►
From YouTube: Urban Computing Foundation TAC Meeting 2019 07 31
Description
Urban Computing Foundation TAC Meeting 2019 07 31
To learn more about critical open source JavaScript projects like Appium, Dojo, jQuery, Node.js, and webpack, and 27 more checkout The OpenJS Foundation: https://openjsf.org/
B
A
C
C
C
C
C
D
C
The
agenda
that
we've
organized
and
planned
for
this
meeting
is
exciting
because
we're
moving
into
some
project
presentations,
so
Javier
from
carto
I
believe
is
going
to
present
on
cardos
technology,
and
we
also
have
Isaac
Zhang
here
from
to
present
on
map
three.
So
I
think
that
that
should
jump
in
and
have
that
as
like
the
core
part
of
the
agenda
and
then
wrap
up
towards
the
end
of
the
meeting.
C
Just
with
like
housekeeping
touch
on
some
of
the
upcoming
events
we
have
planned
and
if
there's
time,
some
open
discussion
about
what
we
want
to
see
out
of
this
meeting
in
the
future
and
what
future
agenda
topics
people
may
have
so
with
that
Javier.
Are
you
I
see
you
on
the
line?
Are
you
prepared
to
join
the
zoom
and
or
take
over
the
screen
share
and
present
on
carto
today.
A
C
A
A
Right
well
so,
well,
first
of
all,
thank
you
very
much
everybody.
It's
a
pleasure
to
be
here
so
I
I
got
this
presentation
and
I.
Don't
know.
I
mean
like
how
much
time
we're
going
to
call
I
can
spend.
There's
a
few
demos.
Also
there
in
case
you
guys
are
interested
but
yeah.
Why
don't
we
get
it
started
so
I,
don't
know
how
many
of
you
know
Carter.
A
So
it's
a
little
bit
of
a
generalistic
highlight
presentation
just
to
give
you
a
little
bit
of
who
we
are,
and
you
know
what
is
our
interests
and
what
are
the
price
that
we're
working
on
and
finally,
I
will
finalize
with
a
little
bit
of
wise
away.
What
is
our
interests
and
the
urban
imprint
foundation?
A
And
you
know
how
we
I
would
think
that
we
can
collaborate
with
you
guys
so
with
that,
maybe
a
little
bit
just
a
little
bit
of
an
intro,
so
I'm
caviar
and
it
I'm
the
founder
of
cartoon
now
she's
a
strategy
officer.
So
my
background
is
on
what
we
call
biodiversity
informatics,
which
is
about
figuring
out
where
species
are
probably
going
to
get
affected
by
climate
change,
and
that's
where
my
interest
on
everything
around
mapping
and
location
really
started.
A
So
I
was
doing
the
spatial
modeling
for
figuring
out
where
the
distribution
of
species,
how
yeah
world
to
place
new
protective
areas,
and
things
like
that.
It
was
on
the
course
of
that
that
student
I
started
developing
some
technology.
That
eventually
became
a
framework
and
eventually
became
a
company
which,
at
the
beginning,
was
called
car
Trinity,
and
now
it's
called
cartoon.
So
so
a
little
bit
of
you
know
like
there's
an
intro
about
you
know
what
culture
is.
A
We
call
it
an
integrated
platform
for
location
intelligence,
but
we
really
are
trying
to
enable
organizations
to
liberate
the
spatial
analytics
to
improve
their
businesses
process,
and
we
know
that
you
know
we
that
means
actually
working
with
different
type
of
audiences.
So
we
work
a
lot
with
data
scientists
problem
spatial
data
scientists,
so
we
have
specific
products
for
that.
A
How
we
work
on
some
with
data
analysts,
who
are
first
started
like
coming
into
the
geospatient
world,
mostly
known
with
GIS
backgrounds
and
finally,
with
developers,
we
have
a
network
of
partners
that
utilize
our
technology
for
developing
multiple
sources
of
projects.
So
actually
we
do
know
all
together
in
the
pieces.
I'll
go
more
in
detail,
you
know
just
for
you
guys
to
get
an
idea
who
are
kind
of
like
the
type
of
users
that
we're
working.
Maybe
it
also
has
a
background.
So
we
started
in
2012..
We
have
around
a
thousand
two
hundred
customers.
A
We
have
a
free
version
on
our
product
that
is
now
subscribed
by
300
000
people
and
we
are
a
company
of
around
140
people.
Most
of
the
engineering
and
development
happens
in
Spain,
where
I
am
from
from
Madrid,
but
we
have
a
pretty
remote
team,
then
I
mean
I
will
commercialize
all
over
the
world
where
it's
our
biggest
Market
being
in
United
States,
where
our
headquarters
so
I'm
I
mean
actually
New
York.
A
So
you're
planning
traveling
a
lot
most
of
the
time
here
in
New
York,
so
for
those
who've,
maybe
heard
of
cars
before
a
little
bit
of
from
the
history
of
a
company.
So
we
started
in
2012.
I.
Guess:
they're
like
coming
out
of
a
research
project
around
biodiversity
and
I
kind
of
like
thing
I
like
to
we.
We
Define
three
different
phases
in
the
company
so
studied
in
2012,
with
the
vision
of
course
CIS
on
the
cloud.
A
So
we
liked
a
lot
post
GIS
and
we
thought
you
know
it
was
just
too
hard
to
actually
develop
projects
and
to
analysis
with
it
at
the
scale
so
so
kind
of,
like
our
color
I,
just
put
the
tags
that
we
used
to
use
after
our
name.
So
we
were
saying
that
we
were
past
the
iOS
from
the
cloud
then
from
2014
we
saw
like
the
opportunity.
A
lot
of
our
users
were
using
us
for
actually
creating
Maps.
A
So
we
started
talking
more
about
like
the
monetizing
maps,
but
under
the
hood
really,
we
were
still
like
an
analytics
platform
which
is
really
where
UC
has
more
Focus
now,
which
is
around
the
spatial
analytics.
So
it's
about
enabling
organizations
to
perform
analysis
with
their
location
data
and
the
data
that
we
provide.
Why
did
we
start
all
this
I
love?
This
will
be
very
familiar
to
you,
guys,
there's
a
question
of
like
a
Revolution
going
on
right
now
with
new
sources
of
data
coming
available.
A
There's
new
form
of
analysis
that
you
know
like
this
type
of
data
enables
and
more
important
for
us.
It's
also
what
we
call
a
new
audience.
You
know
end
users,
so
color
the
space
around
your
space
analytics
it's
just
kind
of
like
expanding
and
there's
never
been
a
bigger
demand
for
many
organizations
to
start
color
like
applying
it
right.
So
we
think
that
actually
provides
an
opportunity
for
the
creation
of
a
platform
like
us.
A
But
if
you
want
to
create
a
model
to
characterize
the
places-
and
you
know
and
actually
figure
out
and
predict,
you
know
like
where's
the
livelihood
of
a
customer
to
buy
your
product,
then
you're
going
to
need
something
like
Carter
right
and
that's
the
space
that
you
know
like
how
we
we
separate
in
there
there's
a
lot
of
people
are
still
kind
of
like
making
the
questions
about.
Where
there's
not
that
many
people
here,
it's
kind
of
like
asking.
A
Why
and
that's
why
we're
really
trying
to
change
also
has
a
little
bit
of
you
know
like
context,
so
just
for
you
guys
to
get
an
idea
of
how
we
talk
about
these
things.
So
we're
trying
to
position
really
a
station
modeling
as
another
type
of
machine
learning.
Although
technically
I
mean
we
can
discuss
a
lot
about
this,
but
you
know
for
the
general
awareness
this
is.
You
know
like
something
that
you
know.
A
We,
like
our
customers
and
our
users,
to
to
understand
that
you
know
libraries,
you
know
spatial
type
of
analysis,
which
is
around
a
spatial
analytics.
It's
based
on
some
fundamentals
that,
if
you're
not
taking
in
consideration,
if
your
data
has
a
strong,
a
location,
color
bias
which,
in
a
lot
of
history,
does
you
are
missing
out.
There's
a
lot
of
this.
Your
models
could
perform
better
if
you
take
in
consideration
of
spatial
modeling,
so
that's
kind
of
like
another
way
that
you
look
at
it.
How
do
we
do
this?
A
So
we
have
essentially
main
four
main
components
so
and
now
I'll
see
if
we
have
time
to
give
you
a
little
bit
of
a
demo
of
a
few
of
them.
The
first
one
I
mean
the
foundations.
We
have
a
product
called
Carter
data
Observatory,
where
we
aggregate
data
from
many
different
sources
and
make
it
available
to
our
platform
in
different
ways
that
you
will
see
later
we
work
with
a
number
of
providers.
A
I
think
some
of
them
are
actually
here
in
the
foundation
and
yeah.
Essentially,
we
we
provide
an
easy
access
to
that
data.
We
have
a
great
table,
transform
it.
We
create
derivative
data,
sets
out
of
it,
essentially
with
the
goal
of
make
it
easier
to
actually
collaborate.
Those
based
on
models
that
we
were
talking
about,
so
that's
kind
of
like
the
data
Foundation,
then
the
physical
main
product
is
part
of
there
is.
A
This
is
a
background
noise
from
someone,
so
there's
parts
of
frame,
so
essentially
it's
a
library
to
to
basically
perform
space
analysis
on
using
python.
Notebooks
I'd
like
to
show
you
a
little
bit
more
about
that,
so
it
has
integrated
visualization
capability,
but
also
analytics
capabilities.
It
talks
to
our
car
like
Cloud
infrastructure,
so
you
can
perform
analysis
with
scales
Etc.
A
You
know
just
with
that
drag
and
drop
interface
under
the
hood
is
used
to
the
same
technology
and
finally
cartoon
engine
is
the
apis
and
sdks
that
we
may
expose
of
as
part
of
our
platform
that
some
of
our
customers
use
for
developing
their
own
the
custom
applications
or
to
embed
the
analysis
or
the
applications
we're
building
the
map
into
their
own
products.
That's
pretty
much.
You
know
like
the
four
main
areas
right,
yeah,
so
I
just
thought
it
would
be
good.
A
You
know
to
collect
with
you
guys
a
little
bit
of
a
demo,
so
so
carto
works
as
a
software
service
on
the
cloud.
So
it's
a
cloud
service
you
can,
you
can
run
it
on.
You
can
run
it
for
you
on
as
a
cloud
service
or
you
can
run
it
as
an
on-prem
commercially
so
or
you
can
just
download
the
Open
Source
version
I'm
just
going
to
show
you
here
a
little
bit.
A
You
know
like
custom
builder,
so
this
is
where
you
actually
can
manage
your
data
sets
under
the
hood
there's
Sports
GIS,
with
a
lot
of
connectors
to
different
things.
We
improve
it
in
many
different
things.
You
also
can
create
like
map
here
you
can
get
API
keys
and
things
like
that.
So
I'm
just
going
to
open
this
one,
for
example-
and
you
can
see-
and
you
have
like
classical
editoring
here,
where
you
know
like
you-
can
essentially
got
like
modify.
A
You
know
the
styles
of
of
the
of
the
of
the
of
the
map.
You
know
like
the
original.
You
can
perform
analysis
on
those
type
of
players
under
the
hood.
All
of
this
is
just
running
out
like
under
the
same
apis
are
available
for
everybody,
but
but
you
know
just
here
exposing
a
simple
way.
There's
also
got
like
here
opportunity
to
do
like
enrichment
based
on
some
of
those
data
sources.
A
If
I
was
meaning,
you
can
find
kind
of,
like
you
know
like
filters,
you
can
personalize
it
from
widgets
and
all
this
stuff,
all
this
type
of
food
right
so
and
then,
finally,
you
can
certain
visualization
or
you
can
build
an
application
based
on
this
type
of
of
interface,
that
you've
created
so
that's
character.
Builder
I
just
wanted
to
also
show
you
here
also
the
part
of
frames.
So,
let's
start
it
I
sort
of
have
it
open.
A
So
quarter
frames.
It
really
is
a
library
mostly
considered
to
be
used
with
python
notebooks,
it's
going
to
open
it
here
in
binder
and
they
provide
you
with
certain
capabilities.
Like
I
said
for
visualizations
we
have
and
you
can
create
Maps
out
of
there
share
them
with
others
using
the
same
underlying
Technologies,
but
then
also
it
has
capabilities
to
process
the
data
on
Carter
run
a
spatial
SQL.
A
A
A
All
right
so
so
I
said
you
know:
I'll
explain
a
little
bit
our
open
source
and
data
strategies.
So
so
we
started
from
the
beginning
being
an
open
source
company.
So
it
means
that
the
majority
of
our
software
is
open
source,
so
the
entire
platform
actually
can
be
download
and
it
can
be
used.
It
can
be
contribute
on
on
GitHub,
there's
certain
pieces
that
we
are
not
open
sourcing
like
those
that
we
consider
you
know
we'll
just
only
provide
essentially
like
competitiveness
to
our
supply
service
that
we
provide.
A
So,
for
example,
we
have
a
scalability
of
infrastructure
in
the
cloud
you
know,
so
those
type
of
things
connectivity
to
external
databases,
that's
kind
of
like
Enterprise
type
of
functionalities,
a
certain
type
of
spatial
algorithms
that
we
keep
you
know
for
ourselves
and
for
our
customers
and
then
on
the
data
part.
Although
we
support
a
very
strong
public
data,
open
data,
type
of
kind
of
like
repo
that
is
freely
available.
A
We
also
have
like
this
commercial
data
providers
connected
to
our
platform,
which
is
obviously
you
know
we
don't
provide
for
free
and
we
also
create
derivative
data.
Sets
that
you
know
from
from
those
providers
so
think
about,
like
someone
like
MasterCard
will
give
us
information
about
like
create
customers,
actions
like
Aggregates,
and
we
will
transform
that
into
other
forms
of
information
that
can
be
used
for
models
and,
of
course,
I
mean
like
we
are.
You
know
under
the
hood
for
CIS
I,
wouldn't
say
password
close
in
more
our
database.
A
I
think
we've
seen
multiple
tries
in
a
way
to
expand
the
adoption
of
Open,
Source,
geospatial
and
open
data
around
us,
peso
and
I.
Think
right
now,
it's
clear
that
you
know
like
the
way
to
push
this
is
going
to
be
through
like
industry
Partnerships.
So
that's
you
know
something
that
I
found
very
interesting
about
the
urban
computer
foundation
and
that's
why
we
want
to
be
part
of
it.
A
So
I
can
elaborate
a
lot
more
on
that
the
areas
of
collaboration
that
we
see
there's
a
number
of
products
already
I
mean
a
particularly
like
Kepler
that
we're
like
looking
at
actually
Ally
liberating
for
carto,
particularly
around
tiling
cartography
data,
credit
cards
and
a
few
other
things.
Thank
you
for
another
talk
we
think
about.
Like
you
know,
it's
Carter
frames
could
be
like
an
interesting
car
like
a
project
to
contribute.
We
also
have
like
a
set
of
solutions
based
on
some
of
data
providers.
A
So
I
talk
about
the
spatial
data
science
conference,
I
think
on
the
last
one
will
be
reminding
you
in
another
email,
but
I
think
you
know
we
are
kind
of
like
creating
that
community
of
spatial
data
scientists,
which
I
think
is
also
a
big
part
of
the
UCF
that
you
guys
are
looking
at.
We
do
a
lot
of
work
from
Angela
evangelization,
there's
not
that
many
spatial
data
scientists
out
there
and
for
us,
it's
very,
very
important
to
connect
to
them.
A
So
so
that's
also
something
that
we,
we
can
collaborate
a
lot
and
then
coordination
on
software
data.
There's
a
lot
of
you
know
like
data
providers,
data
software
products,
you
know
in
a
way
projects
that
I
think
you
know
like
make.
What
is
the
current
modern
architecture
around
location,
intelligence
I
think
you
know
like
this
could
be
a
great
place
for
coordination
around
that
and
finally,
on
data
and
Analysis.
A
We
have
a
lot
of
experience
in
that
for
methods
for
sharing
location
data,
different
type
of
algorithms
methodologies,
but
also
around
privacy
and
regulation.
So
we
have
to
do
a
lot
around
that
which
I
think
might
also
be
a
very
interesting
color
and
I.
Think
I
just
don't
want
to
make
it
much
much
longer.
We're
truly
excited
to
work
with
you
guys
and
and
if
you
guys
want
more
specific
information
about
cartel
I
mean
just
let
me
know
all
right
so.
C
A
I
think
there's
a
question
around
the
license,
so
we
we
try
to
license
everything
as
much
as
we
can
with
the
DSD
lighters.
So
that's
our
overall
call
like
a
strategy,
sometimes
like
in
the
case
of
posies.
As
you
know,
it
has
a
different
type
of
license,
so
we
have
to
just
contribute
to
whatever
it
is,
but
our
products
and
the
whole
platform
is
beautiful.
C
Question
for
me,
I
guess
in
terms
of
like
the
the
geospatial
analysis,
algorithms,
are
they
like
in
the
carto
sort
of
Technology
stack?
Are
they
majority
on
like
a
server
side
or
are
some
of
them
also
done
in
the
client
side
in
terms
of
like
carto
Explorer,
you
know
in
the
web
browser
or
some
combination
of
both.
A
A
I
mean
a
typical
guy
like
customer,
for
you
guys
to
get
an
idea
that
I
had
that
we
have
is
it's
a
retailer
looking
at
words
to
sell
more
solar
panels
and
we
have
to
actually
color
create
a
predictive
model
that
will
tell
us
you
know
what
is
the
likelihood
of
someone
to
to
buy
them
right.
So
some
of
the
things
for
for
doing
this
type
of
operative
models.
A
We
we
create
things
like
you
know,
like
twin
areas,
type
of
analysis
and
and
for
that,
actually
we
keep
that,
as
part
of
you
know
like
in
essentially
on
on
you
know,
it
says:
I
mean
we
just
program
them
python,
notebooks,
and
then
we
make
them
available
as
per
a
sky
like
algorithms,
on
their
own,
that
you
can
call
from
within
the
platform.
So,
as
you
saw
in
one
part
of
the
platform,
one
Builder
where
I
was
there
was
a
list
of
assistant
analysis.
A
We
can
extend
that
and
some
of
those
extensions,
Are
Made
You
Know
by
us,
so
since
Olivia
has
inputs
and
they
produce
an
output
on
the
on
the
client
side.
That's
interesting,
I
think
you
know
like
there
hasn't
been
that
much
I
mean
it
really
depends
on
how
you
go
like,
and
what
how
do
you
find
analysis
right?
Your
spatial
analysis
there's
like
one
layer
which
I
you
can
call
almost
you
know
like
almost
like
serious
Geographic
type
of
you
know,
like
type
of
analysis
or
I
mean
nice
intersections
type
of.
A
A
What
we
think
you
know
like
it's
I
mean
the
next
layer
on
that
that
you
put
on
top
of
it
is,
for
example,
when
you
define
something
like
catchment
area
and
if
you
look
at
you
know
like
how
a
catchment
area
was
done
back
in
the
days
was
by
just
creating
a
buffer
around
a
point.
So
if
you
want
to
know
who
goes
to
your
store,
you
will
say
like
okay,
this
is
you
know,
like
everybody
in
200
meters
around
my
store
is
potentially
my
captain
area
for
customers.
A
Then
we
want
a
little
bit
more
sophisticated
with
the
use
of
things
like
isochrons
so
like
well,
anybody
that
can
drive
up
to
five
minutes
is
on
my
Catman
area.
Now
using
some
other
data
sources.
I
was
talking
with
bullet
point
and
so
on.
We
can
actually
do
we
create
the
real
catchment
area.
So
a
lot
of
you
know
like
some
of
those
functions,
are
now
turning
into
different
type
of
functions.
Powered
by
data
sets
behind
them.
A
So
I
think
you
know
that's
one
very
interesting
approach
and
then,
on
top
of
that,
then
you
have
like
a
algorithm
that
you
can
use
to
actually
elaborate
those
to
create
predictive
models
like,
for
example,
we
have
a
pretty
good
one.
You
know
for
the
technical
solar
panel
because
we
just
happen
to
have
like
many
customers
on
that
space.
So
so
it's
different
levels-
and
you
know,
as
you
go
more
deep
on
the
complexity,
it's
less
likely
so
that
you
can
do
them
on
the
client.
C
A
Platform
and
I
think
you
like
that
self
a
little
bit
about
you
know
the
state
of
some
of
the
open
source
projects
nowadays
in
the
sense
of
I
think
you
know
like
we
just
probably
go
to
I
mean
we
have
a
team
of
engineering
that
they're
just
sometimes
in
I,
don't
have
two
classmates.
You
know
they're,
just
embarrassing
a
lot
on
their
own,
so
I
wouldn't
say
too
much
on
actual
contributions
in
terms
of
of
features
to
our
car
like
core
product.
A
What
we
do
see
is
that
there's,
for
example,
an
interesting
car
like
a
community,
actually
that
made
it
available
for
running
from
scale
on
the
cloud,
so
users
are
actually
like
contributing
that
and
they
contribute
a
lot
towards.
You
know
like
making
Docker
images,
and
you
know
and
like
that's
that's
definitely
for
sure,
there's
a
number
of
those
there's.
A
number
of
you
know
like
open
source
users
and
they
contribute
to
make
you
know
like
that
happens
as
you
see
as
possible,
then
on
the
particular
color
project.
A
You
know,
like
you,
think
about
like
something
like
cartoons
there,
yes,
that
we
can
have.
We
have
more
specific
contributions,
because
you
know
a
lot
of
people
call
I,
find
it
easier
to
contribute
that
in
a
mad
Skyline
project
but
but
I
wouldn't
say
I
mean
like
we
have
a
very
strong
open
source,
color
community
of
contributors
and
the
main
customs
and
there's
a
bunch
of
them.
But
there's
not
that
many
yeah.
D
A
A
great
question:
I
mean
I,
don't
have
a
I,
don't
have
a
definitive
answer
to
say
like
this
is
exactly
the
rule
that
we're
going
to
be
using
on
when
to
open
source,
something
or
not,
because
you
know
it
really
depends
on
the
project
by
project
basis.
In
the
case
of
ways
in
particular
yeah
we
I
mean
the
connector,
so
essentially
our
connector.
If
you
know
the
waste
platform
has
a
waste
API
that
is
accessible
through
the
through
a
particular
program
around
connected
cities.
A
So
it's
not
most
of
the
time
as
far
as
you
know
not
available
for
commercial
call,
I
use
it
and
it's
mostly
available
for
cities
that
want
to
collect,
improve
their
service
and
so
on,
and
in
that
case
then
Google
when
you
become
a
partner
and
we're
like
a
technology
partner
to
waste.
In
that
case,
and
also
we
enable
CDs
to
make
use
of
waste
data,
so
so
they
expose
you
an
API
that
is
is
I
mean
it's
just
very,
very
basic
in
a
way.
A
So
it
only
gives
you
the
feet
of
data,
but
then
it's
all
up
to
you
to
call
like
maintain.
You
know
like
an
historical
data
set
of
it
and
so
on.
So
in
our
case,
what
we
developed
was
a
way
to
collect
a
store,
all
the
data
in
a
in
a
cartel
or,
like
instance,
and
then
we
created
an
interface
to
collect,
make.
A
You
know,
like
historical
call,
like
query
stores
that
data
visually
and
also
called
like
perform
certain
analysis,
how
congested
how
congestion
is
passing
in
is
happening
over
over
time
in
different
parts
of
the
city.
So
some
of
our
customers
wanted
that,
and
in
that
case
we
don't
see
a
value
for
keeping
that
you
know,
like
you
know
like
closer,
so
we
open
source
entirely.
The
value
again
for
us
will
be
to
work
with
the
city
that
you
know
wants
to
elaborate
cartoon
or
knowledge
of
that
to
solve
a
specific
problem.
A
So
one
example
that
we
used,
that
was
for
the
city
of
Madrid,
so
they
wanted
to
essentially
have
a
secondary
way
of
measuring
how
it
is
traffic
changing
in
different
parts
of
the
city,
there's
a
huge
project
going
in
Madrid
around
closing
Street,
and
they
want
to
see
what
is
the
impact
that
is
produces
on.
Other
streets,
so
they
want
to
have
an
monitoring
system
that
could
also
color
validate
what
they
gather
through
other
type
of
sensors.
A
So
so
we
work
with
them
on
that,
so
I
think
you're,
like
those
are
the
those
kind
of
like
building
blocks
of
you
know
you
get
like
waste
data
and
then
you
put
it
into
carto
or
a
postgres
or
an
aurora,
or
you
know
a
bigquery,
whatever
it
is,
and
so
on.
I
think
those
are
the
areas
that
I
think
like
the
urban
computer.
Foundation
should
also
kind
of
like
it's.
D
Sounds
great
thanks
for
that
detail,
because
that
that
reminds
me
so
well
of
the
the
maps
and
proposition
too
that
perhaps
access
to
to
this
Waze
data
to
lie
with
each
one
of
those
City
customers
of
cardo,
but
but
the
the
glue
code.
The
data
functions
could
really
just
be
a
shared
cost
across
all
of
the
consumers.
I
definitely
I'd
definitely
be
glad
to
follow
up
separately
on
that
and
see.
A
You
yeah
and
of
course
I
mean
I
will
I
mean
this
is
a
great
day.
Call
it
perfect
for
that
right.
So
you
have
like
now
Uber
contributing
with
the
data
right.
So
then,
now
you
have
like
ways
it's
different
types
of
data,
but
there's
a
lot
of
stuff
in
in
common
right.
So
we
had
to
do
a
lot
of
like
connectors
for
all
those,
so
yeah
very
glad
to
actually
talk
about
these
things.
C
A
All
right
so
I
just
want
to
again.
You
know,
remind
everybody
to
get
some
ice
cream,
one
more
time
about
the
spatial
data
science
conference.
That
I
talked
last
time.
I
think
that
is,
that
is
a
great
forum
for
it's
a
great
form
to
go.
Sorry!
So
half-
and
you
know
I-
got
this
before
so
so.
There's
this
conference
that
we
organized
on
on
this
is
the
previous
year
in
in
New
York
in
October,
the
16th
of
October.
So
we
expecting
around
400
people
there's.
A
Actually
we
it's
a
mix
between
the
Academia
and
the
and
the
and
the
private
sector.
So
we
also
looking
for
sponsors,
but
also
for,
like
good
speakers,
I
think
like
the
goal
in
there
is,
you
know
to
is
to
gather
you
know
like
a
group
of
people
working
on
just
the
data
science
behind
spatial
is
the
third
year
is
actually
pretty,
it's
been
pretty
pretty
successful.
A
So
if
you
want
to
have
any
details
again,
we're
looking
for
private
sponsors
would
love
to
I
mean
we're
already
quite
a
few
of
them,
but
yeah
spatial
data
science
cover
here
it
is,
and
you
have
like
videos
and
everything,
and
so
on
so
we'll
be,
will
be
great
to
if
you're
interested
just
ping
me
about
it.
Okay,
yeah.
C
A
Yeah,
so
they
there's
they
should
just
contact
I
mean
we
have
like
a
how
you
call
it
like
a
committee
that
is
meant
of
we
organize
it
together
with
the
University
of
Chicago
and
they
and
we
call
I
Define
the
the
the
and
we're
going
to
be
defining
the
content.
You
know
I
mean
we
are
already
College
selecting.
So
no
there's
not
like
we
actually
cherry
picking,
people
that
we
find
you
know
like
have
interesting
stories
to
talk
about
spatial
data
science.
A
So
if
you
have
like
a
specific
car
like
a
story
or
something
that
you
you
think
is
working
on,
just
ping
me
and
I
will
put
you
in
contact
with
the
persons
examples
of
the
type
of
presentations
last
year
that
we
had
where
I
mean
this
year,
for
example,
I
know,
there's
going
to
be
one
about
inequality,
calculating
inequality,
so
some
of
them
that
are
more
around
like
sensor
type
of
demographics,
I,
think
you
know,
there's
gonna
be
likely
warm
from
wework
around.
You
know
like
selecting
locations
or
opening
things.
A
There's
one
probably
around.
You
know
like
Mobility
about
you
know
where
to
place
Target
spaces
things
like
that,
so
it's
normally
the
one
around
biodiversity.
So
overall,
the
whole
idea
is
just
someone
you
know
just
bring
like
10
type
of
like
a
use
case
and
then
go
deep
into
the
data
science
behind
it
right.
So
so
probably
around
80
percent
of
the
audience
is
going
to
be
data
scientists,
that's
what
we're
looking
for
some
of
them
more
familiar
with
the
spatial,
some
of
them
less
familiar,
and
so
that's
that's
idea.
A
C
B
B
Thank
you
for
having
me
very,
very
pleased
to
be
here
so
I
don't
have
a
deck
I'm
going
to
just
to
walk
through
the
content
through
our
website,
it's
to
take
a
developer's
Journey,
so
some
little
bit
backgrounds
by
myself,
so
I've
been
doing
open
map
and
local
locations
since
my
grant
graduate
school.
So
it's
been
a
long
long
time
and
I
also
believe
in
building
for
lasting
impact.
So
I
always
wanted
to
have
a
good
open
map
related
open
source
project,
and
now
we
have
math
three.
B
So
let's
get
started
with
my
memory.
In
short,
it's
a
decentralized
map,
Network,
okay,
on
the
system
side,
it's
a
P2P,
decent
trans
Network
for
maps
and
in
the
network.
It's
hosting
Rich
Community
generated
map
contents.
It's
all
sort
of
layers.
It
can
be
the
crime
map
safety
map,
the
Museum
map,
you
name
it
also.
The
POI
and
all
these
Maps
data
will
be
served
through
very
strong
privacy
and
Trust
proper
goals
and
with
consideration
to
be
compatible
with
existing
map.
B
Technologies
like
the
leaflets
black
box
and
it's
the
whole
thing
is
totally
compatible.
You
can
use
the
tools
that
you're
most
familiar
with
and
with
just
a
plug-in,
a
drop-in
plugin
of
my
free
client
to
help
you
define
the
Define,
the
peers
and
Route
traffic
to
that's
all
about
it.
Let's
talk
a
little
bit
more
about
the
design
philosophy
of
memory.
B
So
this
is
the
graph
I'm
going
to
navigate
the
whole
story
through
the
graph,
so
I
think
in
the
last
two
decades.
In
my
summary,
the
map
industry
has
has
evolved
based
basically
along
to
access
to
the
dimensions.
The
first
will
be
the
map
day.
Data
and
a
second
will
be
the
map
surface,
and
the
underlying
keywords
of
the
trend
is
decent
translation,
because
that
fits
very
nicely
with
the
topology
of
the
map.
B
It's
highly
distributed,
it's
very
local
and
very
Global,
as
well,
so
with
that
Trend
in
behind
and
every
single
map
version
or
map
Dev
Paradigm
transitions
is
driven
by
the
forces
or
advancements
on
these
two
axes
as
well.
So,
oh
by
the
way,
my
three
is
the
coin
that
we
took.
We
at
the
time
that
we
coined
just
to
to
differentiate
different
paradigms
of
maps
and
for
the
first
Photoshop
Maps
I
call
it
the
profit
map.
It's.
B
Basically,
the
map
data
were
generated
by
the
professionals,
survey,
surveillance,
using
purpose
view
devices
and
the
services
are
majorly
host
by
some
very
big
companies,
and
it's
not
open
enough
to
centralize
and,
and
then
that
comes
along
with
many
trends.
Firstly,
on
the
map
data
science
is
the
open
map?
Mapping
osm,
it's
great
five
million
communities
of
at
it's
a
very,
very
Vibrant
Community,
and
then
we
have
the
more
mobile
phones.
B
We
can
work
around
using
our
phone
to
take
pictures
and
to
record
GPS
to
make
the
map
and
also
the
data
Technologies,
and
all
these
two
Trends
are
actually
making
that
data
more
accessible
and
more
accessible
to
individuals
or
small
organizations
to
help
edits
and
on
the
y-axis.
The
map
Services
a
lot
of
Open
Source
gns
stack
developed
by
leaflet,
Nexus
labson,
and
it
covers
basically
all
ranges
from
map
surfaces
and
these
two
Trends
together
help
making
the
map
more
opened,
not
just
by
the
becomes
also
startups
in
the
startup
spaces.
B
And
it's:
the
data
is
not
it's
buying
it's
nice
curated
by
a
lot
of
individual
communities
all
over
the
world
as
well
as
we
can
see
from
the
growth
of
the
osm
right,
but
map
2.0,
open
map
is
I.
Think
it's
still
not
enough.
It's
not
open
enough
and
a
lot
of
underdeveloped
places
under
your
under
different
places.
B
The
map
data
is
not
good
enough
and
the
map
surface
is
too
centralized
and
you
would
have
all
sorts
of
risks
like
the
censorship
Etc,
and
here
we
have
some
new
trends
that
I
think
will
push
map
2.0
to
map
310
the
consensus
map
and
the
first
on
the
map.
Data
side.
I
would
say
it's
a
really
broader
participation
of
the
big
companies
like
Facebook,
Uber
and
Microsoft,
and
also
some
governments
are
pushing
their
open
data
with
the
government,
open
data
license
to
the
osm
and
and
not
just
that.
B
We
have
very
good
fancy.
Ai
mapping
Technologies
now
Facebook
just
released
I
think
the
map
with
AI
right
not
long
ago
and
also
Microsoft
released
the
cntk
unified,
took
it
to
it's
actually
just
machine
learning
extracting
metadata
from
the
aerial
imagery
to
the
Microsoft.
One
is
about
building
footprint
and
the
map
with
AI
is
more
on
the
rose.
B
I
think
all
these
Trends
would
push
provide
a
ground
to
push
the
map
to
a
new
stage
and
that's
hopefully
to
be
that's
also
the
things
that
Matthew
is
trying
to
do
and
we
can
see
along
the
way
the
trends
are
a
snap
industry
of
all.
You
would
have
much
more
higher
level
of
increasing
level
of
decentralization,
which
is
the
cause
and
the
effect
is
increasing
level
of
robustness,
openness
and
efficiency.
B
And
if
you
push
this
to
its
maximum
and
then
you
will
have
the
answer
of
like
one
member
three
10x4
questions,
you
will
be
free
from
control
and
10x
openings
will
be
free
to
participate,
participate
on
all
ends.
On
the
supply
side.
You
can
host
the
note.
You
can
publish
some
permission,
some
map
contents
or
on
the
demand
side.
You
can
use
map
Services
of
all
kinds
and
the
10x
efficiency.
B
It
would
give
you
like
a
really
powerful,
basically
really
low,
Mark
margin,
marginal
cost
map
surfaces,
and
that
would
be
a
very
powerful
I
think
enabler
to
more
build
Avengers
map
services
and
with
the
community
and
with
the
community
donated
resources.
We
can
probably
provided
sufficient
free
base
map
for,
like
for
people
to
use.
B
So,
first,
the
robustness
you
have
the
serves,
continue
the
surface,
the
new
annuity
you're
building,
something
on
you
be
sure
it
would
last
his
future.
So
it's
a
reliable
building
block
and
openness.
I
think
this
part
is
particularly
important
because
nowadays
is
in
the
old
times
it's
a
privilege
for
the
big
companies
to
monetize
to
be
able
to
to
generate
and
monetize
the
spatial
data
data.
B
But
now,
with
the
value
Network
that
people
can,
for
example,
you
have
a
certain
certain
type
of
maps
and
then
you
can
publish
to
the
net
Network
and
if
your
assets
policies
to
to
use
by
pay
and
then
it's
very
easy
for
the
downstream
to
find
your
daily
data.
For
example,
if
I
am
looking
for
a
place
to
rent
on
a
trip.
B
I'm
looking
for
Airbnb
house,
but
I'm
also
concerned
with
the
with
the
safety
with
the
crime
rate
and
also
I
want
to
find
some
museums
in
the
walking
distance
so
that
in
the
math
three
you
can
use
just
one
interface,
one
unified
interface.
To
add
this
all
map
list
and
give
you
a
big
big
one
and
now
I'm.
Just
talking
about
three
basic.
B
We
will
have
like
a
much
more
much
more
data,
intensive,
computation,
intensive
or
latency
in
sensitive
applications,
basic
applications,
AR
or
some
VR
content
rendered
on
the
5G
mini
Data,
Center
and
hdmap,
and
the
network
of
these
devices
can
form
a
very
efficient
way
to
to
let
the
content
can
be
generated
can
build
it
served
on
the
edge
okay,
the
community
is,
the
communities
can
actually
participate
in
join
the
Mastery
Network
on
multiple
fronts.
On
the
supply
side,
you
can
hold
host
a
note.
B
You
can
provide
some
lab
contents
and
on
the
demands
that
you
can
just
use
the
search
surfaces.
Well,
here,
I'm
going
to
demo
demo
on
how
to
host
a
node
we
on
early
stage
with
a
not
much
proper
promotion.
We
already
have
I
think
last
time,
I
checked
it's
a
160
160
notes
from
all
over
the
world
just
to
joining
out
the
network
out
of
Goodwill
to
to
be
a
poor
part
of
it,
and
you
can
find
it's
very
simple.
B
Actually
you
can
also
deploy
it
on
Ali,
Cloud,
AWS
and
other
Cloud
platform
platforms,
so
you
can
also
deploy
the
very
native
just
one
command
line.
This
is
a
Ubuntu
machine
on
aws's
new
one.
Let's
try
to
do
it.
Hopefully
it
doesn't
interpret
me
and
it's
going
to
take
a
few
minutes.
Let's
skip
it
to
to
what
we're
talking
about,
and
you
can
also
publish
some
contents
to
the
map
store.
We
have
a
concert
called
map
store
to
to
let
people
to
find
to
find
the
content
contents
and
to
use
it
right.
B
But
now
the
map
store,
we
don't
have
a
UI
for
it.
I
don't
have
it.
We
don't
have
enough
people
to
build.
You
have
a
better
protocol,
it
works.
So
if
you
want
personal
account,
just
contact
me
and
I
think
it's
still
running
down
downloading
the
dependencies,
or
we
can
talk
a
little
bit
about
how
it
works
as
well
right.
So
it's
just
very
simple
description
of
it.
Let
me
walk
it
down.
First,
on
the
structure
side,
it's
a
decent
trans
Network.
B
The
notes
can
just
join
an
hour
and
discover
each
in
PDP
member
to
slightly
turn,
and
every
single
node
can
register
to
plug
their
cache
certain
map
contents
and
over
time
the
network
learns
where
to
distribute
this
spatial
data
accordingly
to
the
locality
characteristics
of
the
applications.
That's
like
a
really
like,
maybe
the
Google
ways
of
like
building
industry,
to
let
the
network
to
evolve,
to
learn
over
time
to
spatial
shorting
the
day
data
to
giving
the
right
data
to
the
right
right,
plane,
place
and
I'll.
B
Just
the
Mastery
can
also
be
deployed
cross
platforms
and
we
are
all
now
we
support
to
be
to
build
on
the
Linux
machine
and
now
the
next
stage
I
think
it
will
be
released
in
two
weeks.
We
can
support
to
deploy
on
Nas
and
more
and
the
users
can
search
and
find
the
map
contents
and
their
respective
address
on
the
map
store.
All
map
data
are
cryptographically
signed
by
the
owner,
so
that
you
can
easily
verify
the
authentication
of
it
and
the
content.
B
If
it's
private,
we
have
a
network,
it's
called
technical
scheme
to
it's
called
the
proxy
Recon
re-encryption
to
re-encrypt
data
on
the
network.
So
when
I
do
this
on
network
is
in
Pro
equip
you,
but
when
you
use
it,
when
the
client
is
decrypted
and
in
order
to
be
compatible
to
to
our
growth
of
the
battery
surfaces,
it
has
to
be
support.
Yes
to
support
https
right,
so
we
enable
two
modes
of
https.
B
B
Now
you
can
already
use
the
math
resources
through
this
scheme
and
that's
I
think
and
it's
majorly
out
of
consideration
to
be
compatible
with
the
brow
brow
browsers.
Otherwise
it
is
going
to
alert
you
with
a
non-secure
website
alert,
and
this
sounds
really
bad
too,
to
the
new
users
and
also
you
can
do
the
self-site
architecture
to
generate
a
self-signed
certificate
or
you
can
talk
to
the
let's
encrypt,
to
get
a
science
certificate
from
certificate
from
let's,
let's
encrypt
and
on
the
on
the
client
side.
B
It's
just
a
drop-in
library
to
basically
help
you
to
do
all
the
P2P
thing.
To
talk
to
the
tracker
find
the
peers
rather
track
traffic
Etc
and
in
the
future
we
may
build
into
some
token
mechanics
into
it.
Basically,
so
the
math
three
data,
if
you,
if
you're
no
help
to
source
to
serve
data,
you
need
to
crypto
cryptographically
sign
the
data
as
well.
So
it's
very
easy
to
do
a
flaw.
Proof!
B
That's
but
that's
a
remote
plan.
Let's
get
back
to
what
we're
doing
now,
so
it's
still
deploying
so
on
the
who's
using
this.
So
we
talked
about
over
a
hundred
people
hosting
our
notes,
just
out
of
Goodwill
to
try
out
to
giving
feedback
for
for
now,
and
we
built
another
app,
it's
called
the
Titan
either
or
you
can
find
it
on
the
Google
Play
It's
a
basically
it's
a
total
privacy
Preserve
map,
the
location,
never
leave
your
phone.
The
way
to
do
this
to
build
all
the
features
to
make
it
offline.
B
So
the
cool
thing
is
this:
the
best
kind
of
you
user
of
Mastery
is
that
you
can
play
two
rows.
At
the
same
time,
you
can
beat
the
supply
side
and
demand
side,
and
if
you
download,
Titan
and
Titan
downloads
offline
map,
it
caches
on
your
phone
and
potentially
it
can
become
the
node,
the
server
of
another
phone
and
together
they
can
form
a
really
big
Edge
network,
using
the
phone
resources
and
5G
it's
gonna
and
five.
B
When
5G
comes
with
now,
it's
going
to
be
really
cheap
right,
so
you're
welcome
to
download
and
try
to
it
has
the
other
very
cool
like
a
cryptographic,
secure
or
because
cryptography
location
sharing.
You
can
share
your
location
with
others
with
in
encrypted
tags,
and
nobody
can
see.
No
single
platform
like
a
medium
can
see
on
your
message,
so
you're
welcome
to
to
try
it
and
that's
also
a
window
for
us
to
demonstrate
it's
already.
Building
on
map
free
services
and
by
the
way,
I
think.
B
Okay,
nowadays,
like
with
our
routing
engines
like
Valhalla
just
static
type
of
day,
data
can
give,
can
let
us
go
a
really
long
long
way.
A
lot
of
function
is
going
to
be
all
totally
out
offline,
because
most
of
my
computation
is
going
to
be.
It's
very
actually
very
simple,
and
the
data
is
more,
and
here
is
some
testimonial
from
Partners.
So
it's
basically
people
with
experience
it
in
locations
and
Cloud
Technologies.
We
are
activities
seeking
their
advices
on
how
to
build
the
math
free
better
and
how
to
grow
even
better.
B
B
B
All
right
so
we're
working
happily
so
or
you
can
also
have
a
dashboard
built
on
University
DB
and
infused
DB
and
grafana.
You
can
see
the
resources,
the
activities
things
like
that,
if
you
looking
try
to
catch
more
tiles,
hopefully
you
can
find
the
changes
of
your
tiles.
It's
going
to
catch
a
lot
of
tiles,
so
you
can
play
with
it
and
give
us
free
feedback.
That
would
be
very
important
and
I
think
that's
basically
about
about
it.
B
All
on
the
last
thing
is
like
we
I,
really
love
to
hope
to
work
it
with
the
Linus
foundations
on
the
multiple
fronts.
The
first
is
on
the
project
itself
to
be
a
better
open
source
project
to
take
the
architecture,
the
people,
the
developers
right,
the
co-cultures
and
second
thing
would
be
to
to
help
grow
the
ecosystems,
both
the
supplies
and
demand
side.
Let's
say,
Facebook
and
Uber
is
going
to
have
some
a
b
traffic.
B
Five
percent
of
traffic
go
try
on
the
hosting
some
note
or
using
five
percent
of
the
map
user
traffic
to
the
method.
That
will
be
hugely
helpful
and
the
third
part
will
be
together.
We
can
try
to
active
the
very
already
of
very
vibrant
osm
committees
to
to
meet
the
project
better
and
that's
I
think
that's
about
it
and
thanks
for
your
time
and
hope
to
get
good
how
to
get
some
questions
from
from
you
guys.
Thanks.
C
B
Oh
so
the
first
version
of
we
have
a
baby
businessman,
a
basic
Factor
map.
We
have
a
pipeline
to
to
consolidate
many
open,
geospace
or
data
days
and
to
generate
a
very
lean
you
you
use
usable
map
for
people
to
use
on
the
first
stage
is
and
that,
but
that's
all
for
the
base
layers
and
for
the
data
license
for
the
paid
service
proprietary
map
contents.
We
are
still
working
on
the
Sorting
all
the
license
point,
but
for
the
open
data
it
should
be
fine.
B
B
C
Cool
yeah-
this
is
this
seems
really
impressive,
and
you
know
I
love
the
the
vision
that
you've
outlined
in
terms
of
and
that
the
benefit
so
I
think
there's
probably
lots
of
opportunity
like
we
should.
We
should
talk
more
about
if
you're
interested
in
like
formal
proposal
about
this
project,
you
know
joining
the
urban
Computing
Foundation,
potentially
also
you
know
you
mentioned
just
working
with
the
Linux
Foundation.
We've
got
lots
of
sort
of
subject
matter:
experts
and
people
within
the
Linux
Foundation
here
who
can
probably
help
with
that.
C
C
B
C
Just
to
wrap
up
the
meeting
we're
about
a
time
so
great
project,
presentations
from
carto
map.
Three.
Thank
you
Javier
and
Isaac,
just
a
quick
run
through
of
like
upcoming
conferences
and
bookkeeping.
We've
got
two
Linux
Foundation
conferences
coming
up
the
open
source,
Summit
North
America
in
San
Diego,
there's
a
proposal
and
I'm
proposing
that
we
reschedule
our
Tac
meeting
to
be
on
Thursday
August
22nd.
So
if
anybody
has
feedback
about
that,
let
me
you
know.
Please
contact
me,
but
I'll
send
out
some
details
on
that.
C
The
open
source
Summit
in
Europe
in
we
have
like
the
urban
Computing
track
for
that
Linux
Foundation,
open
source
Summit
is
has
come
together
nicely.
We
have
two
days
worth
of
talks
around
Urban,
Computing
and
also
a
final
plug
for
that
spatial
data
science
conference
that
Javier
mentioned
in
New,
York
City
in
mid-october
as
well.