►
From YouTube: IETF115-SATP-20221110-1300
Description
SATP meeting session at IETF115
2022/11/10 1300
https://datatracker.ietf.org/meeting/115/proceedings/
A
A
This
is
the
introduction
we're
going
to
go
through
a
few
things.
First,
of
course,
thank
you
to
Dennis,
who
agreed
to
take
notes.
I
would
love
it
if
somebody
else
would
help
him
out,
because
it's
always
a
hard
job.
The
the
note
taking
link
inside
the
agenda
makes
it
very
very
easy
to
take
notes,
so
please
help
them
out.
If
you're
able,
we'll
have
four
sort
of
major
sections
for
the
the
information
presentation
for
today,
there
will
be.
A
You
know
one
just
about
the
problem:
space
and
goals
by
Thomas
kind
of
describe
the
use
case
with
supply
chain
trade
and
then
there'll
be
another
use
case
with
regulated
Finance
sort
of
transactions
that
the
proponents
believe.
This
was
a
good
use
case
for
and
then
we'll
have
a
status
report
on
sort
of
what's
happening
if
we
had
a
buff
at
our
ETF
114,
and
the
proponents
took
a
lot
of
the
advice
that
came
from
the
room
found
it
very
helpful,
they've
been
meeting
frequently
talking
frequently
and
I've
talked
to
them
as
well.
A
and
then
finally,
there's
a
lot
of
time
for
open
discussion
for
the
problem
space
and
proposed
starting
Solutions
I
do
want
to
hold
the
questions
to
sort
of
clarifying
questions
first
and
then
we'll
actually
have
the
open
discussion
about.
You
know
the
typical
questions
of.
Is
this
a
the
what's
the
right
outcome
for
the
spot,
but
so
for
the
the
middle
part.
Just
please
ask
clarifying
questions
which
I'll
reiterate
as
time
goes
on
I'm
sure.
Most
of
you
are
well
familiar
with
the
note.
A
Well,
since
this
is
now
midday
Thursday
and
we're
not
too
far
from
the
end
of
the
conference.
If
you
don't
know
it
anything,
you
contribute
will
be
used
and
recorded
and
that's
part
of
the
record,
so
there's
a
list
of
BCPS
that
will
help.
You
understand
things
that
include
IPS
and
working
group
processes
and
copyrights
and
all
that
type
of
stuff
Please,
be
aware
of
it.
A
Remember
that
we
do
have
a
code
of
conduct
guidelines
that
includes
wearing
a
mask,
we'll
see
what
happens
at
the
end
of
this
meeting.
But
during
this
meeting
we
have
all
agreed
by
consensus
to
wear
masks.
So
please
everybody
do
wear
a
mask
the
entire
time.
You
can
see
that
I'm
speaking
even
with
it.
You
can
probably
hear
me
pretty
well,
so
you
can
actually
use
one
at
a
microphone,
but
you're
you're
allowed
to
take
it
off
at
a
microphone.
A
B
Can
people
hear
me
okay,
get
a
little
closer,
a
little
closer,
okay,
okay
and
Waze
is
going
to
yeah
run
this
place.
Thank
you
yeah,
so
yeah,
so
sorry
yeah.
So
so
some
of
you
who
attended
the
Philly
ietf
might
recognize
a
couple
of
these
slides
I'm
just
going
to
go
through
and
recap
some
of
the
presentations
and
the
inputs,
the
good
inputs
we
received
in
Philadelphia-
and
you
know,
we've
Incorporated
those
inputs
pretty
much
every
one
of
them
and
made
us.
B
We
rethink
rewrite
some
of
the
slides,
because
I
think
some
of
the
some
of
the
words
were
not
conveying
what
you
know
what
we
meant
to
say
next
slide,
please
so
so
this
is
probably
you
know.
I
put
this
in
just
to
clarify
that
that
you
know
we
have
a
growing
number
of
digital
asset
networks.
They
all
run
on
the
internet.
Many
of
the
next
generation
of
developers
may
not
even
realize
that
many
of
the
you
know
features
and
protocols
actually
run
over
iitf
rfcs
right
ifs
standards.
B
So
this
this
fact
that
that
this
is
just
a
large
growth
and
there's
a
lot
of
proprietary
stuff
out.
There
introduces
a
number
of
interesting
security
challenges
and,
of
course,
data
privacy
challenges,
and
you
know
my
solution
to
this
as
well.
You
know
go
back
to
the
ITF
and
you
know
do
what
the
ITF
is
is
best
doing,
namely
producing
written
technical
standards
and
and
and
goal
is
for
interoperability
of
these
Networks
next
slide
am
I,
am
I
speaking
loud
enough
for
people
online.
B
So
so,
when
we
say
Network
interoperability,
the
goal
really
is
to
move
across
a
digital
asset
and
there's
a
definition
of
a
digital
asset
is
across
from
one
network
to
another,
satisfying
certain
conditions
such
as
you
know,
atomicity,
consistency
and
so
on
today.
What
we're
seeing
in
the
industry
is
that
people
are
doing
just
bilateral
agreements
and
to
me
this,
this
is
like
the
mid
90s.
When
we
have
multiple
isps
and
isps,
you
should
do
you
know
bilateral
peering
agreements,
it's
almost
like
that
today,
and
there
are
a
number
of
other.
B
You
know
proprietary
transfer
protocols
that
they're
being
used,
but
again
it's
only
for
bilateral
connections.
So
really,
this
is
kind
of
a
lack
of
written
standards
that
financial
institutions
and
other
organizations
can
point
to
and
mandate
compliance
of
implementations.
The
lack
of
those
standards
are
inhibiting
the
industry
it's
like.
Well,
you
know
we
don't
want
to
enter
Because.
If
something
goes
wrong,
we
lose
money,
we
lose
assets.
You
know,
let's
just
wait
and
I
think
this
is
a
great
opportunity
for
the
ITF
as
a
Community
to
actually
contribute.
B
You
know
to
the
world,
particularly
in
in
the
context
of
interop
next
slide
next,
so
this
is
from
Philly
Philadelphia,
so
so,
where
there's
so
many
types
and
kinds
of
digital
assets,
so
we're
using
the
term
value-bearing
data
objects.
So
you
have
a
value-bearing
data,
object
in
one
network
and
you're
trying
to
move
it
across
to
another
Network,
so
that
entails
recreating
exact
copy
on
another
Network
and
deleting
or
invalidating
the
original
copy.
B
This
is
your
your
database
textbook,
you
know,
and
freshman
or
Junior
textbook,
and
the
transfer
must
be
verifiable
by
an
authorized
third
party,
because
it's
got
to
be
disputes
and
you
need
to
be
able
to
bring
in
an
auditor
or
something
like
that
that
can
replay
and
look
at
the
traces
of
the
transfer
and
say
yes,
you
know
did
not
happen,
A
Fault
happen
or
a
bug
happened,
or
somebody
was
dishonest
right.
So
that's
that's
a
key
requirement.
B
Next
one,
and
so
this
is
the
proposed
scope-
is
the
yellow
area
in
the
middle.
So
people
ask
me
all
the
time
what
about
the
networks?
Well,
there's
just
too
many
types
of
networks
and
new
ones
are
coming
up
all
the
time
and
to
me
this
is
if
you're
old
enough.
This
is
the
Lan
problem,
if
you
guys
remember
IBM,
sna
networks
and
decnet
and
ethernet,
you
know
who
won
that
battle
ethernet,
but
this
is
also
exactly
like
interdomain
and
intra
domain
routing
right.
B
So
what
we're
saying
in
in
our
language
the
yellow
bit
is,
is
the
bgp
protocol
right?
Are
we
going
to
Define
what
a
router
is?
The
Gateway
is
not
we're,
not
we're
not
going
to
build
gateways;
we
don't
know
how
to
build
gateways.
That's
that's.
You
know,
implement
this
problem.
What
we're
going
to
do
is
Define
the
interaction
between
the
gateways
and
that's
that's
the
yellow,
bit
and
I
think
this
is.
B
This
is
the
best
thing
for
the
industry
as
a
whole,
because
the
industry
itself
is
still
emerging
and
evolving
in
new
types
of
networks
are
being
created
right.
So
we
don't
want
to
touch
that.
Let
the
market
decide
which
type
of
network
wins
wins
the
day.
That's
fine,
that's
fine
with
me,
but
we
want
to
ensure
interoperability
across
these
types
of
of
Networks.
B
Thank
you
so
so
this
is,
you
know
just
some
some
basic.
You
know
assumptions
for
this:
the
yellow
bit
to
run,
which
is
both
that
we
need
to
share
some
common
understanding,
semantics
of
the
data
objects
and
and
validity.
What
does
valid
mean
all
right?
So
so
that's
we
assume
this
exists,
because
we
can't
impose
this
on
the
networks
for
our
own
design
print
principles.
We
assume
the
networks
are
opaque,
so
you
might
hear
the
word.
Permissioned
access,
controlled,
private.
B
Those
are
the
terms
that
sometimes
they
use
I
like
to
use
the
word
opaque
because
it
refers
to
the
data
objects
inside
those
networks.
We
can't
we're
designing
for
the
worst
case
scenario,
when
both
sides
are
opaque.
It's
nice,
if
both
sides
are
open
and
so
on,
but
and
I
when
I
explain
this
to
people,
I,
sometimes
point
to
just
my
own
ISP.
You
know:
cable,
Last,
Mile
provider
in
Boston,
I
can't
see
into
my
ISP
writing
table
right
unless
I
they
give
me
authorization
so
so,
and
I
have
to
explain
to
the
the
undergrads.
B
The
internet
is
is
not
the
one
long
contiguous
public
network,
the
internet
is
a
sequence
of
private
isps
in
this
backbone,
fiber
everywhere
right.
So
you
guys
know.
I'd
have
to
tell
you
guys
this,
but
we're
trying
to
use
the
same
design
principles
that
that
got
us
the
internet
and
the
scalable
scalable
internet
for
sappy.
B
The
third
one
is
gateways
are
assumed
to
be
trusted.
We
have
it.
We
have
to
assume
that
the
trusted
to
even
begin
this
protocol
in
reality,
what's
going
to
happen,
is
Gateway
operators
need
to
take
on
liability
if
they
screw
up
the
thing
crashes
if
they
lose
an
asset
they
have
to
sign
on.
This
is
to
me
this
is
exactly
like
a
CA,
a
PKA
CA
issuing
certs
right
if
they
screw
up.
B
If
you
know
people
here
from
various
and
if
you
screw
up,
you
lose
somebody's
public
key,
there's
a
big
contract
there
that
says
you're
you're
liable
to
a
million
dollars.
10
million
dollars
right
all
that
stuff,
yes,
I,
I
wrote,
I,
wrote
I
was
co-author
of
the
verisign
100
page
certificate
practices
statement.
It's
now
down
to
the
Cincinnati
about
it.
It's
like
80
pages,
and
this
warranties
there's
lines
there
and
in
fact
those.
B
That's
not
that's,
not
the
one
Rat
Hole
here,
that's
not
the
strike,
so
the
other
thing
that's
needed
is
that
there
are
cases
where
you
don't
want
to
move
an
asset.
You
just
want
to
extract
information
out,
so
we've
been
using
the
term
views,
so
somebody
from
outside
might
say:
Hey.
You
know
it
could
be
an
auditor,
it
could
be.
The
government
say
we
want
to
see
this
particular
transaction
on
this
date.
On
this
particular
record,
can
you
extract
it
out?
There's
also
the
function
of
a
grade,
so
it's
not
moving
value.
B
It's
just
extracting
information
out.
We
some
we
call
it.
Data
sharing
or
you
know,
view
sharing
it's
kind
of
the
same
thing
next,
one
yeah.
So
this
is
motherboard
apple
pie,
so
we're
just
we're
just
copying
the
architecture
of
basically
intra
domain
inter
domain,
and
the
last
bullet
is
the
important
thing
that
that,
for
scalability
reasons,
the
Gateway
needs
to
hide
whatever's
behind
it.
Just
like
a
bgp
router
hides
the
fact
that
you
know
a
domain
might
be
running.
You
know
ospf
rip,
one
Ripley,
we
don't
see
it
from
the
outside.
B
So
so
this
is
this
set
of
documents
that
we
propose
as
the
starting
point.
A
lot
of
them
are
version
zero.
A
lot
of
them
have
holes.
It
is
I
like
the
the
the
SAT
core
has
like
blank
sections
with
the
word
TBD
in
it,
which
is,
which
is
fine,
we're
sort
of
taking
our
time
there
and
encourage
you
to
read
the
architecture
document,
because
it
explains
some
of
the
design
principles
that
we're
stealing
from
the
design
of
of
the
internet
architecture.
A
B
Right
so
so
it
needs
to
be
trusted
to
function
as
a
implementer
of
the
SAT
B
protocol.
All
right,
so
it
needs
to
have
keys
it's
going
to
be
signing
stuff
right,
so
we
have
no,
it's
not
in
our
power
to
say
to
an
implementer.
This
is
how
you're
going
to
manage
your
your
private
Keys
you're,
going
to
use
a
TPM
chip
or
you're
going
to
use
whatever
an
sgx
capable
Hardware,
but
we
have
to
assume
that
both
sides-
Gateway
G1
and
G2-
have
done
enough
exchange
of
information
before
the
flow
occurs.
B
That
both
sides
are
understanding
that
so
make
it
even
simpler
gateways
have
say:
certificates,
x509,
inserts,
The,
Exchange
certs.
They
understand
the
search
belong
to
Legal
organizations.
You
know
financial
institutions
branded
financial
institutions
and
they
agree
to
proceed
now
so
so
when
we
say
trust
trust
to
be
running
the
protocol
correctly,
how
how
they
manage
keys,
for
example,.
A
A
We
can't
say
so:
can
you
drop
a
hint
about
I?
Think
it's
in
your
later
slides,
but
one
of
your
assumptions
has
to
do
with
you
expect
some
sort
of
agreement
between
gateways
or
between
networks.
It's
just
like
there's
agreements
today
between
routers
right.
All
bgp
sessions
usually
have
some
sort
of
mou
or
something
at
least
right.
B
So
so
we've
grouped
the
assumptions
into
like
three
groupings.
The
first
one
is
the
identity
of
the
the
Alice
Bob,
the
originator
and
the
receiver
and
the
Gateway
operator.
So
there's
four
four
people
four
entities
there.
We
assume
that
the
identities
have
been
resolved,
that
they
know
each
other.
The
Gateway
providers
know
each
other,
that's
grouping
one
grouping
two,
they
understand
the
type
and
kind
of
the
digital
asset
being
moved,
or
so
they
understand.
What's
what's
happening,
it's
not
that
K.
There's
this
blob
coming
across.
We
don't
know
what
that
means.
B
So
that's
that's
assumed
and
then
the
third
one
is
there's
some
legal
framework
to
guide
them
just
the
way,
and
it's
a
it's
a
bad
analogy:
the
CPS
statement
for
X5
owner
it's
a
legal
document
today
in
many
trading
networks
in
the
financial
institution
they
have
agreements
and
contracts
between
networks,
so
we
assume
I
mean
they
exist,
but
not
in
this
space.
But
for
this
new
type
of
digital
asset,
new
type
of
you
know
Network.
We
assume
they
also
exist.
B
D
B
So
the
idea
of
audit
here
is
that
if
there
is
some
kind
of
suspicious
activity
occurring
right,
this
is
what
happens
today
in
the
real
world.
Okay,
if
if
there
is
a
unusual
transaction
patterns
across,
say
the
Swift
network
and
huge
amounts
of
money
moving
suddenly
and
there's
a
there's,
a
the
trends,
look
there's
an
anomaly
there.
The
the
end
FBI
I,
think
is
whoever
issues
SARS
warrants.
B
They
can
issue
a
war
and
say
we
want
to
look
at
that
particular
transaction
right,
so
so
they
must
be
able
to
come
in
and
the
Gateway
providers
and
network
members
might
be
able
and
Alice
and
Bob.
They
have
to
be
able
to
say
here's
the
transaction
that
occurred
on
November
1st
at
1am.
This
is
what
you
know.
This
is
what
it
is.
So
then
they
need
to
be
able
to
look
at
that.
D
B
D
A
No,
but
one
of
the
the
things
that
I
think
came
out
in
this
box
presentation
was
that
list
of
assumptions
that
was
missing
last
time
and
actually
a
lot
of
it
came
from
your
comment
last
time,
Eric
because
you
said
that
you
weren't
able
to
evaluate
the
nature
of
the
security
of
the
transaction
without
the
Assumption.
So
one
of
the
assumptions
here
is,
if
you're
going
to
transfer
something
from
one
digital
Bank
to
another
digital
Bank.
How
do
you
guarantee
in
the
future?
D
No
like
like
no
it's
like
it's
an
important
Point
like
like
it's
one
thing
to
be
the
case.
That
bank
a
says
you
made
a
transfer
and
Bank
B,
says
I,
didn't
and
that's
a
question
of
like
non-appediation
and
then
another
thing
entirely.
Is
some
law
enforcement
agencies
able
to
go
and
monitor
every
transaction?
There's
like
entirely
different,
contextual
Concepts,
so
which
one
is
it,
which
was
the
requirement.
B
D
But
yeah,
okay,
when
I
ask
you
this
question,
what
you
said
was
the
requirement
was
that
if
the
suspicious
activity
that
they
can
go
and
like
look
at
the
transaction
stream
and,
like
you
said
to
too
many
transactions
like
that's
a
surveillance
property,
it's
not
a
it's!
Not
it's
not
an
important,
not
a
contract
enforcement
property
and
so
like
and
like
am
I
supposed
to
be
very
concrete.
D
B
Yeah,
so
so
our
focus
is
the
protocol
right.
So
how
do
players
play
in
the
real
world
whether
they
misbehave
they
try
to
do
something
illegal,
that's
out
of
scope
for
us
when,
when
an
Engineering
Group?
That's
why
I
can't
answer
it,
because
this
is
a
this?
Is
a
world
problem
in,
like
the
word
surveillance,
I,
can't
answer.
D
B
Authorized
if
you
look
at
the
census
authorized
entities
so
so
both
sides
needs
to
agree
that
you
know
the
government
is
allowed
to
come
in
right.
This
is
by
contract.
So
if
one
side
says
oh
I,
don't
I
don't
want
the
government
to
come
in.
Well,
the
other
side
might
say:
I
don't
want
to
do
any
transactions
with
you.
That's
that's
how
it
works
today
and
almost
you
know,
tangential
to
what's
happening.
F
G
You
quickly
answer
that
Dennis,
maybe
having
two
sides
in
what
you
said.
The
first
is
the
consistency
of
the
protocol
itself,
so
if,
if
we
spot
specific
assets
with
specific
identities,
Dennis
Dennis,
avilians
I'm,
the
note
taker
somebody
take
notes,
please,
if
we
are
addressing
the
problem
of
having
double
spending
and
spotting
the
double
spending,
because
we
know
the
the
identity
of
the
assets.
This
is
part
of
what
we're
doing
now,
having
a
malicious
attempt
of
a
bank
spending
twice
deliberately,
because
of
any
reason.
This
is
outside
the
scope.
A
So,
let's
get
back
to
the
queue
and
and
Eric
if
you
want
to
get
in
the
queue
again,
let's
do
it
that
way,
but
I
think
your
point
is
yeah.
Your
point
is
valid.
That
there
are
different
types
of
verifiability
and
in
the
legal
construct
would
need
to
specify
what
type
it
might
be,
whether
it's
always
open
or
whether
it's
auditable
per
record.
E
Rick
Rick
from
Ryan
oppa
Fortress,
would
you
explain
is
that
you
assume
all
sorts
of
things
from
the
networks
that
you
that
you
connect?
You
want
them
to
Gateway
I
could
imagine
the
gateways
just
be
be
connected
directly
and
do
nothing
in
between
what
I
haven't
gotten
clear
is
what
sort
of
Stadium
what
store?
What
procedures
are
you
trying
to
establish?
What
sort
of
certainty
do
you
want
to
give?
What
what
are
the
problems?
You
are
solving?
That's
not
completely
clear.
B
B
This
is
the
acid
properties
that
that
safety
and
an
acid
that's
kind
of
what
we're
shooting
for
just
just
for
that,
and
that's
why
we
have
to
assume
away
the
all
the
setup
and
security
under
three
groupings
that
that
is
what
we
call
state:
zero
or
phase
zero.
All
of
that
needs
to
have
have
happened
before
we
can
even
run
run.
The
program.
I
hope
that
you
know
it's
primarily
double
spending
and
crash
recovery
is,
is
kind
of
you
know
high
on
our
goal.
B
B
H
Hey
nothing,
so
this
is
a
comment
on
Eric's
point
that
I
guess.
The
thing
I
want
to
say
is
that
it
should
come
as
no
surprise
that
financial
markets
or
financial
industry
is
more
regulated
than
other
types
of
exchanges.
So
it
probably
is
not.
It
wouldn't
be
surprising
to
build
a
protocol
that
contains
much
more
of
surveillance,
even
though
those
properties
might
be
even
be
undesirable
internally
to
some
of
these
networks.
H
That
may
not
be
properties
that
you
can
maintain
when
you
transfer
between
networks-
and
this
also
made
me
think
about
sort
of
the
reason.
I
came
to
take
a
look
at
this
I'm,
a
co-chair
of
the
Tigers
working
group,
and
if
you
look
at
that,
that's
kind
of
also
one
of
these
smooth
stuff
between
wallets.
H
A
A
A
J
A
K
K
So
that
is
a
an
international
standard
which,
as
ISO
is,
is
the
right
venue
for
what
ISO
is
doing
is
at
the
top
level
in
terms
of
governance
in
terms
of
descriptive
standards
and
there's
already
quite
a
few
out
I
actually
chair
interoperability
at
within
ISO
in
working
group,
seven
and
what
we're
doing
is
creating
an
interoperability
framework
and-
and
that
basically
is
what
are
the
building
blocks
that
different
vendors
and
participants
need
to
create
in
order
to
create
standards
that
the
industry
can
adopt?
It's
not
prescriptive.
K
It's
not
saying
the
packet
should
be
64k.
It
should
over
check
some
of
this,
and
it
should
have
a
anak
handshake
of
that.
It's
not
prescriptive
in
that
sense,
and
that's
where
institutions
and
bodies
like
ietf
come
in
and
what
we
kind
of
expect
is
the
technical
engineering
to
be
done
within
engineering
organizations
like
IEEE
and
ITF
and
that
feeds
in
to
the
standard
that's
coming
out
of
iso.
K
Already
so
I'm
very
supportive
of
this
and
I'd
really
like
to
kind
of
bring
some
experience
working
in
banking
and
in
government
in
on
the
cyber
security
decide.
One
of
the
comments
made
is:
is
this
a
type
of
surveillance?
I
would
argue?
No,
it's
not
a
type
of
surveillance.
It's
it's
a
protocol
enabling
interoperability
between
different
types
of
networks
who
uses
that
network
has
to
sign
up
to
some
scheme
rules,
and
this
is
consistent
with
what
happens
in
card
networks,
payment
networks,
Swift
type,
banking
networks.
K
So
if
there's
an
issue,
it's
up
to
the
participant
to
respond
to
that
suspicious
activity
resolve
a
dispute
between
them
and
they're
all
in
writing
in
contracts
within
the
different
participants
within
the
scheme.
So
the
way
it
works
is
you're
not
relying
on
the
technical
protocol
to
solve
your
problem,
you're,
relying
on
human
interaction
and
contractual
rules
to
solve
the
problem.
So
this
protocol
is
not
here
to
solve
that
issue.
It's
here
to
solve
the
technical
interconnectivity
between
these
different
networks,
and
one
final
thing
to
say,
is
digital
assets.
K
K
Now
these
digital
assets
are
here
today
and
they're
not
going
away.
What
we've
seen
from
working
with
central
banks
and
other
banks
that
at
the
company
are
working
at
Quant,
is
there's
a
16
trillion
dollar
opportunity
with
digital
assets
by
2030.,
so
people
are
digitizing
what
they've
some
some
form
of
asset
that
used
to
be
in
in
a
different
form,
into
a
digital
form
and
that's
not
going
away,
but
the
problem
which
I
saw
from
2015
2016
was
people
are
making
these
in
isolation
and
people
are
making
these
in
silos.
K
I
Hi
Roman
denilio,
Carnegie,
Mellon
University,
no
hat
I
couldn't
get
my
laptop
to
connect.
So
Paul
put
me
in
the
queue
so
back
to
helping
tease
out
this
double
kind
of
spend
problem.
We
use
lots
of
different
words,
so
I
want
to
try
my
attempt
at
pulling
pulling
this
apart.
The
question
is:
when
we
talk
about.
How
is
the
double
spending
problem
kind
of
resolved
to
me
is:
it
is
all
of
that
in
scope
or
some
of
it
at
a
scope.
I
So
I
could
imagine
a
world
where
you
could
use
different
fancy
kind
of
cryptography,
different
kind
of
acknowledgments,
where
the
things
that
are
getting
exchanged,
that's
all
packed
in
scope
in
the
protocol
and
that's
how
you
do
the
how
you
address
the
double
spending
problem.
What
I
think
I
heard
you
say
is
that
there
are
part
of
that
verification
process
is
out
of
scope.
I
So
when
you
said
surveillance,
I
think
what
you
were
saying
correct
me:
if
wrong
is
that
the
gateways
have
to
produce
an
audit
log,
so
the
end-to-end
double
spend
verification
is
actually
not
a
protocol
and
not
a
product
of
the
protocol.
There
are
just
some
requirements
about
Telemetry
or
whatever
the
word
you
want
to
use
that
the
protocol
will
squirt
and
to
get
the
double
spend.
Verification
actually
is
something
that's
out
of
scope
of
the
problem
of.
B
The
protocol
and
of
the
working
group
is
that
accurate?
Yes,
so
we've
been
talking
about
two
classes
of
logs,
so
so
the
gateways
need
to
log
for
message
completeness
and
and
proof
that
both
sides
have
done
the
right
thing
per
the
RFC,
the
the
draft
definition
and
then
it
needs
to
allow
for
crash
recovery.
There
might
be
some
overlap
in
terms
of
the
same
data
log
to
be
reused
for
another,
but
we
talk
about
two
sets
but
you're
right,
Roman,
that's
kind
of
you're
right,
yeah,.
L
James,
let's
go
I'm
trying
to
think
whether
Rowan's
just
just
answered
my
question,
but
some
like
so
the
question
I
was
going
to
ask
is
so
is
the
Assumption
here?
Is
that
the
participants
in
a
given
Network
trust
the
Gateway
not
to
double,
spend
due
to
Legal
agreements
or
whatever
else
they
have
in
place
not
due
to
protocol
cryptographics
correct?
Is
that
correct,
correct.
B
A
Thank
you,
Colin.
That's
an
important
point
all
right
with
that
I
think
we'll
shift
to
use
cases.
Next,
there
are
two
again
we
will
ask
for
clarifying
after
each
one.
These
should
have
less
clarifying
questions.
So
really
just
clarifying
questions
about
the
use
case,
not
about
the
protocol
itself.
That
will
come
later.
A
M
The
global
trade
has
has
been
conducted
in
the
past
few
centuries,
and
several
processes
have
cropped
up
in
in
the
space
have
been
built
on
paper.
Naturally,
it's
not
a
computers
until
the
future,
so
these
paper-based
policies
are
quite
manual
and
they're
time
consuming
and
the
any
clearance
or
any
finalization
of
a
transaction.
It
takes
too
long
and
it
can
often
be
susceptible
to
fraud
by
different
participants.
And
there
are
a
lot
of
those
participants
in
you
know
in
a
particular
supply
chain.
M
You
see
that
in
the
subsequent
Slide,
the
shippers
or
the
sellers
and
the
buyers
or
you
can
think
of
them
as
importers
and
exporters
who
are
usually
always
in
different
countries,
they
face
a
lot
of
God
when
it
comes
to
fulfilling
and
international
export
import
transaction.
So
there's
a
lot
of
incentive
to
reduce
the
the
cost
make
this
more
efficient,
because,
finally,
that
will
result
in
benefits
to
all
of
us
as
entrance
the
the
way
trade
happens
in
this
analog
world.
M
Also,
given
that
you
have
a
lot
of
different
players
spread
across
different
countries
participating
in
different
institutions
that
have
their
own
privacy
rules.
Sharing
information
across
these
parties
and
across
institutions
is
a
very
challenging
task
and
digitization
is
something
that
is
like
solely
necessary
for
all
of
these
different
reasons:
an
excitement
so
one
I'll
zoom
into
one
particular
problem
that
is
very
Salient
whenever
any
two
parties
conduct
a
trade
and
imagine
an
exporter
and
importer
or
let's
call
them
your
buyer,
buyer
and
shipper.
M
If
think
of
the
problem,
that
it's
most
basic,
a
buyer
and
a
shipper
or
let
me
call
them
by
himself-
a
buyer
makes
a
deal
with
the
seller
that
if
a
seller
ships
sends
the
buyer
particular
set
of
goods,
then
the
buyer
is
obligated
to
make
a
particular
amount
of
payment.
Now
this
is
challenging
across
two
different
countries,
because
the
question
always
is
who
blinks?
M
First,
if
the
seller
shifts
the
goods
first,
what
guarantees
will
have
with
the
buyers,
go
to
make
a
payment
and
if
the
buyer
instead
makes
a
payment
first,
what
guarantees
does
it
get
that
the
seller
is
going
to
actually
ship
the
goods?
So
what's
happened
in
the
past
few
centuries,
really
a
processes
that
have
evolved
since
the
dawn
of
banking
to
manage
the
property?
M
M
So
the
bill
of
trading
is
a
very
crucial
document,
because
it
constitutes
proof
that
will
conclude
a
an
import,
export
transaction,
good
search.
It
shows
that
goods
are
shipped
and
it
it
obligates
somebody
to
make
a
payment,
so
the
financial
institutions
needs
this.
Bill
of
rating
is
collateral
by
the
way.
There's
a
lot
of
Finance
involved
in
in
this
bill
of
training
can
be
used
collateral.
It
can
be
sold
print
off
as
a
financial
account.
So
we
don't
have
to
worry
about
that.
M
M
Okay,
usually
in
the
the
way
they
had
seen
in
the
process
of
digitization
of
the
philosophy
in
the
past
few
years
is
that
the
financing
part
of
a
trade
is
carried
out
in
one
particular
Network,
whereas
the
shipping
part
and
the
generation
of
the
bills
of
documents
like
the
bills
of
plating
is
carried
out
in
a
different
network,
and
these
two
networks
are
a
big
feature.
They
are
oblivious
of
each
other
and
they
were
built
separately
for
particular
reasons
there
are.
M
M
The
way
we
can
bridge
to
networks
is
why
I
get
the
kind
that
Thomas
talked
about
and
the
and
these
networks
need
to
be
interoperable
for
all
the
reasons
that
Gilbert
so
adequently
put
in
his
comments
shortly.
So
this
bill
of
plating
data
must
be
reported
from
the
network
that
is
surrounding
the
take
Logistics
via
a
Gateway
that
acts
as
a
neutral
service
provider.
On
behalf
of
that
Logistics
Network,
that
is
maintaining
development.
Okay
next
slide,
please
all.
A
H
It
yes
I'm
a
clarifying
question
so
on
this
slide
to
me
it
it
kind
of
looks
like
the
the
what's
really
described.
There
is
sort
of
a
transparency
problem
that
is
kind
of
already
in
scope
for
the
for
the
skit
working
group,
so
it
it
so
what
I
would
like
to
understand
better
is
how
this
really
is
a
as
described
here.
How
this
really
is
a
transfer
of
of
an
asset.
To
me
it
sounds
like
it's
the
the
bill
of
lading
and
you
I.
H
B
Yeah,
so
so
let
me
answer
that
one
Rama
that
there
are
three
types
of
flows
that
we
have.
If
you
look
at
the
draft
Charter,
the
first
one
I
was
talking
about,
was
an
actual
asset
transfer,
and
this
one
is
what
we
call
asset.
Related
data
sharing
I've
been
talking
about
extracting
data.
Now
the
the
problem
is
that
the
other
the
bank
may
want
that
bill
of
lading.
Today,
you
actually
have
to
bring
a
printout.
You.
B
A
And
skit
yeah,
so
let
me
let
me
put
it
another
way
right.
If
we
can
reuse
parts
that
are
in
another,
you
know
ietf
area,
that's
exactly
what
we
always
do
and
the
question
to
me
is
whether
they
need
to
be
bound
together
or
whether
they
can
be
two
different
protocols
and
so
I'll
leave
that
to
the
future.
But.
I
N
Robot
from
you
why
I
think
the
key
thing
here
is
is
that
there
is
an
asset
on
at
least
one
side
of
the
Gateway
and
I.
Think
that's
that's!
The
key
thing
is:
is
that
the
you
know
the
control
of
that
asset,
whatever
that
asset
does,
which
is
the
the
financial
side
of
it,
requires
the
information
on
the
other
side.
So
it's
not
purely
transparency,
I
think
that
might
be
a
distinguishing
characteristic.
M
Sure,
just
to
conclude
that
that
point
for
transparency
I
mean
we
can
probably
go
into
this
later
in
the
presentation
or
during
the
discussion.
Think
of
you
have
two:
you
have
a
network
here
that
is
maintaining
a
document
like
the
bill
of
lading.
The
the
bill
of
reading
now
has
this.
The
information
in
this
bill
of
lading
is
quite
crucial
to
a
concluding
a
process
that
is
beyond
the
scope
of
that
particular
Network.
M
That
is
really
crucial
and,
in
effect,
what
we're
trying
to
do
is
to
figure
out
a
view
of
the
to
provide
a
view
into
the
will
of
leading
to
parties
that
do
not
actually
belong
to
this
network
that
were
not
parties
who
are
not
the
buyer,
not
the
seller,
not
the
financial
institution
back
in
the
buyer,
not
the
carrier
that
is
has
taken
possession
records
that
the
sellers
provided.
So
that's
what
this
example
tries
to
illustrate.
M
You
have
a
financial
institution
that
is
outside
the
this
supply
chain,
intake
logistic
promotion,
Network
and
that
needs
to
know
it
needs
to
have
you
into
what's
going
on
in
the
logistics
part
of
the
tree,
so
it
requests
for
the
bill
of
cleaning
and
it
now
it
can.
In
effect,
what
it's
trying
to
do
is
strength
request.
M
A
Eliminator
2.1
seconds
so
two
things
one:
we
probably
need
to
speed
up
a
little
bit,
I
think
the
slides
are
you're
over
describing
them
a
little
bit
we're
going
to
run
out
of
time
too,
for
the
people
at
the
Queue
at
the
moment.
Is
it
relevant
to
this
slide
or
the
whole
concept
if
it's
the
whole
concept,
let's
take
it
after
the
presentation.
I
I
B
F
M
No,
no,
that's!
Okay!
Okay,
let
me
say
so
you
can.
You
can
combine
this
in
two
different
ways:
you
have
what
is
what
you're
showing
here?
If
you
go
from
left
to
right,
there's
a
bill
of
reading
on
the
network
on
the
left,
which
is
being
exposed
through
gate
to
G1
and
there's
a
letter
of
credit
on
the
network
to
the
right
which
is
being
exposed
by
Gateway
G2.
What
we're
showing
in
this
use
case
is
a
composition
of
both
of
them
and
both
of
them.
D
Okay,
but
I
guess
what
I
guess
what
I'm
being
confused
by
is
like
if
we
have
an
assertion
of
this
bill
of
lading,
that
is,
if
I
understand
it,
being
transferred
from
the
seller
to
the
buyer
like
surely
we
don't
want
to
have
a
bearer
token,
which
we're
then
going
to
apply
to
the
bank?
Who
could
then
take
all
my
stuff
right?
That's
not
that's!
Not
a
desirable
outcome
right.
M
M
So
the
bill
of
leading
on
the
on
the
network
to
the
left,
it
is
actually
something
negotiated
between
the
seller
and
the
carrier,
the
carrier
so
the
when
the
seller,
ships,
goods
via
a
particular
carrier.
They
have
done
that
part
in
the
export
import,
so
the
carrier
is
now
supposed
to
provide
a
bill
of
reading
to
the
seller.
M
This
is
the
seller,
can
then
take
to
any
financial
institution
and
say:
look
not
any
financial
institution,
just
keep
the
bias
financial
institution
and
say:
look
I
ship,
the
goods,
here's
the
proof
you
got
the
bill
of
trading
now
give
make
my
payment.
That
cannot
happen
in
networks
which
are
signed
like
this,
which
are
isolated
like
this,
but
that's
what
we
live
in.
You
have
networks
which
only
run
a
store
plating.
You
have
other
networks
that
only
that
the
web
financial
institutions
have
a
play
part.
M
M
B
It's
it's!
It's
not
a
bear
token,
so
this
is
flow.
Number
two
is
extracting
information
to
show
to
the
financial
entity
that
this
tool
exists
on
the
network
and
so
Gateway
G1
needs
to
sign
this.
This
data,
it's
not
an
asset
itself.
It's
a
report,
it's
a
subset
of
the
fields!
Well,
that's
the
party
okay!
H
H
H
H
Not
in
a
problem
scope
I'm
not
trying
to
change
Hotel
I'm,
going
to
start
trying
to
understand
it,
because
what
my
argument
was
that
it
was
a
transparency
issue
that
it's
about
showing
that
you
have
stuff
on
the
ship
and
the
right
Stones
correct,
which
is
different
from
giving
it
to
the
bank.
You
now
own
it
and
that's
the
thing
all
right,
I'm
getting
pure
answers.
H
L
M
But
what
we're
talking
about
here
is
the
fact
that
you
have
the
financial
part
of
a
trade
of
an
export
import
running
on
one
particular
Network,
whose
activities
are
opaque
to
the
external
world,
and
then
you
have
the
shipment
of
the
logistics
part
of
it.
Turning
on
a
different
network
which
is
again
opaque
to
anybody
outside
of
it,
how
do
you
connect
the
two?
M
They
have
to
make
them
interoperable
the
way
we
make
them
profitable
or
allow
them
to
provide
youth
into
restricted
use
into
their
records,
which
in
this
case,
involves
those
of
trading
on
the
left
Network.
Let
us
read
it
on
the
right
side:
Network
you
do
that
using
these
Gateway,
you
can
communicate
using
a
neutral
protocol.
A
M
M
You
have
a
nautilus
or
document
that
you
see
here
and
if
you
have
networks
that
need
to
incorporate
for
providing
views
into
such
documents,
the
the
same
security
problem
will
arise
in
each
of
the
wheels
cases
and
if
you
go
to
the
next
slide,
you
have
a
lot
of
different
different
parties
that
are
involved
in
this
kind
of
ecosystem,
and
this
gets
very
complex
very
fast
unless
we
have
a
neutral
protocol
that
allows
two
different
networks
to
operate
interoperate.
Thank
you.
A
O
O
I'm
I'll
wait
for
the
slides
a
little
bit
but
effectively
I'll
start
talking
anyway.
So
the
Financial
Services
Market
is
moving
in
the
direction
of
tokenization,
which
is
a
bit
of
a
buzzword
but
effectively
means
digitally
representing
a
variety
of
assets
and
then
trading
them,
particularly
things
that
couldn't
be
traded
before.
O
So
there
are
assets
out
there
which
can
be
traded,
which
represent
things
like
future
Revenue
flows
right,
so
you
can
buy
a
fraction
of
the
income
that
you
will
get
when
this
thing
comes
to
Market,
there's
a
Supercar
manufacturer
that
does
this
that's
how
they
fund
the
build
of
their
cars.
Likewise,
real
estate
is
being
fractionalized,
lots
of
things
are
being
fractionized.
O
This
is
a
big
Trend
in
financial
services
and
these
new
digital
tokens
are
being
traded
on
a
new
class
of
trading
venues
effectively
marketplaces,
so
they
are
not
connected
they're
built
on
a
new
generation
of
Technology.
Some
of
it
is
distributed.
Ledger,
some
of
it
is
not,
but
the
main
point
about
it
is
that
they
are
mostly
not
being
connected
to
each
other
from
a
industry
point
of
view,
there's
quite
strong
Network
effects
in
this
kind
of
thing.
So
you
see
in
the
more
established
markets,
there's
quite
a
lot
of
big
trading
venues.
O
They
connect
to
each
other.
Likewise,
internet
lots
of
isps
connect
to
each
other.
Whether
things
are
proprietary,
you
have
marketplaces
like
eBay,
Alibaba
and
literally
nothing
else
right,
so
Network
effects
can
play
in
your
favor
or
not.
We've
got
slight
awesome,
so
this
is
this
Walled
Garden
problem
that
we
are
looking
at
at
the
moment,
because
there
are
dozens
to
in
larger
economies,
low
digit,
hundreds
of
new
venues
coming
up
and
the
way
I
like
to
think
of
them.
They
are
like
marketplace
right.
O
So
if
you've
created
one
of
these
tokens-
and
you
want
to
sell
it-
it's
analogous
to
something
like
eBay
right.
So
we
were
talking
about
this
before
and
it's
effectively.
If
I
want
to
sell
a
GPU
right
and
I
want
to
sell
it
on
eBay,
then
that's
fine.
Anyone
on
eBay
can
buy
it,
but
if
I've
got
a
buyer
who
is
on
Alibaba
or
something
like
that,
how
do
I,
let
them
see
it.
O
What
we
are
trying
to
do
is
basically
be
able
to
show
it
on
the
other
venue
be
able
to
people
to
buy
it
on
the
other
venue.
The
digital
representation
travels
transparently
from
one
to
the
other,
fulfillment
happens,
and
the
reason
why
we're
trying
to
do
that
is
because
there's
not
just
eBay
and
Alibaba,
but
there's
a
lot
of
these
things.
O
So
as
a
someone
who's
issuing
this
token,
who
wants
to
sell
it
I
want
as
many
people
as
possible
to
be
able
to
buy
it,
and
those
people
are
not
all
going
to
be
customers
of
the
one
or
two
places
that
I'm
a
customer
of.
So
we
want
to
link
all
of
these
things
to
create
a
larger
market
effect,
I
think
so.
Yes,
fragmented
markets,
poor
issue
issue
as
investors
here
are
the
buyers
and
the
sellers.
O
Limited
liquidity
potential
means
there's
less
people
to
buy
it
and
in
the
case
of
these
sorts
of
assets,
that
kind
of
creates
a
lot
of
inefficiency,
which
usually
means
somebody
can
buy
it
for
cheap
on
this
one,
and
they
can
sell
it
for
expensive
on
that.
One
and
all
they're
doing
is
making
a
bit
of
profit
out
of
the
Arbitrage
and
not
adding
any
value,
and
it
isn't
a
very
efficient
market,
next
slide
cool.
O
So
what
we
are
trying
to
do
to
do
this
is
to
connect
all
of
these
networks
right
so
as
they
are
they're
coming
up
at
the
moment.
Some
of
them,
like
Thomas,
said
there
are
some
bilateral
Solutions.
There
are
some
vendor
and
proprietary
driven
Solutions,
and
none
of
them
really
have
any
attempt
at
standardization
really
other
than
we
will
try
and
get
as
many
users
as
possible
and
be
the
next
eBay.
O
O
Excellent,
so
this
is
in
terms
of
what
are
the
things
that
we're
talking
about.
There's
a
few
things
that
institutional
tokenization
is
what
I
was
talking
about.
They
have
a
lot
of
their
own
jargon
in
this
world,
so
issuance
of
interoperable
digital
assets.
I
would
like
to
sell
my
GPU
on
eBay,
and
here
is
the
data
item
that
you
will
buy
and
which
will
give
you
the
right
to
it.
That
includes
things
like
physical
things,
like
I,
said:
they're
real
estate
gold.
D
So
you
mentioned
earlier
that
some
of
these
assets
are
on
you
know,
blockchain
type
networks
so
is
in
order
for
that
to
work
so
because
of
the
structures
I
understand
it.
What
that
would
mean
is
that
the
smart
contracts
associated
with
those
blockchains
would
have
to
require
that
we
have
to
basically
allow
the
gateway
to
override
ownership.
O
What
the
protocol
needs
is
for
the
gateway
to
have
some
form
of
trust
of
the
network
and
it
when
that's
an
implementer
problem
right.
So
if
it's
a
smart
contract
that
provides
a
guarantee
to
it,
that's
fine!
If
it's
something
else,
that's
fine
as
well.
So
the
trust
relationship
that
we
were
talking
about
earlier,
the
Gateway
needs
to
trust
the
network
and
there
will
be
a
liability
relationship
between
them
and
the
gateways
trust
each
other
with
this
separate
liability
relationship.
So
there's
right,
but
it's.
D
O
Excellent
cool,
so
we
will
come
on
to.
We
talked
about
the
size
of
the
market.
Well,
I
didn't,
but
the
slide
did,
which
is
in
the
multi-millions
low
billions
today
expected
to
rise
to
trillions.
So
some
of
the
players
in
this
area
tend
to
be
the
large
Global
Banks
and
which
one
was
it
Gilbert.
One,
though,
starts
with
an
S.
They
have
got
a
about
300
billion,
something
like
that.
Their
plan
is
in
the
next
two
years
to
tokenize
all
of
the
assets
that
they
hold.
So
security
is
everything.
O
O
For
those
that
don't
know,
Swift
is
the
Network.
It's
an
organization,
that's
shared
ownership
organization,
so
all
the
banks
in
the
world
own
it
it's
got
about
22
000
owners.
They
run
across
four
highly
redundant
mpls
networks.
They
back
off
to
a
whole
bunch
of
different
providers.
They've
got
a
lot
of
data
centers
and
they
provide
highly
resilient
messaging
services
for
the
whole
financial
services
industry.
Basically
they're
also
interested
in
this
right,
so
the
hackathon
had
two
things.
One
was
about
custody,
didn't
really
care
about
that.
O
The
other
one
was
about
interoperability
between
networks.
So
we
do
care
about
that.
That's
kind
of
what
we
do.
They
also
have
been
running
their
own
internal
thing
on
this,
so
they
had
their
own
announcement.
They've
been
doing
interoperability
between
two
different
networks
using
Swift's
infrastructure,
they've
spent
10
months
on
that,
and
that
is
their
proprietary
one
built
around
their
message
set
for
this
one.
It
was
more
open,
so
we,
the
the
comp
piece
on
there
is
labeled,
Gateway
and
sapi
is
how
we
connected
these
things,
and
this
was
this.
O
One
was
two
different
networks
representing
effectively
a
source
and
a
destination.
The
use
case
was
around.
It
was
tokimo's
gold,
there's
a
folks
that
are
sitting
on
vaults,
full
of
gold
who
think
we
can
tokenize
this
and
we
can
use
it
as
a
currency,
and
they
are
trying
to
do
that.
So
this
is
what
they're
trying
to
do
in
in
that
case,
there's
about
120
participants
in
it.
Bny
melon
won
it
in
the
end
because
they
throw
a
lot
of
money
at
it.
We
would
join
third
with
NatWest
on
that
one.
O
So
there's
a
lot
of
UK
Banks,
there's
a
lot
of
interest
in
this
area
so,
from
a
use
case,
point
of
view,
I
guess.
The
message
really
here
is
that
the
industry
is
doing
this
and
they're
doing
it
at
scale
and
they're
doing
it
in
a
fairly
proprietary
format.
There
are
other
standards
groups
out
there
that
are
not
doing
the
protocol
side.
So,
as
Gilbert
said,
ISO
are
doing
terminology
framework.
O
That
sort
of
thing
there's
tons
of
projects
that
are
think
their
way
is
the
only
way
right
so
particularly
in
the
blockchain
world,
each
of
the
camps
think
interoperability
is
everybody,
uses
their
system
and
we
were
at
cybus
last
week,
which
is
the
Swift
conference,
which
is
roughly
the
size
of
RSA.
We
spoke
to
a
lot
of
people
there.
O
They
are
mostly
holding
off
because
there
is
no
interoperability.
People
are
doing
a
lot
of
Pilots
a
lot
of
pocs
and
they're
waiting
to
see
if
there's
a
winner
or
if
there's
a
standard
piece
of
interoperability.
So
there's
a
few
people
going
live
q1
at
the
moment,
but
this
is
kind
of
this
is
what
everyone's
waiting
for
and
it's
the
technical
standard
that
we
need.
B
So
we've
taken
a
lot
of
input
in
since
the
last
ITF
in
Philly
and
and
we've
been
having
basically
weekly
meetings
on
this.
So
progress
we've
been
documenting
the
assumptions
because,
in
fact
that
was
the
the
big
question
I
think
coming
out
of
field
is:
what's
the
Assumption
and
we've
been
saying?
Well,
you
know
what
we
call
phase
phase
zero.
What
needs
to
happen
before
this
flow
runs
so
so
been
documenting
that
we've
been
collecting
use
cases.
We
have
a
use
cases
document.
B
In
fact,
you
know:
we've
got
the
digital
dollar
project
GDP
in
Washington,
DC,
that's
going
to
contribute
text
to
our
use
case
document.
I
asked
them
to
give
me
three
paragraphs
and
they
gave
me
four
pages,
so
my
job
is
to
sit
down
with
them
and
and
sort
out
through
four
pages.
You
just
need
three
paragraphs.
We've
been
working
on
a
the
common
message
flow
for
the
asset
transfer.
So
for
those
who
are
looking
at
our
GitHub,
there's
a
figures
folder
and
there's
a
there's,
a
flow
diagram.
B
It's
I
think
where
the
file
it's
like
version
15..
So
so
we've
been
talking
about
this,
probably
since
before
Philly,
but
since
Philly
we've
been
reading
meeting
weekly
and
as
that
particular
message
flow
is,
is
sort
of
the
the
sort
of
Highly
discussed,
subject
or
topic
of
of
our
meeting.
We
also
have
a
draft
Charter
that
that
is
is
up
available
on
our
GitHub
repo
next
next
slide.
Please
thanks!
B
So
so
we
finally
got
agreement
among
ourselves,
okay,
internal
consensus
that-
and
you
could
just
follow
that-
that
link
there.
We
want
to
focus
on
two
key
flows.
The
first
flow
is
that
the
asset
transfer
flow
and
the
second
one
is
the
data
sharing
flow,
more
completely
asset,
related
data,
sharing,
what
I
call
data
extraction
and
we
have
an
architecture
document
and
a
use
cases
document
as
well.
I
know
it's
ambitious.
We
we're
shooting
for
18
months
but
I've
been
told.
B
B
So,
prior
to
Philadelphia
we
would
be
meeting
twice
a
month.
So
it's
a
it
was
a
biology.
Bi-Weekly
meeting,
but
since
fairly
we've
been
doing,
weekly
and
personally
I
was
I
was
surprised.
I
was
surprised,
people
were
willing
to
meet.
You
know
every
week
every
Tuesday
morning
you
know
and
some
people
some
people
like.
Thank
you
Rama
people
in
in
India.
They
were
on
like
8
P.M,
you
know
and
then
and
there's
three
guys
in
Australia
and
it's
like
11
p.m.
B
You
know
in
Australia,
so
they've
been
joining
and
I
I
kind
of
feel
bad
for
the
Australia
crew
and
the
and
the
India
side,
but
but
this
is
kind
of
the
rough
stats
and
and
it's
just
it's
just
Excel
and
I
just
cut
and
paste
Excel.
But
on
average
it's
it's
about
10
attendees,
which
is
you
know,
I
thought
about
this.
Pretty
good
right,
you'll
see
some
dips
there
at
the
end
of
August.
B
You
know
when
you
know
had
like
five
people,
and
that
was
because
it's
vacation
season
last
last
chance
for
a
vacation
before
before
September
it's
open
to
all
the
zoom
is
recorded.
So
you
can,
you
know,
go
back,
you
know
to
this
and
the
links
on
the
mailing
list
I
can
go
back
and
listen
to
the
zoom
recording
next
slide.
Please.
So
so
people
ask
well
how
come
you
don't
have
like?
You
know:
gazillion
mail
items,
I,
guess
you
know
it's
because
we
we
use
the
zoom
session.
B
We
talk
about
stuff,
you
know
and
that's
why
we're
recorded
I
know
it's
not
sort
of
ITF
tradition
to
be
the
recording,
Zoom
I
personally.
Think
it's
for
me,
going
over
start
having
the
zoom
recording
has
been
great,
I
mean
sometimes
I
have
to
like
put
in
the
mid-ins
or
fix
fix
the
notes
and
having
the
zoom
recording
kind
of
helps.
You
know
remember
what
what
I
said.
What
other
people
said
next
slide,
please
so
the
set
of
assumptions
we
have
a
thread
there
on
the
mailing
list.
B
You
can
have
a
look
at
that.
We've
been
collecting
this
and
this
is
I
said
earlier.
There
was
a
question
earlier
about
the
assumptions
we've
grouped
them
into
three
categories:
all
of
them
are
are
out
of
scope,
but
but
we
know
it
needs
to
happen.
So
legal
framework
side,
what
is
the
asset
definition
what's
an
asset
and
in
different
networks
in
different
Industries?
It
might
be
different,
an
entity
I.
We
don't
do
identity
management
when
we
don't
want
to
do
anti-money
laundering
and
all
that
stuff.
B
So
I
also
get
this
question
a
lot
like
why
the
idea
well
I,
my
my
wine
liner,
is
that
it's
the
home
of
interoperability
right.
So
this
is
where
this
is
number
one
number
two.
You
know
technical
review
so
like
guys
like
Ecker
right,
so
you
know
they're
gonna,
they're,
gonna
scrutinize,
our
specs,
which
is
good
I,
would
prefer
to
have
our
drafts
to
be
scrutinized
locally.
B
Instead
of
having
you
know,
mistakes
in
the
draft
and
so
on-
and
you
know
we
did-
we
did
look
at
other
organizations,
so
we
looked
at
that
w3c
they're
primarily
concerned
right
now
about
the
verifiable
credentials,
the
VC
data
structures
and
the
did
so
so
we
kind
of
said:
let's,
let's
try
it
here.
First
ISO,
as
you
heard
from
Gilbert,
is,
is
a
pretty
high
level
organization.
They
don't
really
do
protocol
specs
right,
so
so
I
did
probably
to
explain.
B
There's
a
lot
of
Open
Source
projects
out
there,
but
I
have
to
say
to
many
of
those
open
source
projects.
Running
code
is
not
spec
right.
If
you
go
to
organizations
like
Swift,
so
I
was
in
12,
for
example,
in
you
know,
in
the
cable
open,
cable
industry,
when
they
say
a
spec,
they
were
written
specification
that
members
have
to
implement
and
they
might
have
to
hire
somebody
to
test
compliance
of
the
implementation
right.
So
so
this
is
another
reason
why
the
itf.hf
has
a
history
of
writing
specifications
in
like
in
paper
specifications.
B
Yes,
we
need
running
code.
We
know
that.
But
you
know
more
importantly,
nowadays
I
need
to
explain
to
many
people.
We
want
written
specifications
next
slide,
so
active
supporters,
I've
grouped
them
into
you,
know
ID
and
O.
Implementing
testing
deploying
many
of
those
organizations
say
this
is
great.
B
You
know
when
there's
a
working
group
we're
going
to
come
and
join
right,
so
so
this
is
a
to
me.
It's
a
chicken
and
egg
problem
right.
So
on
one
hand,
we
have
to
show
to
you
guys
here
that
we
have
a
lot
of
you
know
people
interested,
but
on
the
other
hand
they
say
well,
look
hey!
You
know.
Why
should
I
send
my
you
know,
chief
of
innovation,
to
a
meeting
in
IHF
and
we're
not
going
to
get
a
working
group.
So
it's
a
it's
a
chicken
and
egg
problem.
B
But,
however,
having
said
that,
there
are
people
implementing
this
stuff,
you
heard
Pont
at
MIT,
we're
beginning
to
do
an
open
source
implementation
just
of
the
yellow
bit.
So
it's
probably
just
going
to
be
a
client
server
thing.
No,
no
real
Gateway
compelio's
doing
one
on
the
inter
side.
You
have
I
mean
trade,
land
and
trade
lens
and
mercks.
They
have
this
deployed:
okay,
it's
a
proprietary
Network,
and
that
was
this.
The
Second
Use
case.
This
is
running
operating.
I
know
this
personally,
because
I've
been
helping
with
our
project.
B
The
ICF
absolutely
yeah,
they
said
like
great
yeah
yeah,
some
of
them.
Yes,
that's
the
people
who
brought
us
tcpip
and
ipsec
and
TLS.
You
know
so.
Yes,
they
they
actually
would
prefer
the
IHF,
because
the
ITF
were
not
raising
tokens.
When
we're
not
trying
to
raise
funds,
we
write
technical
specifications,
we
don't
own
a
network.
The
ITF
is
a
perfect
neutral
entity.
A
So
there's
been
a
number
of
discussions,
I
think
the
questions
around
the
bill
of
lading
and
the
more
complex
use
case
of
how
do
you
prove
that
you
have
something-
and
you
know
that
can
be
transferred
later
type
of
thing
is:
is
one
discussion
I
heard
I
think
you
know
if
there
is
still
questions
around
what
Eric
was
talking
about
earlier
of?
Do
we
need
to
define
the
level
of
auditability
or
observability?
You
know
it
that
is
important
for
this
work
to
be
considered.
A
A
So
with
that
I
guess
what
the
questions
I'd
prefer
to
hear
now
are
ones
that
are
challenging
in
any
of
these
statements.
Right
is
the
ATF.
The
wrong
place,
you
know,
is
the
ATF
the
right
place.
Do
the
problems
you
know.
Is
this
a
problem
that
needs
to
be
solved?
I'm
not
asking
these
questions
yet
in
terms
of
a
you
know,
a
hum
type
of
approach,
but
I'm
wondering
who,
who
has
challenges
to
you
know
the
typical
reasons
why
we
hold
Buffs
right.
H
So
it's
maybe
not
a
question,
but
maybe
something
that
might
be
missing
and
that
is
kind
of
what
what
are
the
overlaps
I
actually
came
in
here,
believing
there
were
more
overlaps
than
I
I
right
now.
Standing
here,
I
think
this
is
kind
of
fairly
well
scoped,
even
though
there
are
some
also
some
dangerous
sort
of
aspects
to
this
I
mean
the
the
stuff
Eric
pointed
out
that
your
the
gateways
are
kind
of
in
charge
have
to
be
right.
H
H
There
needs
to
be
a
little
bit
better
coordination
among
the
working
groups
and
and
efforts
in
the
iitf
that
work
on
stuff
related
to
digitalization
of
assets,
and
that
involves
Tigris
that
involves
skit.
That
involves
something
like
this
right
and
that
that's
not
an
argument
against
doing
anything
right,
but
it's
an
argument
for
making
sure
that
you're
tightly
scoping
whatever
you're
doing
in
this
working
group
yeah.
H
So
to
specifically
the
use
case
where
the
user
is
not
approving,
stuff,
I
think
that's
kind
of
what
I
gleaned
from
Eric's
statement,
the
user
isn't
necessarily
approving
the
transfer.
The
Gateway
is,
and
that's
a
that's
a
significant
difference
between
this
and
and
tigers,
where
the
user
is
specifically
in
charge
anyway.
B
A
D
D
So
I
saw
a
number
of
companies
on
that
slide.
I,
don't
see
most
of
them
here.
Where
are
you
from
okay?
How
many
people
in
that?
So,
if
we
go
back
to
that
slide,
how
many
people
actually
actually
deploy
shipping
are
on
this
slide
or
actually
here,
I'm
serious,
like
what
I
said,
how
many
people
who
are
not
selling
this
solution,
but
are
consumers?
The
solution
are
actually
here
in
this
business
in
this
room.
B
B
D
A
I,
don't
think,
there's
a
reason
why
so
Eric?
Thank
you.
I
think.
Your
comment
is
very
valid
right.
You
have
to
have
the
people
that
are
willing
to
do
the
work
and
Thomas
talked
about
there.
Are
people
willing
now
I
think
one
of
the
questions
I
can
put
into
the
final
closing
questions
for
you
is
because
we
have
to
include
the
remote
people
right.
So
if
there
are
people
remote
and-
and
you
know,
I
didn't
ask
the
question
in
a
way
that
they
could
raise
their
hand.
A
I
H
You
very
much
disagree
with
you.
Eric
I
think
it's
a
little
bit
unfair
too
characterize
the
turnout
as
weak.
If
that
is
what
you're
doing
I
think
it's
it's
fair
for
a
new
effort
in
a
in
a
in
a
group
that
is
new
for
most
other
companies
involved.
I
think
having
at
least
some
of
them
on
on
site
is
maybe
not
perfect,
but
it's
definitely
an
indication
that
there
is
interest.
A
Q
Thank
you
from
broadcom,
so
yeah
I
would
like
to
support
the
the
point
made
by
Thomas
about
the
difficulty
to
hire
people
in
standardization.
Even
forget
the
ietf
for
a
second.
There
is
a
lot
of
work
going
elsewhere
about
the
lack
of
Industry
engagement.
The
big
difficulty
we
find
is
that
when
you
need
to
make
the
case
as
a
developer
as
an
engineer
as
other
people
to
participate,
you
need
to
have
the
buy-in
from
your
management.
Most
of
the
western
companies
are
clueless
about
standardization.
Q
There
is
another
person
who
mentioned
a
very
good
point
that
open
source
is
not
specifications.
That's
a
very
good
point.
People
have
the
impression
if
you
do
open
source,
that's
good
enough.
No
and
a
lot
of
the
companies
don't
get
that
point
either.
So
for
them,
if
you
tell
them,
I
need
to
open
an
open
source
project.
Yes,
I
need
to
go
to
personalization.
Why
cost
money
your
trips
or
time
you're
not
going
to
spend
on
on
doing
your
day,
life,
job,
etc,
etc.
Q
This
is
just
inaudible
for
the
management
of
most
of
the
companies.
So
I
understand
the
prime
of
qualifying
an
acre
is
right.
In
the
sense
we
have
to
make
sure
that
we
have
enough
critical
mass
I
completely
understand
that,
but
we
have
to
be
careful
about
how
we
manage
the
situations.
Otherwise
we
will
never
start
anything
anymore.
All.
A
I
Roman
nelio
Carnegie,
Mellon
University
I
turned
off
my
video,
so
I
I've
got
in
a
sense
kind
of
a
scoping
question.
I
heard
a
number
of
kind
of
use
cases,
some
felt
very
different
than
the
other
use
cases.
So
I
wasn't
clear
to
me
whether
one
supports
the
other
they're
they're
kind
of
parallel.
So
this
kind
of
made
me
kind
of
step
back.
Is
there
an
MVP,
a
minimal,
viable
product?
We
have
and
are
some
of
these
you
know
to
spin
up
the
MVP.
I
A
I
O
R
And
secondly,
if
you
remove
the
trust
assumption,
you
start
to
go
down
a
rabbit
hole
of
decentralization
and
you
want
to
think
who
is
who
are
we
actually
going
to
now
trust
and
there's
a
famous
paper
famous
to
me
in
the
DLT
interoperability
space
where
there
is
no
complete
decentralization,
you
can
ever
can
never
have
a
complete
decentralized
network.
If
I
give
you
analogy
for
similar
reasons
to
why
communism
wasn't
work
as
soon
as
you're,
completely
decentralized,
no
one's
motivated
to
do
anything.
R
So
lots
of
people
have
been
trying
different
ways
of
experimenting
decentralization
and
if
you
read
the
news
about
blockchain
networks,
you'll
see
that
lots
of
the
asset
transfer
protocols
have
major
hacks
cost
loads
of
money,
big
problems.
So
that's
why
the
two
points
why
I
think
keeping
the
trust
assumption
is
a
good
thing
in
this
protocol.
K
Gilbert,
thank
you.
I'd,
just
like
to
just
Express
a
I
guess
an
opinion
of
what
I've
seen
and
comparing
ITF
to
when
ISO
tc307
started.
It
was
less
than
10
people
who
started
the
initiative
in
2015.
There
was
a
year
worth
of
work
up
front
to
to
create
the
technical
committee,
and
it
actually
started
with
a
group
smaller
than
this,
so
I'm
really
excited
and
surprised
that
there's
so
much
initiative
behind
what's
happening
with
this
potential
buff
and
with
what's
happening
within
iitf
and
I
would
like
to
just.
K
You
know,
cast
your
minds
two
three
years
from
now.
If
you
look
at
the
the
rate
of
of
tc307
there's
over
a
thousand
plus
subject
matter,
experts
across
seven
different
working
groups
across
57
countries
working
on
it.
So
yes
there's
a
representation
of
about
65
plus
whatever
in
the
room
today
about
100.
But
this
could
lead
to
something
much
bigger,
because
we're
here
as
an
engineering
task
force,
is
what
the
ITF
is
to
work
on
these
Solutions
together
over
the
next
few
years
and
doing
it
as
an
RFC
is
the
right
way
to
do
it.
K
A
It
seems
that
if
this
works,
if
this
work
goes
forward,
it
is
likely
a
lot
of
this
work
will
be
done
in
interim
meetings
based
on
how
it's
been
conducted
so
far,
and
it's
actually
more
likely
that
there'll
be
monthly
interim
meanings
now
just
for
for
the
people
that
don't
know
the
ietf
process
in
our
meetings
do
have
to
be
announced
so
far
ahead
and
they
can't
be
too
close
to
ietf,
meaning
there's
a
bunch
of
rules
that
you
will
have
to
start
following.
You
know
other
than
that.
But
there
is
nothing
wrong.
A
A
C
Thanks
Alyssa
Cooper
I
do
think
it's
typical
to
look
at
the
charter,
so
I
was
going
to
ask
you
if
that
was
going
to
happen
or
not.
I
was
just
looking
at
it
and
there
are
I
guess
a
couple.
Questions
well
really
just
one
question
about
this,
which
is
that
it
seems
to
be
missing
the
articulation
of
the
threat
model.
C
I,
don't
know
where
that
would
land,
but
it
to
me
that
that
seems
very
important
for
this
work,
because
there's
a
lot
of
talk
about
assumptions
and
there's
I
understand,
there's
things
being
left
out
of
scope
to
simplify
the
problem
space,
but
a
lot
of
the
threats
are
derived
from
those
things
being
that
are
out
of
scope,
so
I
just
feel
like.
That
is
something
that
needs
needs
to
be
part
of.
A
A
C
The
other
thing
just
to
come
back
on,
like
who
you
know
who
is
here
to
to
speak
about
this.
It's
to
me,
it's
not
I'm,
familiar
with
ISO
and
I,
don't
think.
Actually,
comparing
to
ISO
is
valuable
when
you
come
to
the
idea,
it's
more
about
like
the
the
types
of
participants
and
what
their
respective
roles
are
in
the
system.
So
you
know
sometimes
you
get
like
all
the
people
who
are
there
who
have
a
particular
slant
on
the
problem
and
that
colors
the
design?
C
C
Then
I
think
that
can
be
a
bad
situation
and
I
say
this
as
I
think
the
ad
who
oversaw
the
inter-ledger
buff
that
we
had
some
years
ago,
which
which
very
much
had
that
problem
like
people
who
had
only
one
vested
interest
who
are
trying
to
trying
to
pitch
this
so
I,
don't
think
it's
about
the
number
of
people
who
show
up
I.
Think
it's
about
all
the
entities
that
are
meant
to
interact
in
the
system
being
represented
in
the
design
process.
H
Right
that
goes
to
my
point
of
why
it
might
be
a
little
bit
different
now,
because
we
have
all
of
these
other
efforts
in
the
iitf,
where
some
of
those
groups
are
actually
present
the
software
supply
chain-
people.
Yes,
it's
scoped
to
software,
but
it's
kind
of
it
I
think,
there's
more
there's
more
critical
mass
in
the
organization
now
than
there
was
at
the
time
when,
when
you
were
chartering
the
inter-ledger
thing,
those
were
the
only
people
there
who
was
working
on
similar
things,
I
think
there's
a
better
chance.
I
Romanian
Carnegie
Mellon
University,
no
hat
to
finesse.
What
I
was
asking
earlier
about
the
separability
of
the
use
cases
in
the
poll
a
little
bit
on
what
Alyssa
was
talking
about,
so
I
initially
was
saying
MVP,
so
the
question
would
be
typically,
you
have
an
MVP.
We
heard
folks
are
willing
to
implement
that
who
validates
that
it's
the
MVP
and
that
would
suggest
to
me
the
people
that
want
to
adopt
it
are
going
to
run
it
or
can
say
yes,
indeed,
those
engineering
it
you
hit
the
mark
for
us.
A
S
Hey
Ori
Steele
transmute,
just
to
because
I
heard
the
analogy
to
some
software,
the
skit
working
group
and
the
software
use
cases,
and
over
the
past
couple
days,
there's
actually
been
several
ietf
groups
that
exist
where
they
had
a
very
narrowly
defined
initial
Charter
and
they've
built
very
great
generic
building
blocks
and
they're
talking
about
a
re-chartering
process
for
some
of
their.
You
know
more
extended
use
cases
and
so
I.
You
know
just
saying
that
you
know
when
we
formed
skit.
S
We
talked
a
lot
about
hardware
and
and
physical
supply
chain
use
cases,
and
we
specifically
focused
on
the
software
supply
chain
use
cases,
but
the
building
blocks
that
we've
been
designing
are
with
the
awareness
and
engagement
from
community
members.
Who
are
you
know,
following
those
other
use
cases
as
well
and
I,
think
it's
very
relevant
to
this
and
have
seen
Echoes
of
that
in
suit
and
teeth
as
well,
and
if
anyone
else
would
like
to
sort
of
comment
on
how
how
that's
gone
in
those
other
groups,
I
think
that
could
be
helpful.
B
H
Q
Yep
so
so
I
have
a
question
first
so
because
it's
like
25
years
I've
not
been
in
a
buff
meeting.
So
when
when,
when
you
create
that,
do
you
are
you
doing
some
kind
of
Gap
analysis
versus
other
sdos,
because
I
I
heard
the
iso
was
mentioned,
but
there
might
be
other
groups
that
are
working
on
connected
pieces
to
that
and
like
I,
just
can't
believe
that
that
it
would
be
only
here.
Is
this
part
of
the
of
the
analysis
to
prepare
to
prepare
a
working
above
or.
A
When
I
asked
the
proponents,
why
the
ietf
a
month
you
know
or
two
ago
they
were
overwhelmingly
because
it
is
the
easiest
place
for
membership
to
participate
in
because
there's
not
fees,
there's
not
joining
requirements,
there's
not
all
the
other
things
and
it's
an
open
specification.
So
it
made
sense
to.
Q
B
Q
A
C
A
Q
K
A
With
my
IAB
hat
on,
you
know,
we
run,
we
are
responsible
for
the
the
internet
architecture
board,
that's
responsible
for
reaching
out
to
other
standards,
organizations
where
there
is
overlap,
and
we
need
communication
to
avoid
duplication
and
things
like
that
in
agreement.
That
is
something
the
AV
is,
is
more
than
willing
to
take
on
if,
if
this
actually
brings
in
new
groups
that
we
that
the
ietf
needs
to
associate
with,
there's
absolutely
a
path
to
make
that
happen.
So
that's
that.
A
We
have
a
whole
process
related
all
right
with
that.
I
am
and
chairs
you
not
at
me
if
I'm
heading
the
right
direction
here,
but
my
thinking
is,
there
are
two
aspects
to
the
discussion
today.
The
first
is
this:
a
problem
that
that
needs
to
be
solved,
and
that
is
independent
of
is
the
IHF
the
right
place
to
solve
it.
You
know,
is
it?
A
Do
we
have
the
the
resources
here
to
do
it
so
I'm
gonna
ask
the
first
one
first,
which
is
and
I'm
going
to
constrain
it
to
just
the
the
data
asset
transfer,
so
I
think
the
wording
I'm
going
to
use
is
I
believe
so
this
is
as
if
you
were
voting
right.
I
believe
that
the
problem
of
data
asset
transfers
is
a
problem
that
needs
to
be
solved
somewhere,
so
the
goal
will
be.
You
will
raise
your
hand
if
you
think
it's
a
problem.
A
A
A
A
A
A
A
So
I
think
the
way
I'm
going
to
word
it
is:
is
the
ietf
the
right
place
to
create
the
digital
asset
transfer
protocol?
So
we've
agreed
that
the
problem
needs
to
be
solved
to
be
able
to
transfer
a
digital
asset.
Is
the
ietf
the
right
venue
to
do
that
so
you'd
raise
your
hand
if
you
believe
it
is
the
right
venue-
and
it
should
be
done
here
with
all
of
our
processes
for
working
groups
and
things
like
that
versus
I,
actually
think
it
should
be
done
somewhere
else.
A
A
A
Okay,
so
that
is
that
is
pretty
stable
to
where
it
was
last
time.
So
we'll
note
for
the
minutes
that
there
are
35
people
that
raise
their
hands
and
three
that
did
not
and
I'm
going
to
do
the
same
thing
as
before
for
the
three
people
that
did
not
raise
their
hands,
meaning
you
do
not
think
it
should
be
done
here.
I'd
like
to
hear
and
I'd
love
to
hear
your
opinion,
Richard
Burns,
it
looks
like
you're
first
in
the
queue
go
for
it.
F
E
F
A
F
I,
what
I
would
like
to
be
confident
of
is
that
you
know
if
we
develop
something
here,
we'll
actually
get
usage,
and
so
you
know,
I
I,
think
you
know
by
analogy
to
the
MLS
case,
which
I
think
as
I
could
put
it
on.
The
chat
was
like
kind
of
the
best
case
for
a
buffer.
We
had,
you
know
a
lot
of
of
operators.
A
F
Know
what
I
would
like
to
see
here
is
you
know,
folks
who
are
actually
doing
these
sort
of
transfers.
You
know
moving
assets
around
saying
that
you
know
if
a
tool
like
this
were
available,
and
it
was
a
standard
like
it
would
help
them
do
their
business
better
and
they
would
be
actually
moving
transactions
over
it.
A
D
Eric
go
ahead;
yeah
so
I
mean
I,
I,
I
I.
My
concerns
really
think
the
same
ones
are
terrifically
today,
but
I'll
try
to
sharpen
what
I
would
actually
like
to
see
yep,
which
is
representatives
of
the
networks
of
the
of
the
networks
that
are
being
gatewayed
in
the
scenario
being
here,
because,
like
there's,
a
distinction
between
I
and
the
person
making
the
Gateway
and
I'm
the
people
who
are
operating
the
networks
now
there's
the
people
who
are
operating
the
networks
of
here.
Okay.
So
why
not?
If.
A
A
D
Be
clear
like
I,
hear
Thomas's
point
and
if,
if
if
what
the
outcome
is,
is
that
we
hear
from
those
people
in
email
later
that
they
will
appear
and
we'll
even
review
remotely
that'd
be
great
too,
like
I
understand
that
here,
that's
fine
like
but
like,
but
like
what
I'd
like
to
know.
I
think
that
the
isg,
when
they
go
to
provision
asks,
is
our
critical
mass
of
those
people
are
willing
to
hang
out
and
participate
in
some
fashion.
That's
the
question.
One
question:
I'm
really
interested
in.
A
No
I
think
Eric.
Your
point
is
a
thousand
percent
valid
right
that
they're.
If
you
build
it,
they
may
still
not
come
so
we
need
a
protocol
that
can
both
be
built,
and
you
know
people
are
willing
to
do
the
work
to
build
the
protocol,
but
we
also
need
people
that
are
actually
going
to
use
it
to
say
you
know.
This
is
not
something
that's
just
going
to
be
wasting
paper.
I
have
an
RFC
that
is
only
just
now.
R
A
D
A
Yeah
I
mean
we
have
that
problem,
like
everywhere
in
the
ietf,
unfortunately,
and
with
some
exceptions
in
the
Ops
area,
it
does
have
a
few
implementations.
So
all
right
so
I'm,
going
to
start
with
the
first
question
of
I,
am
a
participant
that
will
actively
implement
or
deploy
the
results
and
we'll
dive
into
later.
Are
you
a
Gateway,
or
are
you
actually
part
of
a
network
that
that'll
be
second
to
this
other
question?
A
A
Do
not
raise
hand
means
I,
am
not
going
to
actively
implement
or
deploy
the
results,
so
yeah
I
know,
but
it
comes
ahead.
I'm
actually
more
interested
in
the
raised
hands
right.
So
I
think
that
the
do
not
raise
hands
is
less
valuable
in
this
particular.
You
know
result
because
it's
not
like
you're
actively
going
to
prohibit
other
people
from
doing
so
right
Roman.
You
have
a
clarific
clarifying
point.
A
So,
as
I
said,
the
the
raw
numbers
of
actively
raising
hands
seem
more
interesting
and
that
that
it'll
be
the
number
of
people
that
are
planning
on
implementing
or
deploying
the
other.
People
might
be
Security
Experts
here
that
have
nothing
no
ties
to
ever
actually
using
the
protocol.
That
doesn't
mean
they
won't
help
develop
it,
because
it
was
clear
that
they're,
worse
all,
right,
I
think
we'll
call
that
good.
So
there
are
12
11.,
11,
12
darn
it
12
people
that
have
indicated
they
will.
A
I
A
I
I
mean
just
practically
I'm
just
trying
to
understand
who
are
those
people
if
they
want
to
I
mean
you
were
the
chair,
but
if
they
want
to
come
to
a
mic
that
would
be
super
if
they
would
tell
us
or
drop
it
in,
can
I
jabber,
or
is
this
double
counting?
You
know
we
have
multiple
Representatives,
which.
J
A
M
M
Last
year
we
solved
an
interoperability
or
set
of
improbability
scenarios
for
the
bank
of
France
and
Agility,
who
are
trying
to
implement
several
asset
exchange
and
as
a
transfer
use
cases
across
their
different
networks,
and
these
networks
are
built
on
different
kind
of
Technologies
like
hypology,
Fabric
and
God,
and
so
they
talked
about
their
experience
and
the
experimental
successes
in
the
recent
hypology
global
forum.
If
anybody
is
interested,
there
are
recordings
available
online.
So
what
I'm
trying
to
say
is
that
people
first
try
to
it's
a
stakeholders.
M
People
are
running
networks,
first,
try
to
seek
out
Solutions
experiment
with
them,
and
then
the
kind
of
people
who
are
providing
Solutions
like
like
we
are
doing.
We
are
maintaining
the
flagship
interoperability
project
under
hyperledger
time.
We
are
the
ones
who
are
participating
in
the
standards
activities
so
that
we
can
take
these
lessons
and
make
sure
that
the
solutions
we
providing
to
those
companies
are
implying
with
the
standards.
N
That's
why
I've
been
participating
in
this
you'll
see
in
the
slide
above
you
know
the
early
on
that
we
were
registered
as
an
observer.
That's
because
ey
is
this
enormous
organization
in
getting
the
approval
to
say
that
ey
is
more
than
an
observer.
Would
probably
take
me
the
next
two
to
three
years.
Yeah.
A
That's
that's
always
a
problem
too.
So
one
last
interest
question
before
we.
We
need
to
start
to
begin
wrapping
up
and
I
want
to
give
the
80s
a
chance
to
to
give
final
things,
but
based
on
Eric's
suggestion
of
wanting
to
know
are
there
actual
operators
of
asset
networks
that
will
use
the
protocol?
That
is
the
next
question.
So
it
is.
I
am
an
operator
of
an
asset
Network
that
will
use
the
SAT
protocol.
A
A
A
A
A
A
Nobody
has
raised
their
hands,
unfortunately,
so
I
I
think
one
of
the
things
that
we
should
ask
Thomas
and
his
crew
to
do
is
potentially
see
if
we
can
actually
get
real
operators
to
to
get
to
provide
a
list
of
people
that
actually
say
that
they
will
deploy
it.
If
it's
as
I
said,
we
don't
attract
operators
with
that.
I
think
that
the
the
only
other
questions
that
I
needed
to
ask
were
the
typical
review
and
edit
type
questions.
Is
that
sufficient
for
you
guys
anything
else?
A
A
All
right,
let's
call
that
good
at
at
28,
just
in
the
interest
of
time.
I
am
will
help
edit
an
author
documents.
So
this
is
for
editors
and
authors.
We
should
not
have
28
okay,
we
don't
need
28
authors.
That
is
not
going
to
happen,
so
we
need
a
few.
But
if
you're
willing
to
actually
do
the
active
role
of
editing
and
authoring
a
document
and
taking
contributions
and
texts
from
people
and
massaging
it
into
the
text,
then
I
I
know
we
have
some
people
doing
that
already.
So.
A
I
I
think
we'll
call
that
good,
that's
12,
which
seems
actually
like
too
many
anyway.
So
we'll
call
that
good
at
this
point
80s
we
did
not
get
the
editing
the
charter.
That
does
not
surprise
me.
We
did
get
a
good
couple
of
good
comments
that
we
will
take
back
and
the
proponents
can
add
things
like
interoperability
liaison
ships
to
other
organizations,
as
well
as
a
security.
A
What's
I'm
blanking
on
the
right
wording,
but
but
looking
at
this
producing
a
document
that
talks
about
the
threat
model,
so
I
think
that
that
will
be
added
to
the
goals.
The
charter
is
fairly
long,
so
I
didn't
think
we'd
get
to
it
today,
but
it
would
be
helpful.
It
is
online.
It's
on
GitHub,
one
of
the
slides
actually
has
a
reference
to
it.
A
A
P
Okay,
so
so
clearly,
there's
a
lot
of
digest
here.
I
think
we
will
take
it
back
and
we
will
do
some
some
long
discussions.
There's
it
doesn't
seem
to
be
like
an
obvious
solution,
one
way
or
the
other
or
sorry
it
doesn't
seem
to
be
a
very
obvious
outcome.
It's
a
it's
an
interesting,
complicated
problem
that
we
will
need
to
think
about
deeply.
A
One
of
the
things
that
the
the
leadership
was
talking
about
the
other
day
is
that
it's
not
obvious
at
the
end
of
a
buff
what
the
outcome
is
and
and
I
I
need
to
apologize
for
the
leadership
that
we
don't
do
that.
Well,
we
take
it
back,
we
think
about
it
and
it's
not
at
the
end
of
the
at
a
buff.
Everybody
knows
the
working
group
is
going
to
happen
or
it's
been
rejected.
That's
there's
discussions
that
happen
afterwards
and
you
know.
That's
definitely
I
think
what
was
happening
here
as
well.
A
I
Sure,
and
if
we're
talking
process,
it's
also
worth
kind
of
recognizing
that
this
is
just
part
of
the
ietf
community
and
buff
is
an
input
to
the
process.
But
we
also
use
the
robust
mailing
list
that
we're
trying
to
kind
of
build
to
help
us
reason
through
that,
and
we
would
check
any
consensus.
We
would
find
there
as
well
yep.
A
Any
I
think
I
won't
ask
for
any
additional
questions,
because
we
are
exactly
at
time
so
I
ran
it
perfectly
to
the
line,
and
that
is
always
the
goal.
Thank
you
all
for
participating
I
actually
found
it.
A
fascinating
discussion,
I
see
where
there's
work
that
needs
to
be
done
and
where
it's
actually
somewhat
decided
so
I
think
there's
decent
progress
here,
one
way
or
another.
Thank
you
all
for
coming.
P
Yeah,
and
also
just
want
to
thank
everybody
for
for
putting
in
this
this,
you
know
a
lot
of
time
and
coming
here
and
efforts,
and
also
especially
West
for
for
doing
a
really
good
job
of
being
a
very
objective
outside
of
the
process
person
doing
the
process
guiding
without
any
stake
in
the
game.
It's
really
appreciated.