►
From YouTube: Approaches to Consortia Governance and Access Control in Hyperledger Fabric Applications
Description
Mark Rakhmilevich, Oracle
“Blockchain is not an island,” is a familiar refrain these days. Practical implications of bringing trading partners into a blockchain-based consortium where they share data and jointly manage the lifecycle of smart contract applications are daunting for many businesses. Who approves new members joining the network? How is chaincode deployment or upgrade are agreed by the members? The need to simplify these processes and automate them leveraging blockchain technology itself is quickly becoming apparent. Is it possible to implement governance policies with sufficient flexibility, while also automating the fundamentals through voting and other means? We’ll discuss practical customer use cases based on Hyperledger Fabric applications and how these issues were addressed. We’ll also cover how business confidentiality needs are met and access control integrations for fine-grained authorizations.
----
Hyperledger – Open Source Blockchain Technologies
Hyperledger is an open source community focused on developing a suite of stable frameworks, tools and libraries for enterprise-grade blockchain deployments.
It serves as a neutral home for various distributed ledger frameworks including Hyperledger Besu, Fabric, Sawtooth, Indy, as well as tools like Hyperledger Avalon, Cactus and libraries like Hyperledger Aries, Ursa. Learn more about Hyperledger projects: https://www.hyperledger.org/use
Case Studies: https://www.hyperledger.org/learn/case-studies
Training & Certification: https://www.hyperledger.org/learn/training
Tutorials: https://www.hyperledger.org/use/tutorials
Webinars: https://www.hyperledger.org/learn/webinars
Events: https://www.hyperledger.org/events
Vendor Directory: https://www.hyperledger.org/use/vendor-directory
Subscribe to the Hyperledger Newsletter: https://www.hyperledger.org/newsletter
Follow-us on Twitter @Hyperledger
Learn about Hyperledger Membership: https://www.hyperledger.org/about/join
#Hyperledger #HyperledgerGlobalForum2018 #Blockchain
A
Welcome
to
the
session
on
approaches
to
governance,
and
so
they'll
go
back
here,
wrong
direction,
approaches
to
governance
and
access
control
in
high
policy
fabric.
My
name
is
Mark
Miller,
which
I
am
part
of
the
development
organization
and
Oracle
responsible
for
our
blockchain
cloud
platform,
which
is
based
on
high
quality
fabric.
A
We
have
announced
general
availability
the
summer
in
July,
and
here
worked
with
a
number
of
customers
in
a
number
of
projects
who
are
deploying
blockchain
networks
in
a
real
business
environment
and
running
into
some
of
those
requirements
that
I'll
talk
about
today,
and
some
of
the
approaches
that
we
are
working
on
with
those
customers
on
addressing
those
problems,
and
there
are
some
if
you
want
to
follow
up.
My
contact
info
is
at
the
bottom
here.
So
we'll
talk
about
some
of
the
items
which
are
kind
of
futures
in
Oracle's
roadmap,
so
the
safe
harbor.
A
What
I
will
touch
on
briefly
is
how
Oracle's
working
with
enterprises
to
help
adopt
blockchain
and
then
talk
about
the
examples
of
the
actual
projects
where
we
are
finding
the
need
for
governance.
We
need
to
be
able
to
solve
those
problems
on
chain
as
opposed
to
off
chain
and
also
access
control.
Our
platform
is
built
to
provide
enterprise,
ready
capabilities
that
allow
customers
to
basically
get
started
in
a
cloud
environment
and
here
the
entire
block
chain
network
provisioned
for
them
in
a
matter
of
a
few
minutes,
with
a
few
keystrokes
leveraging
Oracle
cloud
infrastructure.
A
So
it's
very
quick,
very
easy
to
use
and
it
has
a
lot
of
enterprise
great
enhancements
beyond
the
initial
open
source
capabilities.
That
fabric
provides.
We've
done
a
lot
of
integration,
work
from
REST
API
perspective
as
DK,
as
well
as
various
Oracle
application
integration
points
to
ensure
that
customers
can
very
rapidly
begin
to
use
blockchain
in
real
business
environments
and
integrate
variety
of
data
sources
to
both
drive
transactions
in
the
blockchain,
and,
in
fact
to
where
is
the
data
on
a
blockchain
from
variety
of
back-end?
A
You
know:
systems
of
record
versus
ERP
systems,
human
resource
systems,
customer
experience,
systems,
etcetera
and
learned
a
lot
to
work
with
customers
who
are
in
the
process
of
building
consortium
actually
trying
to
pull
together
organizations
within
the
ecosystem
to
create
those
consortium.
I'll
show
you
some
of
the
examples.
There
is
a
number
of
initiatives
across
various
industries.
Of
course,
everybody's
been
talking
about
financial
sector
and
we're
beginning
to
see
a
lot
of
other
industries
actually
taking
advantage
of
blockchain
capabilities
in
the
last
12
months
or
so
certainly
in
supply
chain
and
related
space.
A
You
can
trigger
events.
You
know
very
simple
example:
when
you
are
in
a
supply
chain,
environment
and
you
have
purchase
of
the
items
verified
against
invoices
again
shipping.
You
can
then
determine
in
fact
that
everything
has
been
sufficiently
mapped
so
that
you
can
go
ahead
and
trigger
payment
event,
and
many
many
other
examples
of
that
right-
around
reconciliation
and
business-to-business
in
the
morning.
A
So
when
you
talk
about
blockchain
models,
you
obviously
see
a
very
broad
spectrum
out
there
right,
starting
with
public
blockchains.
Like
bitcoins
and
the
theorems
out
there
that
are
anonymous
to
the
hermitian
blockchain,
which
is
primarily
what
we
are
focusing
on
and
most
of
our
business
enterprise
customers
are
really
interested
in
permission.
A
How
are
you
going
to
manage
all
of
the
organizational
requirements
around
adding
members
deploying
chain
code,
creating
channels
and
many
other
organizational
things
that
are
important
in
this
environment,
where
multiple
companies
are
working
together
and
there
is
nobody
in
the
lead
right?
That's
ultimately
the
question
here
and,
of
course,
we
see
some
emergence
of
what's
called
public
permission,
blockchain
kind
of
a
hybrid
model
where
typically,
the
people
who
update
the
data,
the
writers
on
the
blockchain
are
permissioned
entities.
A
But
then,
here
also
you
know,
public
access
for
read,
accesses
and
public
is
in
quotes
here,
because
public
would
be
controlled
to
some
other
means
it
could
be
very
open
or
it
could
be.
Some
access
control
system
may
be.
You
know
all
citizens
of
Switzerland,
so
some
kind
of
a
national
system-
or
maybe
you
know
if
you
are
in
financial
services,
maybe
all
participants
in
the
market
or
something
like
that,
but
the
main
focus
today's
consortium
blockchains.
What
we're
finding
is
that
many
customers
are
full,
most
customers
would
fall.
A
I
would
say
in
this
two
buckets
right.
There
are
still
experimenting
and
they
want
to
do
it
fast.
You
know
they
want
to
try
a
use
case
and
move
on
to
the
next
one.
If
this
one
doesn't
pan
out
so
they
need
to
be
able
to
provision
very
fast,
configure,
easily
develop
an
integrated,
and
then
we
have
a
number
of
enterprises
which
are
ready
to
move
forward
into
production
once
they've
determined.
There
is
a
valid
use
case
or
more
than
one,
perhaps
that
they
have
discovered,
so
they
need
all
of
the
enterprise.
A
Great
features
performer
resiliency
security
and
confidentiality
is
extremely
important.
At
the
end
of
the
day,
we're
talking
about
business
processes,
enterprise
integration,
particularly
making
it
easy
plug-and-play.
So
you
here
to
spend
an
inordinate
amount
of
time,
integrating
every
single
system
of
record
day-to-day
supportability
governance
and
so
on.
So
Oracle
platform
provides
a
lot
of
those
capabilities
built-in
in
our
cloud
environment
we
use
Oracle
cloud
infrastructure
to
provide
a
lot
of
dependencies
and
then
hyperedge
a
fabric
with
Oracle
enhancements.
On
top
of
that,
providing
a
great
deal
of
integration
capabilities,
fluoroscopy
is
the
case.
A
Oracle
integration,
cloud
or
C
provides
enterprise
adapters
for
plug-and-play
integration
to
50,
plus
different
applications
in
the
cloud
or
on-premise,
Oracle
and
third
parties.
So
all
this
is
preassembled.
All
of
that
is
available
basically
through
a
simple,
quick
provisioning,
5-10
minutes
and
it's
up
and
running
it's
open
because
it
integrates
not
just
within
the
Oracle
ecosystem,
but
was
external
fabric
nodes.
So
we
have.
He
had
customer
spruces
by
using
their
own
instances
of
open
source
fabric
together
with
Oracle
cloud,
and
in
fact
we
began
to
work
work
with
some
roses
and
members
of
hyper
ledger.
A
That
does
the
same
thing
recently.
It's
an
environment
that
supports
plug-and-play
integrations
as
I
mentioned,
with
European
systems
in
a
cloud
environment,
on-premise,
environment,
Oracle
and
third-party
enterprise-grade
around
things
like
high
availability,
resiliency,
built-in,
backup,
zero
downtime
manage
patching
and
so
on
and
automated
DevOps,
so
that
customers
can
focus
on
the
application,
value
and
Norton's
infrastructure.
Managing
the
platform
to
kind
of
take
care
of
that's
was
a
managed
path.
A
We
have
about
96
or
so
customers
and
a
platform
currently
of
about
400
trials
and
many
of
the
ones
examples
you
see
on
the
right
here
about
to
go
into
production
actually
about.
Half
of
this
are
going
to
be
in
production
in
q1,
so
funds
transfer,
loyalty,
programs,
provenance
tracking
in
food,
in
particular,
maritime
shipping
documentation
exchange
in
maritime
shipping
is
a
very
onerous
task:
franchise
network
between
the
franchise
owner
and
franchisees,
and
the
number
was
examples
in
public
sector,
healthcare
and
so
on.
A
So
this
almost
announced
customers
just
to
share
some
of
the
names
here.
You
might
not
recognize
that
publicly
announced
and
many
others
and
working
through
this.
So
what
we
see
is
that
in
many
cases
we
have
two
types
of
consortium
models
that
are
being
stood
up.
This
is,
what's
called
found:
a
lead
consortium
model
in
the
capital
at
the
fabric
environment
here
of
an
ordering
service.
A
That
typically
is
operated
by
a
founder
somebody
who
might
be
ahead
of
a
supply
chain
consortium,
because
a
major
buyer
you've
all
heard
about-
let's
say
the
IBM
Walmart
food
safety
projects
of
Walmart.
Basically
around
this
consortium,
they
are
responsible
for
the
ordering
service
and
everybody
else
kind
of
joins
in
as
participant
you
might
hear,
of
a
lota
system
led
by
a
large
bank
and
all
of
the
merchants
participating
would
be
participants
as
well,
but
the
bank
would
operate
that
and
would
make
key
decisions,
deploy
the
smart
code,
create
channels
and
set
everything
up.
A
The
participants
here
the
governing
bodies
that
they
want
to
create,
and
then
they
want
to
make
executive
decisions
about
that.
The
second
one
is
a
global
shipping
business
network,
which
is
a
particular
network
that
was
recently
announced
by
some
of
our
partners
on
Oracle
blockchain
in
maritime
shipping
and
I'll
talk
a
bit
more
about
that,
because
it's
a
great
example
of
the
requirements
that
emerging
in
governance
for
the
centralized
networks
and
cover
trying
to
address
them.
A
So
here's
a
two
different
examples
of
solving
similar
problems
in
the
different
way.
Carcass
mode
is
a
partner
borrowers,
customers,
that's
in
maritime
shipping.
They
work.
Where
is
a
very
broad
ecosystem
of
companies
from
shippers
to
freight
forwarders,
ocean
carriers,
non
vehicle
operators,
ports
and
terminal
operators,
and
so
on,
customs
and
such
and
ultimately,
they
have
a
lot
of
documentation
that
has
to
make
it
way
through.
Each
shipment
might
require
20
to
30
documents,
often
multiple
versions,
updated
and
signed
by
different
participants
in
the
system.
A
They're
often
delays
often
disputes
in
the
system
and
at
the
end
of
the
day,
the
ships
cannot
pass
inspections.
Customs
cannot
get
the
container
out
of
the
port
because
on
the
commutations,
missings
and
they're,
paying
fees
and
all
of
that
stuff,
so
cargo
smart
created
this
document
exchange
to
unite
all
these
participants
in
a
single
blockchain
ecosystem.
In
order
to
provide
the
ability
to
trace
all
of
the
documentation
link.
A
All
of
the
documents
on
a
blockchain
provide
a
tamper
proof
audit
trail
so
that
it's
clear
exactly
who
touch
the
document
when
and
why
and
ultimately
collect
the
data
relevant
data
inside
the
document
into
the
metadata.
That's
visible
in
a
blockchain
record
for
context,
based
document
validation.
So
in
this
particular
example,
cargo
smart,
essentially,
is
the
founding
instance,
and
everybody
else
is
joining
and
they're
operating
this.
A
They
are
the
application,
vendors
they're,
deploying
the
chain
code
and
all
the
surrounding
application
components,
they're,
managing
the
identity
and
authentication
they're,
managing
the
plots
of
who's
joining
the
network,
and
so
on.
Now.
The
second
thing
that
they've
done
is
say:
let's
move
completely
to
the
opposite
side
of
this
spectrum
into
the
centralized
model,
and
so,
with
the
help
we've
started,
what
we
call
a
global
shipping
business
network.
It's
a
maritime
shipping
alliance.
A
It
was
announced
at
China's
import/export
in
November
that
involves
five
of
the
major
ocean
carriers
that
are
responsible
between
them
for
about
one-third,
total
ocean
container
shipping
capacity
and
four
of
the
five
largest
terminal
operators
around
the
world.
The
purpose
of
this
blockchain
environment
is
to
be
completely
decentralized,
completely
open,
they're,
going
to
hear
a
variety
of
data
sources
on
this
block
since
starting
this
was
called
titles
and
the
left
there
on
your
left-
and
this
is
documents
all
kind
of
documentation
following
that
putting
variety
of
events
and
tracking
Zatanna
blockchain
as
well.
A
However,
every
participant
the
nine
listed
and
all
of
the
ecosystems
that
they
work
with
can
actually
build
and
deploy,
distribute
applications
on
top
of
this
data
and
throw
add
value
between
themselves.
So
if
you
hear
a
large
steel
company
working,
let's
say
Wisconsin
some
ports
and
they
have
a
particular
problem
because,
as
I
discord,
they
can
schedule
trains
that
takes
the
calls
the
call
delivery
from
the
pork
into
their
factory
because
they
don't
know
exactly
when
the
ship
arrives
and
they
want
to
blockchain
app
to
help
them.
A
Do
that
and
coordinate
the
data
between
their
systems
and
the
portents
are
tracking.
The
data
will
be
available
in
the
blockchain.
They
can
create
an
application
to
access
that
data
and
monitor
it
and
manage
it,
and
so
you
have
an
environment
where
there
is
a
need
for
companies
to
be
able
to
independently
define
how
they
add
members,
how
does
they
deploy
chain
code
and
so
on
in
this
environment?
A
Here's
another
example
in
trade
finance
working
with
a
company
called
export,
which
is
an
Ivy,
that's
building
a
blockchain
for
trade
finance
connecting
into
the
major
ear
piece
system,
starting
with
Oracle
here
feels
like
NetSuite,
which
is
a
small
video
business
ERP
for
around
40
thousand
customers
that
are
using
it
in
small
medium
enterprises
and
Oracle
e-business
suite
at
the
bottom,
which
is
who
had
a
large
customers.
You
know
major
of
Fortune
1000
companies
and
in
between
you
have
a
whole
bunch
of
banks
issuing
banks
advising
banks.
The
process
model
here
is
very
straightforward.
A
When
you
want
to
get
a
letter
of
credit,
you
have
to
fill
in
a
lot
of
information,
send
it
to
the
bank
instead
of
doing
it
manually.
You
can
now
do
it
using
the
export
plugin
written,
that's
with
directly
from
ERP
system,
so
that
is
all
prefilled.
The
protesters
manage
it's
all
sitting
in
the
same
exact
space
where
P
user
is
operating
day
to
days,
and
they
have
to
go,
do
something
separate,
it's
all
built
in
and
packaged,
and
it
can
flow
to
the
bank
ultimately
to
the
seller
and
they
can
manage
this
process.
A
But
the
point
here
is
that
the
consortium
is
going
to
involve
a
lot
of
buyers
and
sellers
and
a
number
of
banks
that
are
going
to
be
participating
here
is
issuing
an
advising
banks.
Again,
they
need
to
determine
how
to
manage
this,
and
certainly
exporters
and
i3
is
willing
to
provide
applications.
They
don't
want
to
take
on
the
role
of
operating
this
and
determine,
in
fact,
who
is
going
to
be
added
and
how
and
here's
another
example
here
in
electronic
distribution
industry.
A
There
is
something
called
sip
and
debit
claims.
Ship
and
Debbie
claims
means
that
when
you
ordering
electronic
components,
you
don't
quite
yet
know
how
many
are
going
to
sell
and
so
you're
paying
for
them
basis
for
the
high
starting
price.
As
you
sell
more,
there
are
all
kinds
of
conditions
that
the
sellers
create
a
circle
K.
A
If
you
sell
this
many
in
this
time
frame
to
this
particular
buyer
will
give
you
a
refund
a
discount
right,
that's
called
ship
and
debit
claim
and
then,
ultimately,
the
data
has
to
be
tracked
back
and
those
claims
have
to
go
back
to
the
manufacturers,
to
the
sellers
of
the
components
so
that
they
can
refund
to
the
distributor.
It's
extremely
arduous
manual
process.
Tens
of
millions
of
dollars
for
a
particular
distributor
is
sitting
in
unpaid
claims.
A
Members
of
this
blockchain
governance
again
becomes
an
important
point.
So
what
we
really
need
here
right,
you
know
when
we
are
moving
from
a
founder,
led
Network
to
the
centralized
governance
model
you
now
have
to
have
each
individual
member
deploy
the
chain
code
determine
how
they're
going
to
work
jointly
on
adding
new
members.
This
is
going
to
be
an
email
sent
around.
Everybody
gets
on
a
conference
call
and
they
decide.
Ok,
here's
a
bunch
of
new
members.
We
want
to
add-
and
you
know,
do
we
agree
with
that
or
not.
A
It's
certainly
not
going
to
be
a
straightforward
process
and
it
will
impose
a
lot
of
administration
costs
on
the
consortium,
because
it's
not
automated.
So
what
we
are
working
on
is
to
propose
an
own
chain
governance
model
that
will
support
voting
based
policies
from
what
we
saw
from
what
we
looked
at.
There
are
three
levels
that
immediately
come
to
mind
at
a
network
level.
You
want
to
be
able
to
control
all
the
process
of
adding
or
removing
organizations
at
the
channel
level.
You
want
to
control
how
you
creating
channels
and
who's.
A
The
members
are
of
the
particular
member
organizations
in
a
particular
channel
and
then
at
the
chain
code
level
of
smart
contracts
deploying
an
operating
chain
code.
If
it
is
a
decentralized
and
democratic
blockchain
environment,
then,
if
somebody
is
going
to
upgrade
to
a
new
version
of
chain
code,
you
want
to
review
process.
A
The
other
process-
that's
very
important
here-
is
how
do
you
collect
the
world's
tabulate
them
and
then
execute
on
the
actions
downstream
actions
required
by
those
words
right,
so
you
want
chain
code
running
that
is
in
fact
going
to
be
able
to
go
through
the
endorsement
process
ordering
process
and
so
that
you
can
get
all
the
policies
defined
unchanged.
The
results
of
the
voting
tabulation
maintained
on
chain
as
well
using
governance
channel
that
will
allow
you
to
then
have
all
of
the
data
available
for
auditing.
A
All
decisions
really
need
to
be
audited
in
order
to
ensure
full
transparency,
and
then,
of
course,
you
also
want
to
have
some
agents
that
allow
you
to
automate
this
process.
If
you
have
a
vote
on
adding
a
particular
members,
there
is
a
transaction
proposal
and
sufficient
number
of
votes
have
been
gathered
based
on
whatever
the
policy
requirements
or
you
want
automatically
go
ahead
and
do
that
you
don't
want
to
then
depend
on
somebody
manually
going
in
and
clicking
through
the
process
of
adding
a
new
member.
A
So
this
is
infrastructure
that
we
are
working
on
and
it's
intentially
going
to
support
some
key
operations
from
the
initial
capabilities
that
we
are
going
to
be
releasing
adding
a
member.
So
there
is
going
to
be
a
transaction.
That's
going
to
be
proposing
to
add
a
member
and
there's
going
to
be
different.
Voting
and
non-voting
rolls
and
they're
going
to
be
policies.
The
tabs
that
will
allow
for
number
of
votes.
A
So
we
are
providing
essentially
three
capabilities
here:
on
top
of
khadija
fabric
environment,
a
governance
model
and
that
governance
module
will
be
based
basically,
as
a
client
application
using
the
fabric
SDK,
it
will
sit
on
a
governance
channel.
It
will
basically
connect
with
the
governance
channel
to
execute
transactions,
and
we
will
hear
of
the
peers
that
here
voting
status
as
read/write
members
in
that
particular
channel,
so
that
they
can
execute
the
chain
code.
That
then,
is
specifically
built
to
handle,
voting
and
handles
the
evaluation
of
policies.
A
The
chain
code
itself
is
going
to
dynamically,
read
the
policy
from
the
governance
channel,
and
then
it
relies
a
trigger
a
volume
and/or
it
will
collect.
It
will
have
methods
that
enable
it
to
actually
collect
the
votes,
tabulate
them,
and
it
will
check
every
time
when
your
board
comes
in.
If
the
policy
has
been
met,
if
I
have
a
policy
that
says
on
the
channel
was
10
members,
I
need
7
votes
to
take
certain
action.
Every
time.
A
A
Basically,
it
Network
initially
method
to
define
the
initial
policies
and
the
voting
members,
starting
with
the
founders
and
others,
could
be
added,
create
proposal
which
is
essentially
creating
a
new
proposal
based
on
the
policy
and
certain
resource
groups,
and
it
will
omit
the
events
so
that
the
client
applications
know
that
there
is
a
vote
outstanding
vote
message
to
actually
record
a
vote
for
a
particular
proposal.
Key
that's
recorded
on
the
chain
and
evaluating.
A
Then,
if
the
policy
requirements
have
been
met
and
if
an
agent
to
automate
the
process
can
be
triggered
and
a
query
function
as
well,
so
you
can
check
and
see
what
the
status
of
a
particular
proposal
is.
It's
kind
of
the
basic
framework
that
we
think
is
going
to
help
solve.
Not
all
the
problems
out
there
but
provides
at
least
an
initial
starting
point
for
dealing
with
governance
on
chain
so
that
in
most
of
the
consortium
environment,
people
can
move
from.
A
You
know
offline
kind
of
out-of-band
processes
to
something
that
is
more
automated
algorithmic
as
brand
said.
Here's
an
example
of
how
this
would
work
when
you're
installing
the
new
chain
code.
So
in
an
environment.
This
is
example
based
on
our
particular
console.
We
have
a
UI
where
you
have
a
chain
code,
tab,
there's
an
action
that
you
deploy
on
your
chain
code.
There
is
a
wizard
that
provides
capabilities
to
specify
all
of
the
parameters
necessary
when
you
installing
a
chain
code
which
appears
you
want
to
turn
with
channel
endorsement
policies,
etc.
A
In
fact,
if
you
want
to
see
this
live,
I
hear
a
demo
tomorrow,
at
2,
o
clock
in
the
theater
to
show
Oracle,
blockchain,
environment
and
console
and
I
can
get
into
that
a
little
more
details,
but
essentially
the
wizard
will
provide
an
option
to
send
a
request
for
vote
right
rather
than
just
install
it
by
yourself
in
your
particular
peers.
So
sending
for
vote
will
then
use
create
proposal,
method
to
be
able
to
put
a
proposal
on
chain
and
the
chain
code
rules
and
triggers
events.
A
The
policy
will
be
defined
on
that
same
channel
and
then
chain
code
will
be
shown
in
pending
status,
while
the
voting
project
is
going
on.
When
the
voting
members
registers
the
world
they're
calling
a
vote
method
and
that
will
be
available
through
an
API
will
extend
the
SDK
to
Dusit,
as
well
as
provide
the
REST
API
and
it's
easier
to
build
some
clients.
If
you
want
to
use
interface
for
that,
ultimately,
when
you
tabulates
official
number
words,
then
an
agent
is
triggered
automatically
install
this
chain
code
on
the
channels
pianos.
A
Now
this
is
quite
a
bit
different
than
what
happens
today
in
the
current
environment.
If
you
want
to
install
a
new
version
of
chain
code,
every
single
member
has
to
have
the
administrator
go
through
the
process
of
installing
chain
code
and
say
node
and
we've
been
hearing
from
I
can't
tell
you
how
many
customers,
how
painful
that
is
at
the
same
time,
the
saying
look,
this
node
is
running.
You
know
we
own
it.
It
might
be
running
in
Oracle
cloud,
maybe
in
their
data
center
whatever,
but
it's
still
the
node
and
they're
responsible.
A
So
they
want
to
have
some
controls,
I,
don't
want
to
say,
blondie
just
go
and
let
somebody
else
install
code.
What
they
want
is
a
process
where
they
have
ability
to
hear
it
done
automatically,
but
after
a
review
and
voting
process,
and
so
what
we've
done
is
automated
it,
but
not
blindly,
we
created
the
process
that
allows
you
to
essentially
take
a
vote.
You
have
ability
to
then
review
it
whenever
you
need
to
and
then
you
can
actually
register
a
vote
and
then
it
will
happen
automatically
once
sufficient
number
of
participants.
A
So
that's
a
limit
on
governance
and
some
of
those
things
that
we've
been
looking
at
and
havoc
planning
to
move
ahead
versus.
There
are
the
areas
that
is
somewhat
different
but
also
related
to
enter.
Presidents
is
confidentiality
and
privacy.
There
is
a
great
tension.
Every
time
we
get
into
conversation
with
an
enterprise
about
how
great
and
valuable
transparency
on
a
blockchain
is
single
source
of
truth.
Everybody
can
see
all
the
data
and
then
you're
sitting
down
with
a
Porsche
sink
office
and
says
Retta
me
what
I
mean
everybody
can
see.
A
The
data
is
purchase
order
for
a
particular
vendor
either
one
does
the
vendor
see
the
supporters
order
I'm
sending
invoices
to
somebody
else.
That's
confidential
right,
I
want
it
to
be
between
the
two
parties
involved,
so
I
don't
know
about
you.
I've
been
in
the
many
of
those
conversations
and
every
single
one
gets
into
this
point
where
ultimately
people
realize
there
is
extension
and
you
have
to
solve
it
now.
There
are
some
approaches
today
in
fabrics
that
try
to
do
that.
There
is
a
channels
mechanisms
where
you
can
subnet
the
ledger
and
participants
by
channels.
A
That
sounds
great,
but
it's
kind
of
coarse
grain.
You
end
up
with
proliferation
of
channels.
If
you're,
not
careful,
you
want
to
have
a
network
of
ten
members
of
you
want
bilateral
privacy,
that's
hundred
channels,
and
if
you
have
a
thousand
channels,
sorry
thousand
members
participating
organizations,
guess
what
you
can
hear:
a
million
channels,
that's
not
really
very
manageable.
A
Since
the
fabric
1-0
came
out,
there's
been
some
additional
approaches
using
access,
control
mechanisms,
attribute-based
and
role
based
access
control,
and
so
we
look
a
little
bit
at
that,
but
force
a
few
examples.
So
we
have
a
customer
of
ours.
Who's
in
electronic
data,
interchange,
EDI
space,
and
they
do
a
lot
of
electronic
invoicing.
In
particular.
Large
customer
base
of
Zaire's
is
franchise
owners
as
their
base
all
over
Europe.
The
implementing
this,
particularly
in
Spain
right
now,
whereas
they
want
to
be
able
to
connect
franchise
owners
and
large
number
of
franchisees
is
in
blockchain.
A
When
the
invoice
is
a
sin
today,
often
there
are
disputes
through
CDI
process.
Did
they
get
it
in
time?
You
know.
Was
it
a
proper
invoice?
A
lot
of
things
get
unpaid,
and
so
there
is
normal
business.
Friction
there
right.
So
the
answer
is
put
it
on
a
blockchain,
so
everybody
can
see
the
same
exact
source
of
truth
or
the
trail
non-repudiation
everything's
time-stamped.
You
can't
have
a
dispute.
Well,
that's
great,
so
they're
putting
it
on
the
blockchain.
A
But
now
the
question
is:
how
do
you
manage
the
confidentiality
I
mean
you
sending
an
invoice
from
a
franchise
owner
to
a
particular
franchisee
company?
X,
you
don't
want
others
to
be
able
to
see
it.
You
don't
want
to
create
a
separate
channel
for
that
pairwise
relationship
either
because
you
go
twos
and
have
a
proliferation
of
channels.
That's
a
challenge.
Here's
another
one
in
healthcare
space.
So
we
have
a
partner
in
US
health,
st.
that
is
working
on
an
application
that
will
support
remote
patient
monitoring.
A
When
you
have
a
patient
who
is
admitted
to
hospital
with
congestive
heart
failure,
heart
attack
for
all
of
us,
laypeople
or
pre
heart
attack
kind
of
symptoms
and
they
get
treated
at
the
hospital,
then
they
get
discharged
home
they're.
Well,
but
often
there
is
a
significant
readmission
rate.
In
fact,
it
looks
like
around
27
percent
of
those
people
who
have
been
discharged
get
readmitted
back
to
the
hospital
because
something
happened
after
the
fact
there
is
billions
of
dollars
in
costs.
But,
more
importantly,
there
is
significant
death
rate
people
who
are
admitted,
but
too
late.
A
They
can't
be
helped.
So
this
particular
company
is
trying
to
solve
that
and
that
Ranta
solves
it
by
having
people
be
discharged
to
go
home
with
a
set
of
devices
and
so
they'll
have
ability
to
collect
the
data
reported
using
IOT
and
blockchain
together
to
a
number
of
participants
who
can
then
monitor
this
data.
So
if
you
have
a
blockchain
that
involves
basically
distributed
care
team,
a
patient,
their
physician,
maybe
there
is
another
attending
nurse
from
another
organization.
A
Maybe
there
is
an
analyst
sitting
somewhere,
you
know,
maybe
somebody
a
company,
that's
providing
a
service,
and
so
on
so
open
is.
There
is
a
great
deal
of
importance
attached
to
healthcare
privacy
and
in
u.s.
the
HIPPA
regulations
are
very
strenuous.
You
have
to
make
sure
that
you
have
patients
concurrence
to
release
the
data
and
also
that
only
people
who
entitled
to
a
particular
data
can
see
it.
A
So
we've
been
working
with
this
particular
company
to
look
at
how
we
can
provide
fine-grained
access
control
mechanisms
to
ensure
that
only
the
relevant
parties
can
see
relevant
data
right.
It's
not
just
about
tracking
the
HIPAA
signatures
and
all
of
that
stuff,
which
is
good,
but,
more
importantly,
how
do
you
track
the
data
that's
being
submitted
and
popped
up
on
people's
dashboards
right
I
mean
a
nurse
sits
in
front
of
what
natural
cooperation
center
right?
It's
what's
called.
You
know
in
medical
terms,
telemetry,
where
we
have
people
sitting
at
home.
A
The
data
is
being
reported,
alerts
popping
up
and
they're
taking
action
based
on
that
at
least
notifying
physicians
and
so
on.
It's
an
area,
that's
actually
a
number
of
hospitals
are
beginning
to
pioneer
in
us.
Not
necessarily
this
blockchain,
but
volscian
certainly
is
an
important
enabling
technology
here.
A
So
the
fine-grained
access
control
is
really
becoming
very
important
and,
as
I
mentioned,
the
existing
mechanisms
like
channels
are
to
coarse-grain.
So
that's
not
going
to
be
an
answer.
Attribute
based
access
control
is
a
good
initial
step
where
you
essentially
hear
x.509
certificate,
optional
fields
used
to
record
certain
attributes
and
then
chain
code
can
query
those
during
execution
and
make
decisions
based
on
the
only
problem
is:
there's
no
standard
definition
of
what
those
attributes
are
you
going
to
hear
an
agreement?
A
That's
required
between
the
users
and
chain
code
developers
to
determine
what
attributes
the
code
for
right.
So
that's
going
to
be
a
little
more
difficult
fabric
actually
introduced
in
1.2
and
as
a
mechanism
for
access
control.
But
that's
management
functions
focused
right,
so
it
provides
access
control
around
things
like
installing
a
braking
chain
code,
getting
information,
getting
blogs,
getting
transactions
getting
channel
information.
A
So
basic
requirement
that
we
see
is
ability
to
have
developer
Shenko
developers
be
able
to
control
what
users
can
access
the
functions
and
resources
related
to
a
particular
chain
code.
You
want
to
hear
of
dynamic
list
of
entitlements.
Basically,
if
any
of
you
familiar
with
authorization,
controlled
access,
control
in
an
enterprise
environment,
given
title
and
management
systems,
that's
what
they
do
all
day
long.
A
The
question
is:
how
do
you
provide
that
in
a
blockchain
environment,
part
of
the
answers
that
we
explored
was
well
here
when
entitlement
management
system,
let's
call
an
API
in
that
entitlement
management
system
from
the
blockchain
chain
code
and
get
a
response
back
and
manage
it
all
externally
in
an
external
security
manager.
The
problem
is
that
there
is
latency
involved,
but,
more
importantly,
have-not
determination
right.
A
You
have
chain
code
executing
on
10
peer
nodes,
and
some
of
them
might
get
a
response
and
others
may
not
because
of
the
state
of
the
network
and
they're,
going
to
make
different
decisions,
which
is
not
really
good
in
a
blockchain
environment.
So
what
we
really
want
is
on
chain
entitlement
mechanism
where
the
changes
are
recorded
and
change.
A
Initial
occurs
when
they
deploy
the
chain
code,
there's
going
to
be
a
matter,
access
control
is
to
determine
which
users
can
be
authorized
to
perform,
a
collaborations
which
is
adding
a
new
article,
registering
a
new
resource,
adding
a
new
user
and
so
on
and
sharing
that
across
channels
is
going
to
be
very
important
because
in
many
cases
many
of
the
same
chain
code
we're
just
starting.
The
ledger
in
terms
of
you
know
short
scaling.
A
The
data
space,
but
using
the
same
chain
code,
so
library
functions
will
provide
the
ability
to
create
and
delete
resources
and
resources
will
be
genetic.
So
it
could
be
a
chain
code
function
name,
it
could
be
a
field
name,
a
key
name
that
you're
updating
there
will
be
a
calls
associated
with
them.
We'll
use
a
concept
of
groups
to
create
and
delete
groups
based
on
identities.
A
So
it's
going
to
be
group
managed,
environment
identities
will
include
5:09,
CN,
sub
common
names,
organizational
units
and
all
starting
with
the
certificate
identities,
and
then
expanding
that
beyond
the
suspect,
ability
to
create
tackles
a
tackle
to
a
particular
resource,
because
the
same
marker
might
be
added
to
multiple
resources.
You
want
to
make
that
orthogonal
and
then
the
ability
to
get
members
get
lists
and
ultimately,
the
check,
access,
API
call
from
the
chain
code
to
be
able
to
determine
was
a
particular
resource
because
particular
type
of
access,
readwrite
change,
delete,
etc.
A
To
a
particular
you
know
for
that
particular
resource.
So
this
is
a
capability.
Is
that
we're
working
on
and
the
things
there's
going
to
be
a
lot
of
value
in
leveraging
this
for
those
enterprise,
customers
that
are
going
to
have
confidential
data
that
they
need
to
protect,
and
they
want
a
chain
code
to
provide
on
chain
entitlement
mechanisms.
A
Now
there
isn't
a
section
here,
that's
quite
interesting
between
governance
policy
and
the
entitlements
or
access
control,
which
is
that
if
you
are
going
to
have
a
chain
code
deployment
process
and
that's
going
to
be
based
on
governance
policies,
how
many
words
do
need
to
the
particular
chain
code?
Part
of
the
developed
deployment
spec
is
going
to
be.
What
is
the
initial
entitlements?
What
is
initial
resources
being
defined
and
who's
going
to
have
access
to
those
particular
resources
in
the
consortium,
and
so
voting
will
in
fact
look
at
that.
A
So
to
summarize,
I
know
we're
coming
up
on
lunch
time.
Here
we
see
this
significant
shift,
I
think
from
blog
change
that
over
the
last
couple
years
were
mostly
found
to
lead
the
emergence
of
this
decentralized
governance,
blockchains
democratic
blog
chains,
and
we
really
need
the
governance
mechanism
so
that
things
like
adding
members
creating
channels
deploying
chain
code
can
be
automated
right.
A
So
you
hear
track
record
if
you
audit
log,
we
all
talk
about
how
great
blockchain
is
in
ensuring
full
transparency,
non-repudiation
tracking
of
everything
that
happened,
but
everything
that
happens
is
implemented
by
the
chain
code
and
the
tracking.
Now,
of
course,
a
chain
code
installed
is
a
system
function
and
there
is
system
chain
code
to
do
that,
but
it's
being
tracked
essentially
on
a
system
channel
and
it's
not
something
that's
easily
accessible.
A
So
the
idea
here
is
to
provide
an
auditable
mechanism
for
on
chain
governance
and
provide
voting
policies
that
allow
customers
to
be
able
to,
in
a
decentralized
environment.
Take
both
of
the
voting
members,
perhaps
subset
of
the
full
membership
list
to
be
able
to
control
all
of
those
operations,
and
then
added
to
that
is
a
fine-grained
access
control
mechanism
that
we
think
is
going
to
help
a
great
deal
to
maintain
business
confidentiality.
But
it's
in
a
regulatory
requirement.
A
In
many
industries,
you
need
to
protect
the
customers,
even
if
you,
in
an
alliance
like
in
the
shipping
global
shipping
business
alliance,
very
strict
requirements
that
vessel
carriers,
freight
forwarders,
can
cooperate
and
exchange
a
lot
of
data.
But
they're
not
allowed
to
share
the
shipper
information.
And,
of
course
you
know
in
financial
services.
You
look
at
proliferation
of
you
know,
particular
type
of
block
chains
that
claim
greater
capabilities
around
data
privacy
like
quorum,
quarter
from
r3
and
so
on.
A
So
our
boss
here
can
take
a
few
questions
and
if
you
want
more
information,
oracle
blockchains
there
are
some
URLs
up
there.
The
first
one
is
just
general
landing
page
was
all
kind
of
useful
information
and
cause
some
examples.
The
second
one
cloud
that
Oracle
accomplished
blockchain
is
actually
the
landing
page
for
the
blockchain
cloud
service,
where
you
can
go
and
sign
up
and
explore
this
life,
and
then
you
have
a
developer
page
as
well.
A
A
So
we
are
going
through
some
process
with
in
Oracle,
legal
and
so
on.
They
got
approvals
to
contribute
and
looks
like
we're
getting
good
progress
there.
So
some
of
this
will
be
contributing
as
well
some
other
capabilities,
because
we're
seeing
this
value
in
you
know
everybody
being
able
to
leverage
this
capability.
If
the
networks
are
going
to
be
decentralized,
many
customers
might
be
using
Oracle,
but
some
may
be
using
other
environments
as
a
deployments,
and
so
you
want
to
make
sure
that
the
governance
capabilities
access
control
can
be
shared.
A
Doesn't
and
fault-tolerance
has
not
been
a
strong
requirement,
I
would
say
maybe
20%
or
so
from
what
we've
seen
so
far
in
many
cases
today,
they're
perfectly
happy
with
the
current
consensus
mechanisms,
Kafka
and
so
on,
but
we're
saying
there's
going
to
be.
You
know
that
number
is
going
to
be
growing
a
suspect
that
percentage
will
grow
as
we
get
more
production
deployments.