how
it
works.
B
B
I
think
we
need
to
start
trying
to
say
well,
let's
collaborate
and
different
people
can
pull
things
together
and
start
working
a
bit
more
as
a
community
and
and
getting
multiple
people
to
contribute,
and
maybe
someone
put
us
like
a
an
outline
or
a
straw,
man
sort
of
in
the
in
the
thing,
and
then
we
use
them
the
github
to
actually
curate
it
maybe
create
a
branch
in
the
okidi.io
regis
repo
that
we
can
then
finesse
it
and
then
eventually
publish
it
when
we're
happy
with
it.
C
I
I
like
the
idea
of
putting
something
that
we
can
collaborate
on
in
github
brian.
I
think
that's
get
a
base
out
there
and
then,
while
we're,
you
know
even
just
amongst
ourselves,
making
it
it's
better
than
a
google
doc
for
getting
this
done,
and
then
we
can,
you
know
once
it's
done,
we
can
get
people's
approval,
push
it
live
and
your
question
jamie.
Where
would
we
put
this
on
the
main
landing
page
for
okd,
where
I
would
put
it
the
big,
bold
lesson.
A
Well
then,
I
think,
let's
move
forward
with
that,
we'll
create
a
discussion
item
for
it
and
create
a
page
for
it
and
start
chopping
away
at
it.
This
is
not
a
time
sensitive
thing,
but
ultimately
I
think
it
would
really
help
us
to
have
something.
That's
like
hey.
This
is
what
it
is
and
generally
how
it
works.
C
I'll
take
and-
and
I
will
volunteer
to
take
a
step
up-
first,
pass
at
it
from
the
high
level.
What
is
okd
riffing
off
the
kubernetes?
I
o
structure
there,
stealing
from
everywhere.
I
can,
then
you
can
all
hack
on
it,
buy
them
by
the
next
dog.
D
D
No,
the
other
guy
who
doesn't
attend
anymore
because
he's
so
busy
he's
in
raleigh.
A
D
A
A
A
D
D
Oh
the
other
thing
to
talk
about
operators
and
how
to
you
know
kind
of
introduce
them
and
get
people
used
to
them.
What
I
found
and
it
works
for
me,
is
you
know
you
find
something
you
give
an
example
like
one
of
the
first
things
you're
going
to
do
when
you
create
an
open
shift.
You
know
cluster
is
do
your
your
authentication
operator
because
you
got
to
be
able
to
have
people.
Log
in
you
know,
walk
that
through
and
people
can
go
and
say
you're
going
to
change
this.
D
A
Yeah-
and
I
I
think
also
it-
I
think
it
was-
was
it
you
mike
that
that
in
the
chat
mentioned
to
someone
about
if
your
author,
if
your
authorization
pods
aren't
coming
up
on
an
install
someone
just
messaged
in
the
chat
a
couple
days
ago
and
within
the
past
couple
days
about
yeah
a
common
problem
with
an
install,
if
you
don't
see
your
auth
pods
come
up
in
your
console,
pods
come
up
or
something
they
made
a
good
point
about
like
this
is
because
of
the
operator,
and
I
don't
remember
who
it
was
I'd
have
to
look
back,
but
you
know
someone
made
a
comment
about
that.
A
I
think
okay,
somebody
did
but
it's.
I
think
that
was
very
valuable
information,
because
one
of
the
biggest
sticking
points
in
an
install
is
getting
the
auth
pods
up
and
then
the
console
is
always
dependent
on
that
and
in
so
many
failed
installs.
I
see
those
cos
failing
and
the
respective
pods
failing.
A
B
B
A
C
A
All
right,
so,
I
think
we're
moving
on
now
to
the
repository
move.
Are
there
any
updates
diane
for
in
terms
of
the
dns?
No.
C
And
that's
it's
kind
of
freaking
me
out
a
little
bit
because
nobody's
responding
to
anything.
So
I
at
I
was
hoping
by
this
morning.
We
would
have
something,
but
after
each
time
we
have
this
docs
meeting,
I
ping
them
again
and
I'm
not
so
I'm
a
little
frustrated.
So
I'm
just
saying
that
out
loud
and
I
will
do
the
same
again
after
this
meeting.
There
hasn't
been
anything
as
well
as
with
the
mx
conversation,
the
mail,
the
mailing
thing,
so
I
don't
think
it's
caught
in
legal
limbo.
B
B
As
so,
if
you
need
people
like
vadim
or
christian
to
be
around
but
at
10
a.m,.
C
C
A
Okay,
all
right,
so
is
there
anything
else
in
regards
to
the
timeline?
Oh
sorry
was
it
did
someone
else
have
him
again.
A
Oh
okay,
I
do
remember
seeing
a
quote
or
not
a
quote,
but
a
quip
somewhere.
That
was
something
like
european
vacations
are
like
yeah,
I'm
gonna
be
going
camping
for
two
months.
Please
leave
a
message
I'll
get
back
to
you
and
then
the
united
states,
one
is
like
yeah
I'm
having
kidney
surgery
tomorrow,
but
I'll
have
my
cell
on
me
so
feel
free
to
call.
A
It
depends
on
where
you
work,
I'm
I
think
lucky
in
that
regard.
Maybe
someone,
okay,
so
is
there
anything
else
for
repository,
move,
timeline
and
steps?
Brian?
Is
there
anything
else
you
want.
B
A
Getting
started
paige.
Well,
we
sort
of
covered
some
of
that
stuff.
Anything
else
you
want
to
add
brian
or
anyone
else.
B
I
mean
the
the
one
thing
I
mean
I've
been
looking
at
the
I'm
tasked
with
getting
the
two
guides
into
blogs
and
what
I've
been
doing
is
I've
been
looking
at
the
guides
and
I'll
see
if
this
is
gonna
work
screen
three.
B
A
B
If
I
go
to
the
guides,
so
I
mean
the
two
that
I'm
actually
is
is
sort
of
vadim
and
sri,
which
is
basically
here's
a
bunch
of
hardware.
And
then
we
run
this
software
and
again
here's
a
bunch
of
hardware-
and
I
run
this
software.
But
if
you
actually
go
through
that's
just
a
list
that
one
there
might
be
something
in
the
wrapper
script
and
the
prerequisites
for
for
the
upi
again.
There
might
be
something
useful
there,
and
this
is
now
crash
my.
B
But
if
you
look
at
a
lot
of
them,
there's
not
actually
much.
I
mean
that's,
not
a
guide.
A
B
That's
not
a
guide,
that's
not
a
guide,
so
I'm
just
wondering
the
the
single
node
one
is
out
of
date
and
it
points
to
a
blog.
But
again
that
doesn't
work
with
the
current
one.
So.
D
A
I
am
working
on
updating
my
vsphere
one,
and
I
do
think
that
that
is
a
useful
guide
if
anyone
thinks
differently
feel
free
to
say
so.
I
just
you.
These
are
the.
B
C
I
would
almost
archive
the
ones
that
you're,
you
know
pointing
out
that
have
so
little
there's
a
there's
another
word
for
it,
not
archiving,
but
just
there's
legacy
or
something
just
the
the
only
value
that
it
has
right.
Now,
in
my
humble
opinion,
is
it
lists
out
the
potential
places
so
people
you
know,
can
see
potential
deployment
approaches
and
if
we're
going
to
do
another
hackathon,
it
gives
us
a
little
structure
still
for
asking
people
to
update
like
in
a
hackathon
on
testing
and
deployment
on
these
other
configurations.
C
But
I'd
almost
put
a
a
header,
that's
archived
or
I
I
can't
or
there's
another
one
when
things
get
decommissioned
deprecated.
Thank
you.
C
I
knew
it
began
with
a
d
and
I
couldn't
think
of
what
it
was
defecated
and
just
not
that
deprecated
support,
but
deprecated
that
they're
out
of
date
or
something
and
so
like,
put
a
subheading
and
then
put
all
the
ones
that
you
think
are
like
really
lacking
in
any
depth
and
out
of
date
under
that
and
just
as
a
holding
pattern
and
then
even
maybe
put
in
like
a
general
surgeon's
warning
box,
saying
if
you're
currently,
you
know
doing
deployments
on
there,
click
here
and
you
know,
will
help
if
you're
willing
to
write
up
your
your
guide,
click
here
and
go
for
it
kind
of
thing.
E
Okay,
I
mean
like
this
is
part
of
the
reason
I
stopped
by
today,
because
I
saw
that
getting
started
guides
thing
there
and
you
know,
as
the
original
author
of
several
of
those
like,
I
think,
like
over
the
year,
you
know
since
they've
been
transferred
and
we've
been
migrating,
like
we've,
we've
gotten
bifurcated
on
some
of
the
initial
intentions
by
which
those
were
created
and,
interestingly
enough,
the
ones
that
seem
least
interesting
now
are
the
ones
that
have
actually
kept
up
in
terms
of
their
accuracy
so
like.
E
When
we
originally
created
those
guides,
we
were
trying
to
create
them
as
like
inventories.
That
would
give
people
an
idea
of
what
kind
of
resources
they
would
need
to
create
a
cluster.
So
if
I
was
just
getting
ready
to
do,
okd
and
I've
got
azure,
I
could
know
well,
I'm
gonna
have
to
get
six
machines
and
they'll
have
to
be
these
sizes
and
I'll
have
to
have
some
networks.
E
E
I
think
some
of
this
depends
on
what
we
want
to
have
here
for
the
guides,
because,
in
my
opinion,
the
installation
material
we
have
in
our
in
our
product
docs-
and
you
know
the
okd
docks.
Those
are
really
good
installation
guides
if
you're
doing
ipi
or
even
limited
upi.
So
the
question
is:
what
are
we
trying
to
tell
people
here
with
these
guides?
You
know
like
I
can
totally
get
behind
dropping
the
ones
that
are
just
like
inventories
at
this
point.
E
If
we're
not
going
to
make
it
clear
that
hey
this
is
just
an
inventory
for
you
to
get
an
idea,
whereas
something
like
vadims
or
shrees.
You
know
those
are
like
here's
how
I
did
an
installation
in
my
lab
or
whatever
so
the
question
is:
are
the
guides
meant
to
be
like
stories
from
users
about
how
they
didn't
install
on
their
bespoke?
E
B
And-
and
I
think
that's
that's
the
perfect
question
mike-
is
we
don't
know
what
they
are
they're
they're
sort
of
they're
not
enough
to
help
somebody
actually
do
it,
but
but
we're
calling
them
guides.
Now
to
me
a
guide
is
going
to
help
me
do
something
so
whether
these
are
just
example
of
people's
installs
to
give
people
an
idea
that
you
don't
need
a
machine
room.
You
people
have
have
done
this
in
in
their
homes
and
and
their
example
in
stalls
are
example
setups,
then
that
helps
people.
B
B
No
one's
sure
what
they're
going
to
get
when
they
look
at
the
next
page,
because
they're
all
inconsistent
they're,
all
very
different.
So
I
think
the
question
is:
what
should
they
be
and
what
do
we
need?
I'm
not
saying
we
delete
everything
and
I
think
you're
right.
If
I
want
to
go
and
install
on
azure,
the
documentation
is
quite
good.
Now
it
tells
you
exactly
what
resources
you
need
and
it
walks
you
through
the
install
so.
C
A
Yeah,
I
was
that's
actually
literally.
What
I
was
just
going
to
suggest
is
like
is
there
a
reason
we
don't
just
create
a
subheader
home
lab
examples
and
then
just
have
those
under
the
home
lab
example.
So,
for
example,
I'm
actually
going
to
be
buying
some
hardware
to
start
doing
bare
metal,
okd
testing,
because
I've
got
access
to
all
the
clouds,
but
I
don't
really
have
a
lot
of
experience.
I'd
like
to
see
what
other
people
have
bought
or
are
using.
So
I
could
assemble
something
like
that.
So
I
would
love
to
see
something.
A
That's
like
home,
lab
examples
that
someone
could
be
like.
Like
you
said,
brian,
like
oh
yeah,
you
don't
need
a
cloud
to
do
this.
You
don't
need
a
big
room.
You
know
a
co-location
facility
or
anything
like
that.
You
can
just
do
this
in
your
home.
Does
that
sound
like
a
reasonable
direction
for
these
ones
that
are
not
actual
guides?
A
B
I
mean
I,
I
think
I
think
that
that
makes
sense
to
sort
of
have
it
a
section
like
like
home
labs,
where
people
can
actually
just
either
link
off
to
their
own.
If
they've
got
their
own
blog,
where
they've
written
them
on
their
own
blog,
they
don't.
We
don't
have
to
actually
have
it
here,
but
just
a
link
of
either
on
our
site
or
linked
up
to
other
places
where
people
can
see.
B
Okay,
so
you've
used
the
free,
vmware
esx
and
someone
else
has
used
overt
and
somebody
else
might
have
gone
pure
bare
metal
and
someone's
using
three
intel
nukes
and
the
raspberry
pi,
while
someone's
actually
using
an
intel,
xeon
and
server
that
they
bought
off
ebay
or
whatever.
E
I've
got
a
slightly
different
way
of
looking
at
this
because,
like
the
guides
to
me
that
the
notion
of
guides
like
doesn't
really
make
sense
there,
and
especially
when
we
bring
into
the
fact
that,
like
we're
talking
about
the
age
of
these
things-
and
it's
like-
if
I
create
like
to
me
these-
this
is
more
like
success
stories
and
I
think
I
might
have
misheard
someone
who
said,
I
think
someone
might
have
said
home
labs.
But
I
heard
home
runs
and
it's
like
to
me.
These
are
like
captured
success
stories.
E
Right,
like
I,
was
able
to
do
an
installation
on
these
10
old
laptops
that
I
found
in
the
garage
and
here's
how
I
did
it
right
like
because
vadim
and
sri,
like
no
one's,
coming
back,
to
update
these
things,
to
say:
oh
well,
I
did
this
originally
with
4.7
and
now
we're
at
4.12
and
here's
what's
changed
right,
so
like
their
usefulness
in
terms
of
their
their
ability
to
be
fresh
and
kind
of
tell
people
what's
happening.
Now.
E
If
I
go
look
at
the
success
stories-
and
I
see
oh
wow
jamie
did
this
with
4.8
and
he
bought
a
bunch
of
old
laptops
from
his
neighbor
for
five
bucks,
a
piece-
and
I
was
thinking
about
doing
something
similar,
maybe
it'll
work
with
4.12.
So
now
it's
not
the
expectation
that
I'm
gonna
follow
your
guide.
You
know,
step
by
step
and
end
up
at
the
same
place.
It's
more
like
here's,
someone
else
who
did
something
interesting.
E
B
E
A
Okay,
I
mean
we
can,
and
at
least
vadim.
We
could
at
least
ask
him
if
what
his
latest
version
is
and
if
that's
still
the
same
correct
setup
and
modify
it
for
him.
If
that's
the
case
right,
but
I
do
like
that
idea.
Yeah
I
mean
we
can
do
the
same
with
shri,
but
if
they're
blog
posts
at
least
yeah,
then
it's
historical
by
its
nature,
yeah
right
right.
E
You
know
vadim
too,
like
if
you
mention
it
to
the
team,
he's
going
to
come
back
and
redo
the
whole
thing
and
brush
it
up
and
whatnot,
and-
and
I
think
the
goal
here
is
not
necessarily
to
create
some
sort
of
social
pressure
or
some
sort
of
idea
that
hey
you
got
to
come
back
and
update
this
thing
or
whatever,
because,
like
that,
like
that's
what
our
documentation
is
for,
like
that's
what
the
installation
docs
are
for.
Those
are
continually
updated
right.
A
D
To
me
I
kind
of
like
you,
I,
like
the
idea
of
having
you
know
a
section
that
has
like
example,
setups
or
something
I
mean
yeah.
I
can
see
a
blog
post
saying
you
know
how
you
did
it,
but
having
something
in
there
where
somebody
can
go.
Looking
like
oh
look
at
they.
They
did
that
with
this,
I
I
think
that
would,
in
my
mind,
would
sit
better
in
a
separate
heading.
You
know
not
necessarily
as
a
how-to,
but
you
know,
as
you
know,
current
running.
You
know
examples
of
systems
that
people
have
built.
B
I
I
think
the
challenge
comes
is
as
the
product
changes,
because
I
mean
we've
got
one
of
the
guys,
which
is
a
single
note.
I
think
charo
chara
did
a
single
note
on
4.8
or
4.7,
where
he
actually
did
some
unnatural
things
to
get
what
what
what
shouldn't
have
worked
as
a
single
note
to
work
as
a
single
node
install
and
if
we
now
say
well,
that's
if
someone
tries
to
follow
that
example
with
4.10,
where
we
now
have
an
official
single
node
in
store
process.
B
D
We
need
somewhere,
maybe
they
time
out
after
a
while,
you
know
if
they
get
to
be
a
year
old,
you
know,
maybe
they
get
reviewed
and
say:
do
we
keep
it
or
do
they
need
to
be
updated,
but
and
going
back
to
the
documentation
yeah
the
documentation
is
good.
D
I
find
that
I
can
read
the
documentation
a
lot
better,
because
I
have
two
and
a
half
years.
You
know
or
more
of
experience
with
the
product.
Now
as
a
newbie,
when
I
read
that
documentation,
I
had
absolutely
no
idea
what
they
were
talking
about,
so
to
be
able
to
look
at
somebody's
setup
and
how
they
set
it
up
and
maybe
example
configuration
files.
You
know
stuff
like
that.
A
I
have
gotten
positive
feedback
on
the
vsphere
upi
document
and
the
script
that
I
wrote.
So
oh
I've
used
it.
There
is
definitely
some
value
in
that
stuff
yeah.
So
what
is
it?
Can
we
in
the
short
term,
agree
to
put
them
as
blogs
just
to
get
them
up
and
in
a
place,
blogs
tagged
as
home
lab
and
then
try
to
solve
this
issue
of
creating
something
that's
sticky
to
the
menus
and
have
a
process
of
knowing
that
they're
going
to
be
updated
or
timed
out
or
whatever.
D
E
Like
I
just
want
to
respond
to
john
for
a
second,
because
I
don't
think
I
don't
think
anyone's
arguing
against
the
usefulness
of
these
things
like
whether
they're
blogs
or
or,
however,
we
decide
to
display
them,
like.
I
think,
you're
you're
absolutely
correct
like
having
this
as
an
extra
touch
point
to
understand
how
to
do
some
of
this
stuff
is,
like
I
mean
yeah,
that's
totally
important
for
me.
E
It's
just
a
matter
of
like
building
the
expectation
that,
like
you
know
like,
unlike
our
product
docs,
which,
as
you
say
like,
can
be
very
complicated
to
read
and
take
like
you
know,
take
some
time
to
spin
up
on,
but
there
are
people
who
are
always
updating
those
things
and
there,
and
you
know,
they're
being
paid
to
like
keep
them
fresh
and
stuff
like
that
and
keep
them
up
to
date,
whereas
the
community
stuff
it's
like
you
know,
I
I
get
nervous
when
we
build
the
expectation
that
someone's
going
to
come
back
and
be
like
well,
I
looked
at
this
thing
like
three
years
ago
and
like
it
hasn't
changed
in
three
years
like
why
not
you
know
it's
like
this?
E
Is
community
stuff
right
like
so
and
which
doesn't
necessarily
mean
it
has
to
be
a
blog
or
has
to
be
displayed
somewhere
else?
I
think
that's
kind
of
orthogonal.
I
just
don't
want
to
build
the
expectation
that
people
like
oh,
like
I
gotta,
go
update,
my
my
configuration
or
whatever.
D
A
All
right:
well,
let's
get
started,
let's
get
started
at
least
with
getting
them
into
blogs
and
then
we'll
go
through
and
define
a
process
later
of
you
know
maybe
surfacing
some
to
the
top
and
then
timing
that
out
after
x
number
of
releases
or
something
so,
let's
create
a
discussion.
I
well
actually
you've.
We've
got
a
discussion
item,
I
think
for
them
already.
Brian
you've
got
that
set
up.
Do
you
want
to
do
it
all
yourself,
or
do
you
want
to
delegate
as
well
to
some
other
folks
to
handle?
I'm.
B
I'm
I'm
happy
for
anyone
to
come
and
help,
and-
and
I
think
this
is
where
the
next
the
next
item
comes
in-
because
at
the
minute
we
don't
really
have
a
way
of.
B
I
think
the
discussion
items
get
lost
because
it's
not
a
to-do
list,
but
I
think
the
next
item
is
really
where
we
need
to
start
focusing
on.
Let's
have
a
to-do
list
where
people
can
actually
go
pick
up,
because
I
think
people
are
willing
to
help
if
they
got
half-hour
spare.
B
B
We've
all
got
family
lives
and
asking
someone
to
take
on
a
big
job,
they're
less
likely
to
volunteer
where,
if
they've
got,
if
they're
doing
something,
they've
got
20
minutes,
they'll
likely
to
jump
on
and
maybe
help
a
little
bit
so
breaking
things
out
and
having
a
way
of
tracking
that
and
someone
saying
oh
yeah,
I
can
help
there,
I'm
not
I'm
not
going
to
own
it
and
do
it
all
myself.
But
I
can
help
out
there.
I
think.
A
B
If
you
don't
want
to
go
to
a
third-party
site,
I
mean,
I
know,
people
use
jira.
I
know
people
use
zen
hub
and
I
know
there
are
other
patello
boards
and
things
like
that
and
I've
been
experimenting
with
the
projects
in
github
and
they're,
not
as
feature
rich
as
some
other
places,
but
I'm
all
for
let's
get
everything
in
one
place.
I
I
hate
having
to
sign
up
to
15
things.
B
If
we
have
one
one
technology
one
place,
then
I
think
it's
easy,
so
I
don't
know
if
anyone's
actually
seen
the
project
feature
in
github
or
played
with
it
yeah
and
it
so
I
I
suggest
we
actually
create
a
planning
repo
which
actually
has
the
issues
that
are
on
the
board,
because
projects
can
pull
in
issues
from
other
other
repos.
B
You
can
create
categories,
you
can
actually
add
fields
to
the
issues
on
the
project
board
and
you
can't
have
things
like
epic,
so
there's
no
embedded,
but
I
mean
it's
just
a
simple
board.
We
can
set
the
the
columns
we
want.
We
can
set
the
fields
we
want.
We
can
set
the
labels,
we
want
so
yeah.
I'm
I'm
happy
to
to
go
along
that
route.
A
I
would
say:
yeah
we
can
go
that
route
and-
and
do
you
want
to
mike,
you
want
to
say
your
comment
over
the
air
that
you
typed,
because
I
think
that
that's
that's
valuable.
E
Yeah
sure
it
was
just
kind
of
a
general
question
I
think
like
if
we
start
to
talk
about
kanban
boards
and
like
organizing
things
into
like
what
needs
to
be
done.
What
is
done,
you
know,
what's
completed,
what's
in
progress
that
to
me
that
kind
of
implies
that
there
is
some
notion
of
like
a
temporal
like
sprint
or
scrum
or
like
there
should
be
some
regular
review
of
that
board
to
make
sure
that
it's
not
just
like.
Well,
we
just
throw
stuff
at
the
board,
and
then
you
know
every
once
in
a
while.
E
Someone
goes
in
and
moves
stuff
around
like
and
I'm
not
necessarily
saying
that
there
needs
to
be
like
some
sort
of
aggressive
timing
or
anything.
But,
like
you
know,
having
like
a
monthly
scrum
where
you
could
at
least
the
interested
parties
could
review
the
board
and
then
you
know,
reshuffle
cards
and
people
would
have
an
office
hours
to
make
their
case
or
whatever
like.
I
think
that
that
kind
of
goes
hand
in
hand
with
the
notion
of
starting
to
do
something
like
this.
B
So
this
is
the
board
review,
call
which
will
include
community
stuff
like
the
documentation,
but
this
call
primarily
becomes
the
board
handling,
call
where
those
that
have
and
again
we
need
to
think
about
who
has
authority
who
can
accept,
pull
requests
if
we're
starting
this
idea
of
having
a
catalog
and
if
we
want
to
then
have
an
idea
of
that.
We
have
some
sort
of
governance,
because
the
minute
the
governance
on
the
website
is
me,
but
if
we
want
to
have
a
process
of,
we
actually
have
make
a
collective
agreement.
B
So
my
proposal
is
that
we
actually
create
a
create
a
project
board
in
github
in
the
okd
project
repo,
and
then
this
call
is
a
priming
mechanism
for
moving
things
through
that
board
and
approving
things
and
moving
the
columns
and
closing
things
off
and
reviewing
them
accepting
pull
requests
and-
and
that's
what
this
meeting
migrates
to
now.
I
don't
know
whether
that's
in
general
sort
of
what
people
think
should
happen
or
whether
we
want
to
keep
this
purely
for
the
documentation
or
we
have
a
different
process,
an
offline
process.
But.
A
I
I
feel,
like
you've,
you've
hit
the
nail
on
the
head.
This
group-
and
we've
talked
about
this
before
actually
recently
over
the
past
couple
meetings
that
this
group
does
more
and
more.
That
has
to
do
with
like
governance
and
process
and
whatever,
and
the
main
meeting
is
really
just
sort
of
updates
from
people.
There
really
isn't
a
lot
of
decision
making
or
process
of
moving
things
forward
with
okd,
it's
sort
of
everyone
checking
in.
A
C
B
A
All
right-
and
that
brings
us
down
to
the
last
actually
that
was
the
last
item
in
the
agenda
I
didn't.
I
did.
C
I
did
sneak
one
more
thing
that
just
a
social
media
twitter
handle
stuff
in,
and
you
texted
me
while
you
were
chatting
here
earlier
and
I
just
wanted
to
let
brian
know
the
same
thing
so
I've
given
the
twitter
handle
con
username
password
to
jamie
and
brian,
so
that
I'm
not
the
bottleneck
for
tweeting
out
things
on
okd,
but
brian
I'm
going
to
also
send
you
turns
out.
You
need
the
gmail
address
that
I
set
up
as
the
new
email
address,
so
I'll.
C
Send
that
to
you
privately
so
that
you
have
that
and
then
that
triumvirate
of
us
can
handle
the
the
social
media
stuff
on
okd
like
when
you
want
to
do
need
help
on
a
guide.
You
know
here.
A
An
actual
release,
I
think,
yeah
when
vadim
cuts
a
release,
let's
get
something
out
right
away,
and
I
my
reason
for
that
is
twofold
number
one,
because
we
want
to
promote
that.
There's
a
new
okd
release.
Number
two
is,
I
think,
vadim
works,
his
butt
off
the
video
man
christian
to
get
these
releases
out
the
door
in
particular
the
team
it'd
be
nice
if
like
they
could
see
that
their
work
is
getting
promoted.
C
Is
that
we're
watching
yep?
The
only
other
update
that
I
would
give
you
is
the
the
s
cost.
Okay,
we
work
has
is
still
being
blocked
by
the
last
meetings,
peter
hunt,
but
there
has
been
a
little
nudge
forward
on
that
and
with
a
a
way
forward
for
that,
that's
an
interim
solution
for
the
packaging
up
of
the
piece
that
we
needed-
and
it
looks
like
the
release-
will
slide
to
probably
the
beginning
of
september
now
because
of
the
delay
in
the
cryo
issue.
D
B
And
it's
probably
just
worth
if
people
haven't
noticed
that
in
the
okd
project
organization
and
the
results
from
the
shift
week,
we
now
have
vladimir's
put
some
work
in
around
building
okd,
but
there
was
also
a
team
that
actually
started
looking
at
pipelines
for
building
operators
to
maybe
use
as
a
foundation
from
community
efforts,
so
they're
the
they're
two
great
projects
to
actually
kick
off,
that
repo
and
and
also
christian,
put
in
the
bass
build
it
builder
image
that
we
want
to
use
and
the
dockerfile
is
there,
so
that
that
is
another
great
thing.
B
That
was
a
mystery
before.
So
these
are
things
that
obviously
want
to
update
the
technical
documentation,
but
there
are
ways
to
that
fans
back.
There
are
ways
to
actually
that
we
can
actually
start
involving
the
community.
A
Yeah-
and
I
think
what
we
can
do
is
start
maybe
finding
little
ways
to
document
these
changes
because
sort
of
everything's
buried
in
the
meeting
videos
and
the
meeting
notes
about
a
lot
of
these
changes
that
are
happening.
I
think
we
should
write
some
little
blurbs.
I
talked
about
this
before,
and
I've
been
sort
of
formulating
some
little
blurbs.
Maybe
I'll
do
discussion
items
for
them
just
little
things
that
we
can
post
like
a
blog
post
or
something
that
hey
did.
You
know
that
there's
a
cicd
process
being
developed
too
blah
blah
blah
well.
B
A
Don't
know,
but
I
mean
like
the
fact
that
these
things
are
happening
like
news
items
to
let
people
know
that
these
things
are
happening
because
we
kind
of
look
dormant
like
we
don't
really
have
a
lot
of
activity
right.
But
if
we
say
to
the
community
by
putting
a
blog
post
up,
hey
we're
moving
towards
doing
this
or
hey.
There's
this
option.
For
you
know.
B
B
What
I'd
like
to
do
is
have
a
custom
panel
that
we
can
populate
from
markdown.
That
says
latest
news
which,
which
is
a
little
bit
flashier
than
what's
on
there
today,
but
if
we
just
have
like
a
our
latest
news
panel,
if
there's
anything
there,
that
it
gets
populated
and
that's
where
we
can
put
things
like
the
new
releases
there.
B
So
when
people
go
to
okd
to
io,
they
see
immediately
on
the
front
page,
the
latest
news,
so
it
could
be,
and
we've
updated
to
see
the
the
pipeline
or
we've
we've
added
a
new
bit
of
automation.
Then
we
can
link
off
to
the
page
that
describes
a
bit
more
detail,
but
it's
somewhere
that
they
don't
need
to
go
and
dig
through
documentation.
C
Yeah,
can
you
send
me
a
direct,
a
little
email?
I
think
you
might
still
have
brandon's
email
address
in
your
inbox
somewhere
from
prior.
B
I've
got
it
on
slack.
It's
sorry
on
yeah,
it's
in
the
in
the
openshift
slack
channel,
yeah.
C
Yeah,
so
yes
send
me
that
I
me
I
meet
with
him
every
thursday,
so
in
the
late
afternoon,
so
you'll
be
asleep
or
doing
something
better
than
this.
Hopefully,
and
then.
B
D
It's
okay.
I
just
want
to
make
sure
so
I
let
jamie
know
this
yesterday
and
I've
talked
to
vadim
a
little
bit
one
of
the
little
side
projects
that
I've
been
doing
over
the
last
week
and
a
half
or
two
weeks
is
working
on
an
external
prowl
configuration.
D
So
I
know
that
you
know
a
lot
of
what
we're
doing
right
now
is
you
know,
based
internally
on
red
hat
prowl
and
honestly,
I've
made
pretty
good
progress,
I'm
probably
about
eighty
percent
right
now
of
having
an
end-to-end
prowl
progress
process
being
able
to
build
things
like
the
installer
and
other
individual
items.
D
I
guess
the
question
is:
is:
is
this
a
worthwhile
project?
You
know
long
term.
I
know
I
heard
that
red
hat
was
looking
at
a
you
know,
a
new
potential
ci
build
process,
but
I
haven't
heard
any
details
about
it,
but
one
of
the
things
that
you
know
has
been
said
passed
is
that
prow
is
like
really
really
convoluted
and
stuff
like
that.
D
It's
not
really
that
bad
once
you
can
get
in
and
understand
what
red
hat
actually
did
because
again,
documentation
is
very
red
hat
specific.
You
know.
So
you
have
to
kind
of
read
between
the
lines
and
there's
pieces
of
automation
that
aren't
really
documented.
You
have
to
kind
of
get
into
the
internals
and
find
out
where
this
automation
came
from
so
right
now
I
have
it
running
from
a
github
repository
called
the
the
okd
experience
it's
connected
to
my
own
okd
cluster,
where
prowl
is
running
and
also
where
all
the
build
processes
run.
D
B
D
Yeah,
I'm
not
sure
proud,
would
run
on
there.
Well,
because
you
have
to
have
you
know
some
of
the
things
have
to
have
admin,
cluster
privileges
and
stuff
like
that,
and
they
probably
don't
want
to
do
that
unless
they
want
to
build
a
certain
cluster.
D
Like
I
said
right
now,
the
cluster
is
running
in
my
basement.
You
know
in
my
own
home
lab
but
connected
through
get
through
github,
so
it
works
end
to
end
with
prowl.
D
But
this
is
kind
of
this
was
an
educational
process
for
myself,
but
the
fact
that
it's
not
quite
as
and
there's
also
the
operator
builds
that
are
in
there
also.
So
we
may
be
able
to
look
at
those
operator
build
processes,
pull
them
out
and
then
run
those
externally
to
red
hat.
You
know
and
use
those
to
build
our
operator
catalog
conceivably
without
having
to
reinvent
the
wheel.
Maybe
that's
all
I
got.
I
just
wanted
to
mention
that.
E
Up
yeah,
I
just
wanted
to
respond
to
john.
You
know,
first
of
all,
it's
excellent
to
hear
what
you're
doing
there
john.
I
think
it's
totally
worthwhile
to
learn
that
and
to
bring
it
to
this
community,
regardless
of
what
red
hat
does
with
like
pipeline
builds
and
all
this
other
stuff
that's
going
on
and
is
all
cool
and
good
and
everything.
E
I
think
it
would
be
really
good
for
the
okd
community
to
know
and
kind
of
match
the
impedance
of
what
the
kubernetes
community
does
and
even
though
proud
looks
weird
and
you're
saying
it's
coming
out
of
red
hat
and
all
this
other
stuff,
and
we
do
have
configurations
there.
This
is
also
what
the
upstream
kubernetes
community
does.
They
they
created
this
tooling.
E
E
D
B
How
do
we
make
it
available
to
the
community,
because
I
think
the
idea
was
obviously
hearing
that
it's
got
to
have
admin
might
sort
of
be
a
bit
of
a
blocker
or
a
bit
of
a
challenge
if
you
want
to
go
into
the
first
community
cloud
or
or
whatever,
but
I
I
think
that
was
the
intention
of
that,
but
we
we
actually
do
have
our
own
build
system
in
there.
So
somebody
who
doesn't
have
enough
of
a
resource
to
run
maybe
a
full
build
system.
B
We
can
still
do
that
and
as
a
community,
so
I
I
think,
that's
what
we
were
hoping
to
do
and
then
this
other
challenge
was
with
pizza.
Was
it
cook
peter
peter.
C
Hunt
yeah
that
that
was
another
which.
B
D
A
I
I
don't
know
that
one
is
that
one
necessarily
surpasses
the
other
right
like
both
can
be
working
in
parallel.
John,
can
you
do
some
documentation
so
that
people
know
what
you've
got
and
people
can
possibly
experiment
and
do
the
same
thing,
etc.
D
D
Right
so
the
other
piece
that
if
we
decide
to
do
this
down
the
road
we
may
you
know
we'll,
have
to
look
at
you
know:
where
do
we
put
a
real
build
environment?
Red
hats
are
all
in
aws,
you
know
so
it
may
come
down,
you
know,
do
we
get
sponsorship
and
somebody
pay
for
an
aws
cluster
for
us.
You
know
you
know
down
the
road
or
something,
but
first
we
have
to
prove
the
technology.
A
All
right
and
so
we'll
look
forward
to
brian
setting
up
our
board
our
project
board
and
then
we'll
start
using
that
as
sort
of
the
foundation
of
these
meetings.