►
From YouTube: IETF-SATP-20230516-1400
Description
SATP meeting session at IETF
2023/05/16 1400
https://datatracker.ietf.org/meeting//proceedings/
A
C
C
C
C
A
A
D
D
E
A
B
A
You
know,
as
as
part
of
the
set
P
output,
we
might
need
to
get
permission
from
the
area
director
to
do
that
and
change
the
charter
a
little
bit
because
he
noted
that
it's
not
in
the
charter
at
the
moment.
That's
really
not
a
big
ask
nobody's
going
to
object,
putting
on
a
vocabulary
document
to
to
get
that
done
and.
A
C
C
C
C
B
B
And
basically
right
now
just
just
collect
them
and
then
we
can
bucketize
them.
You
know
later
on,
but
we
looked
at.
You
know
using
the
model
that
the
TLs
has
for
how
it
describes
error
messages,
and
we
thought
that
would
be
a
good,
a
good
model
to
follow.
And
so
so
that's
that's
not
a
work
in
progress.
F
B
Yeah,
so
so
shipping
is
online,
but
shipping
Denison
myself
are
on
bi-weekly,
like
our
own
little
subgroup
call
if
anyone
wants
to
join
us
most
most
welcome
to
do
that,
but
we're
discussing
a
couple
of
things
first,
is
how,
to
you
know,
address
this.
This
problem
of
get
to
identification
and
Discovery
and.
B
And
so
this
is
by
the
way,
if
you're
wondering
why
there's
this
discussion
on
the
mailing
list
about
you
know
application
to
application
context,
and
so
on.
You
know
it
came
from
this
sort
of
discussion
that
you
know
when
you
talk
about
Gateway
identification,
you
know:
do
we
need
to
also
identify
a
know
or
convey
what
types
of
assets
are
supported
by
a
Gateway,
in
which
case
you
know
we
need
this.
That
definition
and
remember.
B
We
talked
about
asset
profiles,
so
so
I
believe
Dennis
who's,
not
in
the
call
he
wants
to
continue
to
work
on
asset
profiles
as
well,
because
it's
it
is
going
to
be
needed.
You
know
going
forward,
but
that
that's
that's
where
it
is
and-
and
you
know
shipping,
if
you
want
to
jump
in,
please
feel
free.
F
B
Okay,
so
so
that's
that's
on
our
to-do
list
right
shipping
is,
is
provided
next
update.
We
we
know
it's
not.
You
know
it's
not
part
of
the
work
item
list,
but
it's
it's
something
that
I
think
the
group
will
need
down
the
road.
So
in
a
sense
it's
a
we
can
see
it
as
a
side
project,
but
you
know
a
relevant
one
and
wherever
I
go
so
so
there's
been
separate
discussions.
You
know,
with
a
number
of
you
know
what
what's
called
layer.
One
blockchain,
you
know
systems
out
there.
B
You
know
polka,
dots
and
Avalanche
and
so
on,
and
you
know
the
ability
to
number
identify
networks
is.
Is
it's
not
just
our
problem?
Everybody's
got
that
same
problem
and
this
idea
of
gateways
are
slowly.
You
know
you
know
catching
on
so
so
for
folks
who
have
not
heard
of
the
monetary
authorities
of
Singapore,
they
had
a
five-year
project
called
Ubi
and
ubin
and
I.
Think
last
year
or
early
last
year
they
produced
the
the
final
stage
report
and
you
know
they.
B
They
there's
this
massive
diagram,
but
at
the
edge
of
it
there's
this
block.
Is
this
Gateway
right,
and
so
what
they're
trying
to
do
there
is
to
provide
support
for
multi
Network
multi-blockchain
multi-currencies.
You
know
this
is
the
the
ambition
of
Singapore
is
to
be.
You
know,
kind
of
the
next
Switzerland,
but
it's
interesting
that
that
they
understand
that
there
needs
to
be
gateways
to
you
know
be
the
point
where
they
enforce
policies,
whether
it's
asset
transfer
policies,
you
know:
currency,
International,
payments
policies
and
so
on
and
so
on.
B
F
C
C
B
B
You
on
email
is
just
to
go
through
every
you
know
stage
used
to
be
called
phases,
but
every
stage
States,
particularly
one
two
three
and
see
what
things
could
go
wrong
and
and
if
you,
if
you
were
the
I.T
guy,
if
you're
the
engineer
that
was
running
a
Gateway,
a
set
of
gateways
for
your
organization
and
and
one
or
more
of
them
started
misbehaving
or
the
the
failure
rate
of
you
know,
you
know,
interactions
increases,
you
know.
What
would
you
do?
B
C
B
A
So
the
the
simple
Logistics,
having
just
got
it
in
in
back
from
a
heated
discussion
with
the
IAB
and
the
iesg
about
this
particular
topic,
is
that
it
should
be
an
interim
meeting.
So
it
could
be.
It
could
be
a
separate
one
or
we
could
do
it
as
part
of
the
regular
monthly
meeting
and
have
more
of
a
more.
A
Actually
go
over
and
review,
and-
and
you
know
we
have
a
full
hour
and
we
can
spend
an
entire
hour
sometime
on
just
the
error
handling
you
go
through
stage
by
stage.
That's
probably
the
easiest
thing
to
do,
but
it
doesn't
have
to
so
interim
meetings
do
not
have
to
be
on
meet
Echo.
You
know
they
tend
to
to
work.
Well,
too,
we
could
do
Zoom
or
something
else,
but
officially
it
has
to
be
an
interim
meeting
to
to
do
that
level
of
discussion
with
you
know,
conclusions
out
of
it.
B
A
B
That's
that's
good!
So
so
so,
if
that's
the
case
well,
why
don't
we
shoot
for
the
next
at
the
next
interim
meeting
or
or
you
know,
create
one
earlier
before
the
next
date,
just
to
have
a
full
hour
on
the
error
messages
and
just
have
a
just
a
discussion
and
and
I,
don't
know
how
people
feel
about
Zoom
I
can
provide
the
zoo
it's
an
easier.
B
B
A
B
C
So
I'm
reading,
as
I'm
talking,
is
that
it
doesn't
impact
or
doesn't
necessarily
change
what
we're
trying
to
do
it's
more
to
do
with
message
formats
rather
than
the
assets
themselves,
so
John
might
have
more.
He
He
suggests
John
might
have
more
to
say,
but
his
understanding
is
that
we
do
need
to
compare
and
contrast
a
bit
to
make
sure
that
we
are
clearing
how
our
protocol
is
unique,
but
the
iso
isn't
we're
not
duplicating
the
iso
standard
and
that
it
probably
best
seats
situated
inside
the
architecture
draft.
C
So
an
invite
for
comments
on
that
and
then
also
mentioned
Thomas
who,
by
name
in
terms
of
the
multi
cbdc
scenario
in
the
use
cases,
and
that
has
sparked
a
conversation
with
Andre
and
he
will
update
the
document
in
in
near
time
with
that.
Was
there
anything
you
wanted
to
add
to
that
in
terms
of
the
He
suggests,
you
might
have
an
update
to
give
in
terms
of
the
multi
cbdc
use
case.
C
G
Standard
says
more
or
less
that
you
want
to
do
something.
For
example,
I
want
to
send
an
asset
to
to
Claire
on
your
account,
given
my
account,
but
it
does
not
specify
how
this
is
done.
It
doesn't
it's
not
a
protocol
so,
but
it
has
a
wide
wide
range
of
assets
and
facilities.
For
example,
you
can
have
reconciliation
of
bank
statement,
you
can
have
a
payment
payment
orders
into
bank
payment
orders,
customer
payment
orders
and
so
on,
and
so
on.
G
B
G
C
It's
worth
exploring
if
we
can
use
some
of
the
iso
messaging
I
mean.
Why
would
we
not
want
to
use
messaging,
that's
already
been
agreed
and
as
a
standard,
that's
adopted
by
a
lot
of
the
institutions
that
we're
hoping
will
use
our
protocol,
but
it's
good
to
get
the
clarity
that
the
I
said
in
what
you've
said:
John
very
much
mirrors.
What
Rama
seems
to
have
said
here
in
that
it's
not
a
protocol.
It
is
more
of
a
messaging
standard
with
core
messaging
created,
rather
than
a
protocol
for
transfer.
A
Yeah
I
I
think
one
of
John
Levine's
comments
was:
could
that
messaging
structure
be
used
even
within
this
ATP
protocol
right?
Could
that
be
the
messaging
format
so
that
we
don't
duplicate
something
else?
There
are
some
tricky
elements
of
how
to
refer
to
another
document
that
is
normative
from
these.
You
know
the
ietf
doesn't
refer
to
external
documents,
all
the
time
for
stuff
that
is
not
easily
referenced
within.
So
if
it's
a,
if
it's,
we
need
to
figure
out
how
to
so.
Let
me
try
to
enumerate
the
paths
forward
right,
one.
A
B
A
Them
with
ietf
specific
criticals,
the
other
option
is,
we
could
do
something
that
is
at
least
collaborative,
so
that
they
could
interoperate
and
I.
Think
that
that's
always
a
good
thing
to
do.
Even
if
the
messaging
structure
is
different,
we
can
use
the
same
sort
of
Concepts
to
make
sure
that
they
they
can
align,
align
within
an
implementation,
and
then
the
third
option,
of
course,
is
to
actually
use
their
messaging
structure
and
their
messaging.
B
A
A
B
That's
a
good
good
idea:
yeah,
okay,
no,
it
makes
sense,
I
mean
I
I
see
it
was
at
the
very
least
the
at
the
stage
zero.
What
what
triggers
the
whole?
You
know
transfer
unit
direction
from
Gateway
One
Gateway
two
is
this
messaging
thing
has
occurred
at
the
application
layer
and
now
the
gateways
know
exactly
what
to
do,
and
this
is
the
carrying
aside
at
the
very
minimum,
that's
sort
of
one
place
where
we
can
interface.
B
We
could
reuse
what
they
have,
but
but
you're
right
I
mean
if
they
have
a
you
know,
a
message
structure
that
allows
us
to
carry
like
this.
What
is
it
the
the
lock
assertion
and
the
receipt
assertion
coming
back
that
yeah
sure
I
mean
so
so?
Okay,
so
we'll
have
to
look
into
that
I,
don't
know.
If
anyone
is
is
you
know
willing
able
to
spend
some
Cycles
looking
at
that.
B
G
G
No
not
really
on
I
know
a
company
which
would
do
that.
It's
bottom
line,
it's
an
American,
College,
actually
Boston
or
american-based
company
who
yeah
you
will
find
I,
I.
Think
I.
Think
the
risk
of
these
huge
standards
is
that
you
will
find
somebody
who,
who
tweaked
and
twisted
it
to
do
something
similar.
So
but
it's
probably
not
their
real
and
they
have
a
a
completely
different
sort
of
standardization
cycle
and
approach,
which
is
probably
multi-year
and
and
whatever
so.
B
G
E
E
So
if
if
we
were
going
to
do
anything
that
might
make
sense-
and
my
recommendation
would
be
primarily
just
look
at
some
of
the
data
field-
descriptions
that
they
have-
and
you
know
may
as
well
adopt
some
of
them,
but
not
the
message
structure
itself,
but
you
know
they've
got
encodings
for
various
kinds
of
data
fields
that
they've
standardized
on
you
know
whether
it
be
country
code
or
currency
code
or
or
whatever
and
they've
got
a.
You
know.
E
They've
done
a
whole
bunch
and
that's
kind
of
the
approach
that
that
I've
taken
in
a
project
that
I've
been
working
on
is.
Is
you
know
if
they've
got
you
know
everybody
has
to
have
coding
for
some
of
these
fields
and
if
they've
already
decided
on
it
may
as
well
just
use
the
same,
but
I
don't
think
you're
going
to
find
anything.
You
know
in
ISO
that.
F
B
Yeah,
no
thanks
John.
That
was
that
was
my
understanding
that
you
know
this
idea
that
if
you
have
a
tokenized
asset-
and
you
do
you
know,
burning
mint
but
but
I
like
your
idea,
but
just
reusing
the
the
codes
right,
because
you
know
I
I
presume
they
have
codes
to
identify,
not
just
the
bank,
a
bank
but
also
what
network
and
what.
B
E
G
F
G
For
all
sorts
of
stuff
and
they're
also
also
networks,
so
they
are,
for
example,
the
the
Swiss
interbank.
Clearing
networks
uses
the
XML
standard,
the
iso
standard,
but
has
different
codes
for
banks,
but
there's
the
Swift
Network,
which
has
a
standard
identifier
for
banks
and
I
think
they
have
some
semi-standard
identifier
for
companies
and
the
way,
for
example,
to
structure
addresses
street
addresses
which
is
not
in
our
domain
at
all.
But
they
have.
They
have
many
sort
of
codes
because
they
come
from
Swift,
which
is
full
of
Arcane
code.
G
E
A
G
C
C
E
C
C
D
An
open
question,
but
this
is
regarding
the
end
points
of
the
gateways,
so
in
particular
do
we
want
to
keep
the
gate,
the
endpoints
simple,
just
as
in
the
current
drafts,
or
do
we
want
to
propose
a
different
endpoint
for
each
stage
of
satp
I?
Guess:
that's!
That's
one
of
the
options
that
I
was
thinking
of
so,
for
example,
have
an
endpoint
for
the
commencement
have
an
end
point
for
the
lock
have
an
end
point
for
the
lock
assertion
and
so
on
and
so
forth.
B
B
D
Yeah
yeah
because
I'm
just
thinking
about
potential
implementations
and
if
there
is
one
end
point
just
for
the
Gateway,
obviously
that
endpoint
will
have
to
be
able
to
handle
lots
of
different
messages
for
every
single
phase
versus
if
it's
different
endpoints
it'll
be
easier
to
implement
but
yeah.
Maybe
we
can
take
the
one
in
the
main
on
the
mailing
list
and
if
people
are
happy
with
the
suggestion,
then
we
can
look
at
updating
the
drafts
proposed
endpoints
for
each
one.
F
B
G
For
this
we
got
this
review
messages
from
this
Lucas
part.
You
would
I
think
T3
D3
already
discussed
this,
or
so
it
was
a
lengthy
message
on
April
28th
and
one
of
the
points
I
think
where
he's
he's
probably
completely
right,
is
that
there's
considerable
overlap
between
the
architecture
document
and
the
protocol
specification,
but
maybe
I
missed
something,
and
that
was
already
discussed
once.
B
B
A
Things
like
that
and
and
their
job
is
to
review
drafts
from
a
perspective
of
a
particular
expertise.
So
obviously
the
security
ex
you
know
area
directorate
reviews
drafts.
Looking
for
you
know,
issues
with
security,
so
I
guarantee
they're
going
to
look
at
our
draft,
but
it
surprised
me
that
that
one
came
in
so
early.
A
Normally
they
do
not
jump
on
stuff
until
you
actually
sort
of
get
to
last
call
with
you
know:
hey
we're
we're
actually
wrapping
up
this
document,
it's
ready
to
be
published
and
that's
when
most
directors
review
I
think
that
the
HTTP
group
has
learned
that
many
people
don't
get
the
HTTP
semantics
quite
right,
and
so
they
try
and
jump
on
early
to
make
sure
that
people
understand
it.
So
you
know
there
are
comments
around.
This
draft
isn't
ready
yet
I'm
like
well.
A
If
you
know
I,
responded
and
said,
of
course,
not
right,
it
says
brand
new
draft.
It's
just
coming
out,
but
I
would
definitely
you
know
to
my
advice
to
the
authors
would
be
to
to
read
that
and
do
take
it
into
very
careful
consideration,
and
then
there
will
be
multiple
reviews
in
the
future,
and
one
thing
that
we
can
do
as
a
working
group
as
a
draft
is
getting
ready
to
progress.
A
B
A
E
E
G
C
F
A
Zoom
is
entirely
possible
if
we're
more
comfortable
with
it.
There
are
some
advantages
of
this
one.
It's
more
integrated
with
sort
of
how
the
ATF
data
tracker
works
and
things
like
that
it
can
be
I.
This
I
believe
will
be
automatically
published
to
YouTube
and
that
doesn't
happen
with
zoom.
It
can
be
done
sort
of
after
the
fact.
If
we
want
it
done,
but
there's.
A
B
B
A
One
of
the
other
benefits
is
that
meat
Echo
is
also
integrated
with
our
blue
sheet
tracking
system,
which
many
of
you
are
probably
so
know
that
you
don't
know
what
blue
sheets
are,
but
we
used
to
pass
around.
We
had
to
take
attendance
for
legal
purposes
of
of
who,
actually,
you
know,
was
in
the
room
and
because
the
the
meat
Echo
is
integrated
with
the
data
tracker
you
actually
get
an
automated
list
of
of
who
is
you
know,
attending
the
meeting
and
things
like
that.
A
I'll
be
honest,
I
prefer
Zoom
too,
but
but
the
the
the
the
weight
of
some
of
the
other
features
do
kind
of
you
know
this
is
the
ATF
has
spent
a
lot
of
actual
Financial
effort
to
help
boost
the
meat
Echo
team
and
it's
kind
of
unique.
A
lot
of
the
stuff
that
exists
in
medeco
only
exists
because
the
aetf
is
asked
for
it,
even
though
we're
not
their
only
customer
like
the
raising
of
hands
tool
and
things
like
that,
it's
entirely
yes,
so.