►
From YouTube: RATS Architecture Design Team, 2021-11-19
Description
RATS Architecture Design Team, 2021-11-19
B
Can
you
say
something
now?
Okay,
I
can
say
something
now
all
right.
I
disconnected
this
other
speakers
now
and
it
switched
to
the
right
one.
So,
let's
see,
if
it
switches
yeah,
I
switched
back
had
done.
I
plugged
them
back
in
all
right.
I'm
going
to
drop
and
rejoin
okay.
Okay,
be
right!
Back
hi,
hans.
D
A
B
D
You
have
like
a
mixer
app
that
tells
everything
where
to
go.
D
One
of
the
coolest
things
I
learned
how
to
do
was
the
mixer
apps
and
often
have
a
monitor,
which
is
a
an
input
which
is
the
result
of
the
output
of
something.
And
so
you
can.
You
can
make
loops
and
wrap
one
audio
of
one
program
into
the
output
of
one
program
into
the
audio
of
an
another
one.
D
D
D
And
ned
was
a
yes,
it
was
11
p.m.
Yeah
you
might
be
right,
yeah,
seeing
as
they
don't
do
time
zone
chances,
time
zone
changes
which
I
like
apparently
we're
just
waiting
for
new
york
to
decide.
D
D
Okay,
so
first
things
thanks
for
coming
today,
I
I
picked
today
because
we'd
often
used
it,
and
I
wanted
to
get
things
going
again
once
we
saw
what's
his
name,
rowan
roman's
review,
and
I
wasn't
sure
if
I
could
do
next
week,
because
there's
a
ripe
meeting
all
the
all
week.
D
Okay,
so
I
I
set,
I
didn't
think
I
could
do
next
week
on
tuesday,
which
is
why
I
picked
today,
but
then
I
realized
that
actually
it
would
I
I
could
step
away
from
the
right
meeting
and
it
would
be
fine,
so
I
did
put
one
in
on
tuesday,
but
to
be
sure
I
put
one
in,
I
sent
an
invitation
for
tuesday
november
30th
as
well.
D
That
was
our
traditional
time.
Hank
convinced
me
that
that's
probably
a
good
time.
So
if
you
can't
make
it
next
tuesday,
then
oh
well,
but
for
sure
by
the
30th,
I
hope
everyone's
available
for
that
I
can
do
either
or.
B
D
So
I'm
hoping
we're
going
to
do
both,
I
think
we'll
need
both.
I
think
we'll
need
about
four
or
five
meetings
and
I'm
I'm
hoping
that
the
other
sector
reviews
will
begin
to
come
in
as
well.
I
don't
know
if
you
push
the
button
ned
or
nancy.
Did
it
or
not,.
E
And
the
assumption
that
everybody
is
either
off
on
you
are
unripe.
I
I
gave
away
that.
D
Yeah
yeah
yeah,
I
know
yeah,
I'm
sorry.
I
confused
you
that
way
but
anyway.
So
if
we,
if
we're,
if
we're
below
quorum
on
tuesday
we're
below
core
and
between
you
and
me
and
and
us
holidays,
I'm
wondering
ned
did
you
did.
Did
somebody
request
early
sector
reviews.
G
Yeah
nancy
did
yeah
right.
Roman
was
didn't
think
that
would
speed
things
up
at
all.
H
D
G
D
Okay,
all
right,
so
I've
split
up
ned's
comments
into.
I
don't
know
issues
at
least
there's
at
least
12
of
them,
or
I
don't
know
I
didn't
count.
Them
can't
count
that
high
and
I'll
leave
that
there
and
click
on
that
and
hank
started
working
on
some
of
the
responses.
So-
and
I
think
you
marked
some
of
them
as
being
answered.
D
D
Yeah,
so,
okay,
so
I'm
just
going
to
go
through
the
things
once
you
have
commits
to.
You
might
have
to
guide
me
to,
but
I
reviewed
this
one
and
I
saw
what
you
put
up
on
the
screen
of
the
pull
request.
That.
B
Looked
right
to
me,
there.
B
I
B
I
didn't
make
it
through
all
of
roman's
comments.
Reading
them.
I
started
from
the
top
and
I
noticed
they
got
successfully
harder
to
understand
without
looking
at
the
document.
So
oh,
like
okay
in
his
review,
the
first
ones
he
would
like
copy
in
the
text
that
he
was
commenting
on
and
so
on
and
other
ones
like
say.
Okay,
I
think
see
what
he's
getting
at,
but
I
have
to
go
look
at
the
text,
so
I
I
didn't
make
it
to
the
bottom
of
all
of
this
stuff
figure.
We
just
go
through
them
today.
D
B
B
H
There's
that
one,
that's
that
tab
and.
D
E
E
Yeah
I
rent
ifaa
and
epay,
I
think,
or
google
pay
and
no.
B
E
E
D
A
B
D
B
Talking
romano
was
talking
about
the
commercial
ones.
I
wonder
if
what
chinese
ifaa
standard
is.
If
that
really
is
a
standard,
it
should
have
a
stable
or
referenceable
document
from
some
place
right,
and
so
we
could
leave
chinese
ifaa
standard
and
remove.
We've
chat,
pay
and
google
pay
as
being
the
more
aging
ones,
but
I
I
don't
know
what
chinese
ifa
standard
is
to
know
that.
E
E
D
Okay,
so
I'm
going
to
assign
it
to
way,
let's.
I
B
I'm
fine
with
hank's
proposal,
but
if
we
can
find
a
reference,
then
I'm
fine,
leaving
it
open
until
next
tuesday
or
whatever,
as
we
come
back
so.
B
I
need
more
context.
I
need
to
look
at
the
document
to
know
what
goes
right
before
this
paragraph.
B
Okay,
good
there's
no
limit
to
it;
no,
it's
one
about
that.
The
the
one
that
ends
layers
don't
limit
to
okay
in
a
testing
environment;
okay,
yes,
okay!
Now
now
that
I
see
the
context
yeah,
I
agree.
I
just
want
to
make
sure
it
wasn't
gonna
be
incorrect
in
context.
So
yes,
I
I
agree
with
this
with
his
issue.
E
B
You
start
having
a
bunch
of
layers
right,
the
top
layer
could
be
say
the
operating
system
measuring
the
application
running
on
top
of
that,
and
so
the
operating
system
would
be
a
software
piece
of
software.
E
B
What
I
was
looking
for
hank
is
this
is
not
talking
about
the
root
of
trust
right.
This
is
talking
about
every
pair,
and
so
once
you
start
having
layers,
then
the
top
layers
can
be
software
measuring
software
right
as
long
as
the
root
at
the
bottom
is
hardware
right,
but
the
terms
here
are
used
for
every
set
of
layers
going
up.
So
so
I
think
what.
D
B
D
B
It
is
used
in
te's,
too.
There
are
tes
that
can
run
kernels
inside
and
in
some
sense
you
could
argue
that
option
trusts
on
opti
is
the
kernel
and
it
can
measure
trusted
apps
running
inside
options,
so
yeah,
even
in
te
environments.
This
is
true,
but
of
course,
but
yes.
E
B
G
G
B
E
E
B
B
B
E
E
B
B
E
B
I
think
there's
a
term
in
the
windows
world,
I'm
just
looking
up
a
term
to
see
if
it's
I'm
looking
at
the
term
windows
secure
kernel-
and
I
assume
that
there's
actually
some
term
in
the
linux
world.
But
I'd
have
to
do
some
research
to
figure
out
what
it
is.
Yeah
they're.
B
D
B
B
I
don't
think
linux
currently
has
something
in
the
windows
secure,
kernel
category,
but
the
things
like
you
know,
opt
and
so
on
and
arm
processors
actually
run
in
that
same
kind
of
context.
I
just
don't
know
linux.
G
B
G
D
There
was
somebody
who
had
done
something
with
pushing
most
of
the
kernel
into
ring
one.
I
don't
know
if
that
was
linux
or
something
like
aix
or
something
like
that
anyway,.
G
B
In
there,
so
it
doesn't
add
a
new
point
and
right
now
none
of
those
bring
in
anything
that
is
a
vendor
specific
term
and
so
that'd,
be
the
only
that'd
be
the
reason
to
not
say
windows,
secure
kernel,
because
it's
a
windows,
a
vendor
specific
term,
that's
in
there.
So
so
I'm
fine
leaving
it,
as
is,
if
we
want
so.
H
D
All
right,
so,
let's
move
on-
let's,
let's
not,
let's
not
close
this,
but
let's
move
it
on
okay
and
maybe
in
the
response
we'll
find
some
okay.
When
we
reply
we'll
say
well,
someone
will
say
well
what
about
something.
D
So
we
had
these.
I
collected
the
editorial
comments
together.
E
Yeah,
so
yeah
roman
was
surprised
by
the
sudden
increase
in
detail
and
yeah.
That
is
due
to
the
fact
that
we
did
not
want
to
talk
about
rules
of
trust
so
early
in
the
text
and
don't
want
another
glossary
that
wants
to
have
a
separate
section.
We
all
had
the
discussion
so
now
it
is
becoming
more
detailed
when
it
comes
up
so
yeah.
That's
how
it
is.
So
I'm
not
touching
that
again
unless
someone
forces
me
to
and.
B
E
J
B
To
him
what
the
rationale
is,
and
so
I
think
on
the
section
four
one:
would
they
should
they
be
I'm
fine
with
not
doing
a
change
there
you're
hesitant
to
touch
this
again,
I'm
hesitant
to
touch
it
again.
D
D
D
So
here's
a
more
neither
here
nor
there
kind
of
question.
B
On
hank's
last
question
on
the
4.3:
I'm
okay,
with
what
we
do,
I
probably
if
you
went
back
in
the
providence,
I'm
probably
the
person
to
put
those
two
things
in
there
and
so
yeah
keep
them.
But-
and
I
think
there
are
other
documents
that
do
that-
that
say:
here's
our
specific
term
or
here's
the
same
term
and
it's
meant
to
match
this
or
it's
meant
to
be
a
subclass
of
this
other
term
defined
in
this
other
document.
So.
D
Okay,
so
interact
with
roman
to
make
sure
that
yeah.
So
I
I
was
just
saying,
like
I
didn't
understand
why
he
didn't
understand,
but
right
right.
D
E
B
We
choose
not
to.
I
think
I
would
say
that
hank's
point
about.
It
is
neither
a
conceptual
message
nor
a
role.
That's
the
main
point,
because
the
produces
and
consumes
are
what,
with
his
respect
to
a
conceptual
message,
it's
an
element
that
appears
in
multiple
conceptual
messages
and
so
yeah
everything
that
uses
a
claim.
You
have
to
look
at
that
other
thing,
the
conceptual
message
to
say
what
consumes
and
produces
that,
and
so
it's
in
there
by
proxy
by
looking
at
the
message
that
embeds
the
claim.
B
B
Even
though
inside
the
pr
that
pulled
the
request,
description
should,
if
it
doesn't
already,
should
probably
reference
that
without
using
the
fixes,
you
know
one
of
those
magic
keywords.
Yes,.
D
But
but
but
the
pr
doesn't
that
this
pr
at
this
point
doesn't
doesn't
use
the
magic
keywords
anyway.
No,
but
it
should
link
those
two
should
be
linked
somewhere,
so
you
can
find
one
from
the
other
yeah
so
right
now,
they're
linked
by
this
okay,
excellent,
okay,
section
three
to
three
constant.
D
Consequently,
in
consequence,
consequently
single
word
fix
and
then
this
one
is
adding
that
it.
It
was
appraisal
policy
for
testation
results,
yup
good
catch
from
roman
yep.
D
B
F
D
But
yeah
I'm
happy
to
take
that
one
fanway
to
that
did
we
get
all
of
your
changes
here.
E
Yeah
yeah,
but
I
did
them
consecutively
and
if
I
have
don't
have
a
comment
in
an
issue
anymore,
then
we
are
done.
D
B
B
D
B
D
B
D
B
D
Ned's
com
commented
here
what
is
meant
by
having
established
really
I've.
I
kind
of
edited
put
these
in
with
the.
B
Okay,
there
we
go.
I
have
a
view
as
to
what
I
think
relevance
means.
Let
me
read
ned's
response
here.
B
Okay,
ned
might
be
thinking
the
same
thing
as
I'm
thinking.
B
Well,
I
just
based
on
you
know:
evidence
is
appraised
by
a
verifier
to
establish
its
relevance,
compliance
and
timeliness.
I
think
all
three
of
those
are
important
concepts
to
me
relevance.
Let's
say
you
took
evidence
from
one
type
of
thing
that
you
weren't
expecting
and
sent
it
to
a
verifier.
Okay
right,
the
verifier
said
well,
this
isn't
actually
for
me
right,
you're,
sending
me
an
intel
thing
and
I
only
understand
how
to
parse.
B
You
know
arm
evidence
or
you
know
what
you're
sending
me
and
a
dice
thing
and
I
always
understand,
eats
or
something
like
that.
It's
not
relevant
because
I
can't
even
tell
whether
it's
compliant
or
not,
because
you're
sending
it
to
the
room,
verifier
kind
of
thing.
B
B
D
B
D
B
I
think
it's
an
important
word
to
retain
and
I
think
from
ned
your
response.
I
think
you
think
it's
important,
although
I
think
you
and
I
have
overlapping
meanings
of
relevance-
you're
overlapping
example
cases
so.
G
Yeah
it
could
be,
I
mean
I,
I
don't
think
that
this
was
the
word.
Relevance
was
trying
to
be
extremely
precise
right.
It's
just
that.
There's
a
lot
of
cases
where
the
information
is
not
providing.
You
know
any
meaningful
context
for
making
a
decision
about
whatever
what
a
relying
party
is
going
to
decide
on.
D
So
is
there
an
aspect
here
where
you
might
send,
for
reasons
of
you
might
send
evidence
that
you
don't
know
is
irrelevant,
but
you
just
always
send
it,
because
you,
you
can't
know
when
what
you're
supposed
to
say
when.
G
It's
not
one
of
the
practical
cases
that
we've
run
into
is
when
you're
riding
the
a
tester
the
testing
environment
you,
except
for
for
you,
know,
efficiency
reasons,
it's
easier
to
to
report
more
information
than
what
you
actually
need
to
make
the
trust
decision.
And
then
you
rely
on
a
mask
in
the
in
the
reference
values
to
tell
the
verifier
to
ignore
some
stuff.
B
I
I
agree
with
that
point
of
view.
I
think
that's
a
very
common
case
that
you're
going
to
optimize
the
collection
of
evidence
by
reporting
more
than
you
might
actually
need
for
a
particular
appraisal.
B
So
going
back
to
roman's
questions
right,
what
is
meant
by
having
to
establish
that
evidence
is
relevant.
We've
been
talking
about.
That
is
the
concern
that
the
attester
would
provide
evidence.
It's
not
relevant,
it's
not
a
concern,
but
the
answer
is
yes,
it's
just
a
fact
right
and
then
is
that
is
that
attack
or
attempt
a
subterfuge
answer
could
be,
but
not
necessarily
right,
because
I
think
ned
and
peter's
points
is
where
that
would
be
common,
not
not
malicious,
in
any
sense,.
G
The
the
the
point
is
that
there's
something
on
the
on
the
endorsement
side
to
help
the
verifier
do
the
right
thing.
D
D
Hearing
that
we,
like
the
word
relevance
here,
that
it
it
it
doesn't
sound
like
we
have
a
lot
of
easy
change
to
to
clarify
within
this
text.
B
Because
right
now,
I'm
not
sure
what
to
say
if
we
would
say
something
right,
but
if
we
say
here's
kind
of
what
we
mean,
but
there's
nothing,
there's
no
obvious
change
that
we
could
come
up
with
to
make
to
the
document.
Let
us
know
if
that's
sufficient
and
he
comes
back
and
says.
Well
maybe
you
should
say
this
and
then
we
might
actually
have
a
suggestion.
So.
D
B
Sharing
right
here-
and
you
can
go
along
with
that
saying:
okay
in
the
security
considerations
or
privacy
considerations,
we
do
cover
the
issue
of
over
sharing
and
down
in
that
section
right
over
sharing
in
a
privacy
sense.
D
Okay,
all
right
great,
let's
move
on
then.
J
C
Okay,
let's
read
hank's
proposal
here.
B
D
Okay,
so
hank
do
a
pull
request.
I
guess.
B
Ned's
wording
that
hank
incorporated
there
so.
D
D
D
B
Endorsements,
okay,
all
right
gotcha.
Now
we
can
go
by
it's
endorsements
and
reference
values.
All
right,
cool
and
8.3
is
reference
values.
Yes,
I'm
just
trying
to
read
roman's
comment
again.
G
We
define
it
in
multiple
places
if
you,
but
if
you
skipped
over
it
and
didn't
internalize
it,
then
you
read
it
later
and
you're
kind
of
like
what
was
this
thing
again.
Yeah.
B
G
B
G
B
And
so
that
part
was
intentional:
it's
not
a
restatement!
It's
an
elaboration
on
right.
D
B
I
don't
have
a
better
term
for
it.
I
think
that's
the
best
term
that
we
still
have
maybe.
H
E
We
also
very
used
to
this,
so
we
have
experience
now.
B
B
Here
without
reading
ned's
or
hank's
comments,
yet
my
first
reaction
to
this
one
is
all
we
got
to
do
is
add
a
forward
reference
to
the
section
that
covers
the
oversharing
of
new
privacy,
stuff,
okay,
because
he
says
it
also
immediately
suggests.
Well,
that's
fine
as
long
as
there's
a
forward
reference,
so
that
that
would
be
my
simple
proposal
here.
But
now
I
can
read
hank
and
ned
your
comments
so.
B
Okay,
I
think
ned's
comment.
If
I
understand
your
main
point
is
you
can
abstract
out
and
not
mention
specifics
in
this
section
like
passports
and
driver's
licenses,
sorry,
like
name
and
date
of
birth
or
whatever,
that
could
only
be
talked
about
in
the
privacy
section
and
use
some
more
generic
wording
here,
and
I
think
that's
what
you're
proposing
that.
I
think
that
would
be.
I
B
I
would
still
keep
a
keep
a
maybe
a
forward
reference
here,
depending
on
the
wording.
B
I
thought
that,
and
I
read
that
ned
you
were
you
proposing,
that's
a
replacement
for
the
text.
That
gromit
is.
G
Trying
to
remember,
I
think
that
was
being
proposed
as
a
replacement.
Okay,.
B
D
J
D
Yeah,
that's
what
I
suspected.
We
already
said
that
then
in
the
next.
G
We
don't
need
to.
We
can
just
remove
the
controversial
text
if
we're
able
to
you
know
convey
the
idea
that
the
results
can
contain
privacy,
sensitive
content,
but
that
there's
ways
to
deal
with
it
yeah
and
one
we
may-
and
it
may
be,
that
the
tester
is
relying
on
the
verifier
to
protect
some
privacy
through
obfuscation,.
D
B
So
what
I'm
thinking
is
to
replace
that
part
of
the
sentence
with
something
that
makes
ned's
points
about
it
can
contain
privacy,
sensitive
content
as
discussed
further
in
section
whatever
the
one
is.
It's
our
privacy
section.
B
Well,
something
like
that
well,
but
keeping
ned's
working
about
can
contain
something
like
can't
contain
privacy,
sensitive
content
as
discussed,
and
that's
what
goes
in
the
dot
dot
part
yeah.
D
All
right
I'll
sign
myself
to
put
that
in
okay,
all
right
next.
B
All
right,
so
it
sounds
like
ned
and
hank.
You
both
believe
no
change
is
fine
and
I'm
fine
going
along
with
that.
B
If
anything,
and
I'm
not
sure
we
need
to
make
any
change,
but
if
anything
right
we
could
add
a
sentence
or
something
at
the
top
of
the
section
saying
what
the
purpose
of
this
is
right.
This
is
intended
for
those
creating
you
know,
protocols,
whatever
protocols
are
solutions
to
understand
the
answers
available,
or
at
least
some
of
the
directions
available
to
them.
So,
but
I
don't
know
if
we
need
to
do
that.
B
I
just
don't
know
if
that
helps
explain
why
it
stands
out
rather
than
just
letting
it
stand
out.
So
so
yeah
this
section,
the
center
of
those
creating.
B
Don't
remember,
my
point
is
yeah
for
those
creating
protocols
or
solutions
to
it
was
something
along
the
lines
of
making
sure
they
understand
the
options
of
their
the
the
options
available
to
them
to
ensure
freshness.
But
it's
probably
a
better
way
to
phrase
that.
D
So
yeah,
so
I
would
say
exactly
that
and
and
I'll
add
that
sentence,
but
I
would
say
also
in
a
response.
I
would
say
that
we
thought
that
the
freshness
was
the
different
freshness.
D
Decisions
were
in
fact
architectural
dis
choices
in
the
in
the
solution,
and
that
that
the
that's
why
we
went
to
that
level
of
detail
so
that
nobody
in
essence
the
difference,
freshness
solutions,
cause
non-interoperable
solutions
right,
because
if
you
didn't
implement
the
same
freshness
guarantees,
then
you're
not
interoperable,
and
so
it's
an
architectural
decision
as
to
what
you
wanna
do
and
we
want
to
make
sure
that
people
are
consciously
making
this
decision.
Not
just
you
know
stumbling
into
it
right.
B
I
I
think
you
should
take
this
action
this
this
one
and
write
what
you
just
said,
and
I
mean
if
you,
if
you
think
that
that
is
good
to
put
into
the
document
at
the
top
of
section
10.
Then.
B
I
D
With
that
in
mind,
right
and
and
and
I
think
that
that
other
other
thing
that
I'll
in
the
email
is
that
I
think
that
we
needed
that
level
of
detail,
and
I
think
the
reader
needs
that
level
of
detail
in
order
to
be
convinced
that
we've
covered
all
the
options
right
and-
and
I
don't
know
we
might
have-
wound
up
with
like
12
different
freshness
options
or
started
that
way.
But
we
won't,
we
we
reduce
them.
I
think
it's
to
three
right,
so
I.
I
B
Common
approaches,
we
don't
say
that
it
is
complete
right,
there's,
no
there's
only
three,
but
there's
that's
why
I
don't
know
if
options
is
the
right
word,
but
just
those
just
look
at
the
top
of
section
10
right
now
we
use
the
term
common
approaches
and
not
options.
So
sure
sure
we
don't
actually
say
what
you
were
saying
and
we
should
I,
I
think
so.
Yeah.
D
B
E
B
Have
to
have
that
discussion
in
the
document
I
think
just
having
a
generic
summary
like
michael
is
saying
is
fine.
The
freshness
section
does
cover
evidence
and
attestation
results
and
claims
right
now.
All
of
that
discussion
is
kind
of
woven
in
there
in
a
way
that
I
think
I'm
hesitant
to
try
to
change,
but
I'm
fine,
adding
a
preamble
like
michael
is
saying.
I
G
D
B
He
added
this
on
a
privacy
consideration
section,
and
so
I
think
he's
observing
that
in
the
privacy
consideration
section,
we
cover
privacy
issues
with
everything
except
for
the
appraisal
policy,
and
so
maybe
you
want
to
add
a
sentence
about
the
privacy
applies
to
affordable
policies
too,
because
that's
not
mentioned
in
that
section.
B
We
cover
evidence,
manifestation
results
and
so
on,
but
appraisal
policies
aren't
mentioned.
B
G
To
do
that,
I
can.
I
can
look,
take
a
look
at
it
if
you
want.
C
I
C
D
You
go
section,
11
conceptual
net
messages.
B
B
So
meaning
not
necessarily
in
replacement
of
that
text
right,
you
could
say
you
could
augment
that
sentence
immediately
following
you
know.
You
know,
for
example,
if
there
isn't
native
object
level,
confidentiality
protection
in
the
conceptual
mess
in
in
the
consensus,
if
I
prefix
this
with,
for
example
right,
you
just
say
in
the
evidence,
because
it's
just
an
example
right
that
could
the
transport
protocol
should
provide
these
protections.
G
B
Sure
something
like
that,
michael
that's
right!
I'm
just
now
noticing
ned
coming
to
three
days
ago
and
I'm
comparing
the
language
there
with
the
language
there
yeah.
B
B
D
But
I'm
told
this
is
this:
is
why
all
the
all
the
plagiarism
charges
in
the
music
industry
happen
right
is
because
people
didn't
know
that
they
had
actually
read
it
when
they
thought
of
something
right,
yeah.
G
B
G
D
D
Yeah,
so
he
has
suggested
text.
Even
so
you
know
this
should
be
really
easy.
C
With
these
different
subsections,
12.1
is
a
tester
and
attestation
key
protection.
Okay,.
B
And
all
of
his
comments
are
on
under
12.1
right
now,
he's
got
a
2.2
up
there,
yeah
that's
integrity
protection!
Okay,
can
you
add,
like
security
considerations
in
the
title
of
this
one.
B
D
I
I
agree
so,
let's
I
will
break
out
the
the
typos
and
I
suspect,
we're
happy
with
this
suggested
text
here.
B
Essential
value.
Add
provided
by
rex,
is
for
the
relying
party
to
be
able
to
trust
the
tester,
even
if
the
user
or
owner
is
not
trusted.
Yes,
a
foundational
security
assumption
of
a
rat's
architecture
is
that
a
relying
party
is
able
to
trust
the
attester,
even
if
the
owner,
user
or
owner
is
not
trusted.
J
B
B
D
D
Oh
yeah,
there's
more
down
here.
Sorry,
okay
and
more.
D
D
Yeah,
I
agree:
let's
not
you
want
to
sign
it:
okay,
okay,
all
right!
Well!
Thank
you
very
much.
Let
me
know
if
you're
gonna
be
there
on
tuesday
or
not,
and
I
what
I
thought
I
heard
was
everyone
but
hank,
yep,
okay,
great
talk
to
you,
then
all
right.