►
From YouTube: CNCF CNF Working Group 2021-03-08
Description
CNCF CNF Working Group 2021-03-08
A
A
A
A
A
A
A
A
A
A
A
Okay,
so
there's
nothing
else
on
here,
I'll,
just
move
on
jump
right
in
the
pull
request.
Some
of
these
are
already
open,
so
I
think
we've
might
be
able
to
close
them
out.
The
first
one
was
around
the
governance
for
the
voting,
and
it
looks
like
we've
just
really
quickly
think
everything
may
have
been
addressed
here.
A
A
A
A
A
A
A
A
A
I
think
that's
where
this
goes
in
right
now,
we're
not
saying
it's
a
problem
for
so
red
hat
and
ibm
could
have
membership
and
both
could
have
tech
leads,
but
if
it
was
dominated
by
both
and
there
was
only
representation
from
those
that's
the
only
time
this
would
come
into
play
and
it's
also
something
where
everyone
could
go.
Well,
we're
happy
with
how
things
are
being
led.
D
Right,
but
from
a
charter
perspective,
that's
actually
not
how
that
reads.
Your
description
is
fine.
It
just
says
in
the
event
of
a
question
of
company
membership,
so
therefore
anytime,
somebody
could
raise
an
issue
of
company
membership
and
bring
this
up
as
an
issue.
It
doesn't
have
to
be.
You
know
the
scenario
you're
describing
based
on
how
you're,
how
it's,
how
you
just
read
it
or
how
it
reads.
B
F
D
G
Setting
this
up,
I
think
the
spirit
of
this,
though,
is,
is
primarily
to
prevent
things
like
like
it's
easy
to
create
a
c-corp
that
is
a
wholly
owned
subsidiary,
and
if
you
wanted
to
to
basically
try
to
overwhelm
a
board
or
similar,
you
could
easily
establish
a
wide
number
of
these
things
and
then
say:
look
every
single
one
of
them
is
a
separate
entity.
I
need,
I
need
to
be
able
to
put
a
person
in
for
for
each
one,
because
I
paid
the
memberships
on
each
on
the
flip
side.
G
Like
I
think,
the
the
examples
you
gave
would
be
operating
in
good
faith,
but
simultaneously,
if
you,
if,
if
there
was
a
a
thing
where
you
had,
where
the
votes
significantly
mattered
across
the
industry,
it's
easy
for
an
organization
to
set
up
and
and
subvert
some
of
some
of
these
controls,
basically
by
creating
a
whole
bunch
of
subsidiaries
as
well.
So
I
think
the
language
is
put
in
place
to
to
help
act
as
a
trigger
to
prevent
that
kind
of
that
kind
of
thing.
So
it
could
be.
E
C
Well,
to
be
clear,
the
chairs
are
not
from
service
providers
and
from
industry
they're
defined
by
what
you
do
not
where
you
come
from.
So
if
you're,
a
service
provider
who
writes
your
own
cns,
you
have
the
option
of
putting
your
name
up
for
two
chairs
and
since
we
all
use
kubernetes,
then
you
have
the
option
for
putting
your
name
up
for
all
three.
C
The
question
is
probably
more
about
the
consequences
of
what
you're
talking
about.
If
red
hat
and
ibm
were
considered
to
be
one
company,
then
they
would
only
get
what
I
think
we
said
like
30
of
tech
leads.
Didn't
we
or
something
like
that?
It's
you
know
not
as
consequential
as
it
sounds
for
them
to
be
considered.
One.
E
E
So
then,
I
do
think
it's
there
needs
to
be
a
hard
requirement
that,
like
one
company,
is
only
allowed
to
own
one
chair,
and
then
I
hear
where
shane's
coming
from.
But
I
would
worry
that,
like
you
know,
top
level
executives
could
come
down
and
try
to
steer
things,
and
even
if
you
run
as
a
separate
company,
you
could
still
have
pressure
financial
pressure
executive
pressure,
et
cetera,
coming
from
on
high
to
try
to
force
an
agenda
into
this
group.
D
But
I
mean,
let's,
I
think,
we're
being
unrealistic
about
the
impact
of
this
group.
If
you're
talking
about
you
know
to
that
degree
right
I
mean
what
I
mean.
It
is
what
it
is.
If
the
group
wants
to
do
it
that
way,
I
just
think
it.
It
unfairly
impacts
people
who
have
you
know
because
ibm
bought
red
hat
now.
Red
hat
is.
E
Yeah,
I
don't
think
I
have
a
clean
answer
for
this
shane,
like
I
said,
my
worry
is,
is
you
know
red
hat
wants
to
push
tosca
or
ibm
wants
to
push
tosca
other
groups?
Don't-
and
you
know
ibm
says
everybody
on
ibm
side,
everybody
on
the
red
hat
side,
push
tosca,
I'm
not
saying
I
am
for
against
tosca
by
the
way.
E
E
One
of
our
big
fears
is
that
you
know
a
single
vendor
would
come
in
because
they
have
the
capital
and
try
to
dominate
I've,
seen
it
in
other
open
source
groups
where,
like
you
know,
a
a
vendor
specific
model
is
chosen
as
the
quote:
unquote
open
model
just
because
they
have
the
staffing
and
the
resources
to
flood
a
group
and
make
it
happen.
You're,
probably
right
that,
like
this,
is
getting
a
little
bit
overzealous
at
this
point
and
whatnot,
but
I
don't
know
I.
A
D
B
Yeah,
it's
totally
fine
to
say
that
publicly
we
we
really
are
red
hat,
has
its
own
interests
and
they
they
can
conflict.
Maybe
red
hat
and
ibm
aren't
the
best
example
of
this.
Actually,
I
think
we're
focusing
on
that
particular
duo,
but
but
that
is
a
deal
that
I
don't
think
will
be
problematic
in
that
respect.
I
think
we're
more
thinking
about
the
examples.
The
change
set
of
telco
subsidiaries
and.
D
Red
hat
and
and
and
ibm,
I
can
tell
you
from
you
know,
working
in
the
telco
space,
for
example,
I
mean
the
telefonica
example,
I
think
is
actually
perfect,
because
while
they
share
a
corporate
structure,
they
operate
independently
and
have
historically
been
very
vocal
about
operating
independently
and
having
different
opinions
in
fact
have
fought
against
each
other
in
in
standards
bodies.
So
that
makes
maybe
a
more
interesting
and
and
not
personally,
conflicted.
D
D
D
Well,
no,
it's
evaluating
the
independence
of
the
subsidiary,
not
whether
or
not
the
subsidiary
should
have
a
vote
right,
it's
very
different
actually,
so
this
is,
if
there's
a
need
to
val
to
evaluate-
and
this
is
sort
of
what
somebody
else
was
mentioning
right-
is
that
what
we
need
to
evaluate
is
the
independence
of
the
subsidiary.
Not
the
organization
gets
to
this.
The
group
gets
to
decide
that
that
doesn't
give
that
there's
a
conflict
and
doesn't
get
a
vote.
It's
is
the
end
of
you
know.
D
H
H
H
D
G
I
I
agree,
but
it's
it's
also.
It's
also
in
terms
of
defending,
and
I
think
we
need
a
case-by-case
basis
on
something
like
this,
like
I'm
much
more
comfortable
with
ibm
and
red
hat
than
I
would
be
if
you
had
some
company
that
made
subsidiaries
that
were
designed
specifically
to
to
influence
influence,
a
group,
and
so
so,
I
think
that
there
needs
to
be
a
way
to
differentiate
the
both
of
them.
D
B
I
I
I
want
to
add
some
of
my
own
two
cents
here,
so
I
I
I
agree
with
book.
Basically,
you
know
we're
not
a
standards
body
here.
So
if,
if
we
don't
have
a
good
reputation
and
good
legitimacy,
then
people
will
just
not
adopt
us.
You
know
we're
all
spending
a
lot
of
man-hours
here
on
this
group
and
people
will
just
leave
if
the
group
is
seen
as
somehow
dominated
by
a
specific
company
and
not
open
enough.
So
so
there's.
This
is
an
important
discussion,
because
there
is
a
lot
of
stake
here.
B
G
Yeah,
that's
why
I
mentioned
like
short
term,
I
don't
think
it
matters
long
term.
I
think
it
could
matter.
A
lot
is
because
once
it's
developed
that
reputation
and
then
it's
easy
to
to
try
to
push
these
type
of
of
things
forward,
though
I
I
do
think
that
having
the
and
and
that's
that's,
even
even
on
the
on
the
community
side
like
you-
could
have
a
company
that
that
literally
says
you
know
what
I'm
going
to
bring
in
100
people
here.
So
they
can.
G
And
so
so,
I
think
there
are
some
controls
that
that
are
present
in
some
of
the
standards
body
above
now,
because
these
things
have
happened
in
the
past
and
so
there's
there
have
been
things
put
in
place
to
help
defend
against
them,
and
so
I,
like,
I,
don't
think
it's
gonna.
I
think
it's
it's
not
going
to
be
important
in
the
in
the
short
term,
but
it
is
something
that
we
should
be
that
we
should
be
mindful
of
moving
forward.
E
Can
I
tell
devil's
advocate
for
a
second
too
and
just
post
a
counter
argument?
One
thing
I
don't
want
us
to
like
lose
sight
of
is
participation
is
super
important
and
if
we
kind
of
isolate
certain
companies
because
of
like
mergers
and
this
and
that
participation
potentially
drops
outside
of
the
rare
extreme
case
too,
where,
like
a
company,
spends
up
multiple
subsidiaries
to
try
to
like
swing
a
vote.
E
Typically,
there's
not
like
going
to
be
like
you
know
five
companies
under
one
umbrella
that
we
would
have
to
worry
about
to
where
they
have
that
big
of
an
impact.
It's
probably
going
to
be
something
like
a
dell
in
vmware,
a
red
hat
and
an
ibm
I
mean
it's
is
that
one
vote,
because
this
now
may
your
major
corporate
entity
has
two.
My
big
fear
is,
and
this
is
something
that
service
providers
on
my
side.
E
Well,
I
put
that.
I
use
that
example
in
the
chat.
I
don't
know
if
you
saw
it
and,
like
I
said,
softbank
is
weird
right,
because
softbank
doesn't
typically
get
in
to
I
mean
they're
they're,
one
of
the
few
examples
where
they
have
the
capital
to
actually
like
make
an
army
of
subsidiaries.
I
mean
soft
bank
owns,
tons
and
tons
and
tons
of
stuff.
E
They
also,
though,
don't
typically
they're,
usually
like
they're,
usually
like
interested
in
equity.
Right
like
owning
a
stake
in
a
company,
getting
returned.
It's
it's
way
more,
like
investment
banking
than
it
is.
You
know,
I'm
going
to
merge
tech
companies
and
try
to
lead
them.
So
I
mean
I
mean
we,
I
think
you
know
shane.
You
mentioned
that
we
should
have
something
specific
to
subsidiaries
to
make
sure
that
people
aren't
trying
to
like
cheat
the
system,
but
I
mean
I
would
ask
this
group.
E
You
know
the
votes
comment
about
like
you
know
how
severe
is
this
is
if
a
company
you
know
two
companies
merge
and
now
there's
a
potential
that
two
votes
would
swing
in
one
company's
favor.
Does
that
have
the
impact
of
say
something?
Like
you
know,
a
charter
or
an
at
t
coming
with
you
know,
200
developers
from
inside
our
cto
office.
You
know
and
every
single
one
of
us
casting
a
vote
for
you
know
what
we
think
is
best
for
charter
or
atm
t.
I
I
want
to
go
back
to
what
tao
said.
I
think,
there's
a
almost
a
self-destructive
or
self-regulating
effect
here.
If
40
people
from
charter
will
come
in
and
influence
this
group,
I
guess
we
will
all
step
away
from
this
group
and
take
our
business
elsewhere
or
our
knowledge
elsewhere.
So
I
think
there
is
reason
for
concern,
but
I
think
at
this
time
we
prefer
to
encourage
participation
and
if
down
the
road,
we
get
it
a
situation
where
one
company
throws
in
dozens
of
people.
I
H
B
I'll
add
even
more
to
that,
I
think
if
this
group
is
going
to
be
succeed
and
have
some
legitimacy,
it
will
happen
by
consensus.
That
is,
if
we
ever
get
to
a
point
where
we're
really
fighting
over
something-
and
there
are
two
sides
to
it.
Unless
we
find
a
way
to
compromise,
I
mean
this
will
reflect
the
technical
problems
in
the
community
right,
so
so,
in
the
end,
this
is
a
place
where
we're
all
coming
together
from
different
perspectives,
exactly
in
order
to
be
able
to
work
together.
B
A
A
A
D
D
At
all
I
mean,
wouldn't:
wouldn't
the
idea
be
that
some
tech
leads
are
voted
in?
The
real
issue
is
in
how
the
voting
takes
place,
not
in
who
the
tech
leads
or
or
chairs
are
so.
Instead
of
putting
a
limitation
on
who
the
tech
leads
or
chairs
are
we
have
to
figure
out
how
we're
going
to
handle
40
people
from
one
organization
being
involved
in
that
organization
having
more
power
to
vote
than
others
not
who
the
people
are
and
who
the
roles
are.
D
And
other
places
in
cncf
are
handled
by
the
fact
that
there's
only
one
vote
for
per
cncf
membership
here
we're
saying
that
anybody
who's
involved
gets
a
vote
and
that
in
and
of
itself
creates
a
different
problem.
So
I
don't
think
we
have
a
problem
with
limiting
co-chairs
from
one
company
organization
to
to
one
per
company
or
tech
leads
to
one-third,
because
that,
if
people
think
there's
a
conflict
of
interest,
they'll
handle
that
in
the
voting
process.
D
A
We
there
there's
some
language
around
that
that
was
already,
I
guess,
starting
in
the
voting
pr,
but
we
don't
have
all
of
that
in
place
yet
so
there
will
be
a
similar
thing
with
regard
to
representation,
so
that
you're,
if
we're
saying
best
practice,
then
there
will
we're
going
to
have
to
have
similar
rules
on
on
representation
from
a
group.
But
we
don't
have
to
try
to
address
that
right
now.
G
My,
I
think
part
part
of
this,
though,
is
that
I
really
like
the
wording
that
they
have
in
kubernetes.
It's
like
they
from
the
tech
lead
side.
It's
that
there's
a
measure
there
to
try
to
work
out
the
independence
of
of
the
organizations,
and
I
think
that's
really
the
key
behind
it,
because
this
is
something
that
you
cannot
resolve
just
with
verbiage,
there's
a
spirit.
G
D
G
The
reason
why
is
that,
when,
when
you
have
people
voting
both
of
them
are
concerned,
they
absolutely
are,
but
when,
when
you
have,
if
you're
voting
for
a
particular
lead,
there's
also
there's
there's
two
problems
that
appear
so
the
first
one
is
that
if
you
have
them
all
in
the
same
company
structure,
then
you
have
a
single
chain
of
command.
That
says
this
is
how
we're
going
to
vote
so
as
opposed
to
you
have
to
convince
multiple
companies
like
you
still
have
to,
but
the
technologies.
G
A
D
I
guess
that's
my
point.
I
think
we
need
to
resolve
some
of
these
other
issues
before
you
can
put
limitations
in
because
they're
they're
cascading
right
they're.
Not
this
is
we're
we're
dealing
with
these
issues
in
a
vacuum,
but
they
in
fact
we
don't
even
we're
saying
what
the
we're
putting
in
place
policies
and
what
tech
leads
can
be
from
one
company,
but
we
don't
even
have
a
job
description
for
the
tech
leads
yet
right,
because
we
don't
know
exactly
what
they're
gonna,
how
what
impact
they're
gonna.
A
How
do
we
decide
on
updates
for
important
things,
and
probably
the
two
most
important
are
best
practices
that
say
is
going
to
be
the
most
important
and
then
charter
updates.
So
if
we
want
to
make
changes
to
the
charter
like
we're
doing
right
now,
then
how
many
approvals
do
we
need
to
move
it
forward?
Part
of
this
is
just
trying
to
get
pull
requests
to
not
sit
there
forever.
D
A
Right
now,
the
res,
so
we
what
we
don't
have
in
here-
and
this
is
again
a
discussion,
not
a
pull
request.
We
need
to
add
that
other
verbiage
around
for
these,
so
how
how
much?
How
do
we
get
a
pull
request
in?
Well,
it's
probably
the
same
thing.
We
don't
want
a
pull
request.
I'm
sorry
best
practice.
We
do
not
want
to
best
practice
to
be
adopted
fully
and
say.
H
I
I
hear
from
the
standardization
by
the
veterans
who
did
a
lot
of
battles
and
and
their
relevant
concerns,
but
in
this
case
I
think,
if,
if
we
are
producing
best
practices
and
then
assume
that
there
are
standards,
this
will
probably
not
work,
so
there
might
be
for
the
same
thing,
more
best
practices
that
will
get
merged.
So,
even
if
somebody
wants
to
dominate
something
with
whatever
there
might
be
alternative
thing,
that
will
also
be
a
good
practice
or
best
practice.
I
H
I
mean
at
the
end
that's
correct.
I
fully
subscribe
to
that
at
the
end,
we
want
to
probably
live
the
spirit
of
of
cncf
and
and
support
and
and
accelerate
the
the
user
end
user
driven
development.
So
we
need
to
see
who
would
be
the
users
of
the
best
practices
and
what
do
they
have
to
say
on
on
what
we
do,
because
by
this
group,
saying
something
it
will
not
mean
that
the
end
users
will
adopt
immediately.
H
So
we
will
be
measured
here
by
the
adoption
and
by
by
influencing,
in
the
positive
terms,
the
community
and
the
broader
community
to
start
adopting
those
practices
to
start
putting
them
in
the
rfps
say.
Yes,
I
want
that
because
it
makes
sense
for
me
and
so
on
if
it
goes
in
the
other
direction,
as
you
said
that
the
working
group
will
not
produce
anything
useful
and
then
it
will
not
be
adopted,
I
think
there
is.
G
C
I
think
the
problem
would
come
if
we
got
a
good
reputation
and
it
was
subverted
for
what
it's
worth
and
that's
more
the
issue
right,
you're,
absolutely
right.
We
wouldn't
get
adoption
if
this
was
a
lopsided
group
and
more
to
the
point,
even
if
it
was
a
lopsided
group.
If
it
came
up
with
best
good
recommendations,
it
might
still
see
adoption
but
bringing
this
back
to
the
core
of
the
question.
C
C
D
A
C
So
how
about
this
right?
There's
two
parts
to
that?
One
is
the:
are
you
the
same
organization
and
does
it
matter
where
we
could
say
that
a
member
could
raise
a
point
of
concern
that
two
entities
are
the
same
organization
based
on
the
reality
that
there
is
some
ownership,
and
then
the
community
decides
whether
they
want
to
act
upon
that
or
not.
C
G
We
we
could
set
it
up
as
well.
Yeah
like
what
is
the
definition
of
special
election
in
that
scenario
like
if
it's
literally
the
let's
say
that
three
was
the
limit,
but
he
had
five
reps
due
to
a
merger,
and
so
maybe
the
special
election
could
literally
be
among
these
five
people
and
exclusively
these
these
five
people,
who
would
community
like
to
to
to
keep
involved
granted
that
doesn't
solve
your
problem
entirely
with
what?
If
what,
if
one
side
of
it
ended
up
being
more
than.
C
Right
because
then,
okay,
so
so,
firstly,
shane
if
we
did
those
two
things
right.
Firstly,
someone
has
to
raise
it
as
a
concern
and
then
the
community
gets
to
decide
what
the
action
is
and
secondly,
that
if
we
are
over-represented
and
we
need
to
fix
it-
then
there's
a
special
election.
Would
you
be
happy
with
that
and.
C
D
And
just
to
be
clear,
I'd
be
fine
with
the
it's
not
automatic
right.
There
has
the
second
piece
a
special
election,
I'm
not
as
concerned
with,
but
just
bringing
that
in
the
event
of
a
question
to
be
more
definitive
where
it
says
you
know
in
the
second
paragraph
and
the
question,
an
event
of
a
question
to
be
more
definitive
that
it
has
to
be.
Somebody
has
to
raise
it
as
a
question.
C
Yeah
me
would
solve
these
and
somebody
one
person,
perhaps
or
two
people
with
standing
have
to
raise
it
as
a
question
when
it
is
then
debated
and
we
decide
whether
or
not
as
a
majority
decision
to
take
an
action
on
it.
I'm
good
with
that.
Thank
you,
taylor.
How
do
you
feel
about
writing
that
language.
C
Let
me
let
me
work
with
you
offline
and
provide
all
right
wording
for
what
I've
just
said.
I
think
we
could
do
both
of
those
things.
I
don't
think
we
end
up
with
a
charter,
that's
overly
more
complex
or
difficult
to
understand
than
the
one
we
have
right
now,
and
I
think
I
don't
think
we
end
up
with
the
worst
result
than
we've
got
we're
just
basically
putting
stages
in
place
up
front
rather
than
having
to
decide
it
later
on.
When
it's
going
to
be
a
bit
more
awkward.
G
G
A
B
And
you
know
we
can
also
change
this
document
in
the
future.
I
like
to
think
that
everything
we
do
here
is
is
a
living
document
yeah,
we
might
be
bugged
on
this
too
much.
I
think.
I
think
we
we
all
just
want
to
make
sure
that
we're
starting
off
on
the
right
foot
here
and
everybody
feels
confident.
I
I
feel
confident
by
the
this
discussion
itself.
It
shows
that
everybody
is
thinking
very
deeply
about
the
legitimacy
of
this
group
moving
forward
and
we
want
it
to
be
influential.
B
D
And
yeah
just
taylor,
just
your
point,
I
I
tend
to
tend
to
agree
with
you,
but
when
you
work
for
a
company
like
red
hat
who's,
a
subsidiary
of
somebody
like
ibm,
my
concern
is
not
ibm
in
red
hat.
My
concern
is
ibm,
goes
out
and
buys
somebody
else,
and
then
red
hat
loses
its
voice
right.
That's
my
biggest
concern.
A
I
I
would
so
shane,
I
guess
I
what
the
way
I'd
lean
into
this
is,
if
we're
getting
an
overwhelming
number
of
groups
that
are
coming
in
and
they're
all
trying
to
participate
and
contribute,
and
they
go
we're
all
owned
by
ibm,
but
we
all
work
independently.
I'd
love
to
deal
with
that
problem.
We
go.
Let's
go
update
the
charter,
because
everyone
in
the
community
says
all
these
groups
are
helping
so
much
I
I
would
rather
have
that
problem
where
red
hat
goes.
A
Look
there's
too
many
other
groups,
so
we
can't
be
on
and
we've
been
contributing
and
we
go
update
it
versus
we're
going
along
focusing
on
other
things,
and
we
don't
recognize
that
someone
that
had
a
bad
intention
versus
red
hat,
trying
to
contribute
someone
else
they
came
in
and
we
didn't
have
the
protection.
So
that
was
my
my
one
side
on
bias.
A
E
I
think
the
big
thing
we
should
be
striving
for
is
finding
ways
to
increase
participation,
while
also
protecting
the
reputation,
because
I
I
think
we
keep
talking
about.
Nobody
will
use
this
if
our
reputation
is
bad,
but
the
people
who
tend
to
show
up
on
these
calls
are
invested
in
this,
and
I
think
the
other
fear
right
is.
E
A
Whole
thing
about
open
source
right
time,
y'all
we're
at
time
please
go
check
out
the
use
case,
template
so
that
we
can
get
this
closed
out,
opened
this
one
and
there's
some
things
being
split
out
from
this,
but
most
of
it's
been
resolved.
So
please
give
your
feedback
here.
So
we
can
close
this
one
out
and-
and
I
think,
if
you
had
a
different
deployment
use
case-
please
link
so
that
people
can
read
that
and
then
maybe.