►
From YouTube: CNCF TOC Meeting 2023 01 17
Description
CNCF TOC Meeting 2023-01-17
Discussion around https://github.com/cncf/toc/issues/981- Health of the Notary V2 project
Discussion around https://github.com/cncf/toc/pull/991 - Flatcar proposal
B
C
D
D
So
the
first
one
is,
you
know:
we've
had
a
couple
of
turns
at
this.
One
won
a
private
Toc
discussion
last
year
and
more
recently
in
the
cncf
doc
repository
so
hi
Phil,
so
I
would
like
us
to.
You
know
talk
a
little
bit
about
it.
I
want
to
see
what
what
folks
on
the
call
think
about
this,
including
the
TOC
folks
as
well.
D
Let's
start
with
free
flowing.
If
it
doesn't
work,
then
we'll
switch
around
to
you
know,
raising
your
hands.
So,
let's
who
wants
to
get
started
Justin?
Would
you
like?
Do
we
have
two
Justin
so
Mr
capos?
Who
do
you
like
to
get
started
since
you
raised
both
the
email
and
the
issue
in
the
cncft
Repository.
E
Sure,
yes,
I
I
think
this
is
a
pretty
simple
issue
and
I
am
glad
to
see
that
that
some
folks
from
the
notary,
V2
project
did
show
up,
because
there
hadn't
been
any
responses
to
the
thread
and
there
had
been.
You
know
prodding
on
their
SLI
on
their.
You
know
slack
Channel
and
stuff
like
this
to
interact
in
some
way,
but
but
basically
the
the
project
is
entirely
different.
E
D
E
E
We
were,
you
know,
told
a
lot
of
things
were
going
to
happen.
They
were
going
to
change
so
that
we
could
participate.
Others
could
participate
freely
in
the
project.
They
were
going
to
make
changes
to
code
of
conduct
and
all
of
this
and
basically
instead
the
project
really
went
dormant
for
a
very
long
period
of
time,
with
no
visible
progress
of
anything
thing
and
I.
E
So
I
I
think
that
all
of
the
complaints
and
criticisms
are
equally
valid,
both
pre
and
post,
and
none
of
them
were
really
addressed
in
a
way
that
I
think
any
of
the
original
signatories
of
the
letter
felt
was
satisfactory
and
also
I.
I
would
say
that
I
don't
believe
that
by
us
sending
the
letter
privately
to
the
TOC,
we
believed
we
were
trying
to
do
something
private.
E
We
were
trying
to
give
people
a
heads
up,
as
I
did
with
the
message
that
I
sent
to
the
TOC
ahead
of
making
this
public,
which
was
to
just
say,
hey
here's
a
heads
up
of
what
we
want
to
say
publicly
and
we
were
encouraged
to
keep
that
message
private.
While
the
TOC
discussed
it
and
did
things
with
it
and
then
all
we
heard
months
later
was
a
response
that
was,
you
know
effectively
almost
entirely
the
opposite
of
what
we
had
all
wanted
and
thought
was
fair
and
Equitable
and
reasonable
yeah.
D
F
As
for
you
know,
what's
in
that
that
issue,
981
I,
think
you
know
as
being
part
of
that
Community
I'm,
not
a
maintainer
of
notary
V2,
specifically
but
I've,
participated
in
in
meetings.
I.
Think
that
there's
been
a
lot
of
changes
to
the
way
that
that
Community
has
been
run
and
I
would
invite
those
members
to
come
back
and
take
a
look
because
I
think
we've.
F
A
lot
of
those
allegations
in
the
in
the
issue
are
no
longer
valid
and
there
are
other
members
of
that
community
that
are
participating
in
there.
So
I'm
really
trying
to
see
what
are
the.
What
are
the
areas
we
can
improve
and
make
that
Community
better
and
see
what
we
can
put
into
shape
and
what
we've
got
now
is
some
governance,
changes
to
transparency
and
discoverability,
and
then
I
also
think
what
was
called
out
in
there
making
sure
that
our
agendas
are
always
clearly
documented.
F
That
meeting
minutes
are
always
clearly
documented
and
we
could
probably
work
with
the
cncf
to
make
sure
that
videos
are
posted,
I,
think
the
videos
are
there.
We
could
go
and
post
the
recordings
and
do
that,
but
I
suggested
that
we
go
talk
to
Amy,
because
I
believe
there's
a
tool
to
do
that
in
an
automated
fashion.
F
So
I
think
you
know
all
the
things
we
have
a
plan
for
for
the
dangling
pieces
that
need
an
action
item,
I'd
be
happy
to
discuss
them,
but
yeah
interested
to
work
to
make
it
a
better
community
and
and
help
address
any
issues
that
have
been
raised.
That's
where
I'll
stop
at
this
point
take
stems.
Thank.
G
Think
that
also
just
you
know
to
be
clear
that
you
know,
there's,
there's
lots
of
potential
components
needed
and
that
you
know
notation
is
not
a
kind
of
feel
and
all
project
that
replaces
notary
V1
in
any
sort
of
sense,
and
you
know
I
think
there's
ongoing
ongoing
work.
Looking
at
you
know,
you
know,
you
know,
we've
been
doing
internal
draft
work,
which
I
shared
with
Justin
recently
on
like
ways
we
can
incorporate
tough
into
this
ecosystem.
G
There's
a
bunch
of
I
think
about
things
like
in
Toto
and
I.
Think
that
you
know
these
things
are
not
you
know.
These
projects
are
designed
and
Nursery
project
is
designed
to
be
my
general
components
and
not
like
you
know,
complete
kind
of
Replacements
or
or
other
things
there's.
You
know,
there's
a
lot
of
work
to
do
to
actually
put
these
things
together
into
ways
that
people
can
use
them
effectively,
and
you
know
I
think
that
it's
still,
you
know
kind
of
designed
to
be
a
project.
H
Yeah
so
I've
been
doing
a
little
bit
of
Investigation
on
notary
and
notation,
and
what's
going
on
here,
trying
to
figure
out
what
happened
and
I've
learned
a
handful
of
things
that
I
don't
know
how
many
people
know
some
of
the
context.
So
you
know,
like
notary,
V1
is
still
in
use
right.
I
know,
there's
been
talk
of
it's
archived,
it's
not
in
use
anymore.
It's
gone
yet.
I
have
found
multiple
instances
where
it's
still
used
in
production
and
I.
H
Think
Docker
Hub
is
still
one
example
of
that,
so
it
hasn't
gone
away.
Code
has
been
merged
into
it
and
it
is
still
in
use.
So
I,
don't
think
that
needs
to
be
forgotten.
I
also
noticed
that
with
notary,
V2
and
notation
having
a
new
thing
that
comes
up
and
takes
life,
and
then
there
was
the
old
one.
We
already
have
examples
of
this
in
the
cncf
right
where
a
project
comes
in
with
one
way
of
doing
things
and
one
architecture
and
then,
as
things
change
get
re-architected,
the
ecosystem
does
things
differently.
H
So
to
say,
V1
versus
V2
requires
a
new
project.
Is
one
of
those
things.
We'd
really
have
to
dig
in
deep
on
I
think
because
we
already
have
these
examples
of
transformation,
while
in
the
cncf
I
also
noticed
that,
along
the
way,
notary
in
its
new
form,
is
a
collection
of
specs.
In
many
ways,
with
a
reference
implementation
in
notation
right
and
those
specs,
many
of
which
were
worked
out
with
the
oci-
the
open
container
initiative
to
make
some
of
this
stuff
work
together
and
if
I
even
understand
it
right.
H
This
involves
some
of
the
Sig
store
project
which
now
can
use
some
of
those
same
things
and
they
were
worked
out
together.
So
some
of
the
work
that
you
see
in
in
notary,
V2
and
notation,
it's
actually
been
worked
out
with
other
organizations
along
the
way
and
there
are
people
who
are
working
towards
using
this.
H
You
know
I've
gone
around
and
asked
I
said
all
right,
Sig
store,
it's
great,
I
use,
Sig
store
my
company,
you
take
store,
you
can
verify
our
container
images,
a
bunch
of
them
with
Sig
store,
but
I
went
around
and
asking
and
I
said
is:
are
there
places
where
it
doesn't
work?
And
the
answer
I've
been
told?
Is
yes
and
that's
one
of
those
areas
where
something
like
notary
and
notation
may
be
able
to
step
in
and
solve
for
because
it
doesn't
work
in
every
situation
that
you're
going
to
run
to?
H
And
you
know
that's
kind
of
normal.
We
we
have
that
kind
of
thing
in
the
Linux
foundation
in
the
cncf,
look
at
container
D
and
look
at
cryo.
We
we're
okay
with
that,
as
long
as
things
are
able
to
work
out
and
have
healthy
ecosystems,
and
things
like
that
and
so
I
wonder
where
there's
going
to
be
things
that
work
together
using
specs
and
those
same
specs
workplaces
where
certain
projects
are
going
to
carry
themselves
along
further
and
do
really
well
at
it
for
their
own
Niche
use
cases.
H
And
maybe
you
know
a
niche
can
be
a
one
percent.
Niche
can
be
a
graduated
project
because
you
can
have
enough
people
who
are
using
that
and
I'm
curious
to
see
what
projects
align
in
what
ways
and
what
users
and
their
reasons
for
that,
because
you
know,
as
we've
seen
there
isn't
always
one
size
fits
all
Argo
and
flux
graduated,
what
a
week
from
each
other
and
they
both
are
in
very
much
the
same
space.
H
D
I
J
D
E
Okay,
thank
you
for
giving
me
so
much
time
here.
I
appreciate
it.
I
I
wanted
to
mention
a
couple
things
really
quickly
in
response
to
a
bunch
of
this,
so
the
governance
problems
that
likely
mentioned
were
only
kind
of
part
of
the
problem.
It
really
wasn't
so
much
the
governance.
It
was
really
the
people
involved
with
the
governance
and
the
way
that
that
had
worked,
and
so
I
understand
that
some
of
those
personalities
have
moved
on
to
you
know
other
things
like
this
and
and
so
on.
E
Fundamentally
it
it
still
comes
down
to
you
know
this
question
of
of
like
who
created
this.
There
was
an
existing
community
of
people
that
did
something
as
there
was
in
the
case
of
other
projects
like
Argo
and
they
came,
and
that
Community
went
and
created
a
V2,
and
that
was
very
successful.
That's
a
great
starting
point,
okay,
but
this
isn't
what
happened
in
this
case.
There's
zero
of
the
original
maintainers
zero
of
the
original
Code
Zero
of
the
original
design.
E
This
is
effectively
an
entirely
new
project.
That's
come
along
and
is
promoting
itself
heavily
and
in
fact,
like
a
third
of
the
web.
Page,
if
you
go
to
the
notary,
V2
webpage
is
basically
we
are
a
cncf
project.
They've
had
no
Security
review
the
you
know
it's
there.
There
are
fundamental
problems
in
the
security
design
and
other
things
that
have
been
repeatedly
pointed
out
in
the
public
meetings
that
you
can
see.
You
know
time
after
time
that
are
unaddressed
in
the
project
and
yeah
I.
Think
we've.
D
D
So
I
would
like
to
move
to
towards
a
Cooperative
solution
to
like
what
we
are
trying
to
do
here.
So
I
want
to
take
the
things
that
you
pointed
out
in
a
positive
fashion
and
see
how,
like
we
tried
a
few
things
and
the
TOC
ended
up
asking
the
folks
to
do
something.
So
we
we're
going
to
try
doing
it
again
and
see
what
happens.
We
are
not
in
the
habit
of
you
know,
shutting
down
projects
just
because
you
know
people
don't
like
it
right
like
we.
There
is.
D
Possibly
with
totally
different
people
too,
so
I'm
going
to
take
that
input
as
the
input
all
of
all
the
things
that
you've
said
as
part
of
what
the
TOC
will
consider,
but
you
know
we
will
end
up
taking
a
bunch
of
the
things
that
we
heard
here
and
read
on
the
issue
and
whatever
issue
you
want
to
make.
You
want
to
create
we'll
take
that
into
consideration
and
and
then
come
back,
come
back
to
the
team
to
say
hey.
This
is
what
we
think
you
all
should
be
doing.
D
I
I
I
put
it
all
in
the
chat,
but
I
will
reiterate
the
thing.
I
think
the
the
thing
that
I
was
re.
Responding
to
earlier
was
Matt's
comment
about
notary
and
Sig
store
and
some
of
that
work.
I.
Don't
think
that
the
request,
as
as
Justin
laid
out
and
as
I
signed
on
to
was
that
notary
not
exist
or
that
it
like
shouldn't,
doesn't
doesn't
have
a
place
in
the
world
I.
I
Don't
think
anybody
says
that
just
that
as
an
incubating
project,
I
think
the
concern
was
that
it
would
not
meet
the
bar
of
an
incubating
project
if
it
walked
in
off
the
street
today.
The
other
thing
that
came
up
was
the
the
in
the
chat
that
the
security
review
is
a
requirement
for
graduation
I.
Think
notary
being
a
security
focused
project
with
Security
review
security
concerns
being
raised,
sort
of
raises
the
bar,
like
you
know,
notary,
should
have
a
higher
security
bar
than
I
think
the
average
incubating
project
off
the
street.
I
D
What
we've
ended
up
doing
is
we
told
the
project
to
clearly
delineate
which
ones
are
production
ready
and
which
ones
are
not
and
have
enough
language
in
the
repositories
or
in
the
sub
projects
or
in
the
different
repositories
that
it
controls
to
have
enough
guidance
in
there
so
that
people
are
not
LED
astray
by
you
know,
coming
through
one
portal
and
ending
up
somewhere
else
kind
of
thing,
so
we
have
done
this
before
and
the
guidance
the
TOC
has
given
is
hey.
Please
make
sure
that
you
document
these
things
properly
so
end.
D
H
So
I
I
want
to
address
one
thing:
you
know
that
it
would
not
be
an
independent
incubating
project
yet
because
I
I,
when
that
was
said,
I
went
back
and
looked
at
the
incubating
stage
requirements,
and
what
do
we
normally
expect
and
also,
how
does
something
shift
right
and
flux
V1
to
V2?
Looking
at
something
like
that,
right,
V2
obviously
didn't
meet
the
requirements.
Well,
V1
was
around.
H
Initially
it
had
to
pick
up
certain
things
like
your
multiple
production
instances
and
things
like
that,
and
so
when
I
was
looking
at
this
and
looking
over
it
I
couldn't
honestly
answer
all
of
the
questions,
but
it's
at
a
point
where
it
does
meet
multiple
of
the
criteria
and
some
of
the
things
we
would
have
to
do
the
due
diligence
on
to
know
whether
it
met
more
of
them.
But
there
is
a
possibility.
H
V1
still
meets
the
criteria
and
we're
in
that
transition
phase,
and
that's
what
normally
happens
when
flux
V2
came
around
people
didn't,
say:
hey
this
one
doesn't
meet
the
requirements
yet
break
it
off
into
a
separate
project,
and
then
once
it
is,
we
bring
it
in
is
incubating,
even
though
it's
architecture
and
the
Very
way
that
it
did,
things
were
different
and
I
think
you
know
we
try
to
be
consistent
and
I
know
not
everyone's
going
to
agree
with
that.
But
one
of
the
things
that
I
know
I
work
hard
on
is
be
consistent.
H
There
are
things
that
other
Toc
members
will
know
that
I
sometimes
disagree
with,
but
if
it's
an
ongoing
decision
that
the
TOC
consistently
makes,
if
we're
going
to
revisit
that
we're
going
to
revisit
it,
so
we
now
do
it
in
a
new,
consistent
manner
with
a
justification
behind
it
and
until
then
we
will
be
consistent
with
the
way
we're
doing
things.
So
projects
know
what
to
expect
and
I.
Think
in
that
case,
being
consistent
is
okay.
H
Notary
V2
is
a
different
architecture,
different
way
of
doing
things,
but
we
don't
just
jettison
it
right
away
because
it
doesn't
meet
the
criteria
right
away
as
it's
ramping
up,
and
so
that's
just
consistency
in
our
nature,
and
it's
why
I
would
be
tentative
to
go
after
that,
because
then
we're
setting
a
new
and
different
precedent
here
and
does
it
become
now
an
ongoing
existing
precedent
for
everybody?
Is
it
a
one-off?
Why
would
we
do
a
one-off?
You
know
we
have
to
think
about
things
in
this.
H
This
macro
level
way
so
we're
consistent
people
know
what
to
expect
from
us.
They
know
how
things
operate
as
best
we
can,
and
so
that's
just
one
of
my
thinkings
on
this
and
I
did
go.
Look
at
it
and
a
number
of
the
criteria
I
can
tell
you
today
they
already
meet
things
like
the
number
of
production
instances.
I
didn't
go
chase
that
down,
because
that
takes
a
little
bit
more
legwork
than
I
was
going
to
go
put
in,
but
that's
the
kind
of
criteria
where
we're
now
getting
to
do.
They
meet
those
criteria.
D
C
So
there's
a
couple
of
things
that
are
going
on
with
this,
so
the
security
tag
Justin,
who
is
a
member
he's,
a
technical
lead,
we've
gone
through
and
re
we've
revamped.
How
Security
reviews
are
done
in
the
way
that
the
structure
is
set
up,
such
that
a
self
review
by
the
project
is
usually
ideal
for
sandbox
applications
or
really
early
incubating
projects,
and
it's
a
guiding
point
for
them
to
set
up
the
security
of
their
project
from
there.
Between
incubation
and
graduation
The
Joint
review
is
intended
to
be
conducted.
C
Given
everything
that
I'm
hearing
currently
about
the
initial
and
indications
of
notary
V1
and
what
has
come
out
of
as
an
as
a
new
change
within
their
project,
I
would
like
to
caution
folks,
and
this
kind
of
goes
a
little
bit
to
what
Matt
verino
was
mentioning.
We
don't
necessarily
evaluate
changes
in
architecture
and
design
without
giving
them
Runway
to
start
doing
that
development
and
architecture
work
on
and
be
like
saying,
a
Harden,
your
environment.
D
D
D
B
A
B
D
Okay,
I
think
we
are
reaching
the
end.
We
have
one
more
topic
to
cover,
so
thanks
for
all
the
good
conversations
here
and
the
discussions,
let's
try
to
make
each
other
better
for
sure,
and
we
will
we,
as
in
the
TOC,
will
try
to
reach
out
to
more
people
who
haven't
been
on
the
call
like
Justin
Campos
mentioned,
and
we
will
come
up
with
a
set
of
guidance.
For
you
know
the
community
at
large
does
that
sound
okay
to
everyone.
D
L
D
L
L
You
know
core
OS
was
one
of
the
foundational
companies
in
the
cloud
native
ecosystem.
A
lot
of
people
here
have
a
soft
spot
for
core
OS,
and
you
know
core
arrest.
Container
limits
was
the
foundation
of
that
company
and
a
lot
of
the
original
Innovative
crowd
native
work
was
done
on.
You
know,
on
the
basis
of
of
core
arrest
container
Linux,
some
great
work
that
was
done
in
that
project
and
that
kind
of.
L
Obviously,
when
Red
Hat
acquired
core
arrested
company,
core
OS
container
limits
went
on
to
a
new
life
as
part
of
the
Fedora
project,
with
Fedora
core
OS
and
red
hat
core
arrests
as
the
commercial
product
built
on
that
it
kind
of
inspired.
A
lot
of
you
know
a
lot
of
the
things
that
went
on
there,
but
black
car
really
was
the
only
project
that
well
we.
L
So
we
forced
the
original
container
core
OS,
initially
in
a
kind
of
a
friendly
forkway
of
just
we
built
as
a
downstream,
but
then,
when
the
original
chorus
came
to
an
end,
we
took
it
forward
as
its
own
independent
project.
I've
been
running
it
like
that,
for
you
know
getting
on
for
three
years
now,
so
with
with
quite
a
track
record
of
putting
out
frequent
releases
keeping
up
with
security
updates.
All
of
the
kind
of
things
you
want
to
see
around
a
mature.
L
You
know
Linux
project
that
people
are
actually
putting
in
production
and
and
if
I
look
at
the
community
of
users,
we
have
it's
it's
quite
interesting
because
it
splits,
there's,
probably
you
know
close
to
half
of
the
user
base.
That
is
just
folks
that
had
already
built
on
core
OS
that
just
wanted
something
that
they
could
take
forward
and
just
point
there
up.
You
know
point
to
a
different
update
server
and
continue
getting
updates
with
a
compact,
they're
compatible
distro,
and
you
know,
probably
the
other
half
of
the
user
base.
L
You
know
who's
who's
come
to
black
car,
for
new
designs
for
new
builds
and
they're
choosing
it,
as
you
know,
without
without
having
had
that
history,
but
just
because
it's
a
it's
a
great
platform
to
build
containers
on
so
a
lot
of
benefits
to
how
how
flat
car
manages
systems
in
terms
of
security,
it's
an
immutable
OS.
So
there's
a
whole.
L
L
That
way
has
has
a
very
simple
Atomic
update
system
where
you
download
a
new
version
of
the
OS
and
in
a
b
partition
you're
running
an
a
partition
you
boot
over
to
that
when
you're
ready
to
update
if
the
boot
doesn't
take
for
some
reason
it
doesn't,
it
doesn't
system,
doesn't
start
it'll
flip
back
to
the
a
partition,
and
we
allow
you
to
have
a
lot
of
policies
around
how
that
how
those
updates
work
and
it
it
really
is
minimalistic
and
designed
for
containers.
So
it's
a
different
videos.
L
You
know
the
thousands
of
different
distros
and
they
all
have
their
own
Niche.
You
know
the
the
the
the
the
the
raison
Detra
for
flat
car
is
to
run
containers
it's
to
be
a
container
host,
so
it
has
just
the
bits
you
need
to
run
containers
it
integrates
with
container
D.
It
has
etcd
it's
a
great
way
to
run
kubernetes.
L
So
that's
kind
of
black
covered
project
in
terms
of
the
process
and
governance,
and
you
know
community
and
all
of
that,
so
some
some
of
you
may
may
know
the
origins
of
flat
car
came
from
a
small
small
company
called
Kinfolk,
which
was
really
just
you
know.
A
small
team
of
Open
Source
experts
actually
had
done
a
lot
of
work
for
core
OS
at
a
lot
of
the
original
contributors
to
the
rocket
container
engine.
L
So
one
of
the
early
cncf
projects
came
out
of
the
Kinfolk
team
and
it
was
actually
Chris
cool
who's.
The
he
was
the
CEO
at
Kinfolk.
Who,
who
said
hey?
You
know
we
should
we
should
do
this
flat
car
project,
so
we
we
ran
it
independently
for
for
a
while
until
kinfork
was
acquired
by
Microsoft
and
as
you
can
imagine,
a
lot
of
community
had
questions,
and
you
know
a
small
independent,
open
source
company
gets
acquired
by
you
know
one
of
the
large.
You
know
one
of
the
large
vendors.
L
What
does
this
mean
for
the
project,
and
that
was
that
was
nearly
two
years
ago
that
that
acquisition
happened,
and
you
know
there
was
it
was
I
would
say
when
we
went
into
the
acquisition.
The
discussion
was
very
you
know
was
we
were
very
clear
that
supporting
the
community
and
doing
the
right
thing
by
the
community
was
an
absolutely
key
requirement
that
we
had
as
a
team,
but
fortunately
that
was
micro.
You
know
Microsoft's
view
as
well
was
they
want
to
support
the
community
and
to
support
the
cloud
native
community
in
general?
L
Allowed
the
you
know
the
project
to
flourish,
and
you
know,
and
the
best
way
to
do
that
in
our
view
is
as
part
of
cncf,
because
that
provides
this
independent
forum
for
for
us
to
build
governance
and
to
enable
contributions
and
to
really
kind
of
you
know,
put
every
every
participant
on
an
equal
footing,
and
it
also
just
underlines
that
Microsoft's
intent
with
this
project
is
not
to
dominate
it,
make
it
a
Microsoft,
Linux
or
anything
like
that
right.
It's
it's
to
do
the
right
thing
by
the
community.
L
D
Thank
you.
So
thanks
for
the
setup
I,
so
one
of
the
things
that
we
were
trying
to
look
at
or
talk
about
earlier,
when
your
proposal
hit
was,
do
we
open
this
up?
What
are
we
gonna
find
when
we
open
this
up
and
are
there
other
projects
in
this
space
and
who
might
be
interested
like
those
kinds
of
things
like?
Are
we
set
up
to
handle
it?
L
Yeah
Dennis,
maybe
if
I
could
just
address
that
point
before
other
questions
come
in
because
I
think
that
you
know
there
is
an
important
question
and
I
address,
address
it
a
little
bit
in
in
The
Proposal.
But
I
can
imagine
you
know
the
concerns.
Oh
God,
the
floodgates
open
right,
we've,
let
in
one
Linux
distro.
L
L
G
I
mean
just
to
follow
up
on
that
I
mean
you
know
in
the
application.
You
call
it
container,
optimized
operating
system,
I,
think
I.
Think
the
the
name
and
the
description
of
the
category
is
a
little
bit
unclear
in
a
lot
of
people's
minds
because
you're,
whatever
people
say,
Linux
distribution,
but
in
the
classic
sense
it's
not
a
nice
distribution.
G
I
kind
of
think
am
I
in
my
view,
because
you
you
can't
install
that,
like
inside
a
container
to
run
Linux,
for
example,
and
things
like
that,
it
just
doesn't
and
it's
underlying
packages
come
from
Gen
2
and
it
it
so
in
some
sense
it's
a
it's
a
way
of
using
gender.
When
Red
Hat
acquired
core
OS,
they
took
the
technology
and
applied
it
to
you
know
different
to
fedora,
so
but
I'm
kind
of
interested
in
what
you.
How
are
you
if
you've
got
a
better
term
to
describe
the
category
of
thing
that
this
is?
G
L
It's
it's
a
good
point
in
that,
so
there
are
two
very
different
categories
of
container
os's.
If
you
like,
so
there's
container
host-
and
you
know-
and
then
there's
container
base
image
right.
So
where
can
we're
we're
optimized
for
the
container
host
use
case
right?
So
that
means
a
minimal
set
of
packages.
L
You've
got
to
container
runtime
so
that
you
can
start
containers
and
we're,
assuming
that
all
of
the
runtime
dependencies
that
you
have
packaged
in
the
container
you
know
come
back
using
the
container
and
therefore
you
don't
don't
need
them
on
the
host.
So
yeah,
maybe
like
container
host
optimized
OS,
might
be
more
technically
correct.
But
if
I
think
people
understand,
you
know
that
a
base
image
is
is
kind
of
a
different
animal.
D
B
It's
been
things
through
other
cncf
projects,
which
has
been
interesting
and
somewhat
of
a
testament,
because
it's
it
in
that
way.
It
is
somewhat
different
because
it's
not
a
business
driven
development
like
some
of
the
other
ones
in
the
same
similar
category
and
yes,
they
do
all
smell
similar,
but
so
I
understand
the
desire
to
to
distinguish
them
the
hand
gesturing
was
kind
of
in
the
the
path
of
yeah
flat.
Car
is
not
a
container
image,
and
it
is,
you
know,
like
core.
B
Os
was
a
derivative
of
chromium
OS,
which
is
built
from
Gen
2,
but
way
back
whenever
we've
experimented
with
having
it
do
derivatives
of
other
os's.
So
that's
it's
not
it's
not
completely
Preposterous.
So
it
is
kind
of
a
a
derivative
product
or
a
derivative
process,
but
also
even
like
when
Red
Hat
went
to
make
what
they
later
called
Fedora
chorus
or
whatever
Red
Hat
chorus.
B
J
K
K
Keep
going
yeah
yeah,
so
yeah
I,
just
I'll
read
what
I
said.
We
often
refer
to
general
purpose
distributions
versus
purpose-built
distributions.
This
is
squarely
a
purpose-built
distribution.
That's
focused
on
orchestration
data
plane
and
I
was
just
asking.
Does
it
need
to
be
categorized
beyond
that?
There
are
certainly
other
examples
of
that.
As
someone
else
said,
there's
Kalos,
there's
bottle
rocket.
Ricardo
pointed
that
out.
So
that's
all
I
was
saying.
D
Yeah
I
think
when
we
open
up
to
the
host
purpose
will
toast,
like
you
mentioned
I,
think
we
might.
We
might
kind
of
like
the
slippery
slope.
There
is
like
okay.
We
need
to
give
a
similar
consideration
for
the
container
base
images
as
well,
so
I
think
it's
natural
for
this
to
lead
to
that
as
well.
I
think
yeah
go
for
it.
Matt
I
was
looking
for
you.
H
Yeah
yeah,
you
know
in
all
of
this
one
of
the
things
that
sits
out
in
my
mind,
isn't
does
this.
You
know
join
the
Linux
foundation
in
its
family,
because
the
cncf
is
part
of
the
Linux
Foundation.
It's
where
does
it
go?
Does
this
belong
in
the
cncf
or
does
it
belong
in
the
greater
Linux
Foundation
somewhere?
And
so
this
is
one
of
the
questions.
H
That's
in
the
back
of
my
mind
that
I'm
trying
to
answer
with
everything
else
going
on
because
it
may
be
a
purpose-built
Linux
distro,
which
is
a
purpose-built
Linux
distro
belong
in
the
cncf
or
does
it
belong
in
the
greater
Linux
foundation
and
it
I
don't
actually
have
an
answer
to
it.
But
this
is
the
big
question
that
I'm
trying
to
formulate
and
answer
to
in
order
to
know
how
to
respond
to
this
yeah.
D
Perfect
before
I
hand
off
to
I
do
want
to
raise
one
of
the
things
that
we
talked
before
in
the
TOC
was
like:
hey
the
folks
doing.
This
work
are
already
part
of
our
community,
and
you
know
so
it
makes
sense
to
have
them
close
to
the
rest
of
the
things
that
we
are
doing.
So
that
was
one
of
the
thinking,
one
of
the
lines
of
thinking
that
we
had
when
we
were
talking
about
it,
Richie
go
for
it.
A
So,
first
of
all,
I
have
to
agree
that
core
is
kicked
off.
What
we,
what
we
know
called
Cloud
native,
so
there
is.
There
is
certainly
an
argument
to
be
made
for
this
to
be
part
of
cloud
native
I.
Think
it
wouldn't
be
a
slippery
slope
to
to
also
accept
other
base
images.
It
would
be
the
absolute
necessary
logical
step
and
that's
the
thing
which,
which
makes
me
a
little
bit
apprehensive.
Of
course,
we
would
be
increasing
the
overall
exposure
surface
of
the
amount
of
software
which
we
have
Within
cncf
substantially.
D
A
Not
only
this,
but
also
just
like
security
processes
and
and
everything
because,
like
we
are
talking
full
distributions
at
this
point,
and
we
will
not
be
talking
one,
we
will
be
talking
half
a
dozen
except
for
kubernetes.
We
don't
have
any
code
base,
which
is
nearly
as
large
as
a
complete
distribution.
I
think
it's
just
going
to
be
new
and
larger
in
a
lot
of
ways.
At
the
same
time,
yes
from
the
category,
it
makes
sense
because
core
as
everything
like
I,
absolutely
see
this
argument,
yeah.
D
B
Foreign
yeah
sure
to
the
to
the
question,
LF
versus
cncf
or
otherwise
was
was
a
huge
part
of
the
mulling
on
this
aspect
as
well,
and
even
as
Andy
mentioned
earlier,
like
the
kind
of
contributors
and
you
users,
customers,
whatever
you
want
to
say,
of
flat
car,
not
paying
customers
but
still
are,
you
know,
have
have
wanted
to
see
it
in
a
neutral
playing
Ground
so
that
they
can
be
more
involved
as
well.
So
that's.
A
B
Know
as
we've
worked
on
opening
up
the
governance,
that's
been
like
to
get
it
into
an
open
playing
area.
An
open
governance
area
has
been
a
huge
Focus
so
that
we
could
meet
them
where
they're
at
and
just
like,
was
said
it
it.
It's
only
focus
is
cloud
native,
so
that
makes
a
huge
difference
and
a
lot
of
the
all
the
customers
and
of
a
flat
car
already
present
and
involved
in
cncf.
B
That
would
make
it
easiest,
and
if
you
did
end
up
in
some
other
place
like
LF
or
wherever,
then
we
could
try
and
meet
you
where
you're
at
but
like
this
is
Mainline
the
purpose
and
what
you
know,
processes
that
we've
already
put
involved,
we're
literally
trying
to
meet
the
community
where
it's
at
and
to
find
that
that
slippery
slope
or
whatever
of
this
space,
even
from
conversations
had
during
the
week
among
all
the
different
OS
folks
involved
from
like
fedorical
OS,
bottle
rocket
Talos.
Otherwise,
at
the
past,
kubecon
was
really
like.
B
There
was
a
feeling
and
a
sentiment
that
in
all
the
cloud
native-
and
you
know,
the
OS
doesn't
matter
that,
like
the
OS
does
matter,
and
it's
kind
of
a
gap
right
now
like
that
people
are
either
not
talking
about
it
or
it
seems
to
be
kind
of
a
gap
in
the
landscape
of
like
it's,
it's
implicit
and
could.
Could
we
not
just
make
that
part
of
the
conversation
so
I
think
there
is?
B
D
Some
of
the
things
that
we'll
have
to
end
up
talking
about
would
also
include
hey,
you
know
our
own
licenses
and
copyright
and
those
kinds
of
things
we
yes,
the
TOC
has
to
bless.
It
is
agree
that
we
need
to
open
it
up
first,
but
then,
immediately
after
we'll
face
those
kinds
of
questions
that
we
can't
answer
by
ourselves,
we'll
have
to
rely
on
CNC
of
staff,
and
then
we
need
to
go
back
to
the
cncf
GB.
D
There
are
some
things
in
the
charter
that
we
might
have
to
do
something
about.
There
is
a
legal
Committee
in
GB,
so
you
know
have
this
in
the
back
of
your
mind,
so
it's
not
just
the
TOC,
but
there
is
other
things
that
we
need
to
push
and
lead
and
like
prod
and
move
people
towards
to
make
it
happen.
Okay,
we
have
two
more
minutes.
Jesse
did
you
want
to
say
something
quickly
or.
K
I
put
it
in
chat,
I,
just
I
was
being
too
abstract
before
and
I
wanted
to
tie
it
back
to
exactly
the
point,
so
I
think
I,
think
oversight
and
governance
is
best
attached
to
use
case,
and,
and
so
that's
I
guess
to
put
a
fine
point
on
it.
Yet
it
is
an
OS
distribution,
it
could
go
in
the
LF.
The
main
use
case
here
is
data
plane
for
orchestration
and,
let's
not
beat
around
the
bush.
K
Orchestration
typically
means
kubernetes,
so
I
just
kind
of
wanted
to
say
that
I
think
that,
as
you
consider
this,
where
are
the
people
that
are
going
to
be
using
using
it
and
needing
for
it
to
be
a
healthy
project?
Where
are
they
co-located
and
I?
Think
that
that's
sort
of
what
vbas
was
saying
as
well.