►
From YouTube: RATS Architecture Design Team, 2020-07-10
Description
RATS Architecture Design Team, 2020-07-10
B
B
A
B
B
B
C
A
D
D
B
D
The
the
first
one
is
the
second
comment.
I
think
the
original
text
and
read
needs
to
stay
at
the
top
of
P.
I
is
not.
The
only
aspect
of
privacy
is
just
an
important
subset,
as
you
can
see
previously,
you
talked
about
privacy
and
in
the
green
it
replaces
by
introducing
PII
is
the
first
thing.
That's
a
sub
case.
We
need
to
talk
about
the
more
general
case
you
can
see.
D
C
D
C
C
D
I
think
the
you
know
it's
fine
to
have
a
for
example
or
whatever,
but
at
least
for
example,
isn't
the
first
sentence
right,
and
so
you
are
correct
at
the,
for
example,
up
above
is
just
a
sub
case
right,
that's
why
it
starts
with,
for
example,
alright.
So
that's
well,
you
could
say
you
know.
For
example,
you
know
Perkin
identifying
information,
blah
blah
blah,
but
that's
just
for
example,
so.
C
So
get
just
just
to
fill
it
in
for
me
to
be
sure,
I'm
understanding.
So
there's
what
give
me
a
few
examples
of
what
you're
thinking
of
you
said.
I
mean
the
one
example.
You
know
it's
PIAA,
so
it's
identifying
the
person.
That's
one
thing
it's
another.
Another
issue
is,
you
know,
identifying
vulnerable
things
to
attackers
yeah.
D
D
B
D
C
D
E
D
D
Exactly,
and
so
this
is
just
explain-
I
mean
the
question:
what
what's
the
intent
of
the
privacy
considerations
section
is
to
have
some
motivation
as
to
things
that
different
use
cases
would
consider
important
and
why
implementers
should
pay
attention
to
it,
and,
in
particular,
when
encryption
is
not
mandatory.
Here
is
things
that
you
should
think
about
as
to
why
you
should
do
encryption
yeah
it
didn't
stay
that
way.
The
previous
text
did
I.
Think
the
previous
yeah.
F
B
I
took
I
took
the
sum
of
Lawrence's
text
and
I've
added
it
there
that
I
don't
know
it.
You
said
you
don't
think
we
can
deal
with
all
of
it.
I
like
what
you
did
it
Michael
that
looks
great
okay,
so
I
think
just
wondering
if
there's
more
text
I
can
grab
that
improves
the
situation,
remembering
that
we're
going
to
publish
right
after
this
yeah
all,
and
so,
if
Lawrence
wants
to
continue
to
improve
it,
then
I
strongly
encourage
that,
but
I
want
to
kind
of
be
able
to
close
off
at
least
this
116
problem.
B
So
the
next
two
points
where
the
scope
of
access
needs
to
be
emphasized,
and
then
this
is
about
including
administrator
access
to
date.
I,
don't
know
what
to
do
with
that,
and
is
there
a
way
to
make
inferences
but
attestation
from
their
processing?
That
should
be
noted
as
well.
I'm,
not
sure
what
that
means
to
you.
So
by
virtue
of
the
fact
that
you
got
to
enter
the
amusement
park,
the
I
must
therefore
infer
that
you
or
at
least
48
inches
tall
or
that
you
had
to
measure
a
height
measuring
device.
D
So
if
I
understand
your
interpretation
of
Kathleen's
point
okay,
that
is
information
about
the
actions
taken
by
the
relying
party,
not
the
actions
taken
by
attestation,
but
it
sounds
like
what's
going
on
you're
trying
to
say
is
the
attacker.
There
is
trying
to
infer
some
knowledge
of
what
the
appraisal
policy
is
all
right
say.
Maybe
you
knew
that
if
he
knew
its
criteria
was
48
inches
or
taller
to
ride
this
ride,
then
you
can
make
his
met
Frances.
But
nor
do
that
you
have
to
understand
the
appraisal.
Policy
is
48
inches.
B
Yeah,
so
it's
this
point
here:
is
there
a
way
to
make
inferences
about
attestations
from
their
processing
that
could
be
as
I?
Could
that
could
be?
Referring
to
you
know,
timing,
attacks
on
crypto
protocols,
kind
of
thing
or
I,
don't
know
what
right
yeah
I'm,
not
sure
what
to
say
there.
But
if
you
have
an
idea,
no
I
have
no
idea
what
to
say:
yeah.
C
D
We
just
try
to
make
Kathleen
a
happy,
because
your
description
is,
you
know,
maybe
hypothetical,
but
if
we
don't
know
anything
to
say,
then
we
could
just
roughly
repeat
your
sentence.
You
know
without
explaining
why
you
could
say
in
some
cases
there
may
be
you
an
attacker
may
be
able
to
make
inferences
about
attestations
from
the
behavior
of
a
you
know,
from
the
communicate
from
the
behavior
of
a
relying
party
in
response
to
an
ax
tester
or
something
like
that
sure
it
which
doesn't
explain
anything.
D
D
C
D
If
you
in
that
particular
case,
if
you
don't
have
attestation,
you
just
have
authentication,
then
you'd
be
able
to
infer
the
same
thing.
She's
don't
have
an
appraisal
policy.
It's
kind
of
you
know
built
in.
If
you
can
authenticate,
then
there's
a
policy
and
you
still
be
able
to
make
that
same
thing
and
adding
attestation
to
it
doesn't
actually
change
that.
D
B
D
B
Which
case
you
have
a
database
of
who's
who
meets
the
category
and
who
does
not
yeah
I.
Understand
that
that
point
in
that
a
cetacean.
We
always
know
that
that
right,
that
that's
probably
the
case
for
most
devices
and
most
attestations
that
you
could
make
an
extensive
database
of
what
of
of
you
know
what
things
are
in
a
good
state
and
what's
are
not.
D
E
Is
that
privacy
is
something
you
know,
privacy
considerations
apply
every
role
in
the
architecture
and
and
so,
and
we
could
detail
that
with
some
with
some.
You
know
examples,
but
the
general
idea
is
the
the
tester.
If
the
ax
tester
is
privacy
aware,
then
it
will
not
collect
privacy.
Sensitive
claims
in
the
first
place.
D
E
E
C
D
Through
so
maybe
so,
Michael
I
noticed
you
added
the
sentence
from
Kathleen,
although
personally
I
would
prefer
augmenting
it
with
my
observation,
if
you
guys
agree,
which
is
the
fact,
but
such
I
don't
know
how
to
phrase
it
yet
because
my
original
point
is
probably
not
how
I
would
phrase
it
in
a
document.
But
my
original
point
is,
you
could
have
done
the
same
thing
with
with
authentication
just
less
scalable
I.
Don't
know
how
to
say
that
here.
D
D
D
E
B
Well,
sometimes,
the
privacy
considerations
just
needs
to
hit
people
over
the
head
too.
They
members
things,
and
sometimes
it's
just
enough
to
you-
know
it's
just
to
say
yeah
to
the
reviewers.
Yes,
in
fact,
we've
thought
about
this
problem
and
we're
aware
of
it
and
we
don't
have
a
solution
or
we're
not
making
things
any
worse
or
so.
Are
you
okay
with
this
sentence?
You
just
dictated
it
for
me.
Yes,.
B
B
A
Also
I,
like
the
strategy
part,
did
not
like
so
much
was
the
phrasing
like
instruction
on
the
addressing
the
reader
level.
Like
hey
you
this
and
this
we
have
to
do
this.
That
is
I.
Don't
know.
Maybe
it's
just
me.
Maybe
it's
just
me,
but
that
was
a
sign
usual
for
privacy
consideration.
Maybe
it's
I
don't
know.
D
B
D
D
F
C
And
I
think
you
kind
of
have
to
evaluate
every
claim
in
the
use
cases
and
usage
scenarios.
No
I
mean
we
can't
really,
maybe
maybe
you
could
say
many
claims
are
potentially
I
might
be.
Okay
with
that,
but
I
basically
think
you
have
to
look
at
every
claim.
I,
don't
think
you
can.
You
know
because
of
things
like
the
no.
E
It's
just
the
the
wording
is
saying
it's
not
trying
to
say
that
you
don't
have
to
look
at
every
claim.
It's
saying
that
it's
trying
the
the
detect
seems
to
be
asserting
that
that
there
are
some
intrinsic
property
and
claims
that
make
it
such
that
they
all
have.
You
know
privacy
implications.
That's
not
true.
I.
G
D
D
B
D
D
B
B
B
To
get
more
pull
requests
in,
but
I
think
that
we
need
to
simply
say
look
this.
We
can
always
improve
the
document
and
we
didn't
say
okay.
This
is
good
enough
for
the
working
group
to
say
we're
mostly
done
with
this
and
go
and
push
upwards
and,
as
I
said,
we're
going
to
have
lots
of
lots
of
lots
of
pushback
from
from
things.
So
don't.
C
I
still
have
a
lot
of
issues
with
the
endorsed
endorsement
and
he
is
he
gets
in
and
all
that
and
and
my
my
thought
there
was
to
wait
till
this
one
was
published
step
back,
take
a
look
at
it
and
then
create
a
pull
request
against
that.
So
we
have
some
more
clear
text
to
work
off
of
and
because
I
think
I
think
that'll
be
the
more
efficient
way
to
try
and
get
through
it.
D
Mean
some
questions
might
be
more
foundational
as
how
do
people
think
about
things
and
for,
if
that's
the
class
of
things,
then
some
type
of
meeting
presentation
might
actually
be.
You
know:
here's
a
direction.
Here's
another
direction.
People
like
this
direction.
People
have
an
alternate
direction.
Sometimes
that
can
be
more
efficient
than
going
straight
to
texts.
F
D
C
Yeah
one
of
the
trouble
I
keep
coming
up
with
them.
The
endorsement
discussion
is
that
I
tried
to
give
an
example.
That's
illustrative
and
everybody
thinks
that's
what
I'm
proposing
to
standardize
in
detail,
I
see
and
by
just
writing
text.
That's
then
it's
clear
that
I'm
just
this
is
the
text
I'm
proposing.
So
we
don't
get
out
of
that
trap
and
that
death
trap
has
happened
a
bunch
of
times
already
so
I'm
at
least
one.
D
C
I
wouldn't
read
the
Charter
and
and
I
don't
think
it
mentions
the
term
endorsement.
It
does
say
that
I
mean
my
interpretation
of
that
Charter
was
that
we're
not
going
to
write
a
detailed
standard
document
for
what
the
endorsement
or
the
values
or
the
all
that
we're
not
going
to
write
a
detailed
standard
for
that,
but
it
doesn't
mean
we
can't
have
like
two
or
three
sentences
that
describe
it
at
a
higher
level.
Oh
yeah.
D
B
A
B
B
B
C
A
B
C
Case
it's
a
secret
in
which
yeah
and
you
could
have
yeah
millions
or
billions
of
them
and
the
the
way
there
could
be
all
kinds
of
KDF,
s--
and
other
kinds
of
things
going
on
in
the
system
and
just
and
and
I'm
not
making
this
stuff
up,
I
mean
I,
actually
know
people
aren't
really
doing
this.
We
just
do.
C
D
D
C
D
A
Define
this
year
platform
that
they
have
this
endorsement
token
and
that
is
unfortunately
already
set
in
stone
and
we
have
to
disambiguate
that
so
we're
very
aligned
by
using
endorsement
yeah
and
saying
hey.
This
is
not
the
endorsement
token,
but
it's
close.
Here's
our
explanatory
expositional
text
and
I
think
that's
a
good
approach.
It's
actually
very
intuitive
term
and
only
slightly
confusing
platform.
It
isn't
stopped
from
using
it.
C
I
think
it
is
very
interesting,
I
think
it's
very
intuitive
for
some
people
and
I
think
it's
very
confusing
for
others.
I
mean
it's
certainly
confused
me
because
I
I
mean
the
answers,
I
get
of
what's
an
endorsement
and
what's
nice,
but
they
vary
by
who
I
talk
to
and
by
day
of
the
week
yep
you
if
you
show
me
a
written,
a
clear
written
description
of
it,
then
that
we
can
work
off
of
that's
fine,
but
I've,
seen
that
yeah
and
I
just
keep
seeing
different
discs.
Yes,.
A
This
she
takes
a
section
here,
it's
called
it's
meant
and
what
I
hear
from
you?
That
is
not
a
cure,
written
description
of
the
text,
meaning
of
an
awesome
or
endorsement
sorry.
So
there
is
this
section
of
endorsement.
There
is
the
definition
very
small
definition
at
the
top
of
the
text,
and
it's
not
sufficient
to
you.
E
Yeah
I
think
what
we
need
is
to
have
like
a
sidebar
meeting
where
we've
got
a
bunch
of
you
know,
use
cases
and
some
examples
and
look
at
each
one
and
try
and
try
and
understand.
From
the
perspective
of
of
you,
know
the
roles
architecture
what's
going
on
so
I
think
yeah,
it's
sort
of
like
the
elephant
problem
or
anything.
It's
like
everybody's
got.
D
A
different
yeah
I
think
months
is
different
from
may
be
different
from
trust
anchor
provisioning.
You
consider
appraisal
policy
provisioning
to
be
a
bunch
of
information.
One
piece
of
that
information
is
trust.
Anchor
provisions.
There
might
be
a
couple
of
different
elements
of
appraisal
fellows.
That's
one
way
to
phrase
the
phrase:
a
policy,
a
provisioning
which
is
a
bunch
of
different
things,
which
include
trust
anchors.
You
could
also
think
of
those
as
being
disjoint
to
I.
Think
of
one
is
being
subset
of
the
other,
but
you
wouldn't
necessarily
do
them
in
the
same
protocol.
D
You
need
to
trust
anchor
provisioning
mechanism
and
you
need
a
way
to
provision
the
rest
of
the
appraisal
policy,
but
I,
don't
think
we
currently
discuss
in
much
detail
trust
anchor
provisioning,
which
is
necessary,
and
that's
that's
the
part
where
I
completely
agree
with
how
Lauren's
phrase
is
you
got
to
have
some
consideration
of
that
I
think
the
T
Parker
textured
document
did
that
it
could
be
done
in
a
separate
section
next
to
were
so.
The
other
ones
are
I,
don't
know,
but
that's
something
that
I'm
not
going
to
do
in
the
next.