►
From YouTube: Lunch Keynote: Brian Behlendorf
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
Please
thank
you.
So
it's
really
my
pleasure
to
welcome
and
introduce
Brian.
So
Brian
is
the
executive
director
of
the
hyper
ledger
project,
so
we
heard
about
it
from
the
earlier
talks
already.
So
Brian
was
a
primary
developer
of
the
Apache
web
server.
I'm
sure
you
know
well,
most
of
you
know
buddy,
which
is
a
most
popular
web
server
software
on
the
internet
and
a
founding
member
of
the
Apache
Software
Foundation.
A
Your
source
Yaz's
also
served
on
the
board
of
the
Madelia
foundation
since
2003
and
the
Electronic
Frontier
Foundation
since
2013
Brian
was
a
founding
CTO
of
collabnet
and
CTO
of
the
world
economic
firm
and,
most
recently
he
was
a
managing
director
at
midfield,
Capital
Management,
a
global
technology,
investment,
firm,
I'm
sure
there
is
even
more
of
this,
so
Brian.
Thank.
B
B
And
and
I
apologize
I
didn't
present
slides,
because
what
I
wanted
to
kind
of
do
is
just
get
to
understand
the
audience
better.
Get
to
understand
the
other
content
this
this
morning
better
and
maybe
go
in
a
few
directions,
make
some
observations
and
then
open
it
up
for
some
QA.
It
sounds
like
lunch,
hopefully
we'll
be
here
by
about
noon.
So
it's
roughly
the
time
that
we're
looking
at
so
I
wanted
to
tell
you
before
I
dive
in
just
a
little
bit
more
about
what
we're
doing
at
hyper
ledger.
B
It
was
a
501,
C,
6
and
still
is
industry
consortium.
So
the
nonprofit
all
of
the
IP
created
at
the
organization,
has
to
be
published
and
put
out
into
into
the
into
the
public,
and
it's
allowed
to
convene
companies
together
and
others
together,
but
it
has
to
follow
very
strict,
like
everybody
does
antitrust
rules
to
make
sure
that
this
isn't
a
cabal.
B
That
sort
of
thing
and
help
solve
that
question
and
the
other
side
of
that
barbell
is
the
commercial
ecosystem
right?
How
do
you
keep
vendors
from
stepping
on
each
other's
toes
and
trying
to
take?
You
know
a
sense
of
a
proprietary
claim
on
what
is
otherwise
a
mark
that
is
managed
by
the
community
right
and
so
partly
that's
about
getting
developers
together
at
events
running
things
like
training
and
certification
processes
and
trademark
policies,
and
also
helping
the
market
help
them
understand.
B
The
premise
is:
if
there
aren't
people
who
are
going
to
show
up
and
write
your
code
under
this
model,
then
maybe
it
shouldn't
exist
right
rather
than
a
different
approach,
which
should
have
the
Linux
Foundation.
Trying
to
pay
the
salaries
of
everybody
writing
this
code
that
wouldn't
work
so
replicating
this
model
to
cloud
computing
to
Software,
Defined
Networking,
all
these
pieces
of
infrastructure
that
most
of
us
take
for
granted
and
we
should
take
for
granted
the
same
way
we
take
plumbing
and
HVAC
systems
for
granted.
The
same
way
we
take.
B
B
So,
a
year
and
a
half
ago,
a
bunch
of
different
companies
started
to
independently
come
to
the
Linux
Foundation,
saying,
there's
something
going
on
in
this
Bitcoin
space,
this
watch-chain
space
that
seems
like
it
could
use
you
guys
involved
and
after
determining
that
the
Bitcoin
community
itself
wasn't
really
the
right
kind
of
place
for
the
Linux
Foundation
to
play,
because
they
had
already
built
up.
You
know
foundations.
They
built
up
a
governance
process,
imperfect
as
it
is.
B
Sometimes
you
don't
want
to
step
into
a
community,
that's
already
underway
and
try
to
tell
them
what
to
do.
That
doesn't
tend
to
work
very
well,
but
there
was
a
recognition
that
underneath
Bitcoin
was
this
platform
that
the
Bitcoin
instance
of
which
was
one
way
of
doing
it.
But
there
are
different
ways
of
doing
it
right.
There
are
different
algorithms
for
consensus,
as
we've
talked
about
here,
proof
of
a
elapsed,
time,
Byzantine
fault,
tolerance,
etc.
There
might
even
be
different
forms
of
smart
contract
language.
B
The
Assyrian
community
really
advanced
beyond
Bitcoin
in
terms
of
thinking
about.
How
do
you
build
a
global,
unstoppable
computer
right
and
use
a
scripting
language
that
you
deploy
across
this
network?
That
would
embody
this
principle
of
smart
contracts
and
distributed
application.
Something
I
resonated
very
strongly
with
feeling.
Like
you
know,
the
last
10
years,
the
digitalization
of
society
have
been
largely
about
centralization,
rather
than
decentralization.
Why
can't
we
get
to
a
decentralized
uber
or
a
decentralized
air,
B&B
or
decentralized
eBay
right?
B
Why
is
it
so
much
easier
to
end
up
creating
central
monopolies
than
what
I
think
appealed
to
a
lot
of
us
when
we
first
got
on
the
net,
which
was
the
decentralized
kind
of
sovereign,
a
kind
of
point
of
view
right,
a
nature
of
the
Internet
at
the
time
so
I?
So
a
lot
of
these
threads
started
to
come
together
and
they
came
from
sources.
You
might
expect
from
companies
that
had
been
involved
in
open
source.
B
Big
companies
like
IBM
at
Intel,
small
companies
startups
in
the
syntek
space
like
digital
assets
like
intellect
EU,
like
others
who
were
starting
to
work
with
this
new
technology,
as
well
as
banks
who
you
know
they.
Many
of
them
were
using
Linux
deep
inside,
but
many
of
them
had
not
thought
of
themselves
as
open
source
companies,
most
of
which
still
don't,
and
so,
with
all
of
these
kind
of
threads
coming
together,
gyms
Emma
and
the
Linux
Foundation
who's,
the
executive
director.
There
said,
let's
launch
a
project.
B
What
should
we
call
it
and
I
think
it
was
digital
asset
that
it
acquired
a
company
called
hyper
ledger
so
two
years
ago,
there's
a
small
start-up
called
hyper
ledger
with
a
different
technology.
Forget
about
it,
here's
the
brand
and
it
was
generic.
It
was
kind
of
inspirational,
but
it
meant.
Okay,
that's
probably
a
good
good
brand
to
use
I
joined
anime
of
last
year
about
six
months
after
it
got
started
and
and
the
two
projects
that
were
in
a
hyper
ledger
at
the
time
were
fabric
and
sawtooth
and
you've
heard
both
of
them
mentioned.
B
Fabric
originally
was
built
inside
of
IBM,
as
kind
of
an
implementation
of
this
idea
of
a
distributed.
The
centralized
ledger
with
a
smart
contract
engine
on
top
using
the
go
opera
programming,
language
running
inside,
of
a
virtual
machine
container
called
docker
right
and
as
your
smart
contract
engine,
not
perfect,
not
beautiful,
not
necessarily
designed
for
public
networks.
The
way
the
etherium
technology
stack
is
but
an
interesting
start,
especially
if
your
focus
is
on
enabling
these
consortium
kind
of
settings
like
we've,
talked
about
and
then
Intel
brought
some
research
they've
been
doing
internally
around.
B
As
mentioned
the
proof
of
elapsed,
time,
consensus
mechanism
into
a
project,
Sawtooth
very
different
architecture,
written
in
Python
rather
than
and
go,
but
these
two
projects
that
that
hyper
literate
and
part
of
the
thinking
around
hyper,
ledger,
beget
and
continuously
has
been,
rather
than
trying
to
converge
us
all
on
one
single
chain.
The
way
that
say
if
they're,
a
more
Bitcoin
has
with
side
chains
for
scalability,
or
there
are
other
kind
of
things
instead
of
believing,
though,
that
this
was
like
tcp/ip,
where
everybody
had
to
be
running
the
same
protocol
instead.
B
Maybe
this
is
maybe
there's
a
different
vision
of
how
the
landscape
rolls
out,
which
is
one
of
a
lot
of
different
chains
out
there,
where
every
chain
is
essentially
a
pool
of
peers
and
I
mean
that,
in
the
terms
of
business,
peers
and
ecosystem
participants,
stakeholders
in
a
given
industry
or
in
a
given
set
of
use
cases
who
need
to
maintain
a
ledger
and
drive
some
automation
amongst
themselves,
but
when
they
want
to
go
to
other
domains,
what
they
simply
need,
our
bridges.
What
they
need
are
technologies
that
make
it
possible
to
conduct
two
transactions.
B
We
do
here,
a
single
County
where
the
different
nodes
are
like
real
estate
agents-
and
you
know
the
court
system
and
other
good
governance
organizations,
and
maybe
even
folks
who
represent
homeowners
right
or
it
could
be
as
large
as
the
as
an
entire
state
of
the
US
or
even
international.
The
scope
of
a
given
chain
will
be
something
that
industries
will
kind
of
define
and
the
larger
the
better
that
we
know.
B
Whoa
is
that
Metcalfe's
law,
the
the
value
of
a
network,
is
the
square
of
the
number
of
nodes
up
to
some
asymptote
right,
but
but
there's
also
good
reasons
for
chains
to
be
separate,
and
in
my
take
is
that
has
to
do
with
what
I've
been
calling
kind
of
a
spectrum
between
human
governance
and
algorithmic
governance
right.
B
The
enforcement
of
that
is,
we
send
you
to
jail
or
we
kick
you
out
of
the
project
or
we
apply
property
taxes,
or
we
simply
shun
you
right
and
that
sometimes
things
works
most
the
time
that
works
and
there's
there's
obviously
problems
where
it
places
where
it
doesn't
a
lot
of
which
have
to
do
with
imperfect
application
of
those
rules.
But
some
of
those
are
subjective
rules
anyways,
but
the
other
extreme
is
algorithmic
governance.
An
example
of
this
in
Bitcoin.
Is
you
cannot
spend
the
same
Bitcoin
token
twice
right?
B
That's
the
system
prevents
you
from
doing
that.
That's
called
preventing
of
double
spend
that
built
into
the
system.
You
try
to
propose
that
transaction.
The
system
rejects
it
right
that
algorithmic
governance
is
incredibly
useful
as
a
way
to
prevent
fraud.
There
are
other
rules.
We
could
apply
that
might
keep
you
from.
You
know,
selling
a
house
that
you
don't
own,
for
example,
or
trying
to
conduct
some
other
acts
that
you
don't
have
the
right.
Permissions
for
this
network
as
a
whole
can
reject
it.
B
B
This
way
we
could
encode
how
governments
work
this
way
we
could
encode
how
voting
works
this
way,
but
the
reality
is
that
for
any
system
there
will
be
somewhere
in
between
these
two
extremes,
the
more
we
can
automate
the
more
we
can
make
the
systems
able
to
enforce
these
rules,
the
more
scalable
we
get,
but
also,
frankly,
the
more
auditable
and
I,
hopefully
the
more
fair.
We
can
make
these
systems,
so
we
could
fight
fraud.
Maybe
we
could,
by
voting
machine
tampering,
maybe
we
and
voter
registration
problems
right.
Maybe
we
can
fight.
B
All
sorts
of
injustice
is
out
there
by
making
these
systems
more
automatable,
but
will
always
still
need
a
human
governance.
This
is
exhibited
by
perhaps,
but
some
of
you
didn't
follow
this,
but
last
year
there
was
in
the
etherium
community
something
called
the
DAO
hack,
which
was
an
instance
of
a
smart
contract.
B
I
won't
tell
the
whole
story,
but
as
a
decentralised
venture,
capital
fund
that
raised
at
the
time
would
seem
like
a
lot
of
money,
one
hundred
and
fifty
million
dollars
into
this
pool,
but
there
was
a
bug
in
the
code
that
was
used
to
build
this
and
that
bug
allowed
people
to
drain
that
that
fund
and
the
etherium
community
developers
and
miners
got
together
and
said.
This
is
wrong.
B
This
is
systemically
wrong
and
we
need
to
fix
this
and
we've
got
28
days,
which
was
a
characteristic
of
the
issue
to
fix
this
and
they
did
and
they
reversed
that
transaction.
Many
people
thought
that
was
the
wrong
thing
that
this
was
not
unlike
the
bailout
of
AIG
in
2008
or
the
bailout
of
long-term
capital
management
in
ninety
six
or
seven
come
to
correct
me
where
the
people
at
the
who
ran
the
commanding
heights
of
the
economy
said
this.
B
Other
group
of
people
took
a
bad
risk,
didn't
understand
what
they
were
doing
and
we
have
to
bail
them
out
and
that's
a
moral
hazard
kind
of
question,
but
in
almost
every
environment
we
have
to
have
systems
where
the
humans
can
eventually
can
at
time
stepped
in
and
correct.
What
is
an
illogical
extreme?
You
know
outcome
from
an
otherwise
logical
and
ordinary
set
of
rule
right.
So
this
is
the
question.
B
I
think
every
one
of
our
use
cases
will
have
to
address
whether
we're
talking
about
identities
for
displaced
people,
refugees
or
simply
the
homeless
populations
in
the
cities.
What
are
their
rights
around
privacy
and
identity,
matched
against
our
ability
to
audit?
You
know
how
many
immunizations
have
you
had
right
or
how
many
benefits
have
you
received?
There's
a
balancing
act
of
strike
and
and
in
the
social
impact,
applications
or
other
business
applications
we're
going
to
need
the
stakeholders
to
come
together
and
talk
and
figure
out.
B
How
do
we
define
that
balance
and
how
do
we
encode
that
and
rule
that
the
networks
can
enforce,
but
then
also
encoded
in
human
governance
layers
that
can
correct
those
rules
or
recover
from
those
rules
when
a
bug
is
found
right
and
so
part
of
my
appeal
to
folks
who
are
thinking
about
social
impact?
Applications
of
this
is
the
more
that
you
can
tie
these
things
into
the
places
where
the
systems
of
the
world
are
being
reinvented
right,
where
the
food
supply
chain
or
the
diamond
chain
or
the
tuna
fish
supply
chain
are
being
reinvented.
B
Through
these,
we
should
really
look
for
opportunities
to
embed
socially
aware
individuals
and
organizations,
civil
society
for
lack
of
better
word
in
the
design
and
implementation
of
these
chains,
rather
than
coming
in
after
the
fact,
and
trying
to
correct
it
and
trying
to
make
sure
that
the
right
data
is
being
tracked
and
also
that
the
right
observers
are
on
that
chain,
because
it
won't
happen
by
itself.
For
the
good
news,
in
my
view,
is
that
the
right
kind
of
folks
are
paying
attention
to
this.
B
B
A
much
different
view
than
regulators
as
people
who
run
after
the
fact
and
chase
the
bar
to
the
horse
after
its
left,
the
barn,
the
UN
and
the
World
Food
Program
and
the
World
Economic
Forum
all
have
this
buzz
about
it
right,
which
a
lot
of
which
is
ill-informed,
a
lot
of
which
will
be.
You
know
yes,
that
use
case
will
not
need
a
blockchain.
Yes,
there's
a
lot
of
craziness,
and
maybe
it's
about
Bitcoin,
maybe
not
who
knows,
but
I'm,
really
encouraged
by
the
fact.
B
B
What
we're
trying
to
do
at
hyper
ledger
is
build
the
building
blocks
of
these
things
and
build
the
underlying
tools,
but
ultimately
it's
going
to
be
up
to
the
people
operating
in
these
spaces,
to
figure
out
how
to
use
them
to
to
reinvent
the
the
systems
that
they're
embedded
in
so
with
that
I
think
we
have
another
10
minutes
before
food
shows
up.
Hopefully.
D
B
It
takes
more
than
a
one-day
decision
for
a
entire
nation
to
issue
fiat
currency
on
a
chain.
The
Monetary
Authority
of
Singapore
mas
has
been
very
Pro
innovation
for
a
while
and
in
the
last
two
years,
very
Pro.
Let's
figure
out
how
to
use
these
distributed
Ledger's
to
make
something
work.
So
they
have.
You
know
a
lab
where
I
think
I
think
vitalik
has
a
desk
at
the
MAS.
Actually
Vitalik
do
Terra
News
one
of
the
founders
of
etherium
I.
Don't
know
how
often
use
there,
but
it's
like
it's
like
you
know.
B
You
have
a
city,
it's
nice
have
a
desk,
and-
and
so
one
thing
that
they've
been
doing
is
creating
these
sandbox
environments,
where,
like
they'll,
say,
here's
a
real
live
data
set.
Here's
a
live
set
of
partners
who
want
to
try
to
wire
their
systems
together,
and
so
innovative
companies
and
technologists
can
come
in
and
hack
and
figure
out.
B
How
do
we
plug
these
things
together
and
all
of
Asia
frankly
has
gotten
meet
up
crazy
and
hackathon,
crazy
and
FinTech
crazy,
which
is
which
is
really
interesting,
like
the
appetite
for
working
with
nerds
like
us,
is
like
really
rewarding
and
fun
now.
Some
of
that
is
driven
by
this
sense
that
you
know
these
cryptocurrencies
are
going
to
insane
places
and
I
would
have
I
would
encourage
many
of
us
to
have
a
skeptical
view
of
that,
because
that's
not
the
value.
That's
inherent
in
blockchain
technology
is
going
well
bitcoins
2500.
B
Well,
who
says
it
won't
go
to
100k
right.
That
almost
has
nothing
to
do
with
the
value
of
underlying
distributed
ledger
systems.
The
value
does
come
from
reinventing
these
systems
of
the
world.
It
might
come
from
many
of
the
existing
fiat
currencies
becoming
digital
currencies
becoming
as
fluid
as
Bitcoin
is
in
terms
of
being
able
to
move
it
around
the
world
or
move
it
between
partners
with
zero
or
close
to
zero
transaction
cost.
I'll
note
that
Bitcoin
has
a
$3
transaction
cost
now,
because
it's
price
is
so
high.
B
The
fancy
term
is
Center
of
competency,
build
a
center
of
competency
internally
around
where
this
tech
is
heading
and
recognizing
there's
still
a
lot
of
unsettled
questions
about
architecture
and
smart
contract,
language
and
stuff
they're,
realizing,
there's
a
core
set
of
skills
that
they
need
to
build
to
persevere,
no
matter
what
happens
on
the
technology
landscape.
So
that's
all
pretty
exciting.
E
B
Often
I
prefer
the
word
minimum
viable
centralization.
It's
like,
like
the
DNS,
for
example,
right
the
domain
name
system
who
considers
it
centralized
right,
it's
okay,
any
of
us
may
feel
like.
We
don't
really
know
where
to
stand
on
that
question
right.
Dns
is
fairly
decentralized.
There
are
hundreds
of
different
registrar's
out
there
who
can
register
a
domain
name
for
you.
Who've
all
agreed
with
an
organization
can
to
a
set
of
rules
around.
B
You
know:
I
can't
issue
MacDonald
comm
right
to
this
person,
just
because
they're
claiming
to
be
McDonald's
right,
that's
already
registered
to
somebody
else.
So
the
registrar's
have
an
agreement
between
themselves
managed
by
ICANN
international
org,
but
no
one
would
say:
I
can
control
the
internet
or
even
I
can
controls
DNS,
because
much
of
that
isn't
federated
out
right.
So,
by
the
same
token,
a
lot
of
these
consortium
chains
I
mean
you
can
even
argue
that
a
lot
of
the
as
you
are
that
sounds
like
that.
B
The
public
crypto
currencies,
the
public
chains-
are
theirs,
they're,
decentralized
to
a
degree,
but
you
know
the
Bitcoin
developers
and
miners
could
wake
up
one
day
and
decide.
You
know.
Maybe
it's
worth
changing
the
algorithms
to
add
a
whole
bunch
of
new
tokens
right
to
change
that
as
they
do
to
change
the
the
repayment.
What
is
the
repayment
rate
or
the
nature.
B
In
the
rewards
for
finding
a
block
right,
it's
what
is
it
they
head
to
happening
right?
They
have
it
every
time
because
they
want
it
to
get
it
low,
but
that's
a
policy
decision
and
that's
a
policy
decision
not
unlike
the
Federal
Reserve
Bank.
You
know
which
today,
if
you
want,
if
the
Federal
Reserve
Bank
changes
the
interest
rates,
they
can
only
do
that
by
unanimous
consent
of
the
twelve
independent
Federal
Reserve
Bank
governor's.
B
These
underlying
thing.
It's.
Why
I
go
back
to
this
notion
of
don't
think
of
one
global
network
with
one
solid
technology
where
it
takes
where
one
one
one
monolithic,
I,
think
the
term
is
use
technology
that
I
mean
kind
of
like
tcp/ip
s
ubiquitous,
but
then
can
take
twenty
five
years
to
do
a
major
upgrade
to
like
we
had
with
ipv4
to
v6
right
sorry,
John
I'm,
don't
mean
to
look
at
you.
B
That's
what
I
can
say
that
instead
I
think
we're
going
to
see
a
panoply
of
different
technologies,
a
lot
of
different
chains,
and
that
will
give
us
the
ability
to
both
map
it
to
our
human
governance
models
to
the
different
criteria
we'll
have
for
a
global
distribution
versus
high
transaction
rate
versus
close
in
finality
all
of
these
parameters.
We
don't
even
yet
know
what
the
terminology
are
yet,
but
we
know
that
there
are
choices
we'll
be
making
as
we
deploy
these
and
really.
B
The
standards
I
think
will
come
from
cross-chain
technologies
for
conducting
transactions
for
moving
assets
from
tane
to
chain
and
also
for
identity
and
the
ability
to
tie
to
know
if
I'm
going
to
buy
a
house
from
you,
and
that
means
conducting
transactions
on
multiple
chains.
I
want
to
know
I'm
dealing
with
the
same
person
right.
So
that's
where
some
of
this
identity
work
will
probably
be
more
cross
chain
than
than
any
other
thing.
I
went
well
beyond
your
question:
I'm,
sorry
about
that,
but
yeah
minimum
viable
centralization
and.
D
C
That
terminology
I'm
I,
really
appreciated
your
comment
that
we
need
to
have
more
socially
aware
people
developing
distributed
ledger
and
blockchain
technology.
I
am
an
academic
myself.
What
role
do
you
think
academia
can
play,
and
also
centers,
like
citrus
from
the
Bonato
Institute,
and
helping
to
create
next
generation
of
technologists?
Who
come
with
that
socially
aware
consciousness?
It.
B
Means
having
its
stuff
right,
it
means
having
theirs
as
Holy,
Trinity
or
unholy
trinity,
perhaps
between
the
business
world,
the
legal
world
and
the
programming
world
right,
and
we
already
know
that
the
world
of
smart
contracts
means
the
programmers
need
to
think
more.
Like
lawyers
and
the
lawyers
need
to
think
more
like
programmers
right
now,
I
would
add
into
there
that
having
some
business
sense
as
well
is
important
to
understand
it,
what's
the
shape
of
a
market
and
how
could
that
evolve
or
should
that
evolve?
B
I
think
the
right
nexus
point
for
these
are
industry
associations
right
in
the
healthcare
sector,
for
example,
there's
an
organization
called
hims
HIMSS,
which
for
health
IT,
is
largely
defined
kind
of
the
terminology
and
the
standards
that
communities
use
to
interchange,
data
and
necessarily
those
conversations
are
partly
legal,
partly
political,
partly
business
and
partly
technical
right.
I
I
know
that
in
every
other
industry
you
know
take
the
most
boring
or
the
most
they're,
the
biggest
there
is
an
industry
association
somewhere
there.
B
That's
probably
in
the
right
position
to
go
here
is
where
the
industry
is
changing
and
here's
where
we
can
pull
the
stakeholders
together
to
make
those
changes.
So
if
academia
could
identify
those
and
maybe
partnership
and
partner
between
you
know,
organization
like
citrus
and
the
Haas
School
of
Business
right
I,
you
know
to
go
in
with
both
a
social
impact
and
a
business
voice
that
could
be
pretty
powerful.
It.
G
Thanks
speaking
of
the
Haas
School
of
Business
I'm
David
Harrison
with
the
hospital
business
expert,
I
am
really
interested
in
something
that
you
touched
on
a
king
bit,
elections
and
blockchain,
and
if
you
think
that
blockchain
is
an
appropriate
tool
for
providing
auditable
elections
or
will
be
at
some
point
in
the
future,
I
know.
There's
identity
challenge
around
that
I
know
democracy.
Earth
is
one
group
that's
trying
to
work
on
this,
but
I'm
curious.
If
you
or
your
organization's
are
working
on
that
issue
or
thinking
talking
about
it.
So.
B
B
You
know
the
president,
that
sort
of
thing,
because
there
are
interesting
side,
things
like
liquid
democracy,
worth
taking
a
look,
but
it
may
be
a
while
before
those
two
replace
many
of
the
voting
systems
we
have
today
and
for
me
there
is
a
crisis
in
today's
voting
systems,
one
of
growing
lack
of
confidence
in
the
fairness
of
how
we
conduct
elections
and
and
it's
it's
it's
civilization,
threatening.
If
we
don't
solve
this,
we
go
down
certain
dark
paths
like
Zimbabwe
and
Venezuela,
so
I,
here's.
B
What
I
would
not
do
I
would
not
use
it
as
a
direct
replacement
for
voting
itself.
I've
been
a
really
big
believer
in
voter
verified
paper
ballots
as
the
true
path
to
auditable
elections
right
and
there
are
certain
shortcuts,
like
vote
by
mail
and
vote
by
mail
works,
because
the
citizens
in
Oregon
happen
to
trust
the
Oregon
government
enough
not
to
corrupt
the
process
and
okay.
We
can
accept
that.
But
let's
use
that's
recognize
that
as
a
special
case
and
the
main
case
should
be.
B
You
show
up
somewhere
or
you
mail
it
in
as
a
net
as
a
decider
group,
but
you
show
up
somewhere,
you
mark
your
ballot,
maybe
it's
a
computer
that
helps
you
mark
that
ballot,
but
you
read
that
ballot,
you
submit
it
to
a
scanner
and
it's
kept
in
physical
form,
where
one
out
of
hopefully
ten
but
probably
more
like
one
out
of
hundred
get
actually
reviewed
and
audited.
There's
no
role
for
a
blockchain
in
that
process,
where
I
do
think
there
is
a
role
for
for
distributed.
B
Ledger's
is
in
the
part
that
happens
before
in
the
registration
process.
We
do
have
a
crisis
here
where
a
lot
of
people,
especially
in
certain
contested
states,
are
showing
up
and
being
told
you
don't
have
the
right
to
vote
here
right
here
at
either
the
wrong
polling,
place
or
I.
Don't
have
you
on
lived
or
you
know,
or
it's
hard
for
them
to
even
understand
where
to
register
in
the
first
place.
B
There's
a
lot
of
that
part
of
the
problem
that
we
can't
solve,
but
one
thing
we
could
solve
is
having
a
public
registry
a
public
distributed,
ledger
that
tracked
where
everyone
was
registered
to
vote.
We
have
to
manage
some
privacy
concerns
because
we
don't
want
our
home
addresses
necessarily
part
of
the
public
record,
but
it
should
be
public
where
me,
as
a
citizen
where
I'm
allowed
to
vote
and
questions
about
identity
at
cetera.
So
we
could
hopefully
address
that.
But
more
interestingly
to
me
is
the
part
that
comes
after
voting.
B
Where
you
know
you
go
to
a
polling
place.
Maybe
it's
somebody's
garage,
maybe
it's
a
church.
Whatever
you
put
your
ballots
in-
and
you
presume
that
at
that
point
those
ballots
are
counted
and
then
reported
up
into
the
totals
and
and
on
that
doesn't
always
happen,
and
we
have
no
way
independently
of
being
able
to
verify
other
than
we
host.
B
Is
the
person
running
that
polling,
place,
drove
to
the
right
location
and
didn't
have
an
accident
on
the
way
or
their
car
didn't
catch
on
fire
or
something
like
that
right
and
the
ballasts
didn't
get
lost
or
some
new
box
about
didn't
show
up.
Imagine
if,
instead,
every
polling
place
recorded
its
totals
to
a
distributed,
ledger
right.
This
many
votes
for
this
candidate
as
many
votes
for
this,
not
the
individual
votes,
certainly
not
like
the
serial
number
on
the
receipt
that
you
got
but
the
totals.
Then
you
could
go
okay.
B
This
person's
garage
that
I
voted
at
I
saw
a
few
other
people
there.
They
reported
500
votes
over
the
course
of
the
day
that
feels
about
right
and
good
I,
see
it
recorded
in
there,
and
now
it's
summed
up
like
an
Excel
spreadsheet
into
the
total
and
I
can
trust.
That's
now
included
in
the
total,
but
if
you
see
a
like
that,
your
polling
place
wasn't
included
or
that
garage
claimed
to
have
recorded
12,000
votes
or
something
like
that,
then
that's
the
basis
for
going
something
funny
is
going
on
here.
B
H
I
B
Law
and
legal
systems
III,
you
know
everything
exists
in
a
context
of
law
and
legal
systems,
but
I
think
that
has
to
do
with
the
application
of
these
technologies,
rather
than
so
to
make
it
programmable
right.
We're
building
something
at
kind
of
the
operating
system,
layer,
the
kernel
layer
right
or
is
a
programming
language
layer.
What
we're
specifically
not
about
is
a
coin
right
and
so,
or
a
system
of
governance
right.
B
Whatever
governance
rules,
you
feel
should
be
algorithmic,
but
then
a
legal
contract
between
parties
that
reflects
whatever
the
human
governance
relationship
is
right
and
and
so
like
when
we
think
about
the
application
of
this
in
healthcare
right,
there's
nothing
that
really
needs
to
drive
into
the
code,
something
that
makes
it
aware
that
it's
being
used
in
the
healthcare
context.
There
might
be
big
privacy
issues.
You
have
to
be
able
to
handle
big
other
other
requirements
that
are
systemic,
but
but
we
kind
of
leave
laws
and
exercise
to
the
reader,
and
that
seems
to
work.
B
I,
don't
know
if
there's
something
more,
we
could
or
should
be
doing,
certainly
I
think
the
GDP
our
rules
are
something
that's
been
on.
My
mind
is
one
reason
we
launched
the
indie
project,
which
is
a
path
to
being
able
to
implement.
These
are
the
privacy
rules
coming
into
effect
in
the
European
Union
starting
in
May
of
next
year,
which
means
that
any
one
of
you
who,
if
you're
a
European
citizen
you'll,
have
the
right
to
demand
any
company
us
European,
Union
or
not
to
be
able
to?
B
Let
you
know
what
data
they
have
about
you,
let
you
be
able
to
inspect
it,
let
you
be
able
to
correct
it
and
let
you
be
able
to
delete
it
if
it's
no
longer
needed
by
the
business
and
no
longer
needed
is
a
question,
but
all
of
these
are
great
from
the
end
perspective
of
the
end
user
and
horrible
from
the
perspective
of
the
company.
If
you
don't
have
a
system
like
what's
being
builded
Indy,
it
becomes
a
lot
more
complicated
and
hopefully
what
it
means
for
you
all
is
you'll.
H
B
F
The
table
does
arrive.
Maybe
this
is
moot,
but
I
won't
keep
you
from
lunch.
I
promise
just
you
kind
of
mentioned
a
little
bit
about
the
hard
fortune,
etherium
and
I
just
kind
of
want
to
hear
your
thoughts
on
error
correction.
You
know
how
we
go
about
dealing
with
errors
that
get
put
on
to
these
systems.
B
Okay,
so
all
software
has
bugs
right.
The
last
I
subscribe
very
strongly
to
the
belief
the
last
defect
is
fixed
in
the
last
user
is
dead
and
I
had
to
stop
telling
that
joke
when
I
got
involved
in
health
IT
for
a
little
while
that
very
different
implications
there,
but
I
yeah.
We
have
to
have
a
mechanism
for
correction,
and
it's
good
that
it's
hard
to
make
really
important.
B
Corrections,
like
the
big
debate
happening
in
the
Bitcoin
community,
between
seg,
wits
and
and
2mb
I
mean,
like
all
the
different
approaches
to
evolving
that
it's
good,
that
that's
not
a
frivolous
decision
right,
because
these
decisions
do
have
ramifications
and
they
do
benefit
some
parties
over
others
right.
But
we
need
the
ability
to
make
to
have
those
conversations
and
make
those
changes
and
whether
those
are
non
disruptive
changes
that
are
soft
Forks.
B
You
know
that
everybody
can
come
along
and
and
and
those
who
disagree
well,
at
least
they
you
know,
don't
don't
veto
the
decision
whatever,
then
that's
great.
If
there
are
hard
Forks
and
I
mean
this
is
like
in
any
human
consensus
driving
process.
If
you
have,
if
everybody
in
this
room
was
trying
to
decide,
you
know
we're
going
to
order
a
hundred
pizzas,
and
you
know
we
can
only
get
them
of
one
type
right.
Do
we
all
agree
that
it's
pepperoni
or
do
we
go?
B
H
You
hi,
you
just
said
something
very
interesting
before
about
being
the
technology,
one
of
the
things
that
how
are
we
going
to
police
making
sure
people
update
the
soft,
because,
if
you're
delivering
software
and
yes,
you
can
have
the
fork
in
terms
of
a
Tyrian,
but
if
they
don't
do
the
bug
fixing?
What
are
we
going
to
do
about
that?
Because
only
one
software
in
reality
at
this
point
so.