►
From YouTube: COSE WG Interim Meeting, 2020-09-09
Description
COSE WG Interim Meeting, 2020-09-09
A
All
right
welcome
everybody
to
that's
the
the
interim,
just
a
request,
a
reminder
that
this
is
an
official
I
tf
meeting
for
the
working
group.
So
the
notewell
applies.
A
So
some
administrative,
so
we've
already
gone
over
the
note.
Well,
if
anybody
is
willing
to
help
with
minutes,
that
would
be
great
all
we're
really
looking
for
whatever
action
items
or
important
points
might
need
to
be
gutted
down.
A
A
And
then
we
can
move
on
to
some
agenda
bartering.
All
we're
really
looking
at
for
today
is
talking
about
the
various
issues
and
discusses
on
the
documents.
Is
there
anything
anything
else
we
want
to
talk
about.
A
All
right
with
that,
let's
see
we
could
start
with
defense,
discuss
on
on
message
recovery
for
the
struct
document.
C
Yes,
I
was
thinking
about
this
recently
like
in
the
past
day
or
so,
and
I
guess
my
core
question
or
issue
is
really:
what
are
we
actually
trying
to
say
about
signatures
with
message
recovery
because
we
sort
of
say
that,
like
we
would
like
to
do
this,
but
it's
not,
it
doesn't
really
exist
right
now,
there's
there's
not
a
mechanism
to
find
and
so
like.
If
we
just
want
to
call
out
that
it
exists
or
that
it
might
exist
for
cose
in
the
future.
C
That's
one
thing,
but
if
we're
trying
to
actually
specify
what
the
interface
for
it
would
look
like,
that's,
maybe
a
little
bit
different
and
to
get
a
little
bit
more
into
the
details.
You
know
if
we
say
that
the
message
content
that's
transmitted
is
going
to
be.
You
know
some
subset
of
the
original
input
message,
because
the
other
bits
are
in
the
signature
for
recovery.
C
C
C
So
like,
if
we,
if
we
say
that
there
are
two
signature,
all
heard
of
schemes,
the
first
distinction
of
appendix
and
the
second
is
signature
with
message
recovery.
That
sort
of
seems
like
there
are
two
signature
schemes
that
are
two
signature
scheme,
types
that
we
can
deal
with.
C
But
if
we're
really
trying
to
say
that
there's
only
one
that
we
have
written
up
yet,
but
we
want
to
support
the
other,
then
maybe
we
need
to
say
a
little
bit
more
about
how
in
general
and
abstract
there
are
two
types
we
are
going
to
specify
the
first
type
here,
but
we
leave
open
options
for
expansion
and
specifying
the
second
type
is
future
work.
A
D
Okay,
my
computer
told
me
I
lost
it
so
initially
the
idea
was
just
tell
people
that
it
existed
and
then
I
ended
up
tossing
in
a
couple
things
simply
because
I
a
couple
of
things
to
look
at
simply
because
I
thought
of
them
at
the
time
I'm
willing
to
remove
those
out,
although
I
intended
them
to
just
be
these
are
things
you
might
have
to
think
about.
C
Sure-
and
I'm
happy
with
these
are
things
you
might
have
to
think
about.
I.
What
I
was
saying
is:
this
problematic
is,
if
we
were
actually
saying
like
you,
can
just
go
off
and
allocate
a
code
point
first
thing
to
algorithm
and
you're
expected
to
be
able
to
implement
something
and
have
it
interoperate,
because
I
personally
that
I
would,
I
would
be
able
to
do
that
with
just
the
text.
That's
here.
So
I'm
happy
to
see
us
reframe
it
as
well
as
this
is
something
to
be
aware
of.
C
Right
because
I
think
that
at
least
on
my
initial
read
of
it
I
was
talking
about,
we
only
have
algorithm
code
points
as
opposed
to.
We
only
have
the
procedure
specified,
and
maybe
I
was
just
misreading
that
and
I
should
shut
up,
but
that
was
sort
of
where
I
was
coming
from.
D
You
know
I
flipped
my
content,
so
I
was
stealing
the
invites,
because
that
signature,
algorithm
steals
lead
bites
and
there's
also
some
some
lineup
issues
that
need
to
be
dealt
with
by
the
first
person.
Who
does
this
yeah?
C
D
D
C
C
Well,
when
is
the
earliest,
the
rain
is
predicted
to
arrive.
F
D
Okay,
so
I
can
get
some
text
on
the
list
for
that
issue
and
once
you've
said
yes
that
looks
like
what
you
want
to
see.
Then
we
should
be
able
to
get
this
third
and
come
forward.
I
think.
C
A
A
Acceptable,
I
took
a
quick
read
myself
yesterday
and
it
seems
okay
to
me,
but
more
feedback
would
be
welcome.
A
A
All
right
take
that
as
a
no,
I
believe,
the
with
the
hash
alex
document
ben
still
had
a
comment.
D
Right,
I
think
it's
just.
I
think
we
had
just
have
one
thing,
which
is:
I
managed
to
not
pull
one
one
sentence
from
the
document.
I
plan
to
go
over
that
this
afternoon
too,
and
get
that
one
updated.
D
D
And
then
it
should
be
able
to
go
to
the
iesga.
The
last
document
is
the
is
the
initial
algs
document
has
a
whole
slew
of
mail
from
ayanna
that
I
need
to
go
over.
Bin
did
respond
once
to
them
saying
it
may
not
make
sense
to
look
at
this
document
by
itself.
You
may
need
to
look
at
the
structure
document
as
well.
I
have
a
feeling
that's
what
most
of
the
issues
are.
D
C
Yeah
and
just
to
jump
in
this
has
been
again.
It
was
my
sense
that
all
or
almost
all
of
the
ionic
questions
with
alex
document
would
be
would
disappear
if
they
had
read
the
struct
document
first,
but
I
wouldn't
swear
to
that,
because
I
did
not
read
their
email
entirely
carefully
and
I
also
read
it
quite
some
time
ago.
So
my
memory
is
not
perfect.
D
A
C
C
Right,
a
similar
question
came
up
in
a
different
context
and
I
went
and
looked
through
my
mail
archives.
I
believe
it
got
a
review
from
the
cfrg
crypto
panel,
maybe
18
months
or
two
years
ago
roughly,
but
I
have
not
looked
at
how
the
document
has
changed.
If
at
all,
since
then,.
B
C
B
D
C
C
B
D
E
Be
worth
it
so
so
some
people
might
just
say
screw
it
we're
just
going
to
stick
with
warehouse
and
transmit
that.
But
but
many
people
will
say
no,
no,
I'm
just
optimizing.
My
calculation
and
library
use
and
I'll
convert
the
point
and
have
a
standard
have
the
old
standard.
E
Depending
which
form
you
transmit
it
in,
but
if
you
one
side
point,
is
that
one
side
can't
unilaterally
decide
it's
going
to
do
it
in
this
form
without
informing
the
other
one
in
a
protocol.
C
E
A
C
And
if
I
could
back
up
to
a
previous
talk
book
briefly,
I
was
skimming
through
a
bit
more
stuff
about.
I
guess
the
ayanna
comments
and
I
was
noticing
in
my
comments
on
struct,
that
the
ionic
considerations
for
the
which
registry
was,
I
think,
the
kdf
algorithms
or
the
maybe
it's
the
header,
algebra
or
the
register.
I
think
there
was
one
other
registry
that
I
think
we
had
the
registrations
for
in
the
wrong
document
and
that
may
have
also
been
something
that
I
was
confused
about.
D
That
is
extremely
possible.
I
that's
part
of
what
I
was
needed
to
go
through
on.
D
D
E
I
I
saw
that
you
posted
the
outer
sign
document.
I
guess
I
don't
know
if
it
was
yes
last
week.
I
guess
it
was
the
ninth,
the
fourth,
I
guess
did
you
know
that
the
contributing
to
this
document
points
me
at
the
original
document
github,
and
maybe
you
didn't
actually
push
the
github
for
this
one
up.
D
I
haven't
pushed
this
up
into
it.
I
just
haven't
pushed
push
the
document
up
into
github,
it's
going
to
the
same
route,
it's
going
to
be
in
the
same
github
registry.
It's.
E
Going
to
be
the
same,
github
repo
is
what
you're
saying
yeah,
okay
and
in
section.
E
Three,
you
use
the
terms
cosignature
cosine,
one
and
cosign,
and
I
thought
coast.
Signature
is
the
signing
object
that
we
create,
and
so
I
don't
think
I
would
ever
apply
I'm
unclear
if
I
would
apply
a
counter
signature
to
just
a
coast.
Signature.
That's
the
the
third,
the
last
paragraph,
one,
two
three,
maybe
the
fourth
sentence
when
done
on
a
coast,
signature
or
cosine,
one.
E
Okay,
so
so
I
just
was
like
really
confused.
As
I
read
it
like,
I
went
wait.
Maybe
this
works
in
a
different
way
than
I
thought
so
then
I
realized
it
might
be
a
typo
okay.
D
Most
of
those
code
signatures
should
be
signer,
not
signature.
D
B
B
D
E
Did
you
asking,
did
I
read
it?
No,
I
haven't
finished
reading
it,
but
I
am
reading
it
all
right.
A
A
But
do
please
read
it
and
respond
on
the
list
with
any
any
any
comments
or
questions
or
points
of
clarifications.
F
F
A
All
right
anything
else.
D
F
But
it
talks
about
yeah
the
counter
signature
attribute.
D
Does
it
okay?
That
would
definitely
need
to
get
cleaned
up.
F
And
the
last,
the
second
yeah,
the
second
one
is
also
in
the
same
text
and
says
about
two
contexts
are
distributed
as
a
pair.
F
D
A
H
The
topic
for
next
cozy
meeting
is
that
entirely
devoted
to
the
counter
signatures
or
what's
what's
the
plan.
A
That
is
a
great
question
that
was
going
to
be
my
next
question
to
all
of
you
is.
We
currently
do
have
another
hour
scheduled
in
two
weeks.
What
do
we
want
to
use
that
time
for.
A
D
H
So
we
got
a
bunch
of
comments
from
from
various
people
during
the
summer.
We
haven't
really
put
those
together
into
the
into
the
document,
so
that
was
the
reason
why
I
would
ask
if
the
seaboard
certificate
is
coming
up.
We
need
to
prepare
that,
but
we
haven't
done
that
yet.
D
A
B
While
I
think
that
it's
two
weeks
right
not
next
week,
two.
B
You
know
that's
enough
time
for
people
to
read
it.
Maybe
at
least
we
could
raise
issues
even
if
it's
not
fully.
A
Okay,
so
here's
here's
what
I
propose
the
next.
Our
next
meeting
in
two
weeks
will
focus
primarily
on
charter
discussion,
but
if
there's
anything
to
talk
about
with
counter
signatures,
that
can
we'll
definitely
spend
at
least
some
time
on
that
that
does
that
sound
acceptable
to
everybody.
A
A
Okay,
well
with
that,
I
think
we
can
declare
the
meeting
officially
over
I'm
going
to
stop
the.