►
From YouTube: RATS Architecture Design Team, 2020-06-16
Description
RATS Architecture Design Team, 2020-06-16
B
B
So
you
should
cover
this
and
then
I
guess
it
would
be
useful
to
me
if
we
wanted
to
look
at
that
since
I
of
bread,
harnesses
and
Kathleen's
emails,
but
to
figure
out
whether
we
have
extra
action
items
and
what
the
next
steps
are
I
think
would
be
useful
to
all
those
up.
Not
as
most
people
should
have
been
able
to
read
them,
because
it's
been
over
a
week.
C
C
B
A
D
B
C
A
C
B
C
B
A
C
The
hygiene
is
good,
but
at
some
point
you
can
overdo
it.
So
if
you're
living
in
a
clean
room,
your
immune
system
basically
is
not
happy
about
that.
So
you
can
overdo
it
and
I.
Don't
think
you
can
overdo
hygiene
with
adhesives
and
maybe
to
the
matter
of
cost,
maybe
or
run
time,
foot
furred
or
something
but
but
yeah.
Basically,
I.
Think
hygiene
is
always
adding
a
benefits
and
it's
never
really
hurtful
as
it
is
boomin.
C
B
C
B
C
C
B
C
E
B
B
Is
there
I
mean
to
me
I
think
the
text,
what
we're
trying
to
say-
and
maybe
other
people
have
different
interpretations
of
what
we're
trying
to
say,
is
trustworthiness
as
a
quality.
Other
Rand
and
a
trust
and
Trust
is
a
statement
about
yourself,
meaning
the
the
the
receiver
of
the
information.
Trustworthiness
is
a
quality
of
the
center
of
the
information
and
Trust
a
decision
is
a
decision
made
by
the
receiver.
The
information
right
so
difference
is
trying
to
draw
yeah.
F
F
B
A
C
Think
it
starts
with
the
decision
to
trust
and
then
trust,
worthiness
and
then
the
difference.
So
if
the
the
things
that
was
about
I
assume
more
at
the
beginning
of
a
sentence,
it's
easier
to
see
what
difference
we're
talking
about.
Otherwise,
you
have
to
restate
the
Nexen.
This
is
starting
with.
This
is
subtle
difference,
so
I
think.
B
B
A
D
Can
I
interject
something
here
this
notion
of
trustworthiness
in
access
stations?
Affirming
trustworthiness
is
not
quite
right
in
my
opinion,
because
the
system
itself
may
not
be
worth
trustworthy,
you
know
what's
happening
is
is
an
assertion
that
the
system
is
something
and
then
evidence
supporting
the
claim
that
that
is
so.
Perhaps
you
have
more
reason
to
trust
it,
but
whether
it's
trustworthy
not
is
something
totally
outside
the
attestation
mechanism.
I.
C
D
Access
tation
gives
an
assertion
that
something
is
is
something
and
evidence
that
what
you
expect
that
assertion
to
imply
is
indeed
true
and
if
you
can
infer
trust
from
that
at
the
appraising
end.
That's
great!
That's
what
we're
after,
but
it's
not
a
statement
of
trustworthiness
on
the
system
itself
that.
B
D
B
C
B
B
F
B
B
This
one
is
very
early
and
this
is
not
a
sentence
about
the
attesting
or
the
target
environment.
This
is
a
statement
about
Thea
tester
right
in
the
attacker.
Could
that
the
whole
notion
of
target
environment
and
a
testing
environment
and
how
many
layers
and
things
there
are
or
is
all
in
a
much
later
section.
So
this
is
just
a
generic
statement
about
the
ax
tester
as
a
whole.
F
B
B
B
B
I
think
there's
one
subtle
difference
between
the
original
and
Kathleen's
one.
So
now
I'm
going
to
read,
I
think
what
you
said
here:
okay,
so
the
subtle
difference
between
what's
in
the
original
and
what's
in
the
proposed
by
Kathleen,
is
that
a
set
of
rules
to
verify
implies
that
the
outcome
was
boolean.
B
B
They
said,
look
at
the
it
may
be
typical
for
a
relying
party
to
make
a
boolean
decision.
A
relying
party
could
make
a
decision,
that's
not
boolean
and
an
example
of
that
is.
Let's
say
you
have
three
pens
that
you
know
three
conceptual
security
groups.
You
know
anonymous,
authenticated
and
root,
or
something
like
that
right
and
so,
based
on
it,
you
could
make
you
could
put
it
into.
You
know
an
anonymous
if
it's
not
trusted
you
can
put
into
your
root
if
it's
trusted
or
if
it's
I
can
make
a
decision.
B
B
Cleared
for
a
secret,
but
not
top-secret
clearance.
You
know
that
kind
of
thing,
and
so
that's
why
it
could
in
theory,
be
not
boolean
and
that's
why
you
know
to
use
to
verify,
makes
it
sound
like
it's
a
boolean
decision
and
a
set
of
rules
that
control.
How
is
a
little
bit
more
flexible,
so
I
kind
of
like
the
original
wording
better
than
to
verify.
But
if
she
doesn't
like
direct,
you
know,
control
or
something
like
that
and
finally,
I'm
not
quite
sure
what
she
doesn't
think
it.
What
she
doesn't
like
about
direct.
D
B
D
So
as
a
place
where
you
can
take
all
the
information
about
attestation
and
make
a
statement
about
something,
I
sometimes
call
expectedness
of
the
system
you
have,
the
attesting
system
has
proven
or
not
proven.
It's
claims
against
some
appraisal
policy
that
then
becomes
input
into
a
larger
trust
decision.
That's
going
to
be
made
that
may
use
other
things
besides
just
attestation
information,
so.
B
B
B
A
B
A
B
A
B
A
B
A
B
B
B
B
A
B
B
The
production
of
information
does
not
make
you
in
a
tester,
then
a
tester
is
the
role
that
you
take
on
as
part
of
the
larger
architecture
when
you're
fitting
into
talking
to
a
relying
party
for
attestation
purposes,
and
so
you
can
generate
stuff.
But
if
it's
not
used
for
attestation
you're,
not
an
ax,
tester
cuz
you're
not
actually
attesting
to
anything
right,
you're,
you're
reporting,
something
which
are
not
attesting
to
it.
Unless
you're,
it's
the
it's
a
statement
about
how
the
information
is
being
used,
it
actually
makes
you
be
called
an
a
tester.
C
F
C
B
You're
Ned
that
it's
actually
a
role
I,
don't
know
how
to
introduce
that
here
without
making
the
definition
be
harder
to
understand.
But
I
do
agree
that
it's
a
role
but
I
think
the
main
questions
that
I
think
where
I
need
to
see
whether
we
agree
with
is:
do
we
think
it's
important
to
define
an
ax
tester
from
an
outsider's
point
of
view
or
as
Kathryn
and
suggestion
as
suggesting?
B
F
B
F
B
B
A
B
A
B
A
B
A
F
B
B
B
B
There's
no
comma
before
that,
and
then
you
look
at
154,
there's
a
comma
before
that,
and
then
similarly
in
134,
there's
a
comma
before
who's,
and
so
those
three
things
don't
match.
I'm
guessing
it's
probably
more
correct
to
remove
the
commas
and
match
162
as
long
as
it
still
reads.
Okay,
yeah
I
had
noticed
that
until
you
had
the
one
at
the
bottom,
which
has
no
commas
though,
and
then
134
same
thing,.
B
B
A
Yeah,
why
don't
you
yeah
write
something
the
bottom
that
says
that
we
made
a
different
change
and-
and
it
should
refer
here
to
the
changes
that
we
did
make
right.
So
there
did
tell
it
tell
so
someone
she
wants
to
figure
out
what
we
did
instead,
then
she
can
read
that
commit
well.
We
took
her
advice
and.
A
This
happens,
you
know
one
company
in
Ottawa
they
got
a
hundred
million
dollars
of
investment
and
I
couldn't
understand
what
they're
gonna
do
with
100
million
dollars,
because
it
was
too
less
to
build
a
heart
too
little
to
build
hardware
and
too
much
to
build
software
and
I'm
like
either
give
them
more
money
or
less
right.
I
mean
I,
don't
understand
anyway,.
B
A
B
So
I'm
just
gonna
have
my
comments
on
her
suggestions.
Write
about
a
system,
/
firmware
/
software
sounds
like
text
that
we
used
to
have
in
a
document
that
we
didn't
like
and
I
replaced
it
with
the
name
of
the
role
like
we
changed
from
you
know,
to
evidence
and
to
a
tester
and
so
on,
and
so
there
I
think
the
phrase
of
from
a
system
for
a
more
a
software
I,
don't
like
the
things
that
may
include
and
neutral
on
that
last
part.
C
A
B
A
B
A
A
B
D
B
Comments
that
we
haven't
gotten
to
yet
I'm
gonna
paraphrase
cuz.
It
maybe
was
an
exact
wording,
but
he
said
you
know:
what's
the
difference
between
evidence
and
attestation
results,
can
they
have
the
same
format,
in
other
words,
going
to
use
an
e
for
both
and,
of
course
the
answer
is
yes,
because
evidence
is
not
a
definition
of
what
the
format
is
or
what
the
information
is.
It's
a
it's.
The
label
for
the
line
between
two
specific
roles
right,
and
so,
if
you
look
at
Kathleen's
definition
here
that
divide.
D
B
A
C
B
A
B
C
B
A
D
F
B
Agree
with
Peters
point,
because
I
thought
we
actually
covered
that
later
in
the
document
and
not
in
the
definition
of
evidence
which
we're
trying
to
keep
short
I
think
NIT,
it's
correct
for
the
definition
of
evidence,
although
we
do
try
to
cover
a
Peters
point
later
in
the
document.
That's
my
recollection
anyway.
That's.
D
I
think
it's
it's
misinforming
the
reader.
If
we
limit
evidence
to
be
statements
about
the
system
itself
here
in
the
definition
of
evidence,
because
the
evidence
may
be
included
in
a
decision
or
but
it
may
be
coming
from
someplace
else.
What
do
you
mean?
Even
you
trying
to
figure
out
which
case
you're
talking.
D
Claims
can
be
made
directly
or
they
can
be
made
indirectly
by
something
that
you
see
in
the
evidence.
If
you
see
proof
that
a
system
is
something
you
may
use
evidence
that
that
system
has
a
certain
property
that
you
gleaned
from
someplace
else,
either
from
a
some
other
at
the
station
or
knowledge
that
you
have
in
your
at
the
appraiser.
But
that
is
all
evidence
that
goes
into
the
decision
that
you're
gonna
make
about
the
attestation
evidence.
B
B
I,
do
that
I
disagree
with
you
so
either
I
I,
don't
understand
you
or
I
disagree
with
you.
So,
for
example,
the
verifier
uses,
evidence
and
endorsements
and
the
information
that's
in
the
endorsement.
You're
used
to
do
the
appraisal,
but
it's
not
what
I
would
call
evidence,
even
though
it
includes
claims
and
the
claims
might
even
he
claims
about
the
ax
tester,
but
they're
not
evidence
as
we
defined
it.
I.
B
D
All
right,
it's
it's
all
think
of
it
in
terms
of
a
trial.
You
know
the
appraiser
is
a
judge
and
their
point
to
be
a
case
made
at
the
appraiser
and
there's
gonna,
be
things
entered
into
evidence
there
and
that
an
endorsement
is
itself
just
evidence
that
is
being
used
to
determine
the
outcome
of
the
appraisal.
So.
F
C
B
Peters
terminology:
it's
in
the
you
know,
courtroom
analogy
the
since
evidence
was
the
term
that
we
came
up
with
the
have
this
particular
meaning
right
and
Peter's
example.
The
closest
term,
in
that
context,
you
might
be
able
to
come
up
with,
is
testimony
all
right.
If
you
talk
about
the
testimony
of
a
particular
witness
and
using
other
information
to
decide
beyond
that,
witness's
testimony,
whether
that
witness
is,
is
you
know
whether
it's
collaborating
information
or
statements?
I,
have
you
know
whether
that
entity
has
a
criminal
record
or
something
that
would
be
considered
in
door?
B
D
Have
a
question-
and
you
can
tell
me
whether
you
think
this
is
evidence
or
not
evidence,
so
let's
say
that
I
have
a
system
that
is
a
hypervisor
with
some
sort
of
guest
operating
system,
environment
and
I'd
like
to
attest
to
that
operating
system
environment
at
the
appraisal.
Point
I
decide
that
I
need
to
also
have
some
information
about
the
hypervisor
itself
and
even
the
means
and
by
which
I
measured
the
hypervisor
and/or
measured
the
appraisal
are
the
the
guest
operating
system,
environment
and
so
at
the
appraiser.
D
B
That
is
evidence
because
it
all
comes
from
the
overall
system
mean
the
hypervisor
is
on
box,
and
so
it
comes
in
a
single
well
if
the
protocol
is
constructed
to
allow
it
comes
in
in
a
single
message
where
you
have
multiple
testing
environments
all
put
together
into
a
collection
of
evidence.
That
then
comes
from
that
overall
test.
D
B
Can't
answer
that
without
diving
into
a
later
section,
this
is
saying:
if
you
put
a
big
box
around
the
whole
set
of
call
that
the
ax
tester
as
being
the
thing
with
a
protocol
age
right
the
protocol
endpoint
is,
you
know
the
thing
was
say
you
know
a
TLS
session
to
the
to
the
verifier
and
then
what
is
that
entity
is
what
you're
asking
you're
saying.
Well,
it
depends
on
the
implementation.
All
comes
across
one
session
from
something
on
the
Box.
B
D
B
B
B
A
I
think
and
I
think
and
we're
over
the
hour
here,
but
I
also
I.
Think
that
he's
right
that
in
general
that
we're
not
writing
a
PhD
thesis
that
needs
to
be
theologically
correct
about
attestation.
We
need
to
be
clear
enough
to
the
extent
that
it
affects
people,
the
other
documents
and
and
that's
enough.