►
From YouTube: RATS Architecture Design Team, 2021-01-12
Description
RATS Architecture Design Team, 2021-01-12
B
Oh
yeah,
a
lot
of
things
should
be
in
conserve
and
be
ad
hoc
use
items.
B
So
this
recently
webex
does
not
connect
my
computer
audio.
It
does
nothing.
I
have
to
click
on
connect
and
the
screen
opens
that
tells
me
that
I'm
not
connected
yeah.
I
know
that
so
then
I
clicked
on
the
computer
and
it
gives
me
more
options
like
yeah.
Just
just
get
audio
you
know,
so
I
don't
know
why
these
are
jumping
through
loops.
So
now.
A
A
A
A
But
now
now
it's
available
in
the
browser
which
of
course,
you
know,
doesn't
feature
on
unexamined
random,
binaries
right
but
anyway,
so
dave's
built
a
bunch
of
pull
requests
and
reviewing
I
re
them.
I
think
they
will
take
us
the
bulk
of
the
hour
to
do,
and
you
also
updated
some
issues
dave
which
you
said
was
gonna
be
emails.
C
A
C
On
page
one,
it's
the
one
that
has
the
label
that
says
email
only.
I
created
a
new
label
because
I
didn't
remember
that
we
were
using
won't
fix
for
that.
C
A
B
So
there
is
some
idea
here
that
the
relying
party
produces
more
digestible
or
more
or
easier
to
appraise
things,
but
I
think
other
parts
of
the
text
already
highlight
that
so.
C
Ones,
the
the
first
one
is
a
much
larger
one
and
I've
only
reviewed
part
of
it
myself
so
for
the
ones
that
I
generated
read
the
issue
first,
because
I
didn't
edit
any
comments
so.
C
C
Oh
yeah,
so
thomas
had
proposed
something
pink
kind
of
liked
it.
I
said
that
that
proposed
text
had
an
error
in
it,
and
so
I
tried
to
have
the
right
area.
The
error
in
the
text
is
in,
I
think,
thomas's
text
where
he
said
the
same
serialization.
The
attestation
result
the
same
serialization
is
the
evidence.
That's
not
what
was
intended.
It's
the
same
serialization
format
as
the
protocol
between
the
a
tester
and
their
lying
party.
C
The
evidence
is
here,
is
not
relevant
to
this,
so
you
can
go
and
look
at
the
actual
text
and
see
why
so.
This
is
talking
about
the
background
check
model
right.
So,
in
the
background.
C
B
But
I
think
to
thomas
the
point
is:
if
you
build
a
a
complete
ecosystem,
you
can
reuse
code,
so
I
actually
think
that
he
means
evidence
here.
C
Right
and
I'm
saying
that's
wrong:
what
is
important
is
in
a
particular
entity
to
be
able
to
reuse
code.
That
means
you
have
to
be
able
to
reuse
a
parser
that
you
already
have
right.
So
if
you
have
a
protocol
between
a
relying
party
and
a
tester,
you
know
http
or
smb
or
whatever
it
is.
That
does
something
you
know
opc
way,
fill
in
your
favorite
protocol.
You
know,
then,
being
able
to
reuse
the
parser
that
that
protocol
already
requires
is
what
allows
you
to
code
an
attack
service.
I.
B
C
Thomas's
proposal
changes
the
meaning
of
the
original
stuff.
Is
there
that
the
thing
was
filed
to
say
it
wasn't
clear?
It
wasn't
that
that
had
the
wrong
meaning,
so
thomas's
proposal
doesn't
actually
address
the
actual
phone.
Okay
yeah.
So
not
a
separate
issue.
Oh
thomas
is
on
oh,
he
looks
like
he's
in
the
process
of
getting
audio.
So
if
we
wait
a
minute
okay,
yeah.
B
C
A
It
does,
it
says,
he's
unmuted
yeah,
but.
B
C
C
A
So
the
conversation
was
about
the
the
the
proposal
text
and
dave
has
made
a
point
that
the
ability
to
reuse
depends
upon
not
that
it's
the
same
serialization,
but
that
that
you
have
a
existing
parser
for
the
same
serialization.
I
guess.
E
Okay,
yes,
I
I
read
these
comments
and
I
agree
so.
E
C
B
My
my
follow-up
question
so
so
there
is
no
added
benefit
in
tentative
with
the
original
text
that
says
that
evidence
and
attestation
results
have
the
same.
What
you
call
it
now
format
is
of
benefit,
so
that
is
not
a
requirement.
There
was
such
an
accidental
association.
Yes,.
D
C
Original
text
did
not
mean
that
okay,
is
there
any
case
where
that
matters?
I
would
say
only
if
you
were
no
okay
and
here's.
Why
right?
The
attester
has
parsers
for
two
things:
it
has
parsers.
Well,
it
has
a
formatter
for
evidence
and
it
has
a
formatter
and
parser
for
the
protocol.
My
the
protocol,
I
mean
the
resource
access
protocol
right,
the
normal
thing.
A
C
Do
attestation
right,
the
verifier
has
two
parsers.
Well,
when
I
say
parser
a
formatter
is
okay,
I
did
parses
evidence
and
it
formats.
You
know
attestation
results
right,
and
so
maybe
you
can't
reuse
code
between
a
parser
and
a
formatter.
The
relying
party
has
two
parsers
right.
It
has
the
attestation
results
and
the
protocol.
The
only
other
case
that
I
can
think
of
where
you're
parsing
two
different
things
is
inside
the
verifier,
where
you
get
information
from
different
sources
like
endorsements
and
evidence,
and
you
can
make
an
argument
that
says.
C
A
Okay,
so
the
other,
the
other,
the
other
part
dave
which
which-
and
I
completely
agree
with
you.
The
other
part
about
this
is
mind,
share
or
brain
code
print
code
footprint
right,
which
is
that
when
they
are
in
the
same
format,
then
one
savings
is
that
you
probably
have
the
same
person
working
on
it,
because
they
don't
have
to
think
as
much
so
well.
You.
A
Yeah
so
yeah,
but
that's
the
point
is
that
that
that
that
probably
somebody
can
they
can
then
round
trip
things
easier
and
anyway
there
are
advantages
and
but
but
at
no
point
do
we
say
there
was
an
advantage
that
the
well
I'm
actually
trying
to
think
about
that
the
verifier
a
tester
link.
If,
if
it's
similar
to
the
verifier
relying
party
link,
then
that's
that's
not
a.
I
think,
you're
saying
that's
not
necessarily
relevant,
although
in
the
background
check
model,
maybe
it
is.
C
I
guess
you
could
argue
that
in
the
attester,
so
if
I,
if
we
don't
talk
about
a
parser,
if
we
talk
about
a
formatter,
meaning
the
thing
that
that
constructs,
you
know,
cbor
constructs
json
or
constructs
you
know.
Whatever
the
format
might
be,
you
know
xml,
then
you
could
argue
that
the
formatter
in
common
between
evidence
and
their
resource
access
protocol
would
help
on
the
attester
side.
A
A
C
A
Right
exactly
and
it
becomes
a
maintenance
concern
going
forward.
F
A
So
what
was
the
context
of
this?
We
were
just
discussing
the
the
high
net.
We
were
just
discussing
this
pull
request,
which
is
now
closed.
So
I
don't
remember
what
it
is
relating
to
the
to
the
to
the
parser
to
the
code.
C
It
was
just
a
clarity
issue
where
the
issue
filed
was
they
thought
that
in
red
line
706
the
same
serialization
format
was
confusing
and
also
the
the
same
as
watch
right,
but
that
wasn't
the
main
issue.
The
main
issue
was
just
it
looked
like
the
the
filer
thought
that
there
was
a
couple
sentences
that
were
redundant,
and
so
it
could
be
shortened.
C
C
C
F
So
the
word
is
isolation,
isolated
or
isolation
in
tcg.
Contexts
sometimes
has
special
meaning,
because
it's
part
of
a
a
fundamental
definition.
They
have
around
how
the
how
how
what
they
describe
as
a
root
of
trust
is
protected
from
the
rest
of
the
world
and
therefore
becomes
a
root
of
trust
when
they
use
the
term.
C
You'll
see
that
I
added
a
phrase
that
kind
of
defines
it,
and
so
when
we
get
to
the
pull
request,
please
review
that
to
see
if
you
think
that
tcg
has
a
different
meaning,
which
I
didn't
know,
I'm
guessing
it's
compatible,
but
I
may
have
gotten
it
wrong.
So
I'm
trying
to
find
we've
been
getting
the
actual
pull
request.
So.
C
1162
was
the
first
use
of
the
term,
isolate
isolating
isolation
or
whatever,
and
so
I
kind
of
moved
out
and
there's
a
bunch
of
other
uses
of
isolation
after
that.
But
the
first
two
were
in
red
785
and
red
1162,
and
so
I
removed
the
words
from
there
and
kind
of
defined
it
in
context
in
11,
in
green
1162..
F
C
Know
I
didn't
follow
that
what
1161-1163
says
right
now
is
making
sure
that
the
target
environment
can't
tamper
within
testing
environment.
That's
what
I
understand
is
meant
by
isolation.
If
there's
other
meetings,
then
we
can
add
those.
But
here
to
me
isolation
means
target.
A
F
C
The
going
up
off
the
top
of
the
screen,
michael,
if
you
scroll
up
just
a
little
bit
the
context
of
782,
is
maybe
talking
about
that
because
at
least
my
reading
of
782,
maybe
I
got
this
wrong,
because
this
was
my
text.
But
this
is
a
case
where
your
system,
bus
entity
includes
a
verifier
for
other
things
on
the
system,
bus
and
it
may
implement
a
tester
role,
and
I
couldn't
tell
if
that
meant.
F
F
C
So
scroll
down
a
couple
lines,
michael
because
what
you
call
the
domain
isolation?
Oh,
I
guess
you
have
to
expand
and
expand
downward
one,
because
I
think
it's
the
next
paragraph.
A
C
I'm
looking
for
the
next
use
of
isolation-
maybe
it's
not
there!
I
guess
I
have
to
look
to
see
where
it
was
because
I
did
not
change
it
saying:
okay!
Well,
that
makes
sense
in
the
context
of
the
sentence
that
I
added,
I
think
so
let
me
find
where
it
was.
C
C
D
C
I
am
not
certain
that
claims
crossed
the
boundary.
Is
the
environment
measures,
the
target
environment?
So
all
the
claims
are
generated
on
the
testing
environment.
B
This
transversion
between
this
isolation,
somehow
and
and
also
there,
has
to
be
some
understanding
what
the
values
in
the
end
look
like
of
the
claims,
because
otherwise
it's
very
hard
for
the
tester,
for
example,
sorry
testing
environment
to
sign
them.
That
is
not
true
on
the
secure
channel
side,
although
the
testing
environment
probably
would
create
the
sticker
channel.
So
at
the
end
it
has
to
have
get
the
values
in
any
kind
in
any
way,
somehow,
either
by
measurement
or
for
relayed
to
to
put
them
into
this
youtube
channel.
B
C
You'll
see
that
that
the
phrase
that
we're
looking
at
is
unchanged
between
red
and
blue
right,
worthless.
A
F
I'm
trying
to
draw
attention-
and
maybe
it's
pedantic
but
attention
to
the
two
sides
of
the
testing
environment
and
I
know
we
set
a
tester,
but
I'm
just
you
know
digging
a
little
deeper
into
that
and
so
for.
A
A
C
A
F
C
A
C
Just
want
to
pamper
with
the
claims
and
say
that
you're
healthy
when
you're,
not
you,
don't
get
the
key
material,
but
you
can
lie
and
say
that
your
malware
isn't
present.
A
C
I
can
live
with
it,
I
prefer
it
without
also,
but
I
can
live
with
it.
I
think
it
makes
it
harder
to
read,
but
if
you
guys
all
like
it
better,
I
can
live
with
it.
I'm.
B
C
C
A
C
A
As
suggestion
to
text,
I
don't
know
what
happened
here,
I'm
going
to
delete
this.
C
E
A
C
B
I
think
these
are
extended
changes,
so
we
we
reworked
the
story
of
the
diagram
to
be.
Does
that's
where
we
started
and
then
thomas
came
from
an
implementation
point
of
view
and
was
like
yeah.
The
current
example
is
more
create
more
confusion
than
it
creates
clarity.
What
I
would
like
to
know
as
an
implementer
and
then
we
created
basically
addressed
all
these
items.
This
is
new
text.
To
be
honest,.
C
C
A
D
C
I've
only
got
two
comments
and
then
I
had
that's
where
we
ran
out
of
times
and
meaning
the
meeting
was
starting.
So
this
one
grammar,
I
think,
they're
both
correct,
but
the
one
that's
easier
to
read,
is
without
the
ed.
But
the
main
point
here
is,
it
says,
broadcast
to
all
protocol
participants
by
handle
distributor
and
you
can
use
a
handle
distributor.
That's
only
broadcasting
to
two
of
them
right.
Whoever
sends
the
nonsense.
C
Sorry,
whoever
includes
the
notes
in
a
message
and
who
receives
the
nonce
and
checks
that
it
matches
the
one
from
the
handle
distributor.
You
don't
need
to
go
to
all
three
or
four,
so
all
I
don't
think
I
would
say
regularly
broadcast
to
maybe
multiple.
Instead
of
to
all.
A
B
B
For
exactly
that,
all
participants
could
be
not
all.
I
think
that
other.
D
B
To
make
it
simple
here,
yeah
yeah,
I
think
you're.
A
I
I
I
don't
like
multiple,
because
I
I
don't
think
that
clarifies
it
at
all.
G
C
Distributor
but
I
think
to
michael's
point
whoever
said
protocol
participant
is
an
undefined
term
which
protocol
we're
talking
about.
A
Yeah,
hang
on
yeah.
Do
I
want
to.
G
Yes,
I
wanted,
I
would
suggest,
committing
that
one
and
then
opening
a
new
one.
Thomas.
Can
you
hear
us
here?
Oh.
B
So,
are
you
fine
with
removing
the
protocol
texturing
as
well?
Oh.
G
C
A
G
C
Okay,
so
here
we
get
to
the
crux
of
my
comment-
and
this
is
the
last
time
that
I
reviewed
miners,
so
I'm
going
back
to
when
hank
walked
me
through
this,
and
I
thought
that
I
completely
understood
it
here
is
an
epis.
So
an
epic
is
from
the
time
that
you
receive
a
handle
until
the
time
that
you
receive
its
replacement
right.
That's
the
epic!
Yes,
the
overlap.
A
C
C
The
receiver
has
no
clue
what
time
it
was
transmitted
right
and
so
remember,
anytime,
you
use
a
subtraction,
it
has
to
either
be
using
synchronized
clocks
or
it
has
to
be
using
the
same
clock,
and
here
the
overlap.
Time
is
hr.
Prime
minus
time,
hr
is
using
the
receiver's
clock,
and
so
I
believe
that
is
correct
and
the
transmission
time
is
irrelevant
for
this.
B
C
No,
you
are
correct
hank.
My
thing
is
wrong.
Sorry,
it
was
early
when
I
wrote
this,
so
you
are
right
and
so
my
correction
is
not
correct
either.
But
my
first
point
is
the
time
h
t
is
irrelevant
and
we
don't
need
to
add
that
to
the
table
because
it's
not
used
in
any
in
the
in
in
the
overlap,
computation.
B
B
C
Another
one,
so
I
think
hank's
point
is
well
so
first
of
all,
hank's
answer
is
more
general
than
my
suggestion,
which
is
a
special
case.
Mine
could
be
correct
in
an
implementation
where
you
define
the
overlap
as
being
all
the
way
back
until
the
beginning
of
you
know,
for
for
the
whole,
for
a
whole
distribution
period,
but
you
don't
have
to
your
overlapped.
Current
period
could
be
much
less
than
that.
Actually
is
it?
C
B
So
you
that
there
is
a
definitely
a
time
where
you
don't
really
know
that
the
old
handle
is
not
a
stale
now,
because
there's
jitter
on
the
network-
and
you
haven't-
received
a
new
one
yet
and
received
the
oh,
and
so
so
so
this
is
this.
Is
this.
C
A
C
A
No,
but
you
haven't,
no,
you
have
to
speak
the
old
time.
The
point
is
that
you
keep
the
old,
the
old,
the
old
you.
You
have
a
reference
to
your
old
clock,
so
the
time
at
which
you
received
the
handle
the
hra
right.
That's
a
that's
a
that's!
That's
in
the
attester!
That's
a
reference
to
my
own
clock
right!
That's
the
time
I
received
the
handle.
A
So
whenever
so,
as
long
as
the
evidence
is
newer
than
hra,
I
keep
it,
and
even
when
I
get
hra
prime,
I
I
I
keep
this
open
for
an
overlap
period
is
what
I
think.
C
You
should
be
pointing
to
the
verify
or
they're
lying
part
of
it,
because
the
equation
is
what
happens
on
the
receipt
of
the
of
the.
A
C
Yeah,
so
what
happens?
If
you
look
at
the
one
that's
highlighted
now
right,
you
might
be
getting
since.
Let's
see
this
one
is
background,
check
or
passport.
I
think
this
one
is
showing
I'm
sorry.
No,
it's
neither
right
because
it's
the
model,
so
here
you
might
be
getting.
I
mean
I
was
gonna,
say
you're
getting
evidence
or
attestation
results
here,
you're
getting
you
can
see
what
line
goes
off
to
the
right
here
because,
like
in
old
1464,
it's
evidence
going
to
the
right.
There
are
its
attestation
results
going
to
the
right,
but.
C
C
What
does
that
mean?
Is
this
evidence
attestation
result
or
neither.
B
B
This
way-
and
that
is
that
is
another
change
to
the
text.
We
cannot
we've
eliminated
two
line,
edges
and
and
and
created
abbreviations
that
are
shorter.
Rr
means
relate
a
relayed
result,
which
is
a
short
for
attestation
result,
and
we
could
also.
A
B
We're
not
introducing
any
new
term
here.
We
are
introducing
lines
that
are
not
doubled
and
confusing
to
a
lot
of.
D
C
B
C
B
Mean,
but
it
is
different
here
and
let
me
highlight
that
the
handle
here
used
for
relaying
the
rear
attestation
result
is
different
from
the
handle
inside
the
attestation
result.
That
is
the
point
of
the
story.
So
not
highlighting
it
in
the
diagram
would
be
a
a
weird
omission.
So
that's
what
I
would
propose
is.
B
C
B
B
That
is
signed
by
some
entity.
The
testing
result.
I
don't
I,
I
don't
know
that.
There's
not
enough
relevance
here.
The
relevance
here
is
that
you
cannot
from
thin
air
convey
with
magical
means
a
attestation
result
from
the
attester
to
the
relying
party.
There
is
a
message
here
and
it
is
not
the
plain
attestation
result.
It
is
something
around
that
is
so
relevant
to
the
relay.
Otherwise,
implementation
do
not
work,
but
please.
B
A
B
B
Yeah
that
I
I
get
that
point
and
we
can
reverse
it
and
can
make
the
diagram
unreadable
by
introducing
more
lines
again.
So
that
is
not
my
point.
My
point
is
that
I
wanted
to
highlight
that
to
address
dave's
concern
about
the
invention
of
new
things
that
are
complicated
but
are
relevant
to
the
example
eventually
and
and
dave
is
confused
because,
of
course
he
has
not
seen
this
before,
and
it
cannot.
A
B
C
B
F
B
B
So
that
is
an
attempt,
a
proposal,
but
unfortunately
the
people
who
actually
could
not
understand
the
diagram
at
all
were
utterly
confused
by
the
two
lines
and
what
goes
where
and
no
the
exclamation
about
what
the
curly
bracket
means
at
all
in
any
place
in
the
whole
text,
and
therefore
we
simplified
that
we
thought,
but
unfortunately,
thomas
at
the
moment
only
gets
30
percent
of
the
audio
and
is
unable
to
highlight
what
his
concern
was
and
why.
This
is
a
solution
to
it.
B
B
C
C
I'm
listening
to
hank
here's
my
proposal,
then,
is
we
put
this
one?
We
leave
this
until
next
week
and
we
go
through
some
of
the
issues.
A
B
F
F
A
I
I
I
I
have
two
two
take
homes
from
this
and
please
correct
me
if
I'm
wrong,
hank
and
dave.
The
first
is
that
there's
this
there's
introducing
new
terms
that
may
or
may
not
be
useful.
The
second
is
the
presentation
in
the
diagram
of
where
the
terms
are
compressed
into
what
appear
to
be
a
new
term.
B
A
C
By
this
cloud,
yeah,
the
other
thing
is
a
fine
point.
We
can
add
a
verb
at
the
top
of
the
section
before
any
of
the
pictures
or
something
so.
B
C
Michael,
the
only
one
that
you
didn't
mention
is
the
one
that
I
put
into
the
comment,
which
is,
I
don't
think,
there's
any
place
that
you
need
to
use
both
handle
transmitted
and
handle
received,
and
so
I.
C
A
A
semantic,
not
a
syntax
complaint
that
is
true.
B
A
B
C
But
I
said
I
I've
not
read
through
the
rest
of
the
text
after
my
comment
before
the
diagram,
and
so
I
will
do
that
between
now
and
next
meeting
so
that
then
I
can
have
intelligent
comments
in
the
diagram
because
the
diagram
comes
after
the
text
I
haven't
read.
So
I
have
to
read
that
first.
B
C
So
you
could
pick
any
one
of
these
other
ones
that
people
have
already
commented
on.
I
don't
know
if
somebody
finds
a
a
short
one,
because
I
I
tried
to
call
call
on
some
that
had
no
comments,
because
I
figured
the
ones
with
power.
Oh
there
was
one
that
hank
you
had
originally
signed
to
me
and
I
signed
it
to
michael.
Maybe
we
look
at
that
one.
C
A
That
looks
like
it.
That's.
C
C
F
F
C
So
now,
without
the
context,
that's
where
that
paragraph
came
from
was
an
attempt
to
resolve
issue
37.
C
C
C
Because
the
first
question
was,
should
this
add
because
you
can
see
in
the
old
text
the
way
that
they
could
fill
the
verifier
talked
about?
You
know
two
or
three
different
ways
in
which
you
could
fill
out
the
verifier
right,
whereas
the
second
way
did
not
talk
about
two
or
three
different
ways.
It
could
fail
at
the
relying
party.
It
just
said
when
it
fills
the
the
appraisal
policy
for
station
results,
which
could
fail
for
a
number
of
reasons
right.
So.
F
F
So,
are
we
critiquing
a
and
b
and
trying
to
pick
between
the
two
or
are
we.
C
C
There's
the
original
text,
which
is
the
one
that's
quoted
at
the
top
of
the
issue,
there's
a
which
does
something
along
the
lines
of
what
the
filer
was
suggesting,
with
adding
a
reference
to
freshness
and
so
on
as
being
another
reason
that
it
could
fail
at
the
verifier
or
there's
a
b,
which
would
be
my
preferred
way
to
address
it,
which
does
not
add
a
forward
reference
to
freshness.
In
fact,
it
removes
some
detail.
That's
present
elsewhere
in
the
document.
F
And
I'm
trying
to
get
context
for
what
section
this
is
in
good
question.
C
Passport
model
yeah.
Well,
it
is,
although
the
the
particular
paragraph,
the
password
model,
is
the
one
that
appears
first
and
so
in
this
part,
it
has
this
extra
paragraph
that
was
inserted
in
making
like
the
first
instance
of
some
of
a
point.
The
same
paragraph
would
actually
apply
in
other
sections
too,
but
it
would
be
redundant,
and
so,
like
you
know,
when
you
do
the
first
example,
you
do
more
details
in
the
second
example
that
you
kind
of
gloss
over
the
things
that
are
elaborated
in
the
first
one.
C
F
Maybe
that's
fine.
I
kind
of
have
dissonance
with
with
use
of
the
word
way.
Although
the
first
sentence
is
clear,
we
re
we
use
it
to
to
describe
the
three
cases
and
I
think
you
could
just
say
something:
it'd
be
cleaner
to.
If
you
said
there
are
three
failure,
cases
or
three
failure
modes
or
something
like
that,
then
you
can
use
the
failure.
The
term
failure
mode
or
something
makes,
makes
things
readable.
C
C
C
C
A
Sorry,
the
line
breaks
were
so
that
it
would
be
fainter,
but.
B
A
Dave
has
rewritten
this
to
produce
a
what
and
then
again
to
produce
b.
I
think
okay
both
are
from
dave.
No.
This
originally
is
from
thomas
yeah.
B
B
B
So
again,
as
this
was
initially
brought
up
by
thomas,
my,
my
preference
is
a
little
bit
to
the
a,
but
that
does
not
really
yield
any
consensus.
So
maybe
I
should
stay
neutral.
Also
thomas
has
to
weigh
in
here.
I
suppose.
F
B
Came
yeah
that
would
be.
That
would
be
a
interesting
part
that
I
see
at
least.
B
Oh
dang,
it
sorry
yeah.
I
was
a
little
bit
confused
by
silence.
C
A
So
so
I
fixed
the
line
breaks
in
the
original
text
and
I
gotta
remove
the
trailing
space
here.
I
hate
those
trailing
spaces,
guys.
C
The
the
key
change
in
meaning
is
that
the
green
646
to
647
now
encompasses
the
notion
of
freshness
by
reference
to
other
parts
of
the
document.
Right
that
something
isn't
fresh,
it
will
fill
the
appraisal
policy
for
evidence
where
part
of
the
policy
might
be.
It
has
to
be
fresh
within
the
following
period
of
time
right
and
we
don't
say
that
here
it
said
elsewhere,
but
previously
some
error
in
processing
or
some
missing
input.
Well,
neither
of
those
include
freshness,
and
so
that
was
what
the
issue
was.
A
B
A
Okay,
well,
let's
take
that
to
thomas
it's
the
top
of
the
hour.
So
now
you
have
just
the.
C
E
E
E
B
And
I
think
the
the
the
the
issue
is
three
two
three
two
right.