►
Description
How do you begin to build a product that can take any shape, communicate with a user in a number of ways and collect constant streams of data using an array of sensors? You prototype quickly and cheaply to allow for iterative validation. Fortunately for us, rapid prototyping for IoT has never been more accessible. Many tools are open source with vibrant maker community support, but you have to understand what’s available in the your toolbox to know what you can build.
Erica Stanley is a software engineer, researcher and tinkerer. She holds a B.S and M.S in Computer Science from Clark Atlanta University. She has also conducted post-graduate research at the University of North Carolina at Chapel Hill, specializing in graphics, data visualization and telepresence. Always eager to explore the intersection of emerging technologies and social science, Erica has worked with various technologies at the heart of IoT, including: sensor optimization, mobile & embedded development and cloud computing for over 16 years.
She is also passionately working towards gender and multicultural inclusion in tech, via education and exposure to opportunities. Erica co-founded the Atlanta network of Women Who Code, where she leads new developer workshops and organizes monthly tech talks, hack nights and networking events for women technologists. She also sits on the advisory board of 100 Girls of Code where she helps develop new strategies and curriculums to show girls ages 8 - 18 the possibilities of technology.
About Satellite:
GitHub Satellite 2016 was a celebration of the latest and greatest in development. Watch the recordings to learn how engineers, founders, activists, and beyond tackle the latest and most difficult challenges in software.
For more, join us in San Francisco on September 13th-15th for GitHub Universe, three days filled with the creativity and curiosity of the largest software community in the world.
For more information, go to githubuniverse.com.
A
So
my
name
is
Erica
Stanley.
Just
to
tell
you
a
little
bit
about
me
and
why
I'm
speaking
to
you
about
Internet
of
Things
today,
first
I
am
co-founder
and
director
of
women
code
Atlanta.
So
we
do
a
lot
of
things
you
know
from
getting
new
developers
started
to
helping
women
move
further
in
their
careers.
A
We
do
a
lot
of
workshops
and
one
of
the
things
that
I
found
is
that
there
are
a
lot
of
women
who
were
interested
in
getting
started
in
hardware,
prototyping
or
hardware
development,
but
a
lot
of
the
groups
and
in
Atlanta,
especially
we're
very
heavily
male
dominated,
and
so
we
needed
to
find
some
ways
to
kind
of
make.
People
feel
more
comfortable,
I'm
getting
started
with
Internet
of
Things
and
in
hardware.
So
that's
something
that
I
do,
but
also
this
is
kind
of
for
me.
A
This
gets
me
back
into
some
of
my
grad
school
research
and
using
things
like
sensor,
development
and
and
some
of
the
things
that
you
know,
I've
stayed
in
school
for
so
long
for
so
this
is,
you
know
me
getting
to
use
my
grad
degree.
So,
thank
you
so
I'm
an
80's
baby
I,
don't
know
how
many
80s
babies
we
have
in
here,
so
how
many
of
us
are
waiting
for
the
promises
of
the
80s
I
where's,
my
hoverboard,
because
what
they're
calling
a
hoverboard
that
doesn't
hover,
not
a
hoverboard.
A
A
A
So
my
sister
has
this
adorable
Chihuahua
that
thinks
that
he's
like
a
great
thing,
and
so
he
goes
around
policing,
the
whole
neighborhood,
and
so
when
he
gets
lost,
it
would
be
really
useful
to
know
where
he
is
so
that
the
other
dogs,
don't
just
devour
him.
So
you
know
that's
something
that
you
can
use.
A
You
know
that
kind
of
connected
device
for
our
cars
are
going
to
be
more
connected
cameras,
I
all
kinds
of
objects
within
your
home
and
really
all
of
this
is
doing-
is
just
building
on
the
hub
that
we've
created
already
with
with
our
desktops
and
our
laptops
and
our
mobile
devices,
so
using
that
as
a
hub
were
able
to
kind
of
expand.
This
ecosystem
of
connected
objects.
A
So
how
do
these
devices
actually
work?
So
you
have
this
constant
cycle
of
sensing,
something
in
your
environment
and
then
communicating
that
somewhere
and
then
doing
some
kind
of
decision
making
some
kind
of
computing
based
on
that,
and
you
also,
these
are
some
qualities
and
some
characteristics
of
typical
connected
devices,
so
their
hands-free
typically,
so
you
might
be
using
some
voice
or
gesture
control
and
that's
something
we'll
talk
about
a
little
bit
more
into
the
top
is
that
we
have
to
think
beyond.
Your
typical
screen
touch
screen
type
communication,
it's
always
on.
A
So
that
means
there's
something
you
have
to
really
is
power,
consumption
and
connectivity,
its
environment
aware,
so
it
knows
where
you
are.
It
knows
what's
happening
in
your
environment
again,
it's
connected
it's
attention-getting,
but
that's
something
also
that
we
have
to
be
aware
of,
because
let's
say
that
you're
in
your
car,
maybe
you
don't
want
something
getting
your
attention
if
you
have
to
focus
on
the
road,
also
you're
going
to
be
using
a
lot
of
third-party
apps
and
a
lot
of
ways
to
kind
of
connect
this
device
to
the
outside
world.
A
So,
let's
start
to
move
into
the
technologies
that
are
really
building
this
foundation
for
the
Internet
of
Things
and
the
biggest
one
is
networking.
So
here
is
just
kind
of
a
diagram
showing
some
of
our
networking
options
and
they're
growing
in
scale
with
with
the
height
in
this
document.
So
you
know
we
can
go
anywhere
from
NFC,
which
is
you
know,
just
a
few.
A
So
just
looking
at
some
of
the
contenders
for
networking
in
this
space,
and
we
see
that
bluetooth
and
ble
in
particular,
really
wins
in
a
lot
of
cases
because
of
the
low
power
consumption
and
what
is
able
to
to
transmit
well.
Ble
is
great
for
state.
So
keep
that
in
mind
and
a
lot
of
times
when
you
have
a
connected
device,
you
don't
need
to
transmit
video.
You
don't
need
to
transmit
a
whole
lot
of
data,
so
keep
that
in
mind,
vs
bluetooth,
which
can
transmit
all
of
these
things,
but
does
not
transmit
state.
A
So
you
have
your
typical
microcontroller
from
Arduino
and
then,
which
is
much
more
likely
Leonardo
if
anybody
remembers
kind
of
playing
with
Leonardo
kind
of
early
in
arduino
development,
but
it
also
has
a
linux-based
processor,
so
you're
able
a
Wi-Fi
enabled
linux-based
processor,
so
you're
able
to
do
some
things
out
of
the
box
with
the
unit
you
can't
with
a
lot
of
other
microcontrollers.
So
that's
something
to
keep
in
mind,
depending
on
how
important
Wi-Fi
connectivity
is
for
your
application.
A
Also,
you
can
actually
enhance
some
of
the
the
scripts
that
you
might
be
writing
to
communicate
with
sensors.
You
can
kind
of
enhance
that
by
you
know
creating
scripts
that
run
on
a
Linux
machine,
so
you
know
using
something
like
Python
to
actually
do
some
of
the
computing
from
some
of
the
data
that
you're
getting
back
from
the
sensors.
A
Then,
if
you're
interested
more
in
wearables,
which
is
kind
of
a
subset
of
this
Internet
of
Things
space,
then
you
might
want
to
think
about
the
lily
pad,
because
it's
a
fairly
powerful
device
for
being
as
small
as
it
is,
but
one
really
cool
thing:
it
has
slots
that
allow
you
to
sew
it
into
fabric
with
conductive
thread.
So
that's
something
that's
really
interesting
about
the
lily
pad.
Then
you
have
to
be
Google
bone
black,
which
is
always
a
contender.
A
A
So
particle
actually
has
their
own
cloud
platform
that
allows
you
to
kind
of
build
for
the
cloud
right
from
the
beginning,
so
things
like
over-the-air
updates
things
like
you
know,
just
being
able
to
communicate
with
other
REST
API
s
and
and
they've
taken
care
of
a
lot
of
the
things
like
security
and
scalability.
You
always
want
to
be
thinking
about
security,
but
it's
nice
to
be
built
on
top
of
a
secure
platform
from
the
very
beginning
and.
A
So
kind
of
the
extension
of
that
is
the
particle
electron
and
the
particle
electron
is
really
cool
because
it's
based
on
cellular
connection.
So
it's
always
connected
versus,
say
a
Wi-Fi
connection.
That's
only
connected
when
you
actually
make
that
connection.
So
that's
something
to
be
that
there's
something
that
I'm
actually
working
on.
Don't
have
my
particle
electron
gets,
but
that's
something
that's
kind
of
on
my
list,
my
next
toy
and
then
the
tehsil
is
really
cool
because,
instead
of
just
kind
of
figuring
out,
how
do
these
sensors
go
together?
A
So
if
you
look
here,
tehsil
has
all
kinds
of
modules
that
are
built
specifically
for
the
tehsil,
and
so
you
can,
you
know,
connect
an
accelerometer.
You
can
connect
the
camera,
you
can
connect,
you
know,
sensors
for
motion
and
LEDs,
so
it
allows
you
to
kind
of
build
your
device
as
you
go,
which
is
really
important
for
prototyping,
because
you
want
to
have
this
figured
out
before
you
go
into
production.
You
want
to
know
the
the
combination
of
sensors
that
really
help
you
build
the
device
that
you're
working
on.
A
So
next
we
move
into
sensors,
so
the
sensors
are
really
the
nervous
system
of
this.
This
ecosystem
we've
been
talking
about,
so
it
allows
you
to
kind
of
take
in
your
outside
world
by
understanding
you
know,
moisture
in
the
environment
by
acting
as
our
eyes,
by
incorporating
things
like
computer
vision
to
actually
understand.
A
So
things
like
that,
a
few
places
where
you
can
get
some
of
these
sensors
sparkfun
is
always
a
great
place
to
kind
of
start.
If
you're
looking
for
sensors,
even
if
you're
looking
for
microcontrollers
particle,
there's
several
included
with
the
maker
kit,
if
you
do
start
with
a
particle,
that's
my
best
advice
is
to
start
with
the
maker
kit,
because
you
have
a
lot
of
things
that
you
need
to
get
started
right
there
and
tehsil,
as
we
showed
before
they're
sold
separately
as
modules.
A
So
another
part
of
this
this
this
ecosystem,
that
we're
building
is
the
ability
to
be
able
to
host
a
lot
of
the
data
and
to
get
that
data
back
at
any
any
point,
and
also
we're
at
this
point
we're
hosting
huge
amounts
of
data.
If
you
think
about
something
as
simple
as
your
Fitbit
and
how
often
it's
updating
the
data
that
you're
sending
it
with
the
sensors
with
a
pedometer
sensor
that
you
have
that's
a
lot
of
data
and
we're
doing
that
on
a
larger
scale
with
the
Internet
of
Things.
A
So
that's
something
that's
going
to
be
a
huge
part
of
this
and
figuring
out.
Not
just
how
do
we
do
a
host
and
how
do
we
process
this
data,
but
also
what
are
the
stories
that
we're
going
to
be
telling
with
this
data?
You
don't
want
just
numbers
numbers,
don't
mean
anything
to
people,
so
we're
going
to
have
to
find
a
way
to
make
this
data
more
human
and
to
actually
analyze
it,
and
so
that's
something
that's
going
to
be
huge
with
the
Internet
of
Things
and
then
just
a
few
kind
of
auxiliary
technologies.
A
We
have,
as
I
mentioned
computer
vision,
which
helps
me
sheen's
see
the
way
we
do.
It
helps
machines,
interpret
visual
data.
We
have
augmented
reality
so
a
lot
of
times,
people
think
of
augmented.
Reality
is
just
placing
images
kind
of
in
a
projection
manner
and
in
front
of
say
a
camera
or
some
kind
of
visual
display,
but
you
can
augment
reality
in
a
lot
of
different
ways.
For
instance,
plantronics
has
been
working
on
an
augmented
reality:
headset
that
augments
your
environment
with
audio
but
also
kind
of
tracks,
head
movement
and
changes.
A
The
the
feedback
that
you're
getting
based
on
your
your
proximity
to
certain
things
and
based
on
your
work,
the
rotation
of
your
hand,
so
there
are
a
lot
of
different
ways
that
we
can
kind
of
incorporate,
augmented
reality
into
the
devices
that
we
might
be
building
and
then
there's
machine
learning
and
artificial
intelligence.
With
all
of
this
data,
and
all
of
this
that
we're
collecting
from
the
user,
it
might
be
really
useful
to
start
doing.
A
So
here,
if
you
open
cloud
or
open
source
cloud
solutions
for
IOT,
we
have
open
static.
We
have
apache
cloudstack,
which
is
java
based.
We
have
particle
cloud
platform
which
I
mentioned
earlier,
and
some
of
the
really
cool
things
about
the
particle
cloud
platform
are
the
over-the-air
updates,
the
REST
API
that's
available
to
you,
and
that
is
kind
of
already
built
in
with
a
kind
of
a
secure
platform.
A
It's
very
scalable
and
it
includes
real-time
messaging,
a
few
other
software
development
tools
are
eclipsed
ilt,
which
provides
open
source
implementations
for
many
different
ILP
cause
and
also
with
Eclipse
IOT.
You
can
develop
an
even
ojs
or
C++,
so
that's
really
cool
for
some
of
the
boards
that
are
not
some
of
the
older
boards
that
may
be
very
powerful,
but
maybe
haven't
kind
of
moved
on
or
evolved
to
using
some
of
the
higher-level
languages.
A
We
have
Hugin,
which
is
kind
of
an
open
source
rails
based.
If
this,
then
that
we
have
site
where
and
also
can
see,
key
and
I've
used,
Hugin
and
I'm
actually
working
on
a
project
based
on
that
right
now
and
so
far,
I'm
really
loving
it
so
I
know
that
we're
a
bunch
of
developers
and
I
know
we're
a
bunch
of
technologists
in
the
room.
A
But
when
you
think
about
Internet
of
Things,
you
have
to
take
into
consideration
design,
maybe
a
little
bit
more
than
we
have
in
the
past,
which
are
typical
web
or
mobile
app,
because
we
were
thinking
web
mobile,
we're
thinking
of
screens,
we're
thinking
the
kind
of
touch
interaction.
But
what
happens
when
we
don't
have
a
screen?
What
happens
when
the
interaction
is
not
just
touch
base
interaction?
So
here
is
the
plantronics
concept
model
I
mentioned
earlier.
We
have
the
thin,
which
is
this.
A
The
kind
of
white
and
blue
ring
object
that
you
see
there
and
that's
really
just
it
uses
gestures
of
your
fingers
to
kind
of
communicate,
different
things
in
your
environment.
We
have
the
meal,
which
is
a
band
that
you
put
on
your
your
arm
and
it
detects
that
electrical
impulses
on
your
arm
to
navigate.
You
know
different
applications,
different
things
in
your
environment
as
well,
and
this
really
weird
contraption
at
the
bottom
is
actually
detecting
brain
impulses.
A
And
so,
at
this
point
thought
is
your
your
interface,
it's
really
cool
stuff
and
it's
a
little
primitive
right
now,
so
you're
not
going
to
be
able
to
do
some
really
kind
of
complicated
navigation
of
applications,
but
you
can
do
really
simple
things
like
it
will
train
based
on
certain
impulses
that
is
receiving
from
you
and
it
will
based
on
that
training.
You
can
kind
of
connect
to.
You
know
an
action
that
you
want
to
happen.
A
So
now
that
we're
thinking
about
interaction
beyond
your
screens,
let's
talk
about
you-
know
the
transition
of
of
interaction
models.
First,
we
started
with
command
line,
which
is
very
static
and
very
disconnected.
We
moved
on
to
buoys,
which
we're
kind
of
we've
mastered
now
right,
and
that
goes
along
with
a
lot
of
the
screen
based
interaction
that
I
was
just
talking
about,
but
now,
when
we
start
talking
about
internet
of
things,
we
want
to
start
moving
into
more
natural
user
interfaces
or
organic
user
interfaces.
A
That
are,
you
know,
kind
of,
contextual
and
and
fluid,
and
a
user
just
understands
how
to
use
this,
and
so
when
we
start
thinking
about
organic
user
interaction,
we
have
this
organic
user
interface
model,
and
so
a
lot
of
the
things
you
can
do
here
are
our
kind
of
described
in
this
diagram.
So
when
you're
thinking
of
something
that
is
non
screen-based,
what
are
the
things
that
you
can
do?
Well,
you
can
speak
to
it.
A
You
can
manipulate
in
a
different
way,
so
not
just
touch,
but
also
you
know
drawing
so
one
of
my
favorite
I'm,
an
artist
in
my
spare
time
and
I
had
not
at
this
point,
found
the
device
that
made
drawing
as
as
intuitive
as
it
had
been
with.
You
know,
charcoals
and
pencils,
and
that
kind
of
thing,
and
so
getting
a
stylus
and
just
being
able
to
kind
of
have
a
stylish
that
responds
with
pressure
and
all
of
this
on
a
tablet,
just
change
the
experience
for
me,
because
that's
how
people
understand
that
particular
interaction.
A
A
A
It's
going
to
go
through
this
process
of
interpreting
the
meaning,
based
on
the
dictionary
that
you
have
and,
depending
on
how
complex
you
want
the
conversation
between
you
and
the
device
to
be,
you
need
to
have
either
a
really
robust
dictionary
or
you
can
have
a
really
simple
dictionary,
but
one
of
the
things
that
it's
going
to
do
to
determine
what
you're
saying
and
to
kind
of
process.
What
you're
saying
is
that
it's
going
to
call
an
experience
that
experiences
that
you
have
shared
with
that
device.
A
So
it's
going
to
kind
of
learn
gradually
and
and
become
a
better
conversationalist
based
on
your
experiences
and
the
vision
for
your
system
will
determine
how
you
put
these
components
together
and
what
components
that
you
use
again.
For
instance,
how
robust
is
your
you're
a
dictionary
for
determining
the
processing
for
your
natural
language,
and
when
are
we
going
to
be
using
voice
inter
interfaces?
This
is
probably
going
to
be
cases
where
touch
interaction
is
not
desired,
so
you're
going
to
be
doing
kind
of
in-car
interaction.
A
So
now
we
move
into
the
manipulation
portion
of
the
organic
user
interface
model
and
again
these
are
natural,
real
world
interactive
interactions.
So
there's
two
main
types:
there's
deformable
and
non
deformable
non
deformable
is
when
the
object
is
not
changing
shape
because
of
your
interaction
with
it.
So
that's
your
basic
screen
based
touch,
texting,
tangible
handling
of
that
object,
but
when
we
get
move
into
the
formable
user
interfaces,
you're
thinking
more
of
something
that
changes
the
shape
because
you
interact
with
it
and
you
want
to
keep,
for
instance
like
clay.
A
So
there
are
some
studies
where
people
are
using
kind
of
clay
that
has
certain
sensors
in
it
so
that
it
can
respond
to
to
kind
of
the
user,
interaction
and
you're,
seeing
what
you're
building
one
screen
based
on
that
interaction
with
the
clay
that
has
these
sensors.
So
that
would
be
an
example
of
a
deformable
user
interface.
So
the
input
for
these
interactions
are
using
kind
of
thing,
free
hand,
manipulations
that
change
the
shape
and
they
rely
on
skills.
A
A
A
So
some
best
practices
in
2d
spaces.
You
want
to
ensure
that
you
have
about
15,
millimeter
surfaces,
radius
for
interactive
elements,
and
this
isn't
all
that
different
from
a
lot
of
the
things
when
you're
building
interaction
patterns
for
a
mobile,
for
instance,
you
always
want
to
make
sure
you
have
that
about
40
pixel
square
area
for
forefinger
interaction
on
mobile
devices.
So
it's
very
similar
to
that
in
3d
space.
Is
you
want
to
consider
a
visual
feedback
for
from
movable
objects?
A
So
the
user
actually
knows
what's
happening,
and
you
also
want
to
make
sure
that
you
have
something
like
that,
so
the
user
doesn't
think
it's
broken.
So
you
want
to
make
sure
that
it's
kind
of
real-time
feedback
there.
You
may
want
to
consider
haptic
feedback
if
you're
working
with
touch
based
or
deformable
models,
and
you
want
to
again
respond
to
every
contact
with
them
feedback,
and
you
want
to
enable
basic
direct
manipulation.
You
don't
want
to
have,
for
instance,
the
equivalent
for
a
touch
screen
would
be
nested.
A
You
know
pretty
much
nested
menu
hell,
but
you
don't
want
to
do
that
for
these
kinds
of
interactions.
You
don't
want
to
have
really
complex
interactions
that
are
coupled
together
to
do
one
action.
So
that's
not
really
what
you
want
here
and
so
now,
you're
moving
on
to
gestures,
and
so
you
can
have
in
air
gestures,
like
all
of
the
hand,
movements
I'm
doing
right
now.
Speech
like
like
speech
is
one
of
the
oldest
forms
of
communication,
and
this
is
used
to
either
issue
a
command
or
communicate
with
devices.
A
A
Facial
expressions,
body
motion,
I
tracking-
you
would
use
this
when
the
interaction
is
with
a
large
display
or
situations
requiring
you
not
to
touch
anything.
You
will
want
to
avoid
it
where
errors
can
have
life-threatening
impact.
Because,
again
you
have
to
train
the
system
to
understand,
for
instance,
our
hands
move
and
how
eyes
may
be
moving
and
it
may
interpret
that
incorrectly.
So
you
don't
want
to.
A
A
So
if
this
is
your
gesture,
you
know
you
want
to
make
sure
that
you
don't
have
to
say
a
gesture
that
has
just
one
kind
of
enhancement
to
that,
because
that's
going
to
be
a
little
bit
confusing
to
your
system,
so
you
want
to
have
very
clear
start
and
stops,
and
you
want
to
handle
false
gestures.
You
know
kind
of
false
positives
and
false
negatives.
A
Another
recommended
micro
pattern
or
interaction,
pattern
or
micro
interaction,
so
these
are
contained
product
moments
that
revolve
around
a
single
use
case
again,
like
I
said
earlier,
you
don't
want
to
have
a
lot
of
complex
gestures
or
a
lot
of
complex
interaction.
You
want
something
that's
very
simple
and
brings
about
a
very
simple
and
expected
action,
and
so
now
that
we've
kind
of
combined
the
technology
that
is
the
foundation
of
Internet
of
Things-
and
we
talked
a
little
bit
about
how
you
would
consider
designing
for
these
kind
of
connected
devices.
A
Now
we
can
kind
of
move
into
prototyping,
because
this
is
really
a
mix
of
design
and
technology
in
the
thinking
and
the
making
of
it
and
it's
fundamental
to
creating
projects
with
new
features,
something
that
you
really
want
to
kind
of
enhance
something.
You
already
have
it's
really
important
to
think
about
the
tech
as
well
as
the
design.
This
applies
even
more
to
Internet
of
Things,
as
hardware
is
usually
involved,
and
so,
when
you're
building
your
team.
A
This
is
really
interesting
because
in
a
lot
of
cases
you
have
either
a
hardware
team,
either
a
design
team
or
a
software
development
team.
But
this
prototyping,
this
kind
of
prototyping
really
requires
everybody
to
be
working
together.
It
requires
you
know
instant
feedback
from
software
developers
from
you
know,
designers
from
your
hardware
developers,
so
you
want
to
build
a
team
that
has
overlapping
skills,
so
you
can
want
somebody
who,
for
instance,
understands
the
hardware
side
and
the
software
side.
A
So
hackathons
are
a
great
place
for
that,
and
you
want
to
look
for
hackathons
that
are
focusing
on
either
Internet
of
Things
or
wearables,
and
so
really
what
you're
doing
want
to
kind
of
build
this
on
a
ling
collaborative
process.
So
you
want
to
learn.
You
know
from
your
prototype
and
then
build
and
measuring
you
want
to
do
that
in
this
cycle
of
just
kind
of
collaborating
with
these
people
at
the
table,
and
so
how
do
we
combine?
A
You
know
kind
of
design,
thinking
when
were
talking
about
interaction
models
and
kind
of
building
products
based
on
this,
which
is
kind
of
design,
thinking
and
and
kind
of
lean.
How
do
we
can
combine
those
and
from
that
we
get
design
sprints,
and
so
this
is
this
constant
cycle
of
understanding
your
problem,
understanding
your
audience,
defining
what
you
want
to
build,
based
on
or
defining
that
define
and
clearly
your
audience
and
then
the
verge
point
is
when
you're
kind
of
figuring
out.
What
can
we
do?
A
What
are
all
the
things
that
we
can
do
in
this
space
that
will
meet
the
needs
of
this
audience?
That
will
solve
the
business
problem
and
then
you
kind
of
converge.
You
decide
on
several
of
these
features
and
you
start
building
a
prototype
based
on
that
and
then
you
start
validating
it
and
you
just
go
through
this
process
continually
until
you,
you
feel,
like
you
have
the
right
product,
so
the
first
step
is
to
understand,
and
so
this
is
where
we
want
to
figure
out.
Who
are
our
users?
A
What's
the
context?
What
are
their
needs?
So
you
really
want
to
understand
that
very
well
before
you
go
any
further
down
this
process,
then
you
want
to
do
that
step.
I
mentioned
diverge.
This
is
where
you
come
up
with
all
of
your
right.
This
is
the
brainstorming
step.
You
come
up
with
all
of
your
ideas,
anything
that
you
could
possibly
think
of
that
will
meet
these
users
needs
and
it
doesn't,
it
can
be
almost
fantastical.
It
can
be
whatever
you
want
at
this
point.
A
We're
not
worried
about
feasibility
here,
and
so
a
part
of
this
is
when
you
find
that
critical
user
story
for
the
user
that
you've
just
defined-
and
this
is
going
to
highlight
you-
know
the
story
most
critical
to
the
challenge-
that
user
has
and
you're
going
to
make
sure
that
you
kind
of
diagram
where
your
customer
starts
in
this
process
and
where
they
should
end
up
at
what
needs
to
happen
along
the
way.
And
what
this
allows
is
a
kind
of
inspection
of
the
inventive
aspect.
Much
sooner
in
the
prototyping
process.
A
You
don't
want
to
get
all
the
way
to
the
end
of
that.
That's
that
I
showed
you
before
this
validate
step
to
realize
that,
oh
maybe
we
should
have
done
something
a
different
way.
You
want
to
be
thinking
of
your
user
from
the
very
beginning
of
this
process,
and
so
that's
what
that
helps
you
to
do
then,
based
on
the
critical
diagrams
that
you
and
the
team
have
come
up
with
you're,
going
to
decide
on
what
is
the
most
important
user
story
and
you've
got
a
storyboard.
A
That
idea,
once
you
kind
of
have
a
storyboard,
then
you're
ready
to
prototype,
because
you
stopped
this
through
now.
You
thought
through
the
interaction
and
you
can
build
something
kind
of
quick
and
dirty
to
show
your
users.
It
can
be
a
paper
prototype
or
it
can
be
a
little
bit
more
high
fidelity
and
then
you're
going
to
validate
that
you're
going
to
kind
of
test
that
with
people
you're
going
to
be
doing
a
lot
of
user
testing
and
kind
of
user
discovery
here.
A
Well,
you've
already
done
your
user
discovery,
but
you're
going
to
go
around
to
your
potential
users
kind
of
ask
them.
How
would
they
use
a
system
like
this?
How
could
we
make
it
better
and
then
you're
going
to
kind
of
go
right
back
to
the
drawing
board
and
figure
out
how
you
can
make
this
continually
better,
so
that
was
the
typical
process
of
the
design
sprints
for
say:
maybe
a
software
product,
but
now
how
do
we
take
that
and
move
that
to
this
really
integrate
a
team
of
hardware
engineers
and
software
engineers
and
designers?
A
So
in
the
first
step,
we're
going
to
perform
that
together?
We're
going
to
understand
our
customer,
our
users,
we
understand,
you
know
what
drives
them
and
then
what
are
their
problems
and
what
is
the
challenge
and
we're
going
to
clearly
define
that?
But
at
that
point,
during
the
diverge
cycle,
we
can
have
our
hardware
engineers
going
off
to
figure
out
what,
in
the
hardware
space
makes
a
lot
of
sense
to
solve
this
problem.
We
can
have
our
software
engineers
kind
of
going
off
and
figuring
out.
What
did
it?
What
is
it
software
wise?
A
What
you
know
is
there
a
mobile
app
that
can
solve
this
problem?
What
what
can
we
do
with
software,
and
we
can
also
have
our
designers
thinking
about
that.
So
the
verge
process,
you
can
actually
do
separately,
but
then
you're
going
to
come
back
together
for
the
decide-
and
this
is
where
we're
bringing
all
the
ideas
together
and
we're
figuring
out
that
critical
path
diagram,
and
so
once
we
we've
come
together.
We've
decided
that
this
is,
you
know
a
great
idea,
or
we
have
some
really
great
idea.
A
This
is
the
idea
we're
going
to
work
on
then
we
break
up
again
and
we
have
on
the
engineering
side.
We
have
the
breadboarding
and
the
conceptual
hardware
testing
on
the
designers
side
or
in
the
US
side
we
might
have
wireframes,
might
be
developing
wireframes
and
something
that's
really
cool
when
you're
in
this
design
phase.
A
For
the
hardware,
a
really
cool
tool
to
use
is
fritzing,
so
Fritzing
is
open-source
electronic
design,
automation,
software
and
it
aids
in
collaboration,
because
it's
actually
really
simple
to
use
and
you
can
import
a
lot
of
the
those
boards
that
I
mentioned
earlier.
You
can
kind
of
import
them
and
move
things
around
and
see
what
this
does
to
change.
The
the
prototype
that
you're
building
the
hardware
side
of
it
and
designers
can
kind
of
get
involved
as
well
here.
A
A
So
now
that
you
know
we
have
the
step
where
we
have
our
hardware
engineers
kind
of
working
on
their
design,
and
we
have
the
designers
working
on
on
the
wireframes.
Then
we
also
want
our
software
people.
Thinking
about
the
data
modeling,
what
is
the
data?
That's
going
to
be
processed
with
this,
the
solution
that
we've
come
up
with,
and
we
want
to
be
thinking
about
that
because
we
want
to
think
about
you
know
again,
you
don't
just
want
to
throw
data
up
to
the
cloud.
A
We
want
to
create
stories
that
are
actually
going
to
help
people
use
this
product
to
make
this
a
really
usable
product.
So
that's
what
our
software
engineers
can
be
doing
at
that
point.
Software
engineers
data
scientists,
that
kind
of
thing
then
we'll
come
back
together
and
we'll
start
working
on
the
low
fidelity
prototype,
and
so
this
might
be
something
that's
you
know.
Maybe
you
started
with
a
particle
or
maybe
you
started
with
arduino
and
kind
of
putting
all
these
things
together.
A
Some
of
the
things
from
the
wireframe,
some
of
the
things
from
you
know,
data
modeling,
we're
going
to
start
working
on
the
firmware
at
this
point
and
from
that
step
we're
going
to
have
kind
of
an
alpha
prototype
kind
of
the
hardware
and
the
initial
firmware
from
this
step.
For
the
US
side,
we're
going
to
have
visual
design
and
kind
of
the
specs
from
a
visual
standpoint,
and
now
we're
working
on
the
UI
development
and
keep
in
mind
this
is
not.
Does
it
have
to
be
your
typical
screen
base
UI
development?
It
could
be.
A
You
know
some
of
those
other
interaction
models,
I
described
earlier
and
so
now
pretty
much.
We
built
the
system
at
the
step
before
now.
How
does
the
user
interact
with
it
and
we
were
taking
in
consideration?
You
know
things
that
we
came
up
with
in
the
step
where
we
developed
the
wireframes
and
also
the
step
where
we
were
doing
the
visual
design
specification,
and
at
this
point
you
can
put
that
together
and
validate
and
in
the
validating.
A
We
might
also
determine
that
we
want
to
build
a
higher
fidelity
prototype
to
really
test
some
of
these
things
out,
and
so,
when
we
get
to
high
fidelity
prototypes
for
connected
devices,
3d
printing
is
really
helpful
with
that.
So
you
want
to
use
a
really
decent
3d
printer.
You
wanted
something
with
you
know:
greater
than
100
micron
resolution,
preferably
or
less
than
micron
100
micron
resolution,
preferably
50.
You
wanted
to
be
able
to
print
numerous
material
types,
including
flexible
materials
and
and
you're,
going
to
have
this
test
loop.
A
Where
you
identify
candidates
for
field
testing,
you
don't
want
to
get
to
this
high
fidelity
stage
without
really
betting,
some
of
the
low
fidelity
prototypes
that
you
were
working
on
and
because
you
know,
they're
not
terribly
expensive
and
the
material
that
you're
using
is
not
terribly
expensive.
You
want
to
iterate
really
quickly
and
make
any
necessary
physical
changes
here.
Certainly
before
you
decide
to
move
this
into
production.
This
is
where
the
mistakes
are
made.
A
This
is
where
you
can
kind
of
fail
fast
and
just
thinking
of
some
applications
of
connected
device
of
a
connected
device
design
sprint.
We
have
you
know
your
your
basic
hardware,
MVP
I
mentioned
hackathons
and
hardware
projects
involving
teams
and
adding
features
to
an
existing
connected
device,
and
so
that
is
pretty
much
it
it
are
there.
We
see
if
you
have
any
time
for
questions.
We
have
like
five
minutes
so
I
can
take
some
questions.
If
you
have
any.