►
From YouTube: [OCI-WG] Reference Types - 2022-03-01
A
A
Okay,
thank
you
yeah,
so.
C
Yeah,
that
kind
of
just
administrative
stuff-
okay,
I'm
just
going
to
pop
the
agenda
in
there-
sorry
to
cut
you
off
nisha
I'll,
just
pop
the
agenda
in
the
chat
there
for
everybody
following
along
at
home
or
wherever
you
may
be
today-
is
tuesday
march
1st.
It
is
20
days
till
spring,
so
that's
exciting
hope
everybody
is
doing.
Okay.
This
meeting
is
the
reference
types
working
group
meeting
under
the
oci
and
this
call
is
going
to
be
recorded
and
will
be
posted
to
youtube.
C
So
please
treat
others
with
respect
and
choose
your
words
carefully,
because
they
will
be
forever
immortalized
on
the
internet,
the
good
to
see
some
smiles
there
brandon.
I
said
that
one
for
you
calling
it
not
because
you
say
anything
like
that.
It's
just
always
interesting.
I'm
like
this
gonna
be
on
the
internet
and
backed
up
forever.
It's
always
a
fun
one.
We
have
an
agenda
here.
There
are
three
looks
like
two
main
items
here
and
we're
going
to
do
the
governance
one
first,
just
please
before
we
get
into
it.
C
Please
add
your
names
to
the
attendees
list
and
we
have
brandon
as
a
note
taker.
I
can,
I
can
kind
of
grab
silver
place
there
unless
jason
wants
to
beat
me
to
it
silver
medal
and
we're
going
to
go
into
the
agenda
right
now.
Governance
dock
is
approved.
We
have
the
reference
types
paul
request.
11
there
there's
two
pieces
here.
I
won't
steal,
nisha's,
thunder
I'll
pass
it
over
to
tanisha.
Thank
you.
D
I've
got
no
thunder,
you.
D
So,
very
quick,
the
pull
request
is
ready
to
go
for
the
most
part.
One
thing
I
want
to
bring
up
for
this
group
is
where
they
want
to.
Where
would
you
all
want
to
keep
notes?
D
D
There's
I
don't
predict,
I
don't
have
any
preference,
so
those
are
just
two
options.
If
everyone's
used
to
hack
and
b,
that's
fine
with
me,
too
brandon.
D
C
I
think
hackmd
it
sounds
like
I
have
experience
with
hackmd.
I
don't
have
any
issues
with
hackmd.
Is
there
any
like
dissenting
hackmd?
I
think
the
main
thing
is:
do
the
people
that
are
contributing
have
access
to
hackmd,
because
I
know
some,
you
know
different
different
countries
may
not
have
access,
probably
not
an
issue
with
hackmd.
I
haven't
come
across
it
before.
B
C
So
would
we
just
sunset
this
this
google
doc
move
it
to
hackmd
and
then
at
the
end
of
it,
because
the
issue
we're
trying
to
get
to
in
in
the
governance
dock
it
says
that
all
meeting
minutes
will
be
pr
to
the
working
groups.
Repository
and
the
goal
here
is
to
you
know,
cut
down
on
any
of
the
the
work
required
to
copy
and
paste
that
into
a
pr
and
get
that
merged.
Is
that
that's
the
goal
right.
D
C
Okay
next
thing
is:
when
do
we
is
it
from
next
meeting
on
or
I
don't
need
to
be
so
rigorous
about?
When
do
we
want
all
the
notes
today
to
be
over
there
and
should
we
move
over
or
how
about
we
just
say
punt
it
somebody
go,
create
a
hackmd
update,
all
the
links
and
then
from
henceforth
after
this
meeting
we
just
start
printing
the
things
unless
somebody
really
wants
to
go
back
and
and
do
all
the
notes
we
already
have
in
the
google
doc.
C
So
we'll
just
title
this
google
doc
at
the
end
that
is
deprecated
in
favor
of
the
hack,
md
link
it
and
then
don't
don't
remove
this
doc,
just
leave
it
enshrined
as
a
doc.
That's
now
read
only
at
least
in
title:
okay
cool!
Is
that,
okay
with
everybody
I'm
just
like
yeah,
let's
mechanically,
I
know
there
can
be
a
lot
of
moving
parts,
but
is
that
just
okay
after
this
meeting
well
next
meeting
will
be
in
the
hack,
md
and
somebody
will
get
a
location
to
get
and
get
the
links
updated.
E
Yeah
go
with
all
the
heck
md
I
just
wanted
before
we
moved
off
the
governance
just
wanted
to
check
out
through
a
comment
at
the
very
end
there
asking.
If
we
want
to
use
one
sentence
per
line
in
the
formatting
of
our
markdown
file
for
the
governance
it
just
makes
prs
easier.
Somebody
makes
a
change
to
one
line.
It
keeps
the
pr
to
one
line
and
not
a
whole
paragraph.
F
I'm
agree,
I'm
not
disagreeing
with
your
your
comment.
I'm
just
saying
is
that
in
governance,
or
is
that
in
you
know
contributing
code
guidelines
contributing
guidelines
as
opposed
to
governance.
C
C
Comment
yeah,
I
think
so
so
then
just
do
that
and
then
everybody
you
know
if
we
want
to
add
a
style
guide
or
whatever
to
the
to
the
main
repo,
but
I
think
we'll
follow,
because
I
I
agree
multi,
you
know
single
changes
on
a
multi-line
paragraph,
come
out
pretty
pretty
ugly
and
hard
to
see
what
you're
doing
so
lead
by
example.
Reformat
that
thing
and
go
from
there.
I
don't
have
any
issues
with
that.
I
think
it's
a
reasonable
suggestion.
Brandon.
D
D
And
then
I
didn't
see
it.
Okay,
I
think
that
that
completes
this
one
is.
C
There
any
tools
you
use
to
support
that,
or
is
it
just
basically
just
don't
word
wrap,
don't
press
end,
don't
have
a
slash
n
or
whatever
this
is
encoding.
Is.
E
D
I
I
wouldn't
go
that
far,
let's
just
I'll
I'll,
just
update
that
pr
and
ask
for
a
review
later
and
with
that,
I
will
make
a
place
for
all
the
meetings
to
go
and
document
whatever
we
have
discussed
here.
As
far
as
like
process
is
concerned.
B
Sure
so
spent
some
time
getting
our
user
stories
and
requirements.
Thank
you,
nisha
for
getting
our
final
list
into
the
repo.
So
if
you
go
to
the
repo
I
might
as
well
share
at
this
point.
B
If
you
go
to
the
repo
we
now
have,
you
can
see
yeah
okay,
so
we
now
have
this
table
here
which
has
like
kind
of
all
of
our
pages.
So
now
we
have
this
new
requirements
page
and
member
submitted
stories.
Distilled
is
where
kind
of
our
final
list
of
all
the
stories
that
everybody
came
up
with
and
de-duplicated
and
simplified
over
the
last
few
weeks.
So
we
have
this
and
I
think
now
is
probably
the
time
that
we
ought
to
start
talking
about
the
implementations.
B
So
over
the
last
year,
there's
been
a
few
that
have
been
have
come
up
in
within
the
oci
org
and
then
also
niches,
which
is
in
this
google
doc
and
yeah.
I'm
trying
to
figure
out
like
without
diving
right
into
it,
because
that
may
require
that
we
have
like
kind
of
the
owners
here
to
present
and
defend
their
points.
I'm
wondering
if
we
should
talk
about.
First
of
all,
just
enumerate
them
which
lockheed
thank
you
for
putting
all
those
links
there.
B
I
guess
the
first
question
is:
are
there
any
other
existing
proposals
and
then
another
question
to
there
is
beyond
that?
What
is
going
to
be
kind
of
our
framework
for
evaluating
these
and
just
want
to
make
sure
and
I'm
assuming,
if
you're
in
this
group,
the
answer
is
yes,
are:
is
everybody
involved
in
all
of
these
willing
to
flex
on
our
ultimate
results
of
the
working
group?
B
C
Order
it
went
in
sorry,
I'm
also
not
go.
C
D
Okay,
thank
you.
I
would.
I
would
suggest,
making
a
table
start
making
that
matrix
table
over
here
that
we
talked
about
last
last
meeting.
Then
we
could,
we
could
start
documenting.
You
know,
take
a
pro
take
one
proposal
at
a
time
and
start
going
one
by
one.
D
Maybe
does
anyone
care
about
what
proposal
we
choose
like
in
what
order
we
look
at
them.
I
don't
think
an
order
matters,
but
yeah
that
was
suggestion
and
question
and
I'll
give
the
floor.
G
Yeah,
I
was
just
going
to
say
I
agree
with
josh.
I
think
I
think
we
have
a
lot
of
potential
possible
implementations
to
go
through
I
if
we
come
up
with
a
new
one.
That
would
be
fine,
but
I
don't
want.
I
think
that
our
solution
is
somewhere
among
the
list
of
existing
proposals
either.
Some
combination
of
you
know
aspects
of
one
aspect
of
another.
I
think
nisha.
Your
idea
about
the
table
is
good.
G
I
don't
want
to
I.
I
have
seen
this
go
poorly
in
the
past,
where
people
argue
over
whether
it's
a
checkbox
or
a
question
mark
or
a
or
a
red
x
or
whatever,
and
I'm
fine
doing
that,
but
I
think
we
should
try
to
avoid
it's.
Not
it's
not
like
we're
going
to
tally
the
scores
and
then,
whichever
one
has
the
highest
score,
we
just
high
five
and
agree
on
and
go
home.
I
mean
I
wish
that
was
like.
That
would
be
awesome,
but
I
think
we
will
probably
not.
G
It
will
probably
not
come
down
to
a
strict
score
mechanism
that
doesn't
mean
that,
like
doing
the
exercise
isn't
useful,
but
I
think
when
it's
framed
too
much
as
a
scorekeeping
exercise,
people
argue
over
every
point
because
every
point
matters.
I
guess
what
I'm
saying
is
the
points
shouldn't
matter.
I
want
to
tally
the
points,
but
the
points
don't
matter
but
yeah.
I
think
that
could
be
a
useful
exercise.
Do
you
want
to
do
that?
G
E
And
I
think
I
had
something
else
to
say,
but
I'm
forgetting
it,
but
at
least
do
you
remember
we
were
saying:
are
there
any
other
proposals
to
look
at
what
I'm
thinking
of
is
that
we
may
want
to
include
no
proposal
at
all.
Basically,
can
we
implement
reference
types
with
what's
out
there
today
without
doing
reference
types,
and
so
that's
my
only
third
thought
of
you
know.
Maybe
it
just
involves
setting
up
a
higher
level
index
that
points
to
everything
kind
of
solution.
E
Using
the
existing
specs
and
just
pushing
up
an
index
that
points
to
everything
potentially
or
something
along
those
lines,
just
without
adding
an
extra
api
is
there
a
way
to
do
what
we're
trying
to
do
without
adding
something
new
to
our
ci?
And
the
point
that
I
was
forgetting
is
as
we're
reviewing
this.
I
don't
think
we
really
put
it
there
as
a
user
story,
maybe
kind
of
sort
of
we
want
backward
compatibility
and
not
breaking
things,
but
I
want
to
make
sure
we
don't
break
things.
E
I
know
one
of
the
reasons
that
justin
pulled
back
his
proposal
was
that
it
implemented
the
possibility
of
creating
circular
loops
and
in
some
of
the
designs,
and
so
that
didn't
work
really
well
when
you're
trying
to
do
things
like
reference
counting
for
garbage
collection.
So
that's
the
one
extra
piece
to
throw
in
there.
F
F
My
hope
is
we're
finding
we're
seeking
a
way
to
support
the
existing
registries
kind
of
like
old
cars
and
we're
defining
a
way
for
active
registries
that
can
solve
the
gaps
that
I
think
we
identified.
So
I
don't
know.
If
we're
I
mean
look,
we've
always
said:
if
we
can
find
a
solution
that
doesn't
require
any
changes.
Awesome,
we
just
couldn't
find
one.
So
that's
what
we
wound
up
where
we
did
with
some
of
the
proposals.
F
C
Okay,
thank
you
just
a
couple
of
comments
from
my
perspective
and
I
started
just
going
down
the
rabbit
hole
thinking
about
how
we
would
do
a
matrix
or
whatever.
C
So
I
think
it's
I
think
it's
really
important
to
have
you
know
either
a
q
a
or
somebody
present,
maybe
just
a
q
a
but
I'd
like
everybody
to
get
on
the
same
understanding
and
have
read
every
proposal,
the
people
that
are
part
of
it
because
it's
you
know,
I
don't
know
any
forum
in
the
world
where
people
have
taken
a
proposal
and
never
discussed
it
and
ratified
it
off
a
sheet
of
paper.
If
there
is
a
proposal,
there
is
a
way
to
do
that.
C
I
just
don't
think
humans
read
a
piece
of
paper
and
say:
let's
make
that
law.
So
I
don't
know
if
there
have
been
like
here's
the
proposal,
so
I
would
invite,
I
think
you
know
soliciting
all
the
proposals.
You
know
I
put
a
few
in
there
niches,
steve's
and
and
dan's.
C
I
don't
know
if
they're
any
other,
but
I
think
getting
a
list
of
the
proposals
is
good,
whether
we,
I
think,
if
we
to
build
a
matrix,
I
think
it's
really
good
to
evaluate
and
have
some
kind
of
rubric
as
we
go
through,
but
not
what
jason
said:
it's
not
a
score
card.
It's
not
like
whoever
first
gets
to
10
points
wins
it's
like.
C
C
I
think
if
we
could
come
up
with
a
you
know,
a
matrix
where
we
could
in
a
way,
just
you
know
not
score,
I'm
not
going
to
say
score
but
add
a
way
to
distinguish
as
we're
reading
through
everybody
keeping
their
own
kind
of
respective
view
on
it,
and
while
we're
doing
that,
we're
having
we're
inviting
the
proposal
owners
to
come
in
and
talk
about
their
proposal
and
either
we
structure
the
time
and
say
for
each
proposal.
C
We
want
to
know
these
five
things
or
we
all
read
the
proposals
and
say
this:
is
a
q
a
based
and-
and
this
is
you
know,
a
one
hour
time
slot,
where
all
we're
doing
is
answering
questions
about
the
proposal,
because
I
don't
know
again
you
and
me:
I've
read
the
millions
of
comments
on
all
the
proposals.
I
actually
can't
contextualize,
where
the
conversation
is
at
anymore
and
whose
questions
have
been
answered
and
whose
questions
haven't.
C
So
I
just
think
that
time
inviting
each
of
the
proposal,
writers
or
group
of
people
to
come
in
and
actively
answer
questions
outside
of.
What's
on
the
piece
of
paper
to
help
understand
where
we're
at
with
backwards
compatibility,
I
think
that's
the
biggest
thing
in
proposal
in
context
of
all
these
proposals,
so
I'm
just
thinking
about
you
know,
I
think
getting
a
matrix
done
would
help.
C
B
B
F
B
B
Do
you
think,
is
that
a
dupe
of
96
or
the
evolution
of
number
96,
or
should
that
be
the
primary
link
for
tob
96?
C
H
B
Yeah,
it
kind
of
gets
into
what
I
want
to
propose.
So
there's
several
links
here
across
github
and
with
niches.
We
have
google
docs
and
a
lot
of
text
in
different
formats
and
I'm
wondering
for
us
to
like
truly
comprehend
what
is
going
into
here.
I
do
think
the
history
is
important,
but.
D
B
Wondering
can
we
come
up
with
like
a
markdown
template
that
is,
for
example,
changes.
This
would
require
in
image
changes.
This
would
require
in
distribution
and
then
a
bullet
list
of
the
history,
because,
if
I
have
these
things
in
the
same
format,
I
can
really
reason
about
the
differences
or
the
similarities
between
them.
B
Like
I
was,
I
was
just
looking
at
auras
earlier
and
it's
trying
to
you
know.
I
think
it's
well
organized,
but
it's
trying
to
find
where
the
specific
changes
are.
I
had
to
click
down
into
other
places
and
I'm
just
wondering
if
we
want
a
working
group
template
and
then
my
follow-up
to
that
is
like.
B
Can
we
give
these
three
pet
names
or
person
names,
so
we
can
refer
to
start
referring
to
them
versus
just
saying
number
96
like
can
I
we
call
one
or
us
one
misha,
whatever
you
want
to
call,
and
I
can
I.
C
B
Okay
and
then,
but.
C
I
think
you're
right
josh
and
I
see
people,
let's
create
a
format
for
the
proposal
and,
let's
capture
the
proposal
in
the
working
group
thing.
I
know
that's
a
bit
of
work
for
everybody,
but
maybe
if
we
identify
the
bits
that
we
could
categorize
that's
going
to
make
it
easier
for
everybody
to
review,
because
I
found
the
same
thing:
it's
like
I'm,
linking
together,
58
different
things,
code
changes
issues
pr's
to
grab
a
full
context.
What
is
the
context
we
actually
need
for
to
evaluate
the
proposal
and
then
having
a
standard
format?
For
that?
G
Yeah,
I
didn't
know
what
josh
was
still
going,
but
in
any
case
I
think
the
same
advice.
I
would
give
the
same
advice
to
the
requirements
if
we're
going
to
make
a
a
matrix,
having
like
the
proposals
be
across
the
top
and
the
requirements
be
across
the
side.
It'd
probably
be
easiest.
If
we
could
have
a
shorthand
for
each
requirement
like
filtering
a
filtering
b,
filtering
c
rather
than
the
whole
text
of
the
requirement
and
then.
G
D
Okay,
I'll
talk
guy.
D
So
I
would
just
point
out
that
the
proposal
that
I
have
does
not
have
any
specific
code
changes.
D
In
fact,
I
would
almost
go
as
far
as
to
say
that
I
don't
know
whether
the
changes
would
be
complex
or
simple
to
implement,
because
I
don't
actively,
I
don't
regularly,
contribute
to
any
of
these.
D
B
B
You
know
if
you
need
help
from
one
of
us
of
like
where
this
is
where
that
change
would
be
needed,
but
does
that
make
more
sense.
D
Yeah
that
that
would
help
me
at
least
like
to
figure
out.
B
D
C
C
B
B
B
Where
things
will
land,
I
think
actually,
the
dance.
B
Dan's
one
does
a
pretty
good
job
of
like
breaking
this
down
into
kind
of
the
section
so
image,
spec
distribution
spec,
like
the
different
pieces
like
this,
is
what
I
would
like
to
see.
What
are
the
api
requests
that
will
land
into
the
distribution
spec
table
that
need
to
be
changed?
I
don't
care
so
much
that
it's
broken
down
into
these
categories
anymore,
like
I
want
to
know
specific
json
schema
changes
in
image,
spec
and
http
api
changes
in
distribution
spec.
B
Those
two
things,
I
think,
is
a
good
starting
place
and
then
I
would
also
recommend,
based
on
with
these
links,
just
like
a
simple
background
section,
that's
just
bullets,
and
if
people
want
to
go
read
the
history
like,
I
think
they're
it
will
give
them
the
context,
but
I
think
for
someone
just
walking
in
here
like,
for
example,
like
vanessa,
keeps
singing
she
can't
make
this
meeting
but
want
you
know
like.
If
somebody
wants
to
contribute
and
evaluate
these
things,
they
can
look
at
the
technical
specifics.
B
I
don't
know,
but
is
there
anything
else
that
we
care
about?
I
mean.
F
I
think
there's
you
know
this
is
one
of
the
journeys
that
we've
kind
of
been
taking
is.
Do
we
because
there's
a
couple
of
choices
here,
we
can
make
changes
to
the
well.
You
know,
brandon's
proposal
is
what
can
be
done
if
we
can
make
no
changes.
So
I'm
curious
what
that
would
look
like
assuming
we're
making
some
changes.
Is
it
changes
to
the
existing
image
spec
image
index
spec?
Is
it
an
additional
api
in
distribution,
or
is
it
a
new
manifest?
F
B
F
F
F
Per
se,
that's
kind
of
what
I'm
so
you're
asking
a
slightly
different
question.
Josh,
let
me
the
first
part
is:
is
there
a
new
document
format
that
describes
the
way
things
can
be
persisted
in
a
registry
that
is
more
generic
than
just
the
container
image
runtime
format,
because
that's
what
the
image
spec
was
built
around.
It
had
some
good
gene
general
out
generation,
moralities
geez
around
it,
that
we
were
able
to
leverage
and
we're
trying
to
preserve.
That
was
the
multi-layer
thing,
for
instance,
but
there's
some
limitations.
F
We
can
add
some
properties
to
it.
If
the
group
wants
to
do
that,
I
think
we
have
to
figure
out
what
does
that
mean
for
versioning
and
what
does
that
mean
for
down-level
clients?
If
we
decide
that
that
is
too
risky
and
breaking
which
is
kind
of
how
we
wound
up?
Where
we
did,
then,
then
the
proposal's
a
new
manifest
where
the
manifest
lives
is
a
different
question.
Actually
I've
kind
of
been
positioning,
my.
C
F
C
Api
changes,
generic,
and
that
includes
padding,
because
if
we're
too,
you
know
if
we
force
it
right,
I
don't
know
how
brandon's
node
changes
right,
so
it
just
has
to
be
generic
enough.
Are
there
changes?
What
are
they?
Because
I
think
that's
what
josh
is
trying
to
get
at
right?
Are
there
changes
and
lay
them
out
in
the
context
of
what's
there
additions
or
nothing.
F
C
But
in
the
context
of
writing
the
proposal
for
somebody
to
read
like
vanessa
we're
just
using
vanessa's
name
here
as
somebody
who
can't
be
as
active
at
the
meetings.
How
do
we
get
that
information
so
that
vanessa
can
evaluate
the
proposals
anyway?
There's
lots
of
hands
up,
but
I
I
think
stay
on
target
here
of
collecting
the
proposals.
C
G
H
Yeah,
I
think
the
way
that
I
think
about
this.
I
could
see
it
structured
as
sort
of
what
changes
would
this
proposal
make
to
the
existing
image
spec
the
existing
distribution
spec,
and
maybe
what
changes
don't
fall
into
those
buckets
or
we
could
structure
it
more
along
the
lines
of
the
requirements
that
we
had
like.
We
had
a
bucket
for
content
management
and
a
bucket
for
filtering,
and-
and
so
I
think
those
are
the
kind
that's
the
kind
of
conversation.
I
think
that
I
would
like
to
see
is
is
sort
of
with
these
proposals.
H
Do
we
think
it'd
be
more
helpful
to
structure
it
in
terms
of
how
we
want
to
change
the
oci,
specs
and
sort
of
list
out
changes
per
spec,
or
would
it
be
more
helpful
to
say
this
proposal
addresses
our
requirements
in
this
way.
D
Okay,
since
no
one's
talking
I'll
talk,
I
I
think
it
would
be
useful
if
we
go
with
something-
and
maybe
maybe
we,
as
the
authors
are
entering
things,
may
have
some
places
where
they
think
something
doesn't
fit
or
something
needs
to
be
added.
D
But
and
that's
as
far
as
templating
is
concerned,
with
regards
to
whether
we
like
talk
about
the
proposal
and
then
compare
it
with
the
with
the
whether
it
meets
certain
requirements
or
not,
I
think
that's
what
lucky
was
saying.
We
could
have
a
q
a
session
with
the
autos,
and
that
should
like
further
more
solidify
that
conversation.
B
So
I
I
kind
of
don't
know
what
we're
talking
about.
I
I
feel
like
I,
I
I
don't
think,
there's
really
any
contention
on
like
wanting
to
have
a
template
that
people
can
reason
about
these
things.
Like
I
don't
really
think,
there's
an
argument.
I
think
some
things
can
be
bucketed
in
the
distribution
spec.
Some
things
can
be
bucketed
an
image
spec
some
things.
Maybe
they
don't
have
a
bucket.
I
don't
think
like
we
can
just
say
no
bucket.
B
B
I
just
want
to
see
jason
and
I
want.
I
just
want
to
see
jason
and
I
want
to
have.
I
want
to
have
bullets
with
all
the
links
that
we
have
here
with
all
of
the
history.
If
people
want
to
go
and
read-
and
you
know
people
put
a
lot
of
effort
into
writing
very
lengthy
comments,
and
I
think
that
we
need
to
capture
that
into
very
small
documents
for
a
b
and
c
so
people
can
look
at
a
b
and
c
and
they
see
it
in
the
same
format
and
their
brain.
C
Yeah,
I
think
that's
fine
josh,
let's,
let's
do
that.
What
is
what
is
the
most
succinct
set
of
changes
and
and
get
all
the
craft
and
big
blobs
away
of
text
that
can
be
part
of
the
q,
a
conversation
that
we
have
for
each
proposal
and
then
I'm
j.
The
only
thing
I
was
tripping
over
mentally
in
my
head
was
you
know,
nisha's
saying:
well,
I
don't
know
what
changes
my
proposal
has.
C
E
C
Into
it
you
can
fit
into
it.
Okay,
then.
The
only
thing
is
nisha's
requirements,
because
I
think
getting
it
down
to
that
kind
of
level
of
fidelity
is
going
to
make
it
a
lot
lot
easier
for
people
to
pass.
What
exactly
are
the
changes
you're
proposing.
D
So
I
can
jump
in
here
josh
when
you
say
you
just
want
to
see
json
and
http.
D
B
Let
me
give
a
very
succinct
example,
so
this
is
from
the
previously
referred
to
or
us
now
it
is
a
so
everyone.
This
is
a
so
I
this
document
has
a
lot
of
json
and
http.
B
B
And
I'm
not
ripping
on
this
like
there's
all
of
them
like
nisha.
I
want
to
see
this.
I
don't
want
to
see
this,
so
the
closest
we
can
get
to.
B
I
don't
know
if
you
do
we
need
go.
I
don't
think
so.
That's
just
one
language
that
will
implement
these
things.
D
B
D
B
B
C
B
D
C
C
You
know
author
or
authors
to
come
and
present
we
spend
the
next
two
or
three
weeks
going
through
all
the
proposals.
The
only
thing
we
can
do
in
parallel
is
if
we
want
to
get
a
matrix
together
and
and
brandon
actually
came
up
with
a
pretty
good
matrix
last
week.
Verbally,
I
don't
know
if
he
wants
to
just
take
a
quick
stab
at
like
putting
down
his
initial
proposal.
C
I
think,
looking
at
a
sheet
of
paper
with
a
scaffolded
out
idea
is
going
to
be
better
than
us
entering
a
room
with
you
know
empty-handed
and
then
trying
to
figure
it
out
so
brandon.
I
don't
know
if
you
would
be
willing
to
like
put
together
the
first
matrix,
which
is
because
I
think
you
sorry
to
kind
of
point
at
you,
but
you
had
some
good
ideas
last
week
about
yes,
no,
maybe
so
you
know
if
it's
this
day
of
the
week,
you
know
there
were
some
nice
ideas.
C
I
thought
and
then
I
think
the
only
thing
we
need
to
land
in
the
repo
is
the
numbering
of
the
proposals,
and
I
guess
you
know,
does
the
proposal
owner
even
care?
We
can
just
solicit
any
proposals
they
put
it
in
and
then
we
number
them
afterwards.
C
H
To
we
have
owners,
for,
I
guess
specifically
for
proposal
b.
Is
that
is
josh?
Are
you
taking
that
forward
or
is
stan
gonna.
H
B
D
C
C
So
you
get
template
josh
somebody
go
in
and
number
the
user
stories
and
propose
a.
I
don't
know
us.
I
don't
wanna.
What's
another
word
for
scorecard
or
matrix
or
compatibility
matrix,
it's
maybe
better
where
it's
not
like.
There's
no
winner,
because
you
get
you
tick,
all
the
boxes,
green,
it's
more
like
an
evaluation
of
which
use
cases,
your
support,
rubric,
sure.
D
Rubric
sounds
good.
I
can't
do
that.
C
D
No
worries
I
will
make
it
and
brandon
can
comment
on.
C
All
right,
then,
and
if
we
land
these
things,
so
it's
up
the
proposal
owner.
I
guess
if
there's
any
comments,
but
we
should
just
have
a
directory
called
proposals.
I
have
them:
name
numbered
lettered,
sorry,
a
through
a
through
whatever
n,
and
we
can
just
merge
them
in
if
the
proposal
owner
is
there.
If
there
are
any
comments
on
bits
that
are
missing
or
it's
unclear,
let's
just
make
sure
it's
clear
before
it
gets
merged
and
then
but
the
the
merging
of
that
proposal
just
puts
it
in
a
bucket
for
evaluation.
C
B
B
And
then
we
have
proposals,
we
make
a
new
folder
proposals.
D
C
B
B
B
C
I'm
just
going
to
go
ahead
and
say
no
there's
not
only
because
you
can't
fight
over
something
that
isn't
written
down,
so
I
would
just
say:
let's
do
it,
because
we
don't
know
what
we
don't
know.
That
would
just
be
my
perspective,
so
write
these
things
down,
figure
out
the
bits
that
don't
fit.
If
there's
any
contention
over
those,
we
can
talk
about
it,
but
until
we
do
it
we're
just
we
don't
have
contacts.
That's
my.
F
C
D
E
E
B
B
Steve
to
your
point,
my
idea
of
the
template
would
not
actually
write
down
any.
Does
this
solve
this,
these
user
stories
right?
I'm
just
wondering
if
we're
on
the
same
page
with
that
we
would
have
the
thing
without
any
context
and
then
we'd
have
the
requirements
and
then,
as
a
group,
we.
B
Right,
like
I
don't
feel
like,
I
can
submit
a
proposal
and
be
like
all
right
checks,
all
these
boxes,
because
then
that's
just
I'm.
F
C
So
we
just
need
a
place
for
the
proposal
and
the
proposal
is
the
minimum
viable
set
of
you
know
json
or
http
apis
that
we're
getting
chosen
when
we
change.
So
we
just
need
to
take
what
you've
already
documented
in
in
auras,
artifacts
and
and
chop
shop
that
up
into
the
set
of
changes
and
so
steve
you-
and
I
will
work
on
that
for
proposal.
C
A
josh
will
go
and
work
with
dan
on
the
that
for
proposal
b,
nisha's
going
to
do
that
for
c
brandon's
going
to
do
that
for
d
we
come
back
then
everybody
comes
in
discusses,
gives
more
context.
People
ask
questions,
so
we
use
that
you
know
proposal
as
a
way
to
facilitate
the
conversation,
and
meanwhile
we
can
all
go
fill
out
the
rubric
card
of
the
check
marks
as
we
go
through
here
and
then
coalesce
on.
C
B
And
I
was
just
thinking
like
what
you
said:
lucky
the
http
and
json
changes
like
we
can
avoid.
Actually
any
discussion
of
distribution,
spec
image
spec.
I
think
at
this
stage
too
right
and
that
solves
steve's
concern.
We
can
say
the
new
manifest
is
actually
a
json
thing
and
the
rep
refers
api
is
a
http
thing
and
then
later
once
we
have
our
final
thing.
Those
get
bucketed.
C
C
D
Yeah,
this
is
all
going
to
go
on
the
internet.
I
am
not
seeing
anything
more.