►
From YouTube: CNF WG Meeting 2021-08-16
Description
CNF WG Meeting 2021-08-16
C
D
D
D
E
D
D
B
D
D
D
D
A
Taylor-
this
is
oliver,
I
am,
I
think,
I
kind
of
missed
when
you
were
well.
I
was
just
too
slow
to
respond
when
you
were
asking
about
adding
to
the
agenda.
I'm
just
looking
here.
Sorry,
I
was
curious.
If
we
had
this
last
week
we
had
some
conversations
and
I
think
we
kind
of
had
looking
for
it
next
week.
Yeah
like
progress,
I
mean
there
was
some.
A
A
You
know
about
the
potential
for
some
sort
of
high
level
ambition,
slash
goal
slash,
you
know,
maybe
not
objective,
but
I
think
we
all.
I
think
we
generally
know
why
we're
here,
but
I
think,
as
far
as
having
some
kind
of
vision
to
when
do
we,
when
would
we
like
to
have
some
things
done
along
the
way
that
kind
of
gives
us
a
sense
of
whether
we're
making
that
progress
or
or
if
we're
kind
of
just
standing,
still
and
doing
circles?
A
D
A
If
you
want
what
we
could
do
is
when
we
get
to
this
there's,
there
is
actually
a
discussion
that
was
started.
I
think,
back
in
march,
I
I
threw
something
out
there.
I
think
we
just
had
a
lot
going
on.
I
think
ian
might
have
commented
on
it,
and
I
I
just
you
know,
pushed
it
again
with
sort
of
to
ian
saying:
hey.
Is
this
something
that
maybe
we
could
revisit
and
consider?
A
A
D
D
B
D
D
D
D
D
But
we
do
want
to
end
up
with
a
whole
set
of
best
practices
that,
whether
you're
ops
in
an
ops
team
or
a
developer,
you
could
come
and
say
here's
a
set
of
best
practices
and
which
ones
are
useful
for
me,
and
that
could
be
something
there
may
be
we'll
not
just
make.
There's
cncf
will
end
up
with
a
some
type
of
set.
A
subset
similar
to
the
kubernetes
conformance
program
is
a
subset
of
ed
tests.
So
there
will
be
some
set.
D
A
la
carte,
I
guess
for
whoever's
needing
them,
so
if
you're
implementing
great
you're,
building
the
certification,
cool
you're,
not
saying
and
trying
to
look
for
something
whatever
it
is,
we're
trying
to
build
a
set
right
now,
but
we
can
come
back
to
like
some
specific
ambitions,
oliver
and
that
when
we
look
at
that
discussion,
but
so
if
you're
coming
in
the
cnf
work
group
and
you're
saying
where
do
I
contribute?
There's
been
comments
about
this
like?
How
do
I?
How
do
I
know
where
to
get
started,
or
where
can
I
write
stuff?
D
I
think
we
probably
need
to
put
more
in
a
like
a
getting
started
or
it
could
be
a
link
off
the
contributor
or
maybe
a
section
of
the
contributor
guide,
because
this
stuff
will
get
lost
and
it
might
have
been
mentioned
like
a
year
ago.
But
nobody
knows
oh,
we
have
this.
So
that's
what
this
about.
So,
where
can
you
contribute
the
best
practice
proposals
adding
for
any
of
this?
You
can
get
in
an
ad
and
that
might
be
async
right
in
to
content.
D
D
Other
people
are
as
well
so
best
practice
proposals
the
use
cases
and
then
I
think
an
important
thing
is
we
do
want
supplemental
content,
that's
why
I'm
calling
it
supplemental,
so
the
exceptions,
implementation,
examples
tal!
You
point
out
a
lot
of
the
stuff
where
it's
okay?
How
do
I
actually
start
using
this
in
practice?
D
I
liked
the
concept,
but
how
do
I
use
it?
So
we
need
that
content.
I
don't
think
we're
going
to
have
one
document
that
has
everything
it's
probably
going
to
be
a
whole
set
and
then
it'll
expand
and
someone
will
go.
Oh
here's,
my
environment,
so
all
of
that
stuff
is
good
and
if
that's
where
you're
passionate
or
where
you're
needing
to
fix
things,
then
we
want
that.
D
That's
all
related
to
the
context
of
why
something
would
be
useful
and
how
to
use
it
and
then
general
updates.
You
know
if
there's
people
are
coming
in
a
lot
of
you
all
have
contributed
this
way
in
making
things
clearer
in
the
documentation
use
cases
everything
else.
Spelling
grammar
is
always
important,
and
then
I
think
a
big
one
is
reference
links.
So
we're
talking
about
a
lot
of
stuff
but
being
able
to
go
and
read
like
where
did
it
come
from
those
resource,
links
and
stuff?
D
So
writing
up
existing
content.
So
whether
you
want
to
draft
work
on
a
draft,
that's
existing
or
new
content,
we
want
to
add
the
references
to
the
external
content.
So
if
it's
just
something
that's
an
idea,
that's
fine,
probably
better,
for
the
discussion
board,
but
eventually
we
need
to
have
enough
references
that
it's
clear,
that
this
is
a
good
practice
or
that
it's
a
common
use
case
so
that
we
can
make
decisions
on.
Does
this
work
for
me
like?
How
does
it
apply?
D
Don't
feel
that
you
have
to
do
new
content
so
trying
to
make
sure
that's
understood.
If,
if
you
feel
like
you
can
contribute,
but
maybe
not
start
from
scratch,
then
we
can
expand
on
existing
stuff
and
it
may
turn
out
that
it
expands
enough
that
we
need
to
split
it
off
into
something
new,
and
that
would
tie
in
with
this,
like
anything
where
we
feel
like
a
purpose
or
something
is
large
enough
in
itself
that
we
can
dig
into
probably
should
be
split
off,
and
then
we
have
smaller
chunks.
D
So
we
can
complete
opening
pull
requests,
of
course,
to
add
new
or
to
existing
content.
So
then,
I
think
one
of
the
other
items-
and
I
see
your
comment
here
tell
on
hackmd-
is
like:
where
can
we
work
so
the
the
github
repo
itself?
You
can
put
pull
requests
in
against
existing
content
and
the
repo
itself,
and
we
can
continue
to
iterate
on
that,
but
modifying
or
doing
a
direct
commit.
D
D
But
if
you
want
actual
collaborative
editing,
then
we
want
to
support
that
and
we've
created
different
places
to
do
that.
So
in
the
the
telecom
user
group
the
cncf
drive,
we
actually
have
a
a
subfolder
called
cnf
working
group
and
you
can
drop
content
in
here.
So
this
could
be
the
supplemental
content,
new
use
cases
whatever
else,
and
this
will
be
accessible
to.
Whoever
has
a
link
to
that
and
you
can
modify
that
and
we
have
a
hackmd.
D
D
D
If,
if
you
want
access-
and
you
don't
have
access,
then
ping
myself
I'll,
try
to
make
sure
like
jeffrey
and
ian
and-
and
some
other
folks
also
have-
can
make
sure
and
add
people,
but
you
should
have
a
capability
to
go
in
there
and
work.
You
feel
free
to
go
use
whatever
you
want,
I'm
just
talking
about
stuff
that
we've
tried
to
set
up
for
the
group.
D
So
if,
if,
if
you're
wanting
to
use
another
markdown
editor
or
some
other
sharing,
google
docs
type
thing
by
all
means,
do
it,
we've
used
it's
diagramly,
it
had
a
different
name.
I
don't
remember
what
that
was.
I'm
just
gonna
click
go
here,
yeah
whatever
this
is
a
diagramming
software
that
supports
collaborative
editing,
draw.
I
o.
I
think
that
was
what
it
was
originally
called.
Now
it's
diagramly,
but
you
can.
If
you
want
something
besides
say
a
a
google
presentation,
then
you
can.
D
And
share
these,
and
this
actually
can
go
into.
I
don't
know
if
you
saw
it,
I
kind
of
went
through
quickly,
but
you
can
have
it
saved
to
google
drive,
so
you
could
actually
select
like
the
cnf
working
group,
drive
and
then
have
the
diagram
saved
right
in
there
and
then
they
can
be
opened
and
modified
by
other
people
easily.
D
D
So
we
can
keep
adding
here
and
I
think
tal.
One
of
the
things
that
you
were
promoting
was
that
the
discussions
were
not
lost,
like
the
importance
of
the
discussions
happening
in
the
prs
would
continue
like
if
people
are
wanting
to
talk
about
stuff
so
bringing
those
over
into
the
of
discussions
so
that
they
can
be
ongoing.
We
also
slack
in
the
working
group
and
stuff,
but
for
a
collaboration.
D
D
B
D
D
A
Yeah
I
mean
there's
well,
let
me
just
say
two
things.
One
is.
This
is
fairly
old,
so
this
as
you
see
it's
march,
so
I
had
just
at
the
time
when
we
were
you
know
we
were
already
several
months
into
it.
Just
trying
to
for
myself.
Think
about
you
know,
I
think
even
bill
was
you
know,
early
days
we
were
saying
well,
maybe
by
the
summer
or
you
know,
by
kubecon.
We
might
have
you
know
our
first
set
of
best
practices.
A
So
you
know
I
was
just
kind
of
trying
to
play
out
with
the
idea
of
something
lightweight.
You
know
just
to
try
to
give
us
some
some
some
goals,
I'm
not
sure
that
the
right
way
to
do
this
is
to
have
a
you
know.
I
think
we
need
to
either
agree
or
or
disagree
to
have
one
and
then,
if
we
do,
maybe
we
work
on
it.
You
know
offline
because
I
I
think
it'll
be
more
efficient
way
of
working,
but
you
know
the
first
question
is
then
really
do
we
feel
like
we
do.
A
A
I'm
not
sure
if
it's
just
myself,
but
I
sometimes
start
to
lose
the
picture
of,
for
example,
we
you
know,
we've
been
very
focused
on
generating
use
cases.
We
have
some
use
cases
that
are
now
you
know
been
approved,
and
you
know
I
don't
know
how
many
more
we
need
right.
I
know
we
need
more.
I
don't
the
point
is
what
at
what
point
do
we
are?
We
able
to
start
to
extract?
A
Will
we
know
it
when
we
get
there
and
everybody
starts
to
feel
comfortable
or
is
it
something
that
we
can
already
say
today
that
we're
going
to
need
at
least
20
30
use
cases
before
we
can
start
to
extract?
So
it's
a
little
bit
of
what
are
we
trying
to
achieve?
But
it's
also,
how
are
we
getting
there
with?
You
know
just
a
a
high
level
view
for
for
people
who
may
not
really
fully
follow
the
steps
that
were
that
were
taken.
Does
that
make
sense.
B
Yeah,
it
makes
a
lot
of
sense
to
me
as
long
I
I
like
that
you
called
it
ambition
rather
than
roadmap,
maybe
because
it's
going
to
be
evolving,
what
we
don't
need
is
a
gun
to
our
head,
telling
us,
you
know,
get
out
those
best
practices
by
this
date
to
an
extent
it'll
be
ready
when
it's
ready,
but
you
know
we
we
need
to
work
towards
something.
So
we
we
can
keep
this.
B
You
know
I
kind
of
wish.
Sometimes
we
have
so
many
ways
to
get
to
contribute,
but
I
feel
like
the
the
most
natural
way
for
us
at
this
stage
is
something
like
a
wiki.
You
know
it
could
be
google
docs
or
hackmd
or
other
things,
but
you
know
this
isn't
a
discussion,
but
to
an
extent
it
should
be
linked.
B
You
know
up
front,
so
anybody
coming
to
the
work
group
will
immediately
be
able
to
see
the
roadmap
as
long
as
you
make
it
clear
that
this
is
the
roadmap.
Is
something
that's
evolving
and
you
know
we
can.
We
can
keep
updating
it
when
we
feel
we're
not
going
to
reach
that
target
right
now.
So
we'll
just
move
it
forward.
A
Yeah
I
agree
with
utah
as
well.
I
mean
it
and
then
I
was
struck
as
you
can
see.
Even
when
I
wrote
I
was
like
you
know,
struggling
to
call
this
a
road
map
and
ambition.
I
mean
the
point
is
just
to
simply
say:
why
am
I
here?
What
am
I
trying
to
accomplish
until
we
change
it?
You
say
you
know:
okay,
q1,
q2
q3.
This
is
what
we're
hoping
for
and
we
can.
You
know
we
can
always
adjust
it,
but
I
think
it
at
least
allows
us
to
also
reflect
you
know.
Wow.
A
A
B
Yeah
and
to
the
point
you
made
olivier
earlier
that
you
know
when,
when
are
we
actually
going
to
have
enough
use
cases
to
do
something?
Well,
we
already
do.
I
think
even
one
use
case
should
be
enough.
The
the
trick
is,
somebody
needs
to
take
the
initiative
to
say.
Okay,
I'm
gonna
tackle
this
and
start
writing
the
best
practices
for
for
this
use
case,
and
maybe
we're
just
being
timid
about
starting
with
it,
but
I
think
I'm
I'm
always
in
favor
of
just
start
sneezing
things
on
paper.
B
Just
just
start.
Writing.
Nothing
is
set
in
stone.
Part
of
the
reason
why
I
was
not
so
happy
with
the
github
process
of
writing
things
that
everything
feels
just
very
formal,
and
you
know
you
do
a
pull
request
and
then
there
are
different
people
proves
there
seems
to
be.
The
the
barrier
of
contribution
so
far
has
been
a
little
bit
difficult.
B
But
if
you
know,
if
you
feel
you
have
something
to
say
about
a
use
case,
you
can
just
start
a
google
map
on
google
docs
and
add
a
lot
of
to-do's
in
it.
Nothing
has
to
be.
You
know,
I
don't
think
any
of
us
is
expecting
like
a
complete
thing,
like
the
12
factor
website,
that
they
wears
out
right
now,
right
we're
going
to
work
towards
it,
but
we
we
we
should
nothing
is
stopping
us
from
starting
it
right
now
with
just
one
use
case.
B
B
Yeah
an
example
I
give
you
know
the
work
I
did
on
cnf
operators
right.
Definitely
I
need
to
follow
up
on
that
and
and
turn
it
into
something
more
integrated
with
our
work,
but
it
helped
me
a
lot
just
to
start
writing
and
give
me
the
the
topics
you
know
the
the
list
of
of
points
which,
which
are
could
be
pain,
points
could
be
points
of
discussion
of
interest
of
controversy.
D
D
So
if
you
feel
like
where
you
can
contribute
is
more
use
cases,
we
can
keep
adding
use
cases
and,
if
there's
a
use
case
where
you're
identifying
a
best
practice,
then
we
can
write
that
and
that's
related
to
you
know,
tells
saying
you're
saying
we
it's
when
people
step
up
to
that.
So
that's
interest,
you
know,
need
whatever
whatever's
driving
you
for
that.
If
you
feel
like
it
should
happen,
but
you
can't
figure
it
out.
D
That
would
be
where
maybe
you
can
reach
out.
The
principle
of
least
privilege
was
the
the
reason
why
ian
and
I
ended
up
going
this
path
is
we
had.
There
was
a
a
pretty
big
set
of
discussions
already
applying
the
principle
of
least
privilege,
and
we
felt
like
we
could
start
here.
So
there's
a
discussion
around
this
and
there's
best
practices
that
are
listed
around
around
the
principle
of
least
privileges,
there's
already
practices
that
are
out
there
and
including
the
one
that
we
ended
up
putting
forward.
D
So
the
don't
use
root
in
containers
as
a
general
best
practice
and
that's
already
out
there.
So
something
that
we
could
talk
about
and
we
feel
like
getting
to
this
list
of
things
is
what's
important
in
our
mind
right
now,
but
that
doesn't
mean
that
we
want
to
stop
use
cases
or
anything.
So
oliver,
as
far
as
like,
how
do
we
get
to
something
I
think
is
you're
wanting
to
it
seems
like
you're
wanting
to
get
to
something
past.
D
A
set
of
you
know
like
thousands
of
use
cases
which
could
be
interesting,
but
maybe
not
getting
to
a
usable,
a
directly
usable
goal.
So
that's
kind
of
my
driver
as
well-
and
I
think
within
thinking
about
this
and
talking
through
as
well
as
reaching
out
to
other
folks
like
what
could
we
do
so
it
seemed
like
the
least
privilege
as
far
as
ambitions
or
the
words
around
this
roadmap.
So
ambition
sounds
good
and
I
think
we
could
do
milestones.
D
D
So
if
we're
like,
we
have
this
no
root,
I
don't
know
if
there's
an
issue
for
it,
but
you
can
add
them
and
we
could
say,
as
a
this
says,
2021
best
practices,
but
we
could.
You
literally
have
like
a
q3,
and
I
I
personally
am
have
a
goal
of
trying
to
move
forward
some
set
of
some
number
of
best
practices
in
q3
and
q4
to
have
them
in
place
with
enough
content
that
we
go
all
right.
We
understand
it.
D
Maybe
they
don't
all
go
through,
but
I
would
like
to
get
a
set
and
they
may
or
may
not
come
from
existing
use
cases.
There
may
be
new
use
cases
that
are
added
just
based
on
my
experience
and
then
tal
you
were
mentioned.
So
this
is
kind
of
a
side
from
goals,
but
how
do
we
move
forward
on
stuff?
And
I
I
I
definitely
don't
just
go
into
github
and
go?
D
I
must
follow
the
structure,
so
my
just
my
self
tell
I
I
don't
find
it
works
for
me,
as
far
as
like,
I
guess,
the
creative
process
and
and
sourcing
a
lot
of
information.
I
need
to
have
it
a
little
less
structure,
it's
fine
to
have
some
categories,
but
the
when
I'm
working
by
myself
or
collaborating
with
someone
else
like
ian
or
whoever
we
get
on
and
start
working
on
something
this
is
the
hackamd.
That's
had
several
iterations.
D
If
I
go
back
through
the
version,
it
started
with
less
of
the
categories
and
then
we
started
putting
more
and
more
and
moving
stuff
around,
but
the
original
document
and
we're
actually
gonna
end
up
with
a
a
document
I
think
that'll
be
added
to
the
this
drive
was
a
lot
looser,
so
tal.
I
think
this
would
go
with
what
you're
talking
about.
Where
is
saying?
Okay,
so
why
would
you
want
least
privileges?
So
this
is
a
little
bit,
I'm
giving
you
all
a
little
bit
of
the
view
of
how
it
works.
D
For
me,
it
doesn't
have
to
work
for
you
and
pals.
You
know
talked
about
how
he's
worked
through
stuff,
but
whoever
you
know
y'all
on
this
call
or
listening
later.
If
you
have
another
way
of
a
workflow,
that's
fine!
You
know
the
the
github
proposal,
format
that
we
have
for
best
practices
is
kind
of
a
finish
format,
but
that's
not
the
working
process.
D
That's
when
we
get
there.
We
want
to
try
to
hit
some
stuff
that
we
think
is
important
as
a
group,
but
getting
to
there
can
be
all
over.
So
this
would
be-
and
we
I
can
highlight
here
so
why
would
you
want
least
privilege
at
all
and
starting
to
talk
about
this?
So
this
is
just
more
of
conversational.
This
doesn't
have
to
do
with
those
categories,
different
ways
of
helping.
Well,
you
can
see
here
privileged
containers,
so
this
is
not
the
root
in
containers.
D
I'm
sorry
least
privilege,
so
that
principle
of
least
privilege
so
root
in
containers
would
be
one
of
them
access
to
the
kubernetes
api
and
what
you
can
do
there.
Frederick
has
talked
about
the
different
things
you
can
do
with
like
net
admin
and
how
you
can
affect
the
whole
network
and
potentially
cause
massive
problems.
D
So
these
are
a
lot
of
topics
and
we've
had
a
lot
of
different
conversations.
This
is
how
we're
working
on
it.
You
may
need
a
literal
whiteboard,
whatever
it
doesn't
really
matter
at.
You
can
do
whatever
on
that
and
then
eventually,
you
start
just
like.
If
you're
doing
a
draft-
and
you
have
an
outline
and
then
you
start
working
through
it,
we
eventually
started
dropping
in
more
of
an
organization,
and
that
was
the
hackmd
and
then
eventually
we
felt
like
okay.
D
D
Like
I
said
in
in
in
the
github,
where
we
say
what
you
know,
three
practices
could
be
a
milestone
and
then
we
have
another
set
or
whatever,
but
yes
for
having
some
milestones
and
an
ambition
and
then
backing
up
what
tal
said
as
far
as
like
having
we
don't
want
to.
Have
people
feel
you're
forced
to
do
this
and
you
must
ask
time
off
or
whatever
that's
you
know,
it's
we're
trying
to
balance
we're
we're
contributing
in
this
open
community.
D
D
Something
like
I
don't,
and
we
can
always
list
this
in
a
dock.
I
don't
think
we
have
like
milestones
and
we
don't
have
a
ambitions,
roadmap
or
any
type
of
dock
right
now
in
the
repo
you
know
it's
in
the
closest
thing
we
have
is
in
the
governance,
saying
that
the
co-chairs
can
help
build
a
roadmap,
but
the
there
is
no
actual
thing
listed.
But
what
do
you
think
about?
D
A
B
I
have
one
idea
that
you
know
we
have
a
lot
of
participants
here,
some
that
don't
speak
up
speak
out
quite
as
often,
but
maybe
everybody
can
give
themselves
homework
to
to
pick
some
kind
of
contribution.
It
could
really
be
anything
as
you
said
it
could
be,
fixing
grammar
or
things
on
existing
documents
or
commenting
or
contributing
to
discussions
actually
just
going
to
the
discussions
page
choosing
something
that
you
have
something
to
say
about
and
responding
is
is
a
contribution
I
feel
because
they
do
move
us
forward.
B
It's
easy
to
get
lost
in
the
discussions.
I
think,
but
they're
still
there
for
reference
right
whenever
we
end
up
working.
If
somebody,
for
example,
decides
to
go
to
hackmd
and
start
working
on,
you
know
bgp
issues.
Well,
here's
a
discussion
that
they
can
immediately
reference
and
actually
continue
commenting
on.
You
know
the
discussions
don't
have
to
stop.
They
can
continue
growing
in
a
way
yeah.