►
From YouTube: Jason Kridner - Beagleboard Founder
Description
Conversation with Jason.
https://wiki.opensourceecology.org/wiki/Jason_Kridner
-----------------
What you see here at Open Source Ecology is an ambitious open source project for the common good. Join our development team:
http://opensourceecology.org/wiki/OSE_Developers
We run One Day Extreme Build workshops in many locations. Please put yourself on our map if you'd like us to host a workshop near you:
https://microfactory.opensourceecology.org/request-a-workshop/
Take a minute to subscribe to our email newsletter (updates, workshops, etc): http://bit.ly/1LtcM44
A
B
Gerald
and
I
co-founded
BeagleBoard
org
Gerald
Coley,
is
the
has
been
the
he
started
out
doing
all
of
our
hardware
designs,
I
heard
the
specs
kind
of
managed
the
community
did
the
yeah,
the
the
community
facing
side
of
things.
Eventually
he
went
on
to
he
left
T
is
his
day
job.
His
day,
job
became
a
consulting
business
yeah.
C
B
Hardware
design,
so
you
know
basically
he
was
gonna,
have
to
start
charging
me
for
doing
any
other
work
and
he's
expensive,
so
yeah.
So
we
ended
up.
You
know,
kind
of
parting
ways
a
little
bit.
You
know
we
still
interact,
but
we're
the
two
that
started
equal
board
up
and
and
I've
stuck
with
it.
Some
currently,
president
of
the
board
and
my
my
wife
is
the
executive
director
of
the
foundation,
nice.
A
B
A
B
C
B
B
You
know
I
was
you
know
at
the
time
there
were
the
you
remember:
the
One
Laptop
Per
child
yeah.
That
was
a
little
bit
of
an
inspiration,
I
think
some
of
the
stuff
from
from
work
there
was
they
used
to
be
these
DSP
starter
kits,
but
we
kind
of
had
a
little
bit
of
magic.
I
thought
with
the
at
the
time.
Ti
was
big
in
the
this.
The
cellular
phone
business
mm-hmm.
C
B
The
ARM
processors
were
starting
to
get
to
be
fast
enough
to
run
desktop
operating
systems
on
them
and
that
we
could
make
you
know
something
super
affordable,
mm-hm
and
super
hackable
disposable
computer
right.
We're,
like
you,
wouldn't
worry
about.
You
know.
We'd
always
talk
about
it's
times
like
your
hard.
C
B
B
You
know
a
lot
of
this
stuff
that
I
think
later
was
picked
up
by
Raspberry,
Pi
and
I.
Think
that
they've
done
a
really
great
job
at
kind
of
fulfilling
some
of
our
original
mission,
but
but
one
of
the
things
that
they
really
haven't
fulfilled
is
that
it's
not
just
about
you
know
programming,
cheap
computers
that
ultimately
it's
about
building
skills
that
actually,
you
know,
translate
into
individual
entrepreneurialship
right.
B
So
actually
designing
appliances,
not
applications
right,
not
creating
things
that
run
out
of
other
people's
platforms,
but
actually
building
some
electronics
such
that
you're,
no
longer
a
slave
to
technology,
but
actually
empowered
by
technology,
yeah
and
and
I.
Think
that's
the
kind
of
thing
that
we
we
tend
to
want
to
run
counter
to
is
mmm-hmm
the.
It
is
the
idea
that
I'm
going
to
write
a
write,
some
code
or
write
something
for
somebody
else's
platform,
where
I'm
going
to
create
something
that
works
seamlessly
into
my
life
and
doesn't
become
an
extra
distraction.
Ya.
B
B
A
Yeah,
yeah,
and
and
actually
for
the
first
time
like
you
know,
lifetime
design
is
a
big
promise
and
I'm
only
beginning
to
see
that
it's
been
like,
say.
You
know
five
years
that
I'm
driving
the
same
tractor,
but
the
cool
thing
is,
you
know
the
other.
Last
winter
I
swapped
out
the
engine,
the
power
cube
in
like
an
hour
right.
So
in
time
you
can
see.
A
Yes,
the
promise
of
lifetime
design,
which
to
me
is
one
of
the
central
promises
of
open-source
is
being
realized
and
it's
hard
to
see
up
front
because
you
know
we
we've
been
around
for
a
decade,
but
the
value
proposition
becomes
more
clear
as
time
goes
on
for
us
on
the
lifetime
design
aspect.
That's
that
to
us
is
huge.
Like
you've
seen
my
TED
talk,
I
haven't.
B
A
You
got
to
watch
it
too
bad
because
it's
good,
it's
really
good
I'll
leave
it
at
that.
The
idea
was
there
that
the
tractor
broke
down
I
paid
to
get
it
repaired,
then
it
broke
again
and
pretty
soon.
I
was
broke
too,
and
that's
why
I
built
that
first
one
and
yes,
that
promise
is
that
definitely
people
being
in
control
of
their
technology.
Yes,
we're
not
start
serving
it
there.
It's
serving
us
I'm!
Sorry,
you
cut
out
your
voice.
I
can't
hear
you
anymore
there,
oh
they're,
no.
B
C
B
That's
absolutely
you
know
one
of
my
personal
motivations.
You
know
I
easily,
see
us
going
into
a
world
of
I'll
call
them.
You
know
Gandalf's
in
hobbits,
mm-hmm
we're
a
small
number
of
people,
essentially
own
the
knowledge
of
Technology
and
everyone
to
everyone
else
is
just
magic
or
I.
Think
they
can't
even
see
how
things
are
done
right.
It's
just
it's
beyond
them
and
and
I
think
we
could
easily
have
ourselves
fall
into
the
sort
of
world
and
I.
Think
that's
it's
not
the
world
I
want
to
live
in.
B
B
Done
some
things
to
make
a
difference
and
at
the
corporate
level
I'd
say
that
as
I've
gotten
more
Eagle
focused,
my
overall
corporate
footprint
has
shrunk.
My
focus
has
been
much
more
outward
and
really
enabling
people
outside
to
leverage
technology
and
bringing
some
aspects
of
it
out,
but
we
we
we
have,
I
Connor
processor
side,
one
of
the
big
changes
that
we
made.
Is
that
all
what
we
really.
B
That
was
one
thing.
I
was
instrumental
in
changing
at
ti,
so
we
now
have
a
mainline
first
approach.
A
new
SSD
will
be
coming
out
many
many
months
ahead
of
there
being
a
public
announcement
and
and
ship
availability.
There
will
be
patches
on
the
shared
to
the
mainline
Linux
kernels,
to
support
the
processors
ahead
of
released.
A
B
B
B
So
for
a
new
processor
right,
you
have
to
essentially
rewrite
portions
of
the
Linux
kernel
to
have
it
support
the
peripherals
and
capabilities
of
the
hardware,
and
so
we've
made
sure
that
looks
so
so
when
a
bunch
of
builds
bun
mm-hm,
they
take
the
Linux
kernel
from
Linus
Torvalds
right.
So
there
is
a
one
Linux
kernel,
which
is
you
main
tank,
violinist
Torvalds.
You
know
he
handed
off
control
/
correct
for
Hartman
for
a
few
months,
but
essentially
you
know
he
at
that
version
of
Linux
is
analytics.
So
that's
what
we
call
mainline
mm-hmm.
B
So
there's
a
process
of
you
know
full
community
review.
That
happens
in
order
to
make
changes
to
that.
That's
software,
so
different
people
own
different
subsystems
that
essentially
trust,
and
then
they
build
up
patches
and
ultimately
apply
patches
too,
like
he.
He
accepts
those
patches
into
his
version
of
the
software.
B
There's
a
very
open
development
model
for
doing
that,
and
so
and
that's
all
under
public
mailing
list,
so
anybody
can
say
I,
don't
like
that,
you
shouldn't
do
it
that
way:
it's
not
a
good,
it's
not
good
style,
post
structured,
you
know,
I,
don't
you
know,
I,
don't
understand
it.
It's
not
clearly
documented
anything
and
the
person
who
wrote
it
in
order
to
get
it
accepted
has
to
answer
those
has
to
be
accountable
to
those
questions,
and
so
it's
a
it's
a
it's
just
a
very
different
companies.
B
Don't
generally
like
to
do
such
things
for,
like
oh
you're,
saying
my
deadline
is
based
upon
what
some
kid
sitting
in
a
cornfield.
You
know
with
an
internet
access,
it
says
about.
You
know:
I
should
be
writing
into
my
code.
That
doesn't
seem
very
fair,
but
that
is
what's
fair
and
you
know
so
we
have
a
you
know.
The
Linux
kernel
has
that
sort
of
governance
model
yeah
anybody
could
complain
and
say
no
and
anybody
can
say
hey.
It
looks
good.
B
C
B
Developers
the
developers
would
want
to
just
like
okay
I
can
show
to
my
manager
that
this
code
works
and
demonstrate
it,
and
then
let's
ship
it
to
a
customer
right-
and
that
was
just
that
was
the
development
process
right
they
get
things
working,
they
prove
to
their
manager
that
the
code
works
and
you
ship
it
to
customers.
That's
not
the
development
process
anymore.
Now,.
D
B
Development
process
is
they
get
it
working.
They
have
to
submit
it
to
a
public
mailing
list
for
its
review
to
be
included
in
the
mainline,
Linux
kernel
and
they're
not
done
until
that
community
review
cycle
immersed
large
window
actually
completes
it
so
and
that
we,
the
ultimate
selling
point
back
into
the
corporate
management,
is
that
that
greatly
reduces
the
ongoing
support
churn
because,
as
soon
as
your
code
is
accepted
into
the
mainline,
and
if
anybody
breaks
that
code
or
does
anything
it's
their
responsibility
to
fix
it.
B
C
B
B
B
A
Yeah,
interesting,
okay,
so
let's
let's
skip
into
the
steamcast,
it's
I
think
in
education
and
how
that
commit
process.
I
think
I
can
I
think
there's
probably
some
stuff
I.
Could
we
could
learn
for
open
source
ecology
of
how
to
manage
a
project
for
real,
so
I
think
so
part
of
this
whole
effort
of
the
team
is
is
to
also
push
that
forward,
but
for
now
so
would
you
actually
be
interested
in
running
the
steam
cams?
Do
you
have
the
time
to
do
that
or
that's
you
think
we
can
help
in
other
ways
or.
B
B
C
B
C
A
Me
yeah,
so
tell
me
more
about
the
distinction
in
practice.
How
do
you
see
the
you
know,
working
practically
with
raspberry
pie
versus
BeagleBone,
the
difference
that
the
so
you
you
have
the
blueprints
open
so
that
if
you
went
out
of
business,
somebody
else
could
produce
it
or
things
like
that
in
practice.
Like
what
do
you
tell
me
more
about
what
you
see
the
difference
between
the
two.
B
B
We
have
a
lot
of
people
who's
who
that
sourcing
and
longevity
really
matters
to
mm-hm.
You
know
we
have
people
that
put
our
boards
up
into
satellite
networks,
there's
one
of
them,
that's
their
their
claim
to
fame.
As
this
every
every
square
meter,
every
hour
of
the
earth
is
photographed
by
BeagleBone
of
using
x-ray
x-ray
cameras
and
satellites.
C
B
B
B
Think
they
care
about
it,
there's
also
some
hardware
capabilities
that
tends
to
make
it
better
for
more
electronics
and
mechatronics
types
of
applications.
We
have
some
microcontrollers
inside
the
SOC
need
that,
while
doing
hard,
real-time
control
a
lot
of
a
lot
of
folks
there.
Certainly
you
know
that
they're
kind
of
skewing
professional,
more
just
based
on
robustness.
B
B
You
know
they
kind
of
people
supply
chain
right
and
get
full
documentation
on
all
the
components
used
on
the
board.
They
just
feel
more
in
control
plus
they're,
more
capable
of
doing
the
hard
real-time
control
things
that
you
do
and
like
a
mechatronics
sort
of
environment
we
need
used
and,
like
gem,
all
uses
us
in
their
newest
laser
cutter.
The
first
ul-listed
laser
cutter,
which.
A
B
A
B
So
thermal
is
named
a
laser
cutters.
There
I
remember
who
they
hang
fine
off
of.
They
bought
the
design
off
to
somebody
that's
more
common
and
laser
cutters,
but
anyway
it's
the
first
ul-listed,
which
is
like
a
safety
qualified
yep
laser
cutter
and
they
just
they
put
a
BeagleBone
inside
the
product
right.
It's
actually
driving
the
controls
through
the
laser
all
right
so
moving
the
motors
it
produces.
The
stepper
pulses
for
the
stepper
motors,
as
well
as
you
controls
the
electronics
port
of
the
laser
strength.
A
B
Yep,
okay,
so
machine
control
type
things
are
something
we
do
really
well.
You
know
see
I
you.
B
B
C
Know
but.
A
B
B
C
Yeah
yeah,
what's.
B
A
And
in
practices
Oh
so,
for
example,
for
practical
considerations
on
our
projects
say
we
wanted.
We
already
said
you
pointed
out
to
us:
okay,
raspberry
pi,
tablet
kind
of
deal.
Can
we
do
the
same
I'm
sure
we
can?
But
the
question
is
how
much
prior
art
there
is
an
open
source
to
make
it
happen.
I
mean
that's
the
question.
We're
gonna
be
up
against
or
do
you
know
any
any
good,
raspberry
BeagleBone
tablets
that
that
exist
yeah
so.
B
B
I
mean
that's
the
easiest
thing
to
do.
We
have
some
companies
start
up
like
there's
one
company
always
innovating
that
had
this
kind
of
interesting
tablet,
design
where
they
took
one
of
our
boards
and
it
it
could
be
a
tablet
or
you
can
slide
it
into
a
keyboard.
It
could
be
a
laptop
or
in
three
different
modes.
I,
don't
know,
but
I
think
there's
a
lot
of
people
that
have
done
like
custom
tablets,
but
you
want
a
continuity
supply
right.
That's
critical!
B
If
you're
going
to
invest
educational
materials
around
it,
you
need
a
real
continuity
of
supply
and
I
wouldn't
steer
you
towards
any
evil
specific
laptops.
I'm.
Just
looking
for
someone,
that's
open
hardware!
That's
for
a
party
that
we
can
just
reuse.
There's
not!
You
know
if
you're
talking
about
the
meets
USB
keyboard,
USB,
touchscreen,
yeah,.
A
Essentially,
you've
got
a
touchscreen,
so,
for
example,
I've
got
my
I
mean
ideal
for
me,
I've
got
my
little
cell
phone
I'd
like
to
replace
that
with
a
tablet
that
I
could
also
put
a
phone
on
yeah
they'll,
be
the
dream
product,
okay,
but
the
first
step
is
just
to
replace.
Like
you
know,
sometimes
in
the
morning
I
don't
go
on
a
computer.
I
go
on
a
tablet
to
do
some
things.
I
want
to
get
rid
of
the
proprietary
stuff
that
is
throwaway.
I
want
to
have
a
lifetime
product
so
shares.
C
B
Yeah
for
sure
yeah,
because
we've
had
so
many
of
these
kind
of
like
different
little
fringe
products
with
you
being
offered
Hardware
people
make
certain
things
like
one
of
the
one
of
the
interesting
is
the
open,
moko
phones.
We
actually
have
people
doing
open
hardware,
phones
based
off
her
design,
all.
B
B
B
It's
very
expensive
to
make
the
design
I
think
the
other
reticles.
They
say
that,
like
the
the
the
blueprint
for
a
piece
of
silicon
is,
you
know,
could
be
millions
of
dollars
to
develop.
So
you
have
to
sell
a
lot
of
them
even
though
individually
they
may
not
cost
much
mm-hmm
after
so
a
lot
of
them
in
order
to
make
your
money
back
so
and
there's
interesting
efforts
here
on
risk
dog,
yep.
C
B
We
shouldn't
definitely
be
you
should
be
paying
attention
to
that.
Helped
reduce
some
of
the
development
costs
that
ultimately,
like
the
what
they
call
takeout,
or
you
know
that
that
process
of
actually
making
the
first
ship
is
very
expensive.
You
know
there.
There
are
a
number
of
mostly
Chinese
manufacturers
that
make
processors
that
used
in
a
very
high
volume
in
little
Android
tablets.
Some.
B
C
B
C
C
Yeah
and.
A
B
Si
sort
of
so
what
I'm
just
trying
to
pull
it
down
to
is
you
can't
have
to
set
a
performance,
go
away.
What's
the
software
that
you
need
to
run
on
here
really
well
and
let's
make
sure
if
you
can
run
that
software
really
well,
if
you
want
like
this
continually
upgrade
to
run
the
latest
and
greatest
software
like
the
always
chasing
the
latest
Android,
for
example,
you
know
yeah
three
for
three
years
or
two
three
years
from
now,
you're
gonna
have
to
redesign
it
again
and
that's
probably
not
something
you
want
to
do.
Yeah.
A
B
B
You
know
there
are
people
certainly
focused
on
on
unstable,
for
phones,
I,
think
the
some
stuff
in
that
like
well
used
to
be
the
high-low
text
stuff
out
of
the
Media
Lab
at
MIT
and
Dave
MELAS
was
working
on
a
phone
learning.
He
was
going
back
to
essentially
like
right,
really
simple
phones
that
didn't
run
genron
gooeys
at
all,
there's
just
a
phone
just
a
phone,
but.
B
Still,
we
still
constantly
see
Nicolas
maintaining
the
patches
for
it,
but
it's
just
it's
stable,
but
I,
don't
know
if
he's
gonna
like
what's
this
plans
for
doing
additional
production
runs
and
that
sort
of
thing
that's
just
and
I,
don't
know
who
else
is
kind
of
picked
off
the
mantle
for
that
development?
Okay,.
C
C
C
B
C
A
B
B
B
B
B
The
the
server
just
needs
to
give
a
shared
documents
do
the
the
physical
interactions,
so
we
can
run
web
servers
very
quickly
and
then
the
Chromebooks
you'll
run
their
browsers.
We
have
the
development
environment
entirely
upper
of
cognizant
Co,
cloud9
IDE.
We
don't
use
cloud
servers.
The
servers
on
the
board
boards
run
no
js'
and
they
run
they
serve
up
cloud9
IDE
and
you
get
to
the
Linux
command
prompt
and
and
have
a
nice
interactive
environment.
There
I'm
happy
to
send
you
some
boards.
B
If
you
wanted
her
buddy
and
that's
I,
didn't
we
didn't
do
our
library
examples
in
secretary?
How
do
I
have
I
promote
or
how
do
I
read
a
GPO?
How
do
I
one
of
the
things
we
have
analog
to
digital
converters
on
their
boards?
All
reports,
every
converters,
so
you
can
actually
read
physical
handlock
values,
yeah.
A
That's
right
right
now,
for
example,
we're
using
the
very
basic
system
of
RepRap
Arduino
mega
Pololu
shield
I
mean
that
that
stuff
for
3d
printers,
so
we've
got
stepper
drivers
that
we
plug
in.
We
can
also
use
external
stepper
drivers.
Is
there
a
case
for
going
for
doing
a
advanced
controller
with
BeagleBone
yeah?
If
you
want
to
do
higher
performance
stuff,
but
we're
trying
to
say,
okay,
here's
the
basic
minimum
hardware,
you
need
to
even
control
a
laser
cutter
in
a
simple
way,
just
push
the
limits
of
the
simple
wrap,
Arduino
mega.
You
know
or.
B
It's
hard
to
say
so,
like
like
some
of
the
so
the
something
like
machine
kit
is
the
way
somebody
would
like.
Like
one
way,
somebody
might
build
a
3d
printer
out
of
people
ball,
and
you
know
you
start
with
this
environment
and
you
there's
a
there's
a
what
they
called
a
hal
layer
or
hardware
abstraction
layer.
So
you
write
little
Python
snippets.
B
You
know:
math
geometry,
to
stepper
motors,
so
you
know
you
have
you
have
your
G
code
right,
which
kind
of
describes
movement
within
the
physical
world
along
x,
y
and
z,
axes
right,
mm-hmm
commands
whenever,
but
but
you
got
just
like
go
to
this
point.
Yep
go
to
this
point.
This
path.
Look
at
this
point
that
fast
and
something
like
machine
fit,
allows
you
to
translate
that
into
different
geometries.
So
you
can
have
like
I
can
RepRap
world
you
can
have
core
XY
systems.
B
C
B
B
C
B
A
B
Like
machine
kit
or
you'll
have
something
like
lead,
scape
is
one
hour
not
let
scape
of
has
Falcon
Christmas
players
is
really
cool
one
because
you
can
do
all
these
complex,
LED
lighting.
So
like
you
put
and
you
go
through
a
web
interface
and
you
define
where
the
different
LEDs
are
connected
up
to
the
board,
and
then
you
just
Street,
like
it
handles
stringing
of
protocols
that
you've
run
a
tool
called
X
lights.
B
B
So
you
can
make
a
music
synthesizer
or
from
physical
events
like
maybe
a
push
button,
maybe
an
accelerometer.
Maybe
you
know
a
PITA
piezo
and
looks
like
we're
going
into
the
next
meeting,
so
they
can.
They
can
it's
half
a
mill
association
that
to
synthesize
audio
out
footage
and
they
have
a
nice
web
GUI,
for
you
know
creating
the
sounds
and
then
experimenting
and
they
interface
with
things
like
supercollider
and
pure
data
right.
So
the
beagle
world
is
it's.
You
know
both
simple
and
complex.
You
know
at
the
same
time
right
because
people
then.
B
B
A
B
B
You
know
it's
hard
to
because
you
know
like
like
you're,
you're
kind
of
crippling
yourself:
it's
lower
performance
and
it's
like
you
get
and-
and
nobody
wants
to
have
you
know
frustrations.
You
know
frustrated
kids,
because
it's
too
slow
they
use,
they
use
their
picked
up.
It
especially
depends
on
who
you're
targeting
right.
That's
why
I
say
the
performance
understands
if
what
they
have
is
that
cheap
little
Android
tablet
that
they
play
with
and,
like
you
know,
what
are
you
doing
in
this
in
the
class
right?
What
are
you
doing?
C
A
C
C
I
mean
arirang
yeah
and,
like
it
craps.
C
B
If,
if
you're
building
a
tablet
right,
you
can
sort
of
replace
like
if
you
have
a
screen
that
has
an
HDMI
input,
you
can
have
any
sort
of
like
if
it's
a
touch
grade
with
HDMI
and
USB.
Any
computer
can
be
the
base
for
that
right.
It
doesn't
need
it.
It
can
be
a
beagle.
It
can
be
a
PI,
you
know,
don't
make
it
dependent
on
one
right.
Just
just
use
a
generic
okay,
I've
got
HDMI
I've
got
a
standard
interface.
I've
got
a
USB
touch
screen,
that's
a
standard
interface
right
anywhere.
A
B
D
A
B
A
Are
you
getting
cloned
by
anybody?
We.
B
C
B
B
B
C
B
So
we've
had
a
number
of
those
that
are
authorized
clones
a
lot
of
times
for
TI
custom
products,
so
they'll
send
in
boards
for
review
before
they
go
into
production.
They'll.
Do
him
send
them
into
TI
for
review
a
large
proportion
of
those
like
80%
plus
of
the
designs,
using
that
same
processor
coming
in
and
they're
very
they've,
taken
the
BeagleBone
design
and
modified
it
for
their
purpose
yeah,
and
that
includes
everything
from
point-of-sale
tablets.
C
B
So
yeah
there's
a
lot
of
there's
a
lot
of
clones
out
there,
probably
10
times
as
many
clones
as
there
are
beagles.
Do
you
guys,
like
that
I
do
yeah
I
like
that?
A
lot
I
wish
that
we
find
better
ways
again,
I
get
them
to
pay
back
to
the
community,
but
it's
not
worth
viral
licensing.
In
my
opinion,
like
I,
don't
want
to
prevent
Commerce
and-
and
you
know
it's
it's
very
satisfying
to
me
right-
that
people
are
actually
building
products.
Oh
you.
B
C
D
B
C
B
But
you
said
you
don't
have
the
articles
yeah,
so
our
viral
clause
doesn't
only
impacts
the
design
materials,
not
the
manufacturing,
so
so,
like
you,
can
be
a
little
bit
more
like.
If
you
look
at
gplv3
like
versus
gplv2
right
with
with
v2,
you
have
a
publishing
requirement,
then
that's
the
difference
right.
So
we
don't
have
any
publishing
requirements
and
with
you
know,
with
like
gplv3,
in
addition
to
having
a
publishing
requirement,
there
is
a
a
no
locking
out
requirement
right,
so
we
don't
have
either
of
those.
So
there's
like
we
use
GPL
v2
codes.
B
So
if
they
ship
that
product
to
somebody
they
do
have
a
share
in
requirement
pay.
If
you
ship
a
product
to
somebody
and
say
you'll
here
is
your
product
that
uses
Linux
on
it.
They
are
required
to
provide
a
copy
of
the
source
code
good.
If
it's
just
using
our
Linux
kernel,
then
big
deal
it's
already
public.
They
don't
have
to
share
anything
new.
C
B
And
it
has
no
impact
on
the
user
space
or
their
application
code,
so
they're
not
required
to
share
those
things
and
they
can
potentially
lock
it
down.
There's
a
there's.
A
term
in
the
industry
called
TiVo
ization,
so
TiVo
ships,
the
Linux
kernel
on
there
they're
boxes,
but
they
signs
the
kernel
so
that
there's
no
way
to
replace
the
kernel
with.
B
With
your
own
and
you'll
find
also,
this
sort
of
holds
true
on
many
Android
phones,
it's
very
hard.
You
could
route
them,
but
it's
very
hard
to
go
and
replace
the
kernel.
So,
even
though,
there's
a
sharing
clause
that
says
you
have
to
share
the
source
code
back,
you
know,
there's
no
requirement
that
it
be
up
streamed
or
that
it
be
actually
replaceable
on
the
hardware
where.
C
B
C
B
You
actually
have
to
be
able
to
replace
it,
so
it
closes
that
the
loophole
that
that
gets
called
tebow
ization
because
they're
the
first
ones
to
kind
of
like
make
it
big.
You
know
case
about
it
that,
like
there's
no
way
to
replace
the
Linux
kernel
on
a
TiVo,
even
though
it's
running
Linux,
so
we
don't,
we
don't
have
any
that
sort
of
viral
stuff
right.
B
So,
yes,
sir
design
materials
are
CC,
BY,
SA,
so
there's
a
share
alike,
Claus
and
a
attribution
clause,
but
that
only
applies
is
if
they
publish
the
design
materials,
if
they
build
a
board
and
they
put
it
in
a
production.
There's
there's
there's
nothing
limiting
the
saying
that
they
have
to
publish,
we
don't
have
a
publishing
cause
and
we
don't
have
a
you
know.
As
chair
I
like
that
they
have
to
make
their
hardware
hackable
right.
That's
not
a.
It
was
neither
one
of
those
are
requirements.
Yeah.
A
A
Did
yeah,
let
me
just
point
you
to
it
real
quick,
because
maybe
you
can
point
can.
C
A
B
A
D
A
A
A
A
There
so
we
make
that
we
make
a
little
battery
pack
out
of
18
650
cells
and
putting
if
there's
12
people
participating.
We
stack
allowed
each
together
to
make
a
cordless,
well
they're,
once
again
using
the
same
Universal
controller,
so
kind
of
like
have
this
whole
crazy
ecology
of
many
many
things
you
can
do
using
a
very
small
tool
set.
A
D
A
A
And
the
action,
the
first
guys
that
I
made
the
first
manned
multi,
rotor,
that
guy
from
Germany
so
I'm
trying
to
get
him
to
teach
about
motors
I'm,
trying
to
put
together
a
team
of
powerful
people
that
can
produce
the
best
curriculum
and
we
all
go
about
creating
it
together
and
it's
open
source.
And
then
we
pay
people
to
run
these
camps.
The
instructors
that
we're.
A
D
A
D
A
D
B
I
think
there's
I
mean
there's
quite
a
few
people
on
our
hacker
space.
That
I
think
would
have
some
of
the
the
knowledgebase
behind
this.
But
you
know
the
my
network,
I
think
stems
you
know,
you
know
quite
a
bit
beyond
that
being.
A
B
C
B
Yet
another
three
axis
gantry
like
a
3d
printer
but
they're
doing
you.
B
B
B
A
B
B
C
C
A
D
A
A
B
There's
another
esta
to
another,
like
the
first
Linux
battery's
running
low.
Now,
the
first
Linux
autopilot
was
actually
done
on
our
boards.
Okay,.
C
B
C
D
C
B
C
B
C
B
Communicating
through
mavs
link
to
the
autopilot
right,
so
they
would
essentially
give
GPS
waypoints
to
the
autopilot
and
tell
it
to
go
there
or
whatever
other
you
know.
Kind
of
you
know,
navigation
mechanisms
that
they
would
have
right.
So
their
logic
tends
to
sit.
You
know
above
the
autopilot
layer
right
and
they're
just
running
on
the
same
computer,
the
Beagle
bone
as
the
autopilot
and,
like
so
people
doing
like
a
fire
detection
right.
B
C
B
That's
and
that's
that's
tends
to
be
how
people
utilize
our
our
boards
right,
so
they
they
tend
to
do.
Those
can
more
full-stack
integrations,
which
makes
it
a
little
bit
harder
to
build
the
understanding,
sometimes
for
as
an
entry
point,
because
there's
a
lot
to
understand.
There's
many
layers
to
the
onion.
Unfortunately,
where
we've
been
trying
to
we've,
been
working
with
the
linux
foundation
to
try
to
build
up
some
more
fundamental
trainings
around
linux
and
just
like
it's
sensory
integration
like
just
just
talking
to
things
at
low
levels.
B
B
B
A
B
Folks,
because
I
think
that
there
would
probably
be
some
some
overlap
in
the
community,
you
know
I
think
that
we
I
don't
know
what
your
qualifications
end
up
being
too
kind
of
roll
people
in
a
row.
People
out,
but
certainly
you
know
getting
some
attention
of
volunteers
based
on
your
curriculum
I
think,
would
be
fairly
easy.
A
D
C
B
A
B
A
A
D
B
C
C
B
B
D
C
C
C
B
B
B
B
D
C
B
B
Important
work,
/ki
is
currently
a
very
tiny
preview,
but
you'll
find
out
about
it
in
in
a
couple
weeks.
Mm-Hmm,
oh.
B
C
C
B
B
C
B
B
You
know
moving
to
a
certain
XY
based
on
something
that's
seen,
but
it
just
depends
more
on
your
on
your
under
logic.
Right,
there's,
there's
there's
a
few
different
aspects
that
the
framework
one
is
just
an
autopilot
with
Ross
that
the
interface
is
up
to
remapping
and
then
there's
you
know
kind
of
different
levels
of
slam
that
you
can
do
or
slide
to
simultaneous
location
and
mapping.
B
B
You
know
the
drone
to
go
to,
like
you
know,
almost
like
your
g-code
thing,
mm-hmm
right,
where
you
want
to
go
to
a
position
and-
and
you
essentially
get
a
little
bit
of
sensor
fusion
going
on
in
order
to
try
to
build
up
the
control
system,
because
you
might
you'd
likely
have
some
type
of
sort
of
downward
facing
sensor.
Mm-Hmm.
C
B
Tell
the
distance
from
the
ground
in
conjunction
with
the
GPS
or
whatever
other
sensors
you
I,
be
adding
to
it.
So
our
G
pilot
has
some
degree
of
sensor
fusion
that
it
will
use
to
generate
at
the
control
loops,
and
then
you
just
have
to
build
kind
of
your
your
logic
around
the
hey.
You
know,
yeah,
so
tell
the
drone
to
move
where
this
object
is
this
much
in
the
frame
and
then
tell
it
to
do
something
else.
C
B
It's
got
the
best
community
around
it.
There's
a
little
bit
more
of
a
professional
developer.
Community
may
be
around
px4,
but
the
hobbyist
community
around
our
G
pilot,
makes
it
most
attractive
to
me.
C
B
D
B
B
C
A
C
A
B
Raspberry
pie,
playing
in
a
skill
to
you
know:
they've
already
had
a
number
of
application
spaces
within
the
field,
even
though
it's
not
something
that
they've
been
specifically
targeting,
but
the
performance
is
a
little
a
little
anemic
compared
to
something
like
the
people.
Gonna
hire
the
Jetson
nano,
but
the
I'd
say
the
you
know
the
my
phone's
about
to
die.
Okay,.