►
From YouTube: [OCI-WG] Reference Types - 2022-01-11
A
B
Taken
about
20
feet
that
direction.
C
C
A
D
Yeah
and
I
unpause
the
recording
right
as
everybody
was
saying,
happy
new
year
whatever
so.
A
Excellent,
thank
you,
vincent
okay.
I
will
kick
us
off
here,
but
very
excited
to
see
everybody.
This
is
the
inaugural
working
group,
reference
type
meeting
so
welcome
everybody.
Today
the
agenda
is
mentioned
in
the
meeting
minutes
there
that
I've
just
linked
so
feel
free
to
go.
Take
a
look,
and
please
add
your
name,
but
what
I'd
like
to
do
to
start
with
is
do
some
welcome
and
introductions.
A
So
today
I
think
the
goal
was,
as
I
thought
about
it-
and
you
know
this
is
an
open
discussion-
was
just
setting
the
stage
here
for
the
work
that
we're
trying
to
achieve
in
this
working
group,
and
I
thought
it
might
be
good
just
to
go
back
to
first
principles
and
and
do
some
introductions
and
and
see
what
everybody's
hopes
and
desires
are
for
this
working
group,
and
then
we
can
set
up
the
different
administrative
pieces
of
how
we
can
communicate
and
move
things
going
forward,
and
then
I
also
through
kind
of
a
a
fun
group
activity.
A
Well,
I
think
it's
fun,
I
guess
the
proof's
in
the
pudding
and
I
thought
it
would
be
fun
to
put
together
a
mission
statement
as
a
group
activity
just
a
couple
of
sentences.
I
always
like
to
say
if
you
got
on
an
elevator,
and
somebody
said
what
are
you
trying
to
do
with
this
thing?
You
could
give
a
kind
of
a
two
or
three
sentences
about
what
it
is
you're
trying
to
achieve,
and
I
think
that's
incredibly
easy
to
say
and
difficult
to
do
so.
A
A
Keep
it
brief
because
there
are
a
few
peop,
fair
few
people
here
and
just
what
you're
hoping
to
achieve
in
this
working
group
and
why
you're
here
so
I'm
gonna
do
set
a
little
protocol
up
here,
one
that
I've
used
in
the
past
where
everybody
raises
their
hands
using
the
digital
hand,
raising
which
is
under
reactions,
and
you
click
raise
hands,
can
everybody
see
and
then,
if
everybody
raises
their
hands,
when
somebody
passes
you
the
ball,
to
give
your
introduction,
you
lower
your
hand,
so
that
way
people
know
who's
already,
given
their
introduction
and
who's
left.
A
So
it
also
gets
everybody
knowing
where
their
raised
hand
function
is
because
it
it's
buried
under
the
reaction.
So
if
everybody
can
raise
their
hand
digitally
or
physically,
then
we
know
who
to
pass
that
sound
okay,
I'll
get
I'll
get
us
kicked
off
just
with
a
quick
icebreaker.
My
name
is
lachlan
evenson
I
go
by
lockheed
lachlan,
hey
you,
hello,
mate.
I
work
at
microsoft
on
our
cloud
native
upstream
team,
where
we
work
in
open
source
communities
on
cloud
native
tech
really
excited
to
be
here.
A
I'm
excited
about
all
this
work
in
the
security
space
that's
happening,
and
what
the
role
of
container
registries
and
oci
registries
are
in
that,
and
I
think
I'm
really
excited
to
work
with
you
all
on
fleshing
out
what
reference
types
could
be.
So
in
that
space
I
will
pass
it
to
josh
and
lower
my
hand.
E
I
can't
see
myself,
but
I'm
assuming
I'm
here,
hey
great,
I'm
josh
jolitsky
recently
started
at
chainguard
and
at
chainguard
a
lot
of
stuff
with
registries
going
on
so
but
prior
to
this
was
involved
with
oci
getting
distribution
to
1.0,
as
well
as
some
of
the
auras
and
artifacts
stuff.
So
I
will
pass
it
to
brandon.
B
Yeah
brandon
mitchell
here
with
box
boat
and
ibm
company,
soon
to
be
properly
transitioned
over
to
ibm.
Also
do
some
client-side
registry
tooling,
and
just
interested
in
making
sure
that
I'm
kept
up
to
date
with
what's
going
on
with
trying
to
make
this
stuff
happen
on
the
registry
and
whatever
I
can
do
to
help
out
and
I'll
go
ahead
and
pass
over
to
vincent.
D
Hey
very
excited
to
see
reconciliation
of
references,
and
otherwise,
so
we
can
move
this
forward
and
get
kind
of
the
generic
feature
out
into
the
space
for
everybody.
That's
looking
forward
to
use
it.
Yeah
jason
go
for
it.
G
All
right,
I'm
lowering
my
hand
before
I
forget,
I'm
jason,
I'm
at
red
hat.
I
don't
operate
a
registry.
I
don't
work
on
clay,
but
I
am
involved
in
a
lot
of
tools
that
build
images
and
push
images
and
want
to
do
that
securely
and
well
and
efficiently
and
blah
blah.
So
I'm
very
interested
in
the
references
work
as
it
relates
to
how
we
attach
s-bomb
signatures.
Anything
resumes
cat
photos
whatever
and
interested
to
work
with
all
of
you
here
to
do
that
better.
Let's
go
to
marina.
J
I
work
on
tough
and
a
lot
of
kind
of
other
projects
related
to
secure
software
updates
and
I'm
here
because
I've
been
working
on
integrating
tough
into
the
noted,
v2
work
and
specifically
to
get
it
working
natively
on
registries.
So
yeah,
I'm
really
interested
in
seeing
this
work
and
see.
If
I
can
use
this
to
to
make
that
that
effort
work
and
I
will
pass
it
to
steve.
K
Thanks
so
yep
steve
lasker,
I
work
at
registries
at
microsoft,
for
customers
and
our
services
that
we
ship
for
customers
for
customer
software
and
basically,
I'm
just
hoping
to
minimize
the
number
of
services
users
have
to
care
for
for
managing
and
storing
these
types.
So
everything
from
s
bomb
signatures
and
results,
helm,
dependencies
and
there's
just
a
bunch
of
stuff
that
users
are
trying
to
distribute
and
having
to
configure
all
of
them
in
their
v-net
and
security.
And
everything
having
consolidated
has
been.
Something
makes
a
lot
of
sense.
K
So,
with
that
tina.
L
Sure
hi
I'm
tiana
gravy,
I'm
with
a
company
called
infosifter.
We
do
a
lot
of
consulting
there's.
A
lot
of
everything
been
in
the
container
space
for
a
long
long
time
started
back
when
docker
was
called
cloud
yeah,
I'm
excited
to
see
references
moving
forward.
I
would
love
to
figure
out
what
what
changes
we
need
to
make
to
image
and
distribution
specs
and
get
those
moving
forward.
I'm
excited
I'm
actually
going
to
pause
and
ask
derek.
Who
is,
I
think,
on
the
phone
so
probably
can't
raise
his
hand
derek.
L
M
M
Jason
works
on
and
we'll
continue
to
do
that
for
the
foreseeable
future.
I
will
tag
in
vanessa.
C
Hey
everyone,
I'm
vanessa
socket
just
forget
about
my
last
name:
you'll,
never
need
to
pronounce
it
again.
It's
awful
I'm
a
computer
scientist
at
lawrence
livermore
national
lab
the
stuff
I'm
doing
now
is
not
super
relevant
to
containers.
I've,
sort
of
worked
on
binary
analysis
and
stuff
with
avi,
but
historically
I've
just
been
very
involved
with
container
stuff.
So
I
was
one
of
the
original
developers
for
the
singularity
project.
I
ran
my
own
container
registry
for
about
five
years.
It
was
not
oci,
but
you
know
it
was.
It
worked.
C
I've
tended
to
be
interested
in
oci,
and
I
guess
after
that,
just
kind
of
the
weird
artifact
types
so
like
sif,
I've
recently
been
playing
with
auras
and
s-bombs
and
I
don't
really
know
the
extent
which
I'm
going
to
be
involved,
but
I
want
to
come
here
because
I
feel
like
the
communities
that
I
represent,
so
academics
national
labs.
We
don't
really
do
a
good
job
of
being
involved
in
these
discussions.
C
So,
like
I'm,
gonna
try
to
do
that.
I
really
want
you
know
my
vision.
Is
that,
like
a
you
know,
you
can
work
at
a
national
lab
and
be
more
using
kind
of
stuff
that
I
think,
is
more
modern,
so
like
not
to
not
to
harp
on
hpc,
but
there's
a
lot
of
really
cool
stuff.
Going
on
in,
like
the
cloud
native
space
that
I
think
we
could
use
too.
C
I
No,
I
work
at
microsoft,
doing
container
runtime
stuff
and
also
producing
artifacts
for,
like
like
container
images
for
aks,
and
what
have
you
in
packages
for
distros
to
run
container
runtimes,
all
that
fun
stuff
kind
of
hoping
to
get
two
things
out
of
this,
mostly
what
I
want
to
be
able
to
experiment
with
new
image
formats
that
we
could
that
we
can
kind
of
upgrade
to
like
gracefully
and
on
top
of
that,
interested
in
stuff,
like
s-bombs
and
all
the
stuff
we
have
to
do
to
for
all
the
artifacts
we
produce.
I
So
I
think
I'm
the
last
person
so
back
to
you.
I
forgot
who
was
hosting.
A
Thanks
brian
I'll
pick,
it
up
appreciate
it
that
went
surprisingly
smooth
thanks
for
doing
the
hands
and
participating
in
that,
and
it's
great
to
you
know,
get
a
background
on
everybody
and
and
welcome
again,
and
I'm
thankful
everybody's
here
and
excited
to
work
with
you
all.
I
thought
it
might
be
good
just
next
to
I'll
shoot
the
link
out,
but
here
is
actually
the
proposal
as
it
was
checked
into
the
main
branch
of
the
tob
repository
in
the
oci.
A
So
I
think
I'm
just
putting
it
out
there
so
that
everybody
has
a
reference
point.
It's
also
in
the
meeting
minute
so
that
you
can
go
over
it
and
take
a
look
at
the
proposal
as
it
stands
today
and
has
checked
in.
But
obviously
you
know,
working
groups
is
a
new
concept
inside
the
oci.
So
I
think
we're
we're
you
know
working
through
this
together
and
and
excited
to
see
the
outcomes.
A
If
anybody
has
any
questions
about
that
proposal
feel
free,
there's
some
communications
channels
which
I'm
going
to
go
through
now,
but
on
the
administrivia
pieces.
I
thought
I
would
outline
just
some
different
communications
channels
and
pieces
that
have
been
set
up
at
the
moment
and
then
go
into
a
repository.
That's
also
been
set
up
where
we
can
document
all
these
things,
but
I'll
verbally
go
through
them
and
they're
in
the
meeting
minutes
as
well.
So
as
for
communication
channels,
we
have
a
slack
channel
specifically
under
the
oci
slack.
A
I'm
just
grabbing
the
name
of
it
because
it
escapes
me.
I
think
it
is
wg
dash
reference
dash
types
feel
free
to
join
that
channel
and
that's
for
any
discussions
or
information
pertinent
to
this
working
group
in
the
operation
of
the
working
group.
Instead
of
having
a
specific
mailing
list.
If
you've
worked
in
the
oci,
there
is
the
dev
mailing
list
and
there's
a
few
others.
A
Chris
anachik
from
oci
had
suggested
that
we
use
github
discussions
for
any
long
running
mailing
list
type
activity
and
if
you
go
to
the
specific
repository
which
is
under
open
containers
I'll
place
a
link
in
the
chat,
you
should
see
in
the
middle
there
there,
the
middle
tab
in
the
github
interface,
there's
a
discussions
tab
and
you
can
go
under
there
and
start
discussions
as
you
would
like
on
a
mailing
list
for
any
long-running
things.
So
I
think
I
don't
know.
Has
anybody
used
discussions
before
any
positive
negative?
I
haven't
personally
used
them.
A
Yeah,
so
I
think
that's
up
for
grabs
tianan
and
whether
that's
a
good
forum,
I
think
it'll
be
interesting,
is
what
constitutes
discussion
over
issues
and
comments
on
issues
but
we'll
give
that
a
shot
and
and
provide
feedback
there.
As
for
the
cadence
of
this
meeting
amy,
thank
you
very
much
for
getting
this
together
has
set
it
as
a
weekly
meeting.
That's
up
for
grabs
whether
it
needs
to
be
weekly,
bi-weekly
monthly.
A
Let's
just
feel
that
out
for
now,
but
it
is
at
this
time
weekly-
and
I
know
I
did
see
in
the
doodle
that
this
didn't
work
for
many
many
people.
So
I
think,
as
we
get
this
up
and
running,
there
is
opportunity
to
have
a
subsequent
time
for
those
folks
who
this
time
doesn't
work
for,
and
we
can
make
amendments
to
that.
So
I
guess
we'll
just
lead
as
as
this
is
convenient
or
inconvenient
to
folks.
A
Please
just
bring
that
up
and
we
can
figure
figure
it
out
because
it
didn't
work
for
everybody
and
then
finally,
there
is
this
repository
under
the
open
containers.
Github
org
called
wg-reference
types
that
is
the
home
of
any
output
work
that
we're
going
to
do
here.
A
There
is
nothing
in
that
at
the
moment,
so
I
actually
took
a
look
around
at
some
other
working
groups
in
the
ecosystem,
namely
kubernetes
working
groups
and
how
they
operate,
and
they
actually
place
a
lot
of
the
information
that
I've
gone
through
in
a
readme
to
begin
with,
so
I
was
thinking
you
know
in
the
next
week
I
would
put
put
a
pr
in
to
actually
just
document
the
communication
channels.
A
The
people
involved
when
the
meeting
cadence
is
so
that
it
is
discoverable
at
an
organizational
level
and
then
any
you
know,
work
outputs.
We
can
go
in
there
as
well.
So
I
think
that
that
repository
at
the
moment
is
completely
empty
and
we
need
to
build
that
out,
but
that
is
a
working
space
that
has
been
created
for
the
output
of
this
working
group.
So
we
need
to
make
sure
that
we're
putting
everything
there.
A
At
this
point
I
will
pause
I've.
I've
spoken
a
lot
before
I
I
had
suggested
a
group
activity
in
the
agenda,
but
is
there
anything
else?
That's
burning
people
want
to
bring
up
right
now
that
we
should
discuss.
A
Nope,
okay,
excellent.
I
think
I
will
say
if
you
take
a
look
in
the
proposal
down,
I
think,
halfway
down
the
proposal,
there's
been
a
lot
of
work
in
this
space
already.
So,
if
you're
not
familiar
with
anything,
that's
happened
already.
I
would
like
you
know,
there's
an
opportunity
to
go
and
have
a
look
at
any
prior
art.
That's
been
brought
to
the
table
and
I
think
in
the
proposal
there
is
a
good
cross-section
of
links
and
we
should
get
it
updated
if
we're
missing
anything
there
of
different.
A
You
know
ideas
that
have
already
been
brought
out.
I
wanted
just
that
to
set
the
stage
here
and
then
obviously
we
can
take
that
all
those
things
as
input
to
you
know
how
this
working
group
works
on
things
together
as
well.
A
So
the
next
thing
was
that
I
had
put
here
was
elevator
pitch
working
group
mission
statement,
so
I
thought
it
would
be
interesting
to
us
all
to
come
together
and
write.
You
know
in
in
a
simple
statement
a
paragraph
or
a
couple
of
sentences,
what
we
would
like
the
charter
or
the
mission
of
this
working
group
to
be,
even
though
it's
it's
loosely
defined
in
the
proposal.
A
So
if
somebody
was
to
stumble
across
or
look
for
this
working
group,
they
could
look
at
it,
read
a
couple
of
sentences
and
know
exactly
why
it
is
that
we
meet
and
whether
they
want
to
participate
and
exactly
what
we're
trying
to
achieve
here.
Has
anybody
done
an
exercise
like
this
before
or
have
any?
You
know,
feelings,
good
or
bad
about
doing
one
of
these,
I
thought
it
would
be
good
just
to
break
the
ice
between
us
all
and
and
do
something
like
that.
A
But
not
live
now.
I'm
thinking
about
this
a
little
does
it
make,
I
don't
know,
do
we
have
breakout
rooms
here?
Does
it
make
sense
to
split
us
up
into
like
three
or
four
and
have
a
couple
of
groups
come
back
with,
or
does
it
should
we
try
and
work
on
it
together
and
and
flesh
it
out
any
suggestions
on
that
front.
I've
done
both,
but
I
don't
I
don't
mind.
Either
way.
G
In
the
absence
of
breakout
rooms,
we
could
just
like
take
three
minutes,
write
down
what
you
think
in
one
sentence.
It
should
be
see
what
everybody
see
where
people
agree
or
disagree.
I
mean
it's
possible,
we
all
completely
agree
and
there
will
be
no
bike
shedding.
I
I
don't
anticipate
that,
but
just
at
least
getting
an
idea
of
like
what
do
you
think
we're
doing
here
could
be
useful.
I
hate
breakout
rooms,
I'm
trying
to
stave
off
breakout
rooms
as
much
as
I
can.
I
I
I.
A
Okay,
so
how
about
we
take?
Let's,
let's
call
it
five
minutes.
The
goal
of
this
five
minutes
is
to
write.
You
know
a
couple
of
sentences
to
a
paragraph
and
I
put
a
little
ideas
there
about.
You
know,
purpose
and
primary
motivation,
and
you
could
also
put
you
know
some
deliverables
as
well,
but
individually
do
you
want
to
go
and
well
how
about
everybody
go
and
write
up
their
idea
of
a
mission
statement
and
no
pressure?
You
can
participate
in
this
if
you'd
like
you
can
obviously
spectate
as
well.
A
That's
just
fine
and
then
we'll
coalesce
and
try
to
try
to
get
them
into
the
meeting
minute
stock
at
the
end
of
this
and
then
see
if
we
can't
pare
down
and
and
either
agree
on
something
real
time
or
let
have
it
being
starting
point
to
agree
on
something
asynchronously
as
well.
Does
that
sound
all
right?
Okay,
it's
it's
22
past
the
hour!
So
let's
just
do
till,
let's
say
half
past
the
the
hour,
so
we
can
sit
here
and
and
type
it
out
feel
free
to
put
it
in
the
notes.
A
If
you
want
that
to
be
your
scratch
pad
or
just
put
it
somewhere
and
then
at
the
end
of
it,
I
will
collect
all
these
things
in
in
the
notes
and
we'll
go
from
there
sound
all
right
cool.
A
I
know
the
silence
will
be
awkward,
but
in
in
in
many
cultures
silence
means
you're,
giving
things
considerable
thought
and
it's
time
to
be
respected.
So,
let's
think
about
silence
like
that.
A
Okay,
we
are
at
time
a
little
over
how's,
everybody
feel
do
they
need
a
few
more
minutes
or
are
they
feeling
good?
If
you've
already
finished
up,
I
see
a
bunch
in
the
in
the
notes
feel
free
to
read
through
them
or
do
we
want
to
press
on.
A
Press
on
okay
for
the
next
part
of
this
does
it
make
sense
to
go
and
review
these
things
and
see
how
they
sit
and
bits
of
the
missing.
Actually,
how
about
I
just
read
through
them,
each
individual
read
through
them
and
with
no
response
at
this
point.
I
just
want
to
hear
everybody's
feedback
and
see
it
out
on
the
on
what
everybody's
come
up
with,
and
then
we
can
coalesce
from
there.
A
Okay,
so
steve
in
the
elevator
pitch
section
of
the
notes.
Here
we
have
steve
steve.
Would
you
like
to
read
yours
first.
K
A
Thank
you,
steve
brandon,.
B
Yeah
mine
went
kind
of
technical,
but
I
just
said
creating
a
way
for
a
dag
plus
a
cast
to
attach
additional
child
child
content
to
an
existing
record
in
a
registry
so
kind
of
focusing
on
the
very
lowest
level
of
you
know.
If,
if
we
didn't
have
to
attach
external
child
content,
we
just
modify
the
original
thing.
This
would
be
easy,
but
I
didn't
want
to
get
into
what
the
content
was
or
anything
else
like
that.
L
L
Sure,
sorry,
no
it's
okay
figuring
out
the
minimal
acceptable
set
of
changes
necessary
on
the
image
and
distribution
specifications
to
enable
things
that
refer
to
slash
attached
to
other
things.
I
love
using
the
word
things
in
a
professional
context
because
it
makes
a
lot
of
people
uncomfortable
things
and
stuff.
A
E
Mine
also
had
stuff
in
it.
What
we
have
created
with
oci
distribution
is
the
world
wide
web.
Without
hyperlinks
we
could
not
have
search
engines
without
hyperlinks
search
engines.
Let
us
know
things
faster
more
efficiently
than
we
could
do
ourselves.
F
Cool
stuff,
better
yeah,
excellent
cool
stuff
love,
it
john.
M
My
noun
of
choice
is
objects,
so
so
my
goal
is
to
define
the
semantics
and
a
mechanism
for
describing
relationships
between
objects
and
then
also
a
mechanism
for
querying
those
relationships.
I
guess
implied
here-
is
oci
objects.
A
Okay,
I'm
really
enjoying
this
thanks
for
all
your
creativity,
everybody
so
far,
so
I
really
enjoy
going
through
these
things
and
see.
I
think,
expressing
you
know
seeing
everybody's
how
they're
thinking
about
it
and
putting
it
out
really
helps
us
come
together.
So
excited
love,
love
all
the
creativity
here.
So
thanks
for
participating,
I'm
up
next,
let's,
let's
see
which
noun
I
chose.
A
If
you're
australian,
it's
probably
going
to
be
mate
because
mate
is
our
thing
so
define
the
relationships
between
objects
stored
in
an
oci
registry,
for
the
emergent
need
for
a
centralized
point
to
provide
a
response,
so
I
feel,
like
I'm
a
little
cryptic
in
that,
but
that's
what
I
came
up
with
in
the
time
that
was
allotted.
A
Thank
you,
marina,.
J
Yeah,
let's
see,
I
have
proposing
a
change
to
the
oci
specification
that
allows
artifacts
to
be
discovered,
discovered
and
used
based
on
the
relationship
to
other
artifacts.
So
I
kind
of
took
it
from
the
the
user
side.
I
guess
rather
than
the
registry
side,
but.
A
Excellent
love,
it
jason.
G
I,
like
everyone
else's
more
than
I
like
mine,
but
mine,
was
evolving
oci
apis
to
allow
clients
to
express
relationships
between
oci
resources,
finance
resources
have
those
stored
in
a
registry
and
made
available
to
other
clients.
Resources
refer
to
other
resources
by
digest,
producing
a
dag.
I
think
that
might
be
a
little
in
the
weeds
of
technical
detail.
G
A
Excellent
jason,
I
just
want
to
let
you
know:
you're,
not
alone.
I
read
ever
I
read
yours
and
I
like
yours,
better
than
mine,
so
just
just
let
just
letting
you
know
that.
Well,
I
think
we're
probably
all
in
a
similar
vote,
while
we're
going
through
these
I'm
just
going
to
share
what
I'm
thinking
and
just
maybe
you're
all
thinking
the
same
thing.
What
are
the
set
of
nouns
verbs
and
and
things
that
everybody's
expressing
that
are
similar,
and
I
kind
of
want
to
do
you
know
maybe
word
cloud
as
a
little
bit.
A
You
know,
but
what
are
the
things
that
all
these
ideas
have
in
common
and
what
can
we
draw
from
them
all
and
create
a
single
statement,
because
I
do
see
a
lot
of
relationships
about
whether
it's
a
noun,
the
verb?
You
know
child
relationship
registry,
just
if
you
can
start
to
distill
those
things,
and
maybe
we
just
start
to
list
them
out,
but
before
I
go
any
further
I'll
give
everybody
else
an
opportunity.
Vincent.
D
Yeah
leveraging
a
container
registry
to
push
arbitrary
content
along
with
typed
associations
to
another
to
other
content
by
digest
reference.
So
that's
and
see.
Yeah
like
you
said
that
if
a
human
could
chew
this
up
and
spit
it
back
out,
I
think,
like
your
word
cloud,
this
thing
references
this
other
thing,
they're
associated
with
your
type.
You
can
look
them
up.
You
can
index
them
whatever.
N
Sure
my
thought
was
drafting
a
set
of
oci
reference.
Spec
changes.
I
picked
those
you
know
specifically
to
highlight
that
really
the
output
needs
to
go
into
the
oci
spanx
if
it
doesn't
right.
Why
are
we
here
in
osci
right
to
enable
artifacts?
I
I
picked
that
word
specifically,
because
that's
what
we've
been
calling
these
things?
You
know
for
the
last
year
and
a
half
so
keeping
that
name
for
for
now,
including
new
types
of
artifacts.
N
I
think
the
the
new
types
is
required
because
we
have
non-oci
image,
artifact
types
that
we
need
to.
You
know
proof
it
out
in
this
process
and
also
artifacts
that
refer
to
existing
oci
images.
Again
refer
referencing
urls.
That
sort
of
thing
right
linking
these
things
together,
has
become
primarily
important,
because
we've
got
to
extend
the
oci
images
at
a
minimum
out
of
this
process.
A
Thanks
mike
just
putting
the
call
out
to
anybody
else
who
would
like
to
share
me,
go
ahead,
vanessa.
C
Alrighty,
so
mine
is
kind
of
different.
I
feel
like
most
people's
are
very
technical.
On
a
specific
thing,
I
tried
to
focus
on
like
starting
with
community
needs,
and
I
wrote
it
out
because
I
have
I'm
a
very
verbose
person,
so
I'll
just
read
it.
The
reference
types
working
group
should
identify
and
keep
abreast
of
community
needs
for
interacting
with
different
kinds
of
artifacts
in
the
context
of
the
registry.
Once
use
cases
are
identified,
working
groups
should
explore
different
avenues
of
approaching
solutions.
C
C
To
make
it
really
easy
to
do
all
of
the
above
templates
for
commonly
used
services
interact
with
registries
and
community
friendly
documentation
and
tutorials,
and
I
think,
as
we
do
work
one
thing
that
I
have
a
hard
time.
Some
with
is
sort
of
always
kind
of
checking
back
in
with
like
the
problem
that
I'm
trying
to
solve.
I'm
working
on
something
really
technical
and
I
get
very
stuck
in
the
technical
work,
because
it's
just
really
cool
and
I
sort
of
drift
away
from
the
original
problem.
I
was
trying
to
solve.
A
Yeah
vanessa,
excellent.
I
I
I'm
super
impressed
that
you
managed
to
do
all
that
in
seven.
Odd
minutes
makes
me
question
my
life
choices,
but
that
I
think
that
was
a
really
good,
a
really
good
synopsis
of
the
the
problem
area
and
what
we
could
try
to
do.
Yeah
and
that's
that's
great
vanessa,
not
not
having
a
go
at
you
at
all.
A
I
think
that's
fantastic
and
I'd
love
for
you,
if
you're
comfortable,
to
share
that
in
the
the
meeting
notes
because
yeah,
I
think
it's
extremely
valuable
and
we
should
have
that
somewhere.
So
at
this
point
you
know,
we've
all
laid
out
different
ideas
for
a
mission
statement
here.
A
A
What
are
the
main
themes
we're
hearing
between
all
these
things
and
how
could
I
just
capture
those
specific
words,
because
I
think
what
we
can
do
is
given
the
time
left
capture
all
the
different,
nouns
and
verbs
that
are
in
here
put
them
down
on
paper,
and
then
we
can
try
and
see
if
we
can
coalesce
around
a
mission
statement-
and
at
least
you
know
have
something
to
work
off
before
we
meet
next.
Does
that
sound
reasonable?
Any
other
proposals
we
have
about
seven
or
eight
minutes
left.
G
See
you
hand
raised
yeah.
I
think
that
sounds
good.
I
think
one
thing
that
stuck
out
to
me
in
a
positive
way
about
all
of
these
things
was
that,
basically,
I
think
one
thing
I
think
we
all
agree
on,
please
throw
something
at
me.
G
If
you
disagree,
is
that
our
goal
is
a
set
of
recommendations
to
api
existing
oci
api
maintainers
and
not,
for
example,
new
apis
like
new
a
new
top-level
thing
aside,
they
have
on
the
level
of
image,
background
level
of
distribution,
spec
etc,
which
I
happen
to
agree
with,
but
also,
I
think,
is
a
good
idea,
because
we
don't
want
a
new
thing
and
our
output
is
recommendations
to
existing
groups
to
comment
bike
shed
and
change,
but
not
to
unilaterally.
This
group
is
not
going
to
like
produce
the
api
spec.
G
This
group
is
going
to
produce
a
recommendation
for
the
api
spec
owners
to
ratify,
validate
throw
tomatoes
out.
Is
that
more
or
less
how
people
see
it?
That
is
a
thread
that
I
saw
through
everyone's
answers
to
the
to
the
elevator
pitch.
Just
want
to
make
sure
that
that
is
what
I
was
hearing.
G
D
G
You're
saying
or
I'm
not
I'm
not
being
prescriptive
about
it.
I'm
I'm
trying
to
say
I
was.
I
noticed
a
through
line
of
people,
people's
elevator
pitches,
that
that
sounded
like
in
general,
a
place
where
we
read
was
that
our
output
is
not
a
fully
formed,
ratified
spec,
because
we
have
no
power
to
ratify
a
spec,
but
that
the
output
should
be
recommendations
to
the
groups
that
can
to
hey.
We
talked
about,
we
sat
in
the
room
and
talked
about
it
for,
however,
many
months
or
years
we
came
up
with
this
output.
G
D
Yeah,
no,
I
think
maintainers
maintainers
should
be
constantly
you
know
brought
along
in
that,
so
that
it's
not
right,
but
yeah
there's
there's
secondary
conversations
we
had
there
of.
Like
you
know,
that's
yeah.
I
think
we're
in
agreement
with
that,
because
I
think
there's
a
lot
of
work
to
do
and
a
non-blocking
thread
of
this
project
regarding
other
like
existing
specs
and
so
forth.
So
I'm
in
agreement
with
that
statement.
A
I
I
added
just
at
the
bottom
of
the
the
dock
here
I
can
share
my
screen
if
it
would
help
a
kind
of
a
word
cloud
from
the
above
and
I'm
starting
just
to
take
a
pass
through
different
pieces.
I'm
seeing
that
in
everybody's
responses.
Let
me
just
go
ahead
and
see
if
I
can
share
my
screen.
A
Do
you
all
see
that
I
don't
know
if
it's
the
right
size,
but
I
was
trying
to
trying
to
grab
specific
things
that
were
in
everybody's
output?
So
does
everybody
want
to
start
just
giving
me
words
and
I'll
put
them
here,
because
I,
I
think
you
know
the
who?
What
when,
where
and
why
the
construct
a
paragraph
is
is
something
we
could
at
least
coalesce
on
and
make
some
recommendations,
or
at
least
take
this
back,
and
we
can
all
individually
do
that.
So
are
there
any
other
words?
I've
found
object,
artifact
content.
A
A
Not
alone
you
you're,
not
alone,
I
think
you've
got
to
have
stuff
stuff.
Okay,
there's
no
incorrect
answer
here.
C
A
B
A
C
H
I
E
Lockheed
I,
like
the
word
link
for
relationship
linked
yeah
linked
stuff.
K
K
The
other
one
was
separable
objects,
several
artifacts.
K
A
A
All
right,
given
that
we're
at
the
top
of
the
time,
I
think
what
what
I
would
put
out.
The
call
to
do
is
to
take
these
words
that
we
have
in
common
and
for
those
who
are
interested
in
participating,
see
if
you
can
re-take,
given
the
the
information
you're
given
now
on
the
on
the
mission
statement
and
then
present
those
back,
and
we
will
add
that
as
an
agenda
item.
But
what
I'd
like
to
do
is
lock
on
a
mission
and
then
I
vanessa.
A
I
really
want
to
review
what
you
wrote,
because
I
think,
like
a
more
detailed
synopsis
that
would
go
with
the
original
proposal
would
be
really
nice
too
to
have
where
you
dig
into
each
each
of
these
components.
So
what
my
homework
is
going
to
be
is
to
start
to
get
a
pr
ready
to
lay
down
what
the
proposal
says
and
leave
space
for
a
mission
statement
and
maybe
a
more
detailed
synopsis.
A
And
then
what
I
would
also
like
to
put
a
call
out
is
for
the
next
meeting.
What
are
the
building
blocks
of
the
conversation
that
we
want
to
have
to
start
kicking
off
the
the
work
process,
but
I
really
wanted
to
take
some
time
with
everybody
here
to
make
sure
we're
all
in
agreement
for
what
this
working
group's
setting
out
to
do
and
lock
that
into
the
repository
so
that
we're
aligned
there.
So
I
think
we're
we're
well
on
the
way
to
that,
and
then
we
can.
A
A
Yes,
I
think
so
for
this
stage
until
there's
more,
you
know,
proposals
and
and
conversation
that
needs
to
happen
where
we
could
break
out
into
that
github
discussion.
But
I
I
agree
with
what
you
said
vanessa,
so
I
will
summarize
and
put
the
the
output
of
this
meeting
and
next
steps
and
actions
on
the
like.
I
like
the
cake
reaction
in
in
the
in
slack
and
then
you
know
we'll
see
you
all
next
week.
A
I
really
appreciate
everybody's
participation
today
and
making
this
a
fruitful
conversation
and
kicking
off
this
working
group
and
feel
free
to
reach
out
to
me
or
anybody
else.
If,
if
you
have
any
questions,
otherwise
we'll
see
you
all
next
week
and
I
look
forward
to
continuing
getting
this
off
the
ground.