►
Description
Our co-founder Alex Skidanov (https://twitter.com/AlexSkidanov) in conversation with Ryan King (https://twitter.com/frothcity) from FOAM (https://twitter.com/foamspace) during the New York Blockchain Week discussing FOAM's decentralized location services based on proof of location.
Location: FOAM HQ, New York, NY
Download the video here: https://near.ai/41513
------- About Whiteboard Series ---------
The Whiteboard Series with NEAR is a set of video interviews with Founders who are building scalable solutions for Blockchain. We deep dive into their story, what they're developing and how it is helping the ecosystem.
Follow the latest from NEAR Protocol on,
Discord: https://near.ai/discord
Medium: https://near.ai/medium
Twitter: https://near.ai/twitter
GitHub: https://near.ai/github
#WhiteboardSeries #FutureIsNEAR #Consensus2019
A
B
My
name
is
Ryan
I'm,
a
co-founder
and
CEO
foam
foam
is
a
protocol
for
building
a
consensus
driven
map
of
the
world.
So
we
build
a
lot
of
geospatial
tools
and
features
for
blockchain
developers
to
work
on
if
they
want
to
have
a
geolocation
in
their
smart
contract
or
way
to
visualize
it,
and
then
we
use
our
token
to
kind
of
verify
and
curate
this
data
to
make
sure
it's
accurate
and
can
be
relied
on
by
smart
contracts
yeah.
So
how
should
I
start.
B
Carrier
industry
sure
looking
good
yeah,
we
so
the
currently
what's
live
for
the
foam
product
is
called
the
foam
map.
You
can
find
it
at
basically
map
dot,
foam
dot
space,
and
this
is
a
live
web
3
application.
You
need,
like
meta,
mask
or
foam
to
basically
interact
with
it,
and
the
point
of
this
application
is
to
build
up
a
token
curated
registry
of
verified
points
of
interest
data.
So
currently,
today
points
of
interest.
B
Data
goes
out
of
style
really
quickly,
because
the
business
might
have
closed
or
moved
and
there's
a
lot
of
open
source
mappers
with
like
Open
Street
Map
contributing
data,
but
there's
no
one
really
checking
or
verifying
that
data,
and
so
a
company
has
to
then
do
it,
and
then
they
sell
this
open
source
source
data.
So
we're
trying
to
involve
an
incentive
mechanism
to
actually
curate
the
data.
So
that's
why
it's
called
a
token
curated
registry.
B
So
the
kind
of
primitive
is
called
the
TCR,
and
the
idea
is
that
you
have
this
registry,
which
is
a
list
of
things
so
normally
like
it
would
be
like
top
movies,
and
the
idea
is
to
get
on
this
list.
Somebody
has
two
staked
oaken's
for
you
to
get
onto
the
list
and
so
there's
all
these
different
parameters
around
a
token
curated
registry
and
these
kind
of
parameters
are
like
min
deposit.
B
So
like
what
would
be
the
minimum
amount
of
tokens,
you
have
to
stake
to
add
this
information
to
the
list,
and
so
the
reason
you
have
this
min
deposit
is
it's
like
an
anti-spam
mechanism.
So
you
have
to
have
some
skin
in
the
game,
to
like
add
information
to
the
list,
and
it's
also
so
you
make
sure
that
information
stays
up-to-date,
because
if
it's
out
somebody
can
come
and
challenge
you
so
in
our
case
we're
building
a
registry
of
locations.
B
So
it's
about
building
up
a
map
of
things,
and
so
the
way
that
it
works
is,
if
you
want
to
add
a
point
to
the
map.
Let's
say
I
want
to
add
Times
Square
of
New
York
City.
It's
similar
interface
to
like
a
Google,
Maps
or
open
trip
map,
but
as
a
web
3
application,
you
basically
can
open
it
up
and
you
say:
ok,
I
want
to
click
here
on
the
map
and
I'm
gonna,
add
a
point
and
essentially
like
a
menu
card
will
come
up
and
you
essentially
fill
out
the
name
at
the
information.
B
The
phone
number
the
tag,
and
then
you
decide
how
many
tokens
you
want
a
stake
and
right
now
the
minimum
is
50.
So
let's
say
you
say
50,
and
so
you
don't
actually
spend
your
tokens
or
your
money.
It
essentially
goes
into
the
protocol
and
it
goes
into
the
map
and
your
tokens
are
kind
of
backing
the
truthfulness
of
this
data.
So
people
can
say:
oh
there's
like
tokens
behind
this
and
it
looks
pretty
accurate,
so
I'm
going
to
be
able
to
trust
this
data.
A
B
So
foam
tokens
are
used
to
contribute
work
to
the
network.
So
if
you
want
to
be
a
cartographer,
you're,
basically
staking
tokens
and
they're
very
slowly,
getting
taken
off
the
market
and
actually
backing
the
data
set
of
the
map.
But
we've
seen
about
5%
of
the
points
so
far
I've
actually
been
challenged.
So
this
is
another
feature
of
the
token
curated
registry
is
that
you
have
the
contributors
which
we
call
cartographers,
but
then
you
also
have
the
curators
and
curators
or
people
who
actually
come
and
investigate
this
information
and
see
if
it's,
okay
or
not.
B
A
B
A
theorem
it's
on
the
live
aetherium,
and
so
the
metadata
is
stored
in
ipfs
and
there's
like
some
hash
and
then
basically,
it's
in
the
registry,
so
they'll
be
like
a
listing
hash
for
it
in
the
phone
registry
and
it's
on
aetherium.
And
so,
if
this
point
you
know
three
days
go
by,
then
the
point
will
turn
green,
which
means
nobody
found
a
problem,
but
at
anytime
somebody
can
come
and
initiate
a
challenge
against
any
point
on
the
map.
B
B
So
now
they
also
take
50
tokens,
and
so
now
what
that
gives
you
is
you
have
this
kind
of
reward
pool
of
100
tokens
that
are
now
locked
up
and
then
it's
open
for
all
other
cartographers
and
curators
to
kind
of
weigh
in
I'm
kind
of
like
a
vote,
and
it's
kind
of
deliberately
that
you
know
the
minimum
is
kind
of
low.
It's
like
enough
that
you're
not
going
to
spam
the
map,
but
it's
not
that
high.
That
now,
like
the
incentives
to
go
after
points,
are
also
meant
to
be
kind
of
lower
staked.
A
B
So
you
can't
challenge
more
than
an
estate
in
it,
but
if
the
point
has
a
thousand,
which
many
do
you
can
stake
a
thousand,
so
then,
in
that
case
the
reward
pool
is
2,000
great
and
so
then,
basically,
what
we
have,
then
is
a
five
days
period,
and
this
is
another
parameter.
So,
just
like
the
minimum
is
fifty.
How
long
is
the
voting
period
is
another
parameter,
so
these
are
all
can
be
voted
on
to
change
the
parameters
currently,
but
right
now
you
have
a
five-day
period
and
it's
then
open
for
a
vote.
B
We're
all
cartographers
can
kind
of
weigh
in
and
vote
cast
their
votes
and
we're
using
partial
lock.
Can
it
reveal?
So
that
means
that
the
votes
are
hidden.
Nobody
knows
which
side
you
voted
for,
and
so
after
this
five-day
period,
there's
a
three
day
reveal
period
and
that's
when
people
have
to
manually
essentially
reveal
their
votes
and
I'd
say
that
guy
revealed
that
guy
revealed,
maybe
that
guy
forgot
to.
B
B
Times
Square
should
look
this
way,
even
though
it's
not
essentially
because
the
channel
fails
60%
of
this
50
go
to
here
and
right
now,
40%
go
to
the
voters
who
are
on
the
right
side,
and
so
basically
he
had
to
put
his
tokens
at
stake
to
initiate
the
challenge.
People
voted
and
in
this
case
the
Challenger
lost.
What.
B
B
Of
the
reward
fool
goes
to
each
challenger
and
40%
of
the
ward
pool,
so
we
just
had
a
point
basically
where
the
point
had
27,000
right
and
the
Challenger
came,
and
this
was
just
last
week.
So
that's
actually
starting
to
be
a
decent
amount
of
money.
Maybe
like
a
thousand
bucks
for
800
books,
they
came
and
they
also
locked
27,000.
So
now
we
have
this
like
pretty
big
reward,
pool
right
and
the
Challenger
lost.
So
then
you
actually
see
when
you
go
back
to
this
point.
It
no
longer
says
27,000.
B
B
So
if
the
point
let's
say
has
27,000
and
you
see
an
error
and
you
don't
have
27,000,
you
can
stake
only
50,
and
so
then
that
takes
only
50
of
these,
so
the
reward
pulls
still
hundred
and
it
will
show
that
the
point
then
has
26,000
something
and
during
the
vote
the
person
could
withdraw
up
to
leave
behind
50.
So
you
guys
have
to
leave
50.
So
if
they
weren't
challenged
for
the
full
amount,
they
could
use
those
tokens
in
the
vote
right,
but.
B
A
B
A
B
So
you
can
do
things
like
punishing
the
wrong
side
of
the
vote
as
well,
but
I've
dated
them
state
they're
not
punished
if
you're
on
the
wrong
side
of
the
voice
you
right
now,
but
the
reason
that's
also
challenging
is
because
with
the
partial
lock
can
MIT
reveal,
you
can
use
the
same
voting
credits
across
multiple
polls.
If
you
start
getting
punished
all
of
a
sudden
you're
at
this
one
and
you're
now
like
in
debt,
because
so
there's
a
lot
of
unsolved
problems
around
how
to
punish
the
voters.
A
B
No
because
the
way
it
works
is
you
can
lock
tokens
in
the
voting
contract
and
then
you
get
these
voting
credits
and
you
can
use
the
same
credits
across
multiple
votes.
I
see.
So
then,
if
you're
getting
punished
across
them
and
then
all
of
a
sudden
you
only
locked
up
27,000,
you
use
27,000
across
ten
votes.
Now
you
like
lost
in
ten
votes.
Now
you
like
lost
more
than
you
put
in
so
I
kind
of
need,
a
different
voting
module
or
a
clever
way
to
avoid
this
kind
of
edge
case.
But.
A
Now
another
question:
so
let's
say
that
the
point
of
interest
is
some
coffee
shop
on
an
intersection
of
the
streets
yeah.
So
in
general
it's
a
very,
very
small
percentage
of
the
cartographers
is,
can
properly
validate
the
exams
of
the
coffee
shop
and
unless
they
go
to
Yelp
but
I
guess
the
entire
point.
Yeah.
C
B
I
mean
it's
kind
of
like
we've,
seen
versions
of
this
already,
it's
kind
of
like
a
grief
attack
and,
of
course,
if
you
have
a
lot
of
tokens,
you
can
kind
of
run
the
show
you
know
with
sense,
but
this
just
happened
in
a
way
where
kind
of
activated
the
community
more
than
ever,
we're
kind
of,
like
factions
formed
of
people
to
kind
of
like
vote,
instant
and
also
things
then
like
the
parameters
come
into
play.
So
we
have
governance.
So
let's
say:
there's
a
guy
with
Auto
tokens
challenging
all
these
coffee
shops
for
fun.
B
Well,
we
can
say:
let's
have
the
challenge
period
last
a
month,
so
we
have
now
a
month
to
rally
voters
or
say
change
the
voting
quorum
to
90%
so
that
it's
much
easier
to
defeat
the
Challenger.
So
because,
there's
all
these
like
levers
that
you
can
like
change
in
the
TCR,
you
can
like
react
to
that
and
because
the
stalkers
have
to
be
locked
for
a
period
of
time,
there's
like
a
period
of
time
for
people
to
see
what's
happening
and
react
to
it
and
the.
B
Per
vote,
it's
kind
of
been
varying
based
on
the
challenge
and
the
parameters,
but
I
couldn't
say
we
had
like
400,000
tokens
in
the
voting
contract
and
then
it
went
to
800,000.
Then
it
was
1
million,
2
million
and
then
for
this
contentious
vote.
We
had
over
eight
million
tokens
in
the
voting
contracts.
Over
8
million
tokens
in
the
voting
contract
was
the
total
number
of
tokens
in
the.
A
B
Yeah
so,
basically
now
we
have
people
debating
these
kind
of
parameters
of
like
how
we
should
change
them
and
coming
up
with
like
different
KPI
metrics.
So
if
we
wanted
to
change
them
in
deposit,
we
need
to
really
take
into
consideration.
What's
the
gas
cost
of
voting,
because
we
can't
have
the
min
deposit
mean
that
if
you
vote
the
rewards
so
small
that
you
lose
money
with
gas,
so
we're
kind
of
like
trying
to
look
at
all
these
KPIs
and
analysis,
and
we
also
like
have
people
building
on
top
of
the
map.
B
Now
so
building
like
analytics
tools
and
like
curator
dashboards,
to
see
how
people
are
voting
more
intricately,
and
we
just
let
out
like
a
grant
program,
so
people
could
build
on
our
API
and
so
hopefully,
more
comes
out
of
the
map
and
new
specifications
will
come
of
how
to
improve
the
incentive
models
or
make
it
more
challenging
to
be
a
challenger
or
you
know,
harass
the
map
in
some
sense.
So
that's
kind
of
what
is
live
now
and
yeah.
We're
working
on
more
broad.
B
Is
this
proof
of
location
to
verify
things
in
real
time
with
these
radios?
So
we
can,
you
know,
transition
to
that,
but
if
you
had
a
system
like
that,
you
would
be
able
to
actually
in
this
system
enforce
that
you're
not
able
to
challenge
a
point
in
San
Francisco.
If
you
can't
prove
you
were
in
in
San
Francisco
recently
or
that
you
can't
vote
on
these
points.
If
you
can't
show
you
like,
don't
live
here.
B
B
If
you
find
an
error
and
you
can
challenge
it
and
succeed,
you're
winning
tokens
so
right
now,
there
is
no
incentive
to
really
check
any
of
these
maps,
and
so
the
incentives
are
a
little
bit
less
on
the
actually
cartographer
side,
because
you
don't
get
direct
payouts
for
just
simply
adding
a
point.
You
can
earn
tokens
if
your
point
survives
a
challenge.
So
if
you
are
confident
in
your
point,
you
might
survive
and
we've
seen
this
point
got.
B
You
know
almost
half
their
tokens
just
added
to
it
just
by
existing,
but
we
kind
of
found
that
because
it's
ultimately
subjective
work,
having
like
an
inflation
or
mining
for
adding
points,
is
very
game
Abul
because
it's
not
computational,
it's
not
objective.
So
it's
kind
of
hard
to
have
this
like
a
infinite
inflation.
Just
for
adding
points.
We
think
it
actually
would
make
the
incentives
more
perverted
than
they
are
today
right.
All.
A
B
So
we
had
a
challenger
that
was
actually
started
being
very
first,
very
nitpicky.
So,
like
oh,
it's
lower
case
instead
of
upper
case,
and
then
they
just
started
lying
and
that's
when
it
kind
of
activated
the
people
to
rally
against
them.
And
at
that
point
they
were
like
the
whale.
So
they
could
easily
win
the
votes
but
because
they
like
we're,
challenging
falsely
so
many
people
then
got
involved
to
like
fight
back.
It's.
B
A
B
And
there's
about
like
30-something
of
them
right
and
so
the
way
GPS
works
is
that
they
use
a
time
synchronization
protocol
to
synchronize
all
of
their
clocks,
and
there
are
these
stations
on
the
earth
that
will
synchronize
them
for
you
and
then
the
GPS
is
simply
just
broadcasting
out
to
earth
all
the
time.
What
time
it
is
right,
that's
the
sole
purpose
of
GPS,
and
so
us
on
the
ground
as
a
receiving
it
standing
here,
and
you
hold
your
phone.
B
If
you
can
get
at
least
four
signals,
your
phone
can
do
trilateration,
which
is
a
way
it's
like
a
form
of
triangulation
that
it
can
actually
through
knowing
all
the
time
figure
out
the
distance,
the
differentiation
between
the
how
long
it
took
to
get
all
the
signals
and
can
then
figure
out
your
location.
So
that's
currently
how
it
works.
Today,
when
you
pull
out
your
phone
and
you're
waiting
for
this
blue
dot
to
appear
it's
listening
to
at
least
four
radios
right
and
everyone
would
agree,
GPS
works
pretty
good
right
yeah.
B
You
can
figure
out
how
to
get
where
you're
going,
but
this
is
only
a
one-way
communication.
So,
as
the
receiver,
you
don't
speak
back
to
GPS
GPS
has
no
idea
where
you
are
right.
It's
all
assuming
is
broadcasting
the
time,
and
so
we
actually
find
is
that
because
of
that,
it's
extremely
easy
to
lie
about
your
location,
because
there's
no
like
history
or
place
where
these,
like
your
location,
exists
other
than
like
your
apps
logging,
it
after
the
fact.
B
B
So
this
kind
of,
like
whole
market
of
location
kind
of
activity
is
suppressed
because
there's
no
way
to
verify
the
data
and
when
you
look
at
things
like
uber,
for
example,
they're
spending,
hundreds
of
millions
to
try
to
prevent
the
GPS
spoofing,
because
the
drivers,
if
they
make
the
ride,
look
longer,
they
can
make
a
little
bit
more
money.
And
so
this
is
a
problem
from
everything
from
supply
chain.
B
Internet
of
Things
gaming
mobility
to
problem
for
everybody,
but
everyone's
kind
of
solving
in
their
own
weird
band-aid
way,
and
so
we
think
that
there's
a
chance
to
launch
a
full
alternative
to
the
GPS,
not
necessarily
for
pinpoint
navigation,
but
for
this
new
service
of
verifying
location.
And
so
what
we
want
to
do
with
this
system
is
basically
have
a
terrestrial
based
system
made
of
low-power
radios.
I
owe
tea
based
ones
that
are
quite
small.
B
Maybe
I
can
just,
for
example,
something
like
this,
and
so
these
are
radios
that
can
go
really
long
distance,
but
they
also
have
a
really
long
battery.
So
it's
not
like
bluetooth
where
they
would
die
quickly
and
you
put
them
on
the
places
and
so
in
our
system.
We
call
these
zone
anchors
and
we're
trying
to
incentivize
people
to
be
network
operators
and
offer
a
zone
of
coverage,
and
so
what
I
mean
is
instead
of
satellites
in
space.
B
And
so
the
idea
is
that
in
our
system
are
going
to
have
these
radios
running
this
time,
sync
protocol,
and
so
that
protocol
is
like.
So
you
have
time,
is
running
over
radio
right
and
then
they're
keeping
the
logs
and
sharing
a
state
machine
and
with
that
right
now
we're
using
a
tender
mint
based
consensus.
B
So
you
have
basically
like
this
zone
of
four
radios
is
covering
an
area
over
radio,
and
then
they
have
a
shared
state
machine
and
they're
keeping
kind
of
a
consensus
on
what
happened
in
the
logs
using
ten
Derman
and
then
ultimately,
for
this
zone
to
really
exist.
They
first
would
have
to
stake
foam
on
the
root
chain
and.
B
Could
be
a
cosmos
zone,
it
could
be
Definity,
but
right
now
imagine
it's
a
theorem
whatever
would
happy
the
safest
place,
and
so
they
would
stake
there
put
in
collateral
and
basically
enter
into
an
agreement.
Saying
we're
gonna
follow
the
protocol
rules.
We're
gonna
offer
location
services
in
this
particular
area.
If
you
then
catch
us
lying
or
inserting
fake
data,
I'm
gonna
lose
my
stake,
but
if
not
I
can
get
these
mining
rewards
and
I
can
get
these
fees
from
customers.
B
They
don't
have
to,
but
if
they
are
near
each
other,
they
will
automatically
hear
each
other's
messages
and
so
they'll
each
be
their
kind
of
own
plasma
chain.
So
they
they.
They
don't
speak
to
each
other
via
the
blockchain
directly,
but
they
have
overlapping
messages
and
then
they
get
into
each
other's
logs.
So
if
all
of
a
sudden
in
this
zone,
if
they're,
actually
overlapping,
starts
saying
oh
I
got
Jonny's
messages,
he
was
here
and
in
all
the
zones
around
it,
those
messages
don't
than
they
were
probably
inserting.
Like
a
fake
message,
I
see.
A
B
Because
they
can
hear
each
other
over
if
they
can
hear
each
other
over
radio
and
then
when
you
have
ones
that
are
completely
isolated,
they
will
like
have
like
a
trust
level
of
zero
and
you'd
have
to
like,
and
there
will
be
a
class
of
users
called
like
mobile
verifiers,
and
it's
like.
If
they've
already
been
in
a
trusted
zone,
you
have
to
like
attract
them
to
your
zone,
to
like
a
test
that
you're
a
real
zone
or
something
so.
B
Yeah
I
mean
it's
basically
then
like
the
first
come
the
first
to
sign
it
back
will
be
the
ones
on
the
route
chain
that
get
your
payment,
but
it's
possible
that
it
hits
this
zone
and
they
approve
it.
And
then
it
travels
further
and
further
and
they
all
try
to
approve
it.
They
don't
wind
up
getting
the
payment,
but
they
do
keep
the
history
of
your
message.
So
that's
like
use,
then
for
the
validators
to
go
back
and
look
for
fraud.
B
So
it's
very
easy
to
check
fraud
if
it's
like
all
about,
if
you're
violating
the
laws
of
physics.
So
if
like
someone
from
near
team,
has
all
of
a
sudden
a
presence
named
in
China
and
in
New
York
within
the
same
day,
same
12
hours,
it'll
be
automatically
caught
by
the
validators
that
reading
all
the
logs
mm-hmm.
A
C
A
B
But
the
idea
is
that
you
have
all
this.
If
the
protocol
is
working
improperly,
you
have
a
mesh
of
zones,
and
so,
even
if
you
corrupt
this
zone,
the
idea
is
that
this
will
then
be
the
only
zone.
That
has
your
message
and
it
should
have
showed
up
in
everybody
else's.
So
then
you
have
to
bribe
this
on
this
on
this
on
this
on
this
on
this
on
this
on
yeah,
but.
A
B
C
A
I'm
actually
walking
into
this
place,
and
at
this
point,
I'm
willing
to
pay
the
fee
began
to
listen
to
the
laws,
and
only
only
four
of
them
here
include
me
into
the
logs
here,
but
after
that,
I'm
not
paying
anyone.
Yeah
I'm
still
like
how
was
it
different
from
the
case
where
I'm
not
physically
there?
Why?
Why
would
other
zones
know
that
I
was
well.
B
It's
time-stamped
right
so
you're
saying
I
was
in
this
vicinity
at
this
exact
time
and
it's
are
then
a
receipt,
and
so
it's
ultimately
then
lives
on
the
route
chain
as
a
first-class
object.
So
you
can
then
let's
say
you're
the
uber
driver
and
the
app
says
for
you
to
get
paid.
You
have
to
produce
seven
receipts
from
your
route
right
at
the
end
of
your
route.
You
would
automatically
reveal
those
to
the
app
and
that
app
has
all
its
own
logic
of
what
they
want.
B
C
B
Completely
up
to
the
application,
so
the
protocols
job
is
to
serve
one
single
purpose,
which
is
to
offer
fault-tolerant
time
services
that
you
can
derive
location
from
and
if
it's
producing
those
correctly
and
incentivizing
network
coverage.
The
idea
is
that
any
application
can
then
define
their
own
logic.
So
the
pokemons
mom
might
say:
hey.
B
You
have
to
get
to
presence
claims
within
a
one-hour
window
at
this
part
of
town
in
this
part
of
town,
and
then
you
unlock
Mewtwo,
but
the
uber
one
might
say
you
need
fifty
to
get
paid
because
it's
more
important
to
us
to
verify
your
every
step
for
something.
So
every
application
can
kind
of
define
how
they
want
to
consume
presence
claims.
So
our
protocol
just
allows
the
infrastructure
for
customers
to
generate
them.
A
B
To
every
single
zone
is
it's
like
own
state
machine
and
the
reason
is
because
they
also
could
all
have
different
time
periods
that
they
commit
to
the
protocol.
So
you
might
have
thousands
of
zones,
but
one
zone
is
committed
to
one
month.
One
zone
is
committed
for
three
months,
and
so
they
each
can
have
their
own
kind
of
engagement
with
the
protocol,
and
each
individual
case
is
its
own
plasma
chain
or
its
own
tender.
Mint
consensus
and.
A
B
Not
by
necessarily
like
so
to
form
a
zone,
you
need
at
least
four.
So
that's
four
XYZ
and
a
fourth
for
time
same
with
EPS.
You
need
at
least
four
signals,
and
so
the
idea
is
that
you
can
blindly
stake
tokens
and
try
to
find
peers
to
join
his
own
or
yourself
setting
up
for
and
like
creating
his
own
I
see.
B
But
then
the
idea
is.
They
could
also
be
open
for
new
notes
to
pro
in
that
zone
or
something.
So
it's
like
one
business
like
let's
say
my
company
set
up
most
of
the
zones,
so
we
have
a
ton
of
coverage,
but
then
individuals
could
come
online
and
contribute
to
those
zones
and
join
that
agreement
or
something.
B
Know
it's
meant
to
be
completely
GPS
free,
so
there
probably
has
to
be
some
like
trust
zones
that
have
some
sort
of
ceremony
where
they
can
be
established
so
that
you
have
like
the
coordinates.
You
know
you
can
believe
the
coordinates,
and
you
know
it's
there.
So
let's
say
here
in
the
new
lab,
we
made
the
first
trust
zone
through
some
process
that
everyone
was
comfortable
with.
Then
all
the
zones
that
appear
around
it
can
like
cross-reference
that
they
know
where
they
are
and
eventually
with
enough
coverage.
You
don't
need
this
trust
own
anymore,
right.
B
B
B
B
A
A
B
We're
not
trying
to
say
like
where
you're
competing
with
GPS
three
that
you're
not
going
to
not
use
GPS
like
you
might
be
using
GPS
to
drive
your
car
around,
but
you're
using
foam
to
actually
prove
where
you've
been
right.
So
the
idea
is
you
have
this
trust
anchor
situation
like
how
do
I
know?
I
am
I,
know
I'm
at
the
new
lab.
There's.
No
one
I
don't
need
GPS
to
know
where
I
have
it
yeah,
so
you
know
where
you
are
and
you're
self
reporting
it.
A
C
A
B
A
A
B
Basically,
the
time-sink
message:
I'm
a
cryptographic
signature,
so
it's
like
you
know
full
the
cap
of
the
message
is
some
bytes
and
like
90%
of
the
message,
is
a
cryptographic
signature
and
then
it's
just
the
timestamp
and
they're
just
like
doing
it
over
and
over
again
and
then
there's
like
a
reset
function
if
they
like
drift
too
far
apart
and
so
they're,
basically
only
broadcasting
I'm
happy.
This
is
what
time
it
is
or
hey
guys.
We
should
we
sync
and
they're
just
keeping
a
log
of
that
all
the
time.
B
If,
like
two-thirds
of
the
nodes
are
malicious,
even
one
third
right
I
mean
yeah
one
third,
so
then
the
zone
would
basically
be
off
and
then
there's
like
ways
to
exit
a
malicious
plasma
chain
and
every
all
the
other
zones
would
still
be
running,
and
so,
by
the
time
a
verifier
reads
all
these
logs
and
like
post
something
back
to
the
root
chain.
It
would
be
determined
that
those
guys
got
slashed
if
these
guys
didn't.
B
A
A
A
B
B
A
A
B
Not
because
I
have
to
put
out
the
same,
like
kind
of
like
collateral
to
be
a
verifier
was
someone
that,
and
also
we
didn't
code
this.
So
it
could
be.
It's
like
ran
your
like
randomly,
provided
you
don't
get
to
choose
which
tones
your
validating,
so
they
would
you're,
basically
putting
your
own
collateral
at
risk
to
like
fake
the
verification.
So.
B
B
B
The
idea
is
like
you
can
receive
those
messages
for
free,
so
you
can
say:
hey
open
the
phone
app
I'm
here:
okay,
I,
open
the
GPS
app
okay,
they're,
similar
cool
I,
don't
want
to
buy
a
receipt.
I'll
just
gonna
shut
it
off,
and
so
you
can.
Anyone
can
hear
the
messages,
but
then
the
kind
of
value
add
over
GPS
is
that
you
can
actually
do
this
cryptographic
handshake
and
produce
this
receipt.
What
we
call
a
presence
claim
that
doesn't
exist
today.
A
How
will
they
know
I
didn't
think
personally,
imagine
imagine
I'm
here
right,
I'm,
a
validator
I'm
here,
I'm
listening
to
the
to
the
messages
from
these
two
zones
and
let's
say
that
this
zone
is
going
malicious,
but
the
only
thing
they're
doing
is
that
they
just
reporting
time.
Like
you
know,
half
a
second
earlier
than
it
is
right.
So
I
can
clearly
see
there's
a
discrepancy
between
the
two
but
I.
Don't
know
which
of
them
is
wrong.
So
this
point
it's
got
a
more
subjective
right.
I
will
just
try
to
sing
cries
with
another
zone.
B
B
The
career
crystals
will
get
a
little
bit
colder
and
vibrate
a
little
bit
slower.
So
it's
like
not
about
there
is
no
absolute
time,
so
you
can't
really
go
to
any
single
reference
point
to
compare,
and
so,
if
it's
still
synchronizing
and
it's
still
accurate
just
because
it's
slightly
slower
doesn't
mean
it's
necessarily
wrong
because
they
could
actually
vary
by
environmental
conditions.
B
B
A
A
Here's
a
client
right.
The
way
it
works
is
it
receives
four
messages
from
GPS
or
from
film
those
messages
like
effective
the
moment
they
receive
the
message.
I
can
see
the
timestamp
I
know
how
long
it's
been
traveling
and
they're.
Having
four
of
those
messages,
I
can
I
can,
but
all
those
four
messages,
the
the
the
meters
of
those
messages
they
had
to
have
very
the
clocks
in
a
very
synchronized
State
mm-hm.
So
in
this
case
it
feels
to
me
that
if
one
of
them
is
actually
start
slowing
down,
yeah.
B
A
B
A
B
If
they're
so
they're
gonna
be
operating,
one
way,
which
is
like
the
happy
path
and
so
they're
gonna
be
posting
logs
every
day
and
it's
better
checked
by
a
verifier
every
day
and
if
all
of
a
sudden
by
on
Friday
they're
like
logs,
are
not
being
reported
correctly.
Now,
there's
like
a
reason
to
you,
know:
well,
I
guess,
watch
them
so.
A
A
very
specific
example:
I
took
these
four
notes
that
every
day
they
were
operating
properly
and
one
of
the
days
they
intentionally
made
one
of
the
notes
to
report
time
a
little
bit
wrong,
so
that
people
who,
within
the
zone
their
location,
is
very
often
you're
you're
a
verifier.
You
walk
into
this
zone.
You
can
definitely
say
that
they
are
your
location
is
over.
You
have
all
those
logs,
but
how
do
you
present
them?
How
do
you
prove
to
the
to
everybody
else?
Who's
not
in
this
location
that
that
they
is
reporting
yeah.
B
A
A
Me
thanks:
you
can
afford
for
five
dismissed,
including
a
pipe
five,
but
this
one
says
it's
intentionally
sending
five
a
second
to
wait
and
everybody
else
all
of
those
four
nodes.
If
they
were
owners,
they
would
just
say
well.
Your
time
is
completely
wrong.
We're
gonna
drop
you,
but
they
do
not
drop
them
right.
So
then,
the
user
hearing,
those
for
me.
B
B
B
Protocol
will
still
other
work
correctly
because
it's
fault
tolerant,
even
if
this
guy
is
sending
the
wrong
message
and
at
the
end
of
the
day,
he'll
lose
his
stick.
So
even
if
it's
fake,
even
it's
on
purpose
or
if
because
his
hardware's
broken
he's
still
not
following
the
protocol
correctly,
but
but
if
they
all
then
switch
their
time
to
15
10
15
10
then
still.
A
A
Still
a
little
confused
is
that
the
log,
because
energy
like
that
I,
could
look
like
five
seconds
away
right,
but
in
reality
it's
going
to
be
like
two
nanoseconds
to
Dinah
seconds
can
like
bring
you
like
30
meters,
away
right,
which
is
a
pretty
long
distance
in
the
city
right
in
the
logs.
You
can
easily
hide
the
two
nanoseconds
to
skip
and
so
like
in
the
logs,
you
will
say:
yeah
yeah.
That
message
was
sent
at
the
proper
time.
You
cannot
prove
that
it
was
sent
to
nanoseconds
later
right,
I
guess.
B
But
it's
important
that
you
have
four
in
the
first
place
for
XYZ
and
a
fourth
to
check
the
time
variance.
So
that's
why
you're
using
four
you
can
do
actually
the
trial
iteration
with
three.
So
that's
why
you
have
a
minimum
before,
but
ideally
your
zone
is
15
and
then
you're
gonna
be
able
to
drop
the
nodes
not
performing,
and
if
more
too
many
are
acting
malicious,
the
whole
thing
will
go
down.
A
Yeah
I
don't
need
to
think
more
about
it.
I
guess
like
on
the
very
high
level.
What
seems
a
little
hard
to
achieve
is
that,
even
if
you
have
a
doughnut
15,
but
they
cooperating
right,
the
intention
of
accumulation
see
it
could
be
that
they
will
make
like
seven
of
them
or
like
even
all
15
of
them,
to
introduce
some
random
discrepancy
of
like
five
nanoseconds
right
that
will
create
almost
gibberish
locations
for
everybody.
B
A
B
Just
like
I'm
sorry
I
just
think
this
is
like
not
valuable
path
to
go
down
because
for
the
time
synchronization
protocol
accounts
for
clock
jitter.
So
every
time
there's
jitter
it'll
like
call
for
a
resync,
and
if
somebody
is
introducing
to
interested
or
all
the
time
and
it
has
to
keep
resyncing
well,
then
it's
not
synchronized
and
therefore
the
zone
will
be
slasher.
B
That's
an
unknown
unknown,
we're
working
now
and
like
a
demo,
that's
gonna
have
the
plasma
demonstration
and
have
the
radios
to
the
time
sink
and
then
move
towards
kind
of
a
reference
design
for
a
developer
board,
and
then
we
need
kind
of
mass
testing
to
see
how
it
performs
and
go
across
all
these
unknown
unknowns.
So
then
it's
be
more
of
like
a
grass
roots
kind
of
effort
to
determine
when
it's
time
to
activate
I
mean
that
well
exciting.
Yeah
thanks.