►
From YouTube: CNB Core Team Sync: 2021-12-08
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
A
It
is
bumpy
over
here,
especially
in
a
house
with
two
different
people
doing
streaming,
video
conferencing
all
day,
nice.
A
I
used
to
joke
that
the
internet
just
works
better
in
the
bay
area,
like
you
can
literally
call
someone
to
deliver
you
a
six-pack
of
beer,
and
it
can
be
here
in
10
minutes,
unlike
almost
anywhere
else
in
the
world,
and
yet
I'm
having
very
jittery
internet
today.
A
Cool
don't
forget
to
sign
to
the
duck.
I
feel
like
that's
the
steven
line
I'm
used
to
hearing
over
the
years.
A
Looks
like
we're
live
thanks
for
doing
that,
javier,
I
don't
think
we
have
any
new
faces.
A
A
B
A
At
this
point,
my
my
home
office
feels
like
a
bit
of
a
sell
sometimes,
and
I
just
need
a
change
of
scenery.
It's
where
you
do
your
other
meetings
right.
B
I
have
a
a
potentially
contentious
point
on
the
file
system.
One
of
the
comments
that
I
made
about
order
tamil
and
how
it
seems
a
little
bit
off
to
require
that,
given
that
most
platforms
override
it
and
or
can
create
it.
B
B
B
Well,
the
depends
on
what
you
mean
by
it's
a
change
from
the
world
I
made
I
made
mention
of
what
we
do
already
and
so
like
pack
already
adds
this,
when
you
add
build
packs
k
pack
does
you
know,
creates
an
order.
Tumble
dynamically
and
techton
could
create
one
as
part
of
the
repair.
B
A
B
When,
in
pack
as
a
utility
that
creates
builders,
you
have
a
builder
tamil,
the
builder
tumble
will
have
an
order,
yeah
define
there.
We
translate
that
over
into
json
and
stick
it
into
that
label
there
into
label.
A
B
Yeah
and
then,
what's
again,
what's
even
worse,
is
we
do
it
again
if
the
user
specifies
let's
say
a
different
order
through
command
command
line
argument
or
through
the
project
descriptor
we
overwrite
the
order.
Tamil,
in
that
case,.
A
A
Well,
it
seems
reasonable
on
me,
but
I
also
did
not
have
I
think,
heavy
opinions
on
the
original
builder
stuff
from
back
in
the
day.
B
A
Or
something
yeah,
that's
fair,
I
will
at
least
ping
jesse
at
least
is
on
the
team
that
I
work
with
and
talk
with
every
day
and
then
joe
javier.
I
think
you
should
also
ping
like
just
in
get
in
here,
ref
emily
and
let
her
decide
if
she
wants
to
come
back
to
weigh
in.
A
Cool
that
sounds
good.
A
Yes,
I'll
follow
up
with
them.
I
don't
think
I
have
a
strong
opinion.
Removing
order
thomas
requirement,
but
still
being
optional,
sounds
like
a
reasonable
suggestion
to
me.
I
assume
that
also
has
no
qualms.
A
None
cool.
A
I
don't
know
if
we
have
a
lot
to
say
here.
It
seems
like
we're
just
trying
to
allow
the
specs
of
seems
to
be
just
moving
distribution
of
three
four.
B
I
do
have
a
question
about
project
descriptor
02,
given
that
certain
projects
are
starting
to
implement
it.
B
I
do
wonder
if
we
want
to
kind
of
necessarily
time
box
it,
but
like
feature
box
it
and
say
like
okay,
we're
done
with
this:
let's
ship
it
as
is
right.
I
know
that
there's
a
lot
of
other
conversations
happening
in
relation
to
it,
but
I
think
that
those
might
be
out
of
the
scope
of
this
upcoming
release.
A
A
Thanks,
xavier,
let
me
circle
back
with
joe
to
make
sure
he
doesn't
have
anything
else
that
I'm
missing
and
then
we
will
cut
a
thing
for
o2.
Doesn't
pack
right
implements?
I
felt
like
one
of
the
blockers
was
like
making
sure
this
was
not
unreasonable
and
pack
or
something
that's
what
we
said
like
I
feel
like
four
months
ago.
B
Yeah,
I'm
not
sure
what
the
state
of
the
world
is
like.
I
said
because
I
feel
like
there's
partial
implementation
in
different
projects.
Okay,
I
think
one
of
the
things
that
I
was
wondering
about
in
relation
to
the
distribution.
Spec
changes,
we're
making
and
now
project
descriptor
is
if
we
have
more
of
a
holistic
overview
of
the
thing,
that's
going
to
be
released
and
making
sure
that
all
that
is
accurate
and
appropriate,
and
the
reason
I
guess
I'll
bring
some
cases
to
light.
B
We've
gone
through
and
released
specs
with
the
wrong
version
of
itself
being
defined
so
like
we'll,
we'll
go
and
create
prs
for
individual
chunks
changes,
but
we
never
look
at
the
entire
thing
and
ask
ourselves:
does
this
make
sense
for
us
to
actually
go
ahead
and
ship?
It.
C
B
Yeah
and
make
sure
that,
in
you
know,
part
a
makes,
makes
sense
with
part
b,
because
I
feel
like
that
might
be
a
a
situation
where
certain
cases-
that's
not
always
true.
A
Okay
and
that's
just
like
an
ask
for
when
we're
cutting
this
next
thing,.
B
Yeah
I
mean
maybe
at
minimum
to
get
started,
would
just
be
like
hey
we're
about
to
release
this.
Please
give
it
a
thumbs
up
or
something
similar
to
how
we
do
for
release
notes.
A
I
think
emily
did
that
for
her
big
merge
pr
of
the
last
her
like
last
platform
thing.
There's
that
one
big
pr
that
was
like
merge
into
master.
B
A
Yeah,
I
think
we
did
that
and
still
missed
the
wrong
version
number
somewhere.
But
you
know
human
error
is
also
a
thing.
Fair,
fair,
yeah
I'll
make
sure
we
do
that,
for
when
we
do
project
descriptor
stuff.
Maybe
I
should
also
release
something
release
empty.
A
A
We
should
be
doing
them,
I
don't
know
the
last
one.
We've
done.
A
I
was
not
around
for,
I
guess
the
stuff
post
emily,
because
I
I
think
emily
did
the
last
one
and
then
joe
did
the
last
two
or
something
while
I
was
on
vacation.
So
I
actually
am
not
sure
what
happened
during
that
time.
I
think
he
just
asked
me
to
prove
something
verbally
or
asynchrony.
While
I
was
on
vacation.
A
But
if
we
aren't
doing
we
should,
I
guess
your
point
javier.
So
if
we
did,
that
was
mistake.
A
No
support
dockerfiles.
I
proved
this.
I
think
the
only
change
I
wanted
to
make
was
just
to
capture
joe's
specifier
at
the
bottom
of
for
spec
changes.
A
A
So
when
this
gets
merged,
you
can
actually
follow
through,
because
no
one
I
feel
like
it's,
no
one
actually
goes
through
most
times
and
opens
the
old
pr
that
this
rfc
was
in
to
can
find
all
that
stuff,
but
other
than
that,
I
think,
can
I
even
just
commit
my
own
suggestion
to
steven's
branch.
I
guess
I
can.
A
I
saw
stephen
and
if
I'm
okay
moving
this
to
fc,
I
guess
he
won't
be
around
next
week.
B
Has
the
issue
with
gen
pack
been
resolved
or
I
think
it
has.
C
C
B
I
I
I
think
so
right,
especially
when
we
talk
about
the
what
we
defined
for
the
run
image
s
bomb
structure.
I
think
how
gen
pack
would
work.
I
think
it's
still
problematic
and
very
ill
defined,
but
I'm
not
sure
exactly
how
to
proceed.
B
C
It
seems
like
they
I
had
to
think
about
it
in
the
the
thing
that
makes
sense
is
if
the
gen
package
is
going
to
take
the
standard
thing
like
either
a
registry
reference
or
a
local
oci
layout
of
the
image
it's
trying
to
scan
and
if
it
ran
that
on
the
build
image
that
would
work
so
restricts.
What
the
gen
package
is,
how
easy
it
is
to
implement
that
gen
packages
thing,
but
yeah.
C
A
Okay,
just
to
get
movement
on
this,
you
still
you
do
still
want
changes,
though,
into
this
rfc.
C
A
A
Oh
thanks
for
that
call
out
here,
system
bill
packs.
A
I
see
most
people
have
approved
this,
or
at
least
ben
steven
did
so
I
guess
I'm
the
only
blocking
person
it
seems
like
there's
just
some
open
discussions
unresolved
threads
like
one
of
the
ones
that
sam
commented
on
on
kind
of,
I
think
the
build
plan
of
like
what
we
want
to
do
with
that
and
detect
for
system
bill
packs.
A
And
like
how
hidden
do
we
actually
want
the
system
build
packs
like?
Are
we
not
gonna?
Do
we
want
to
include
them
in
kind
of
the
s-bomb
metadata
stuff?
Even
if
we're
not
displaying
output
to
the
users
like
if
I
inspect
an
image
I
I
should
know
that
a
system
built
back
was
executed,
or
is
that
still
the
result
into
that
implementation?
Detail.
A
The
I
just
had
when
I
went
through
there
was
still
like
two
open
issues
on
this
particular
rfc
yeah
one
was
the
bill
plan
stuff
that
you're
you
commented
on,
I'm
just
like
how
should
that
interactive
work
and
there
was
discussion
there,
but
it
seems
like
it
was
not
resolved.
So
I
don't
feel
okay
moving
forward
with
this
until
we
resolve
that
issue,
and
then
stephen
also
asked
about
like
why
is
this
hidden
from
users
of
like,
I
think,
there's
some
questions
like
how
hidden
is
it
like?
C
A
Yeah,
I'm
fine
with
that
too.
I
I
just
said:
should
we
just
yeah,
like
I
even
said,
should
we
just
not
have
built
plan?
Be
part
of
system
build
packs,
not
folks
for
me,
since
we
don't
have
use
cases
like
you
said
right
like
that,
makes
it
a
lot
easier,
but
as
the
rfc
is
written,
that
is
not
true
right,
so
that
would
still
be
a
change
in
the
rfc.
C
A
C
C
A
It
says,
block
or
doesn't
say
blocked
this
one
blocked.
Listen,
it's
not
blocked
anymore.
Actually,.
C
B
Yeah,
terence,
this
might
be
one
of
the
ones
where
you
might
want
to
ask
joe
if
we
want
to
hold
up
project
descriptor,
2.0
or
0.24.
Okay,.
B
A
Just
to
understand
that
this
is
the
ability
to
push
the
same
image
to
multiple
registries.
B
It's
to
define
the
image
parameter
in
the
project
descriptor,
since
it
doesn't
exist
right
now
and
so
they're
they're
also
adding
to
be
able
to
do
it
to
multiple
registries
right
but
like.
A
Has
stats
fields,
so
this
just
goes
through
and
updates
all
the
stuff.
I
recall
right
right.
This
is
not
an
rc
proposal.
This
is
literally
just
implementing
that
the
old
one.
A
Awesome
thanks
for
doing
this,
anthony
I'll
go
through
look
through
that
after
this
meeting
and
add.
C
C
This
is
something
that
I
added
and
we
discussed
in
the
implementation
game
meeting
today.
The
proposal
is
that
I
think
it
links
to
one
of
the
older
issues
that
they
linked
over
there
at
the
very
bottom.
Is
this.
C
It
we
I
had
to
introduce
some
api
changes
in
there,
but
like
from
the
implementation
team
meeting
today,
we
agreed
that
I
can
just
remove
all
the
api
changes
and
we
can
still
implement
this
so
that
tools
like
dive
or
registries
can
still
show
something
useful
for
buildbacks.
And
then,
if
people
want
to
customize
whatever
is
shown
for
each
of
the
layers,
we
can
add
something
to
the
api
in
the
future.
A
I
guess
do
you
want
it
to
be
like
do
you
want
to
do
you
want
to
have
api
changes?
You
want
to
remove
them.
C
C
If
people
are
fine
without
like
having
custom
ways
of
displaying
it
and
they're
fine
with
the
default
for
those
things,
then
yes,
the
idea
is
that
when
you
do,
when
you
use
dive
or
when
you
go
to
docker
hub,
each
layer
shows
like
the
command
or
something
so
it's
a
way
to
visualize
what's
happening.
This
proposes
that
we
add
some
default
values
for
the
buildback
layers,
so
something
similar
happens
for
buildback
images.
C
It
doesn't
allow
buildback
authors,
for
example,
to
change
what
that
message
looks
like.
I
know
there
was
some
discussion
on
that
issue
that
I
linked
below,
where
people
wanted
to
customize
that
so
maybe
that
can
be
a
future
rfc
that
actually
warrants
an
api
change,
but
right
now
this
was
just
to.
B
Yeah
I'm
wondering
if
I
know
in
the
past,
I
forgot
who
it
was,
but
they
recommended
a
structured
format,
yeah
consumption.
I
do
wonder
if
it's
something
that
should
be
project
wide
and
not
implementation,
specific
yeah.
C
A
C
B
Should
we
talk
about
this
tomorrow
because
I'm
curious
why
platforms
need
to
implement
some
of
this
stuff
as
opposed
to
the
life
cycle,
taking
ownership
of
it.
C
Yeah
it's
we
can
talk
about
it.
The
main
issue
is
like
there's
a
file
called
project,
metadata
tommles,
which
is
populated
by
the
platform
and
read
by
the
life
cycle.
The
platform
needs
to
fill
the
metadata
table
in
that
file
from
project
normal
or,
like
other
places
where
it
can
find
the
information.
A
B
A
Makes
sense
and
then
move
cncf
slack.
C
C
C
B
A
A
C
I
definitely
want
to
talk
about
the
course
and
integration
and
there's
a
separate
topic
which
I
wanted
to
bring
up,
which
is
like
the
daemon
versus
ocl
layout
use
case,
which
relates
to
all
of
my
rfcs
and
a
bunch
of
other
things
in
the
future.
So
I,
if
possible,
I
want
to
talk
about
the
cosine
stuff
and
the
oc
layout
thing.
C
C
Yeah,
I
think
the
annotation
stuff
is
like.
Oh
it's
two
minutes,
and
we
can.
We
can
just
do
it
asynchronously.
I
think
people
might
have
more
points
of
contention
on
on
the
cosine
one
and
the
oc
layout
one
so
that
might
be
useful
in
the
working
group.
A
Okay,
I
would
like
to
insert
probably
just
the
slack
thing,
even
if
it's
a
footnote
at
the
beginning
of
or
somewhere
in
the
meeting
yeah,
just
because
people
should
provide
feedback
and
put
on
it,
because
it
probably
impacts.
B
What
if
we
put
anthony's
demo
before
the
slack
conversation,
just
in
case
we
run
out,
we
could
do
it
later
as
well.
A
Well,
I
don't
necessarily
want
to
have
a
conversation
about
the
slack
thing.
I
just
want
to
make
sure
that
people
know.
A
C
A
Cool
sounds
good
to
me.
Is
there
anything
else
that
I'm
missing
with
us
being
19
minutes
over?