►
From YouTube: Michiel Leenaars - The Significance of Reproducible Software in International R&D (SoN2022)
Description
Technology is impacting or even driving ever more aspects of our society, and publicly funded research was behind many of the key breakthroughs in recent decades: the internet, the web, and of course Nix itself. From a policy angle "public money" may officially mean "public code", but that copyright licensing imperative says nothing just yet about deployability or sustainability. Lots of technology developed with public funding ungracefully floats into oblivion, simply because it is buried inside a fragile technical context and was never made easy to deploy. Even if people find out about some interesting technology through luck or perseverance, bitrot often sets in almost immediately - change of a single dependency turns the prospective user into an amateur software archeologist by necessity even if they just want to have a look how it works.
Nix offers a different path. By its own or combined with open data, it allows for reproducible research in a cost-effective and non-discriminatory way. For development, it makes collaboration and usage easy - making technology something you can share, adjust and build on at scale. In this talk, Michiel Leenaars from NLnet will explain how Nix is part of the larger ambitions of the Next Generation Internet (NGI), and how nix helps to bring reusable technological building blocks towards the most exotic corners of the internet.
Special thanks to the NLnet Foundation, the European Commission, the NixOS Foundation, and Tweag for making this event a reality!
The continued discussion for the lecture series is happening over here:
https://matrix.to/#/#son2022-lectures:matrix.org
More information about the Summer of Nix can be found on the website:
https://summer.nixos.org
A
A
Of
course,
I
will
again,
as
always,
you
guys
are
used
to
the
drill
now
I'd
like
to
thank
the
people
that
made
it
possible
the
participants,
the
nl
net
foundation,
which
you
know,
there's
a
special
guest
from
the
nlnet
foundation
today,
which
is
nice,
the
european
commission,
the
nexus
foundation
and
tweak
so
without
much
further
ado.
Let
me
pull
michael
in
here
as
well:
hey
michael.
C
B
So
hi
everybody,
my
name,
is
michilinas.
I
work
for
an
organization
called
a
nelnet
foundation
and
my
talk
today
will
consist
of
three
parts.
First
I'll
I'll
do
an
introduction
of
an
elnett
and
ngi
zero
I'll
do
a
little
two
of
the
tableau
of
the
kind
of
projects
that
we
fund,
then
I'll.
Go
into
detail
how
reproducible
software
and
mix
fit
into
our
work
and
that
of
our
peers
and
well.
B
If,
if
this
is
all
you'll
get
because
you're
bored
of
my
voice
already,
then
the
too
long
didn't
read
this
reproducible
packaging
across
the
board
for
a
research
program
is
really
a
best
practice
so
about
an
elnett
foundation.
So
an
elnett
is
not
a
household
name.
It's
it's
a
small
public
benefit
organization,
but
it
has
pretty
pretty
proud
history
because
back
in
1982
so
40
years
ago,
those
were
the
people
that
basically
bootstrapped
the
internet
in
europe
or
and
actually
the
first
public
internet
connection
outside
of
the
us,
was
made
to
nail
net.
B
So
it's
a
it's
a
a
a
pretty
amazing
history
and
the
foundation
has
been
focusing
since
1997
as
a
grant
maker
solely
focused
on
independent
researchers
and
developers
to
work
for
the
internet,
so
we're
a
registered
public
or
acknowledged
public
interest
organization
and
everything
that
we
do
is
open
standards,
open
source,
open
hardware
and
open
data.
So
we
have
a
history
with
with
many
projects
that
you
may
or
may
not
know.
B
No
need
to
delve
into
that.
I
will
we'll
get
to
that
later
so
about
the
next
generation
internet.
That's
a
something,
an
initiative
that
we
mean,
quite
literally
so
for
us,
the
internet
is
well
to
some
extent
broken.
If
you
look
at
its
original
design,
it
wasn't
meant
to
do
the
kind
of
things
that
we
are
now
doing
with
it
and
with
privacy,
security
scalability
all
lacking.
B
If
you
want
everybody
to
use
the
internet
to
its
full
potential
for
the
next
thousand
years
without
holding
back,
then
we
can't
leave
it
the
way
it
is
so
somebody
has
to
do
something
now
back
in
2016
to
2018.
We
did
this.
The
study
for
the
european
commission
called
next
generation
internet
2025,
so
that's
only
three
years
away
together
with
gardner,
and
this
basically
established
the
vision
for
the
ngi
program
and
and
established
that
we're
we're
talking
about
increasing
resilience
because
well
the
internet
is
moving
into
space.
B
B
That's
also
no
good,
so
those
are
the
three
pillars
that
that
we're
striving
towards
this
is
this
is
not
a
place
where
people
sit
sit
idle.
So,
meanwhile,
we
have
over
a
dozen
research
and
innovation
actions
that
have
been
started,
and
these
support
hundreds
and
hundreds
of
projects
with
with
with
grants
on
all
layers
of
technology,
and
there
are
some
support
actions
as
well.
One
of
them
is
called
ngi0
review
and
actually
the
nexus
foundation
is
part
of
that.
So
we'll
get
back
to
that
later.
B
So
ngi
distinguishes
10
layers
of
technology,
and
this
really
goes
from
well.
The
top
level
stuff
like
that.
The
server
that
we're
now
using
called
owncast
is
part
of
it,
but
we
also
have
search
engines.
We
have
open
data
tools
like
open
food
facts,
going
all
the
way
down
to
actual
networking
infrastructure
and
even
lower
to
the
actual
hardware
and
how
it's
manufactured
and
designed.
B
So
it's
it's
an
immense
spread.
But
then
again
the
internet
is
is
a
big
thing.
So
as
an
el
nino,
we're
involved
with
ngs0
programs
and
with
ngi
sure
those
are
the
two
programs
but,
like
I
said,
there's
many
more
now
what
is
unique
about
ngi
in
in
public
funding,
because
the
the
money
comes
from
the
european
commission,
it's
a
unique
and
that
it's
entirely
focused
on
technology
commons.
B
So
it's
all
about
free
and
open
source
software,
about
open
hardware,
about
open
standards
and
about
open
data,
and
that
follows
the
mantra
of
public
money,
public
code,
which
also
means
that
everything
comes
with
the
license
that
you
can
run
the
software
without
any
restrictions.
You
can
study
the
source
code,
you
can
study
the
behavior,
you
can
adapt
it
to
meet
different
needs
and
you
can
distribute
any
derivatives
under
the
same
conditions.
B
And
of
course,
if
you
only
look
at
say,
an
iphone
you'll
understand
how
different
it
is,
and
also
it's
a
it's
a
big
step
for
the
european
commission
to
actually
even
do
this
kind
of
thing.
So
we're
very
much
applauding
their
their
move
towards
this
direction,
because
what
it
enables
is
permissionless
innovation.
It
allows
you
to
to
pick
up
any
topic,
any
piece
of
technology
that
you
want
and
think
with
it
until
it
does
what
you
want.
B
It
also
means
that
you
can
make
a
limited
use
of
every
all
of
it
in
the
research
of
this
research
and
development
in
education,
because
there's
access
all
areas,
there's
no
ndas,
which
is
in
many
areas
of
technology
like,
for
instance,
in
in
the
silicon
domain
of
silicon.
It's
ndas
all
over
sometimes
there's
three
layers
of
non-disclosure
agreements
that
you
have
to
sign
in
order
to
just
be
able
to
use
technology
so
with
ngi.
We
we
don't
have
to
deal
with
that,
and
people
can
actually
see
everything,
use
everything
and
and
change
everything.
B
So
what
we
try
to
do
is
try
to
be
almost
invisible
because
good
ground
making
isn't
about
us
as
an
organization
it's
about
enabling
the
people
that
do
stuff
at
the
end
of
our
lives.
When,
when,
when
the
people
that
work
in
our
foundation
are
on
our
death
beds-
and
we
look
back
what
have
we
done
then
all
we
can
say
is
we
found
the
right
people
and
made
them
do
the
right
things,
but
it's
other
people
that
have
to
do
the
work.
B
We
tell
them
how
to
improve
their
usability,
how
to
make
sure
that
things
are
accessible
to
people
with
disabilities,
make
sure
that
people
can
have
localized
and
internationalized
software
so
that
people
that
don't
speak
english
can
also
still
use
it.
We
help
them
with
standardization,
because
sometimes
just
doing
it
by
yourself
isn't
good
enough.
We
help
them
with
community
building
and
diversity
and
with
software,
packaging
and
distribution.
B
This
is,
of
course,
where
nexos
and
nix
and
the
whole
ecosystem
comes
in,
also
because
we
want
every
technology
building
block
to
reproduce
to
actually
be
useful
and
be
usable.
That
means
they
have
to
be
secure
and
trustworthy.
So
we
have
independent
experts,
look
at
every
project
and
again
that's
a
different
thing
from
grab
the
money
and
run
while
you
can.
B
So
we
really
want
to
take
care
of
all
these
projects
and,
of
course
it's
not
just
us
that
can
moderate
monitor
things
because
well
everything
is
published.
Everything
is
public,
so
the
source
code
of
every
project
is
available
for
everyone
to
study.
You
can
use
it,
you
can
improve
it,
you
can
redistribute
it.
So
if
you
can,
if
you
want,
you
can
join
the
effort.
B
B
B
A
4g
network
also,
but
going
down
to
the
realm
of
semiconductor
development
in
electronics,
because
our
goal
is
basically
to
achieve
the
kind
of
freedom
and
openness
that
we
have
in
the
software
stack.
We
want
that
to
be
the
same
in
hardware.
You
know
we
know
it's
different,
because
obviously
the
cost
of
doing
something
and
setting
it
to
a
fad
is
is
much
higher.
B
But
this
collaborative
bottom-up
approach,
where
everybody
can
do
everything
it's
getting
closer
and
closer
and
closer,
and
so
some
of
the
things
that
we
do
are
create
open
designs
for
chips,
make
sure
that
we
have
trusted
platform
modules,
various
kinds
of
peripherals
and
other
building
blocks,
but
we
also
make
sure
that
we
have
design
tools,
new
design
tools
and
new
design
paradigms
versus
doing
with
formal
verification.
So
we
can
create
technology
commons
also
at
the
silicon
level
and,
of
course,
the
tool
chains.
There
are
some
tool
chains
that
are
horribly
expensive.
B
So
you
can't
expect
what
it
does
and
we
want
fully
open
verifiable
tool
chain
so
that
the
next
generation
of
of
hardware
can
also
be
confidently
built
using
that
so
and
well,
as
you
see,
we
want
to
connect
all
the
dots
and
innovate
from
the
lowest
layer
of
actually
producing
the
silicon
all
the
way
up
to
to
the
user
and
have
something
that
they
can
use
from
from
the
alternative.
Silicon
fabrication
and
hardware
design
languages,
fpgas,
of
course,
grade.
B
B
So,
of
course,
packaging
makes
a
big
difference
there
and
ultimately,
in
in
the
far
far
future,
we
want
all
of
these
open
instruction
sets
to
be
able
to
be
combined
so
that
we
can.
We
can
use
them
wherever
the
best
use
case.
Is
there
because,
as
we
as
I
explained
before,
we
also
have
this
longevity
and
sustainability
goal,
and
every
instruction
set
has
a
strong
and
its
weak
cases,
so
by
being
able
to
mix
and
match.
B
Obviously,
we
will
get
the
we
hope
to
get
the
best
of
best
of
all
worlds,
and
that
includes
heterogeneous
cores.
We
we
already
have
some
efforts
in
that
space
and
ultimately
well,
why
not
aim
high,
ultimately
we'd
love
to
have
a
high-end
processor
that
can
actually
take
care
of
your
day-to-day
business,
as
as
your
your
daily
driver?
B
There's.
No
reason
why
this
shouldn't
be
because,
ultimately,
it's
the
combination
of
technology
with
the
user
base
that
wants
to
pay
for
it
and
the
user
base
is
already
paying
through
the
notes
for
other
options.
So
hijacking
that
and
making
it
move
to
the
open
domain
is
something
that
we
think
is
within
within
reach.
B
So
some
of
the
projects
that
we're
funding
just
because
you'll
you'll
you'll,
you
may
have
seen
these
in
the
list
of
projects
that
were
that
are
being
packaged,
libre
sock,
is
making
a
libra
system
on
a
chip
and
it
it's
it's
dogmatically
using
a
fully
open
tool
chain
so
which
is
which
is
awesome
because
it
actually
makes
that
this
tool
chain.
Then
it
becomes
silicon
proven
and
that
people
can
can
trust
that
it
actually
results
in
a
working
working
technology,
cg
arrays.
B
We
have
projects
that
are
trying
to
combine
the
best
benefits
of
the
benefits
of
using
an
fpga
with
a
normal
asic,
and
that
is
to
define
parts
of
the
the
silicon
in
in
in
a
in
a
in
a
definitive
way
burn
it
into
the
silicon.
For
instance,
cryptographic
functions
that
you
don't
want
to
have
reconfigurable,
because
then
they
could
be
weakened
artificially
without
seeing
it,
but
at
the
same
time
have
the
the
programmability
of
an
fpga.
B
D
B
Open
wi-fi
is
is
another
awesome
project
if
you
want
to
do
research
on
on
wi-fi,
typically
historically,
you'd
have
to
wait
for
some
vendor
to
produce
a
chip
and
then
you'd
get
the
chip,
but
you
couldn't
tinker
with
that
chip.
What
open
wi-fi
has
amazingly
achieved
is
to
have
an
fpga
with
software-defined
radio,
which
means
that
you
can
actually
completely
redesign
the
protocol
as
an
individual.
B
B
So
and
of
course,
this
is
also
something
that
you
can
then
actually
burn
on
a
chip,
because,
historically,
if
you
buy
some
component
and
stick
it
on
on
on
on
on
your
in
a
laptop
or
in
a
phone,
you
have
to
trust
the
people
that
made
it.
The
nice
thing
about
having
a
gateway
you
can
inspect
is
that
you
can
actually
see
if
it's
trustworthy,
because
it's
just
now
an
antenna
and
software
instead
of
a
black
box
component.
B
Another
project
that
we're
finding
is
the
be
trusted
project
which
is
done
by
bunny
wang,
and
this
is
a
an
accelerator
for
curve.
2
5,
5
19
operations
and
it's
a
super
small
component.
But
it's
really
awesome.
You
can
now,
if
you're,
making
your
own
chip,
you
can
actually
include
that
and
and
just
reuse
it.
So
you
don't
have
to
build
it
again
over
and
over,
because
that's
such
a
an
often
used
technology.
B
There's
a
tool
called
mosaic
which
is
moving
into
the
space
slowly
moving
to
the
space
that
cadence
occupies
and
other
tools.
The
tool
that
I
just
mentioned,
that
costs
a
whopping,
90
million
dollars
to
pay
and
ideahub
is,
is
a
tool
to
like.
If
you
want
to
do
hardware
design,
you
don't
want
to
use
something
like
github,
which
is
meant
for
well,
essentially
a
text
level
software.
B
Now,
like
I
said
for
us,
it's
really
important,
given
this
enormous
horizon
of
of
all
these
other
architectures
on
the
horizon.
All
these
possibilities
to
achieve
architecture
agility,
because
the
open
architectures
that
we
that
we
produce,
even
if
they
have
known
use
cases
like
open
powers,
powered
many
supercomputers
scientific
experiments.
Some
generations
of
the
the
famous
play
con
game
consoles
actually
contained
the
open
power
instruction
set,
but
they
can
only
flourish
if
there's
enough
software
to
run
it.
B
C
B
Gnu
mess
project
if
people
are
aware,
but
also
have
icebreaker
projects
like
the
the
the
geeks
and
nyx
import
them
to
risk
five,
so
that
all
these
applications
become
available
in
a
reducible
way
and,
of
course,
create
devices
that
use
these
these
technologies,
so
that
people
can
actually
get
them
under
in
their
hands
and
not
just
on
the
development
board
that
they
have
lying
somewhere
on
their
on
their
desktop
on
those
devices.
Of
course,
you
need
an
operating
system,
because
vanilla,
android
or
ios
won't
be
running
on
risk.
5.
B
B
For
measurement
monitoring,
analysis
and
abuse,
handling
for
middleware,
for
decentralized
solutions
and
alternative
internet
protocols
for
data
and
ai,
and
lots
and
lots
and
lots
of
applications
as
you
would
have
spotted
versus
in
the
field
of
activity,
pub,
a
search
engine
and
so
on.
Well,
okay,
I'll
quickly
skip
this
too.
We
have
lots
of
useful
outputs
and
I'll
put
the
link
up.
So
you
could
you
can
have
a
look
at
it,
but
there's
so
many
components,
even
though
there's
a
it's.
B
A
relatively
small
program
of
these
are
relatively
small
programs
compared
to
to
the
to
the
tens
of
billions
that
the
european
commission
puts
out.
I
think
it's
still
amazing
to
have
like
so
many
different
global
firsts,
even
in
the
short
span
of
just
a
few
years,
but
that's
because
we
have
amazing
amazing
people
inside
our
portfolio
from
from
various
open
hardware,
open
software
standards,
communities
and
well
it's
it's
a
it's.
B
Because-
and
I
think
that's
what
attracts
people-
the
goal
is
really
to
have
a
real
world
impact
because
we're
well.
Everybody
else
is
allowed
to
fund
piles
of
paper
or
produce
stacks
of
bits,
just
because
that's
cool,
but
we
really
really
want
to
have
solutions
that
are
that
solve
problems
that
society
has
and
and
technology
is
actually
a
big
part
of
it.
So
technology
should
be
solving
technologies
problems,
so
we
want
to
actually
fill
gaps
in
our
capabilities
as
people
as
you
as
mankind.
B
Why
would
we
go
to
the
trouble
of
packaging
everything
because
we
mandate
we,
we
ask
the
the
package
inside
the
sum
of
renix
to
spend
months
of
their
lives
packaging,
the
technologies
that
I
just
showed,
but
why
should
that
be
our
responsibility?
Well,
isn't
that
the
responsibility
of
the
project
shouldn't
they
be
taking
care
of
this?
Well,
even
if
they
have
the
time
do,
do
they
have
the
expertise.
B
D
B
Steep
learning
curve
of
something
like
knicks
or
geeks
is
actually
asking
a
bit
too
much,
and
if
we
turn
the
question
around
well,
we
see
that
they
don't
pick
up
the
responsibility.
We
see
that
most
people,
most
programs,
most
research,
doesn't
deliver
in
this
in
this
field
and
who's
actually
the
victim
of
that.
B
So
we
looked
at
this
with
a
sort
of
a
real,
real
political
approach.
Saying
well,
surely
everybody
should
have
done
this,
but
let's
do
it
for
let's
do
it
for
them
simply
because
it's
something
that's
orthogonal
enough
to
the
development
and
we
notice
that
build
it
and
they
will
come
isn't
true.
We
notice
that
there's
lots
of
awesome
code
that
it's
just
lying
about,
because
the
license
itself
doesn't
say
that
it
needs
to
be
deployable,
needs
to
be
sustainable
and
it's.
B
It
is
a
fact
of
life
that
lots
of
useful
technology
is
lying
around
in
drawers
waiting
to
be
forgotten
because
they're
buried,
while
inside
a
vm
or
inside
a
scientific
publication,
and
nobody
ever
took
the
effort
to
deploy
it
at
scale.
So
so
so
it's
an
alternative
future
that
never
happened,
and-
and
we
intend
to
change
that
because
we
find
that,
even
if
first
through
some
miracle,
people
are
able
to
find
some
some
interesting
technology
and
and
and
have
a
look
at
it.
B
B
So
if
people,
if
you
expect
people
to
have
a
look
at
how
how
you've
done
something
and
and
they
they
should
be
looking
into
a
software
distributions
from
from
from
even
a
couple
of
months
ago,
that
is
just
pushing
it.
That's
never
going
to
happen
and-
and-
and
I
think
we've
been
proven
correct
by
by
history,
so
we
can't
afford
unless
a
fair
attitude,
we
really
have
to
do
something
and
of
course,
there's
nixos.
That
offers
us
a
different
path,
because
its
reproducibility
helps
us
to
tackle
the
reproducibility
crisis
in
science.
B
I
don't
know
if
people
know
that
term,
but
the
original
idea
behind
science
was
that
you
can
make
something
that
other
people
can
then
verify,
because
they
do
the
same
thing
and
when
descriptions
were
as
simple
as
they
were
in
1600,
when,
when
the
first
scientific
papers
were
written,
replicating
experiment
was
was
feasible.
But
these
days
that's
just
not
terrible.
B
People
have
to
produce
like
crazy
and
even
setting
up
the
environment
of
the
original
research
is,
can
be
well
weeks.
Months
of
work
and
in
the
case
of
software,
that's
just
not
necessary,
of
course,
outside
of
the
domain
of
of
academia.
Developer
environments
are
also
the
thing
where
you
make
or
break
collaboration
with
outsiders.
B
Because
if
something
is
packaged
and
you
can
try
it,
you
can
also
try
it
in
anger.
You
can
try
it,
but
you
can
use
it
for
real
world
use
cases,
so
that's
kind
of
blurring
the
lines
of
research
and
development
in
the
way,
in
the
proper
way
that
devops
has
become
a
popular
term.
So
and
of
course,
as
I
explained,
nix
is
to
a
very
large
degree
platform
agnostic,
which
again
helps
us
to
to
achieve
this.
B
So
nick's
makes
the
technology
that
we
fund
something
we
can
share
and
build
on
at
scale
now
a
question
that
we
often
get
is
why
not
docker?
Why
not
stick
everything
into
containers
and
and
that
that
could
easily
work?
And
our
answer
is
always
that
containers
are
great
for
demos
and
for
certain
types
of
production,
but
there,
if
you
look
on
the
inside
they
they
are
often
horrible
and
fragile,
and
so
in
six
months
from
now,
they're.
B
If
you
put
yourself
in
our
in
our
position,
we
have
people
that
are
working
on
trying
to
to
change
the
whole
tcp
ip
udp
internet
stack.
We
also
have
people
that
are
trying
to
replace
the
socket
pc
mechanism,
the
historical
socket
mechanism.
We
also
have
people
that
are
trying
to
structure
the
system
layer.
We
also
have
people
working
on
applications
and,
if
you
try
to
just
to
to
cram
all
the
you
can't
have
separate
containers
that
solve
each
of
these
things,
they
have
to
be
composable.
B
B
B
So
and
of
course,
we
want
the
technologies
that
we
produce
to
actually
be
deployed
in
the
backbones
in
the
with
the
registries,
with
the
hosting
companies
with
isps
and
there
people
are
not
eager
to
to
take
on
the
burden
of
maintenance.
So
nyx
is
ideally
suited
to
to
take
care
of
this.
I
often
compare
this
to
to
if,
if
we
give
everybody
a
digital
copy
of
a
small
wooden
house,
that's
where
everybody
lives
it.
If
we
give
them
a
little
stone
house,
that's
where
everybody
lives
in.
B
If
we
give
them
to
taj
mahal,
then
that's
what
everybody
lives
in
the
the
replication
of
something
that's
good
and
that's
that
works
with
nick's
because
of
his
reproducibility
is
actually
something
that
we
can.
We
can
count
on
and
we
can
we
can
lever
to
to
to
have
a
massive
impact.
We
believe
so
for
us
if
something
isn't
packaged,
it's
not
really
to
be
used
in
the
actual
internet,
and
it's
also
almost
impossible
for
future
research
to
actually
benefit
from
from
from
from
that
work.
B
This
involves
something
where
pretty
much
all
of
mankind
is
dependent
on
and
it's
hugely
complex
and
all
the
components
are
actually
lots
of
people
are
real-time
dependent
on
it.
So
we
have
to
hot
swap
the
technology
out
from
under
the
feet
and
replace
it
with
something
better
without
them
noticing,
and
just
to
on
top
of
that,
everything
that
everybody
else
has
done
so
far
has
typically
mostly
run
aground.
So
it's
it's
a
big.
It's
a
big
burden
to
solve,
because
it
is
the
biggest
thing
that
mankind
has
ever
made.
B
So
so
we
we
have
to.
We
have
to
go
and
take
take
the
extra
mile
and
for
us
nyx,
is
that
extra
mile,
because
we
we
couldn't
find
any
other
technology
besides.
Obviously
it's
sibling
geeks,
which
which
is
of
course
equally
likely
to
succeed,
no
other
technologies,
but
apart
from
these
two
seem
to
be
able
to
deliver
what
we
need
at
scale.
So
because
we
need
that
low
barrier
to
entry.
B
We
need
to
increase
the
systemic
responsiveness,
which
means
that
we
can't
allow
any
broken
thing
to
take
days
or
weeks
to
fix.
We
need
something
that
can
be
almost
instant
and
and
the
declarative,
reducible
approach
of
nyx
actually
allows
for
that
plus.
We
need
reliable
and
scalable
shared
procedures
across
all
these
different
research
efforts
and
again
nyx
is
for
us
that
that
method
so-
and
I
just
want
to
conclude
with
a
with
a
short
advertisement-
there
is
much
more
that
needs
to
be
done.
B
If
you,
if
you're
interested
in
applying
yourself,
then
you
can
actually
propose
something
to
us.
So
in
case
the
summer
of
nix
ignites,
your
your
your
appetite
to
to
do
something:
meaningful,
we've
actually
already
fun
we're
funding.
Quite
some
members
from
this
community
two
different
to
do
different
things.
B
A
D
A
To
remove
my
heel
man
to
remove
my
heels
to
slide,
I'm
sorry
there
you
go
mothers
in
here
so
yeah.
The
first
question
coming
from
matthew,
krogen
of
coben's
rhys
over
on
youtube,
is:
are
there
any
examples
of
the
european
commission
stating
that
next
slash
nexus
is
a
good
thing
that
anyone
can
read
or
cite
online.
B
B
And
I
guess
it's
the
same
kind
of
thing
that,
for
instance,
a
couple
of
months
ago,
in
the
same
week
that
elon
musk
announced
that
he
would
buy
twitter,
which
turned
about
to
be
a
bit
of
a
dude.
But
in
the
same
week
the
european
commission
adapted
one
or
two
of
our
projects,
one
called
macedon
and
another
called
pier
tube
for
its
for
its
first
communications,
and
this
went
through
our
basically
through
the
pipeline
of
being
funded
through
us.
B
B
B
B
A
B
B
They've
grown
so
big
and
successful
that
so,
yes,
we're
we're
working
with
them
and
in
fact
the
group
behind
scion
or
some
of
the
people
behind
scion
are
not
not
people
that
we
funded,
but
other
people
are
actually
part
of
the
ngs0
patent
discovery
program,
so
they're,
the
ones
that
help
people
standardize
for
the
next
two
months.
After
that,
it
will
be
picked
up
by
toronto
networks,
which
is
from
ireland.
C
D
D
B
Yeah,
it's
it's
it's
fast,
it's
hundreds
and
hundreds
of
projects
and
and
and
every
every
two
months
like
new
projects,
come
in,
and
then
we
think,
but
but
there's
that's
why
open
calls
are
so
important
we
couldn't
predict.
We
couldn't
like
put
out
a
call
saying
this
is
a
problem
we
have
so
much
money
to
to
solve
it.
No
you
think
of
the
solution.
You,
you
identify
the
problem,
you
think
of
the
solution
and
then
we're
there
to
recognize
that
it's
indeed
a
something
that
worthwhile
we
helped
well
to
achieve.
It.
D
So
along
this
point,
I
I
have
a
question,
so
I
I
think
so
you
are
in
direct
contact,
for
example,
with
some
politicians
probably
who
decide
about
you
know:
where
does
the
money
flow,
and
so
how
do
you
see
this
interaction
between,
let's
say,
an
individual
developer?
It
works
on
one
single
project,
all
the
way
up
to
the
politician.
Who
has
these
big
societal
issues?
I
I
guess
one
could
just
see.
B
So
they
were
asking
people
to
install
cisco
and
and
some
other
tools
to
to
to
to
to
to
do
this
and
it
broke
and
at
the
same
time
we
were
funding
wireguard,
which
was
formally
proven
by
the
prosecco
group,
was
formally
proven
by
by
oxford,
which
was
like
won
all
kinds
of
prizes.
So
we
say
we
have
this
free
thing
that
you
can
download,
which
is
proven
by
some
of
the
best
groups
in
the
world,
to
actually
be
precisely
reliable
and
do
what
you
need.
So
how
can
we?
B
How
can
we
close
that
gap-
and
I
well
that
and
the
question
wasn't
answered
at
the
time,
but
basically
it's
it's
it's
impossible
for
them
to
to
ignore
the
fact
that
what
what
is
being
done
here
is
awesome
and
what
we're
seeing
is
that
it's
step
by
step
people
are
like,
for
instance,
we
have
a
project
that
is
working
on
firmware
to
replace
a
proprietary
firmware
on
motherboards,
with
with
something
that's
open,
also
for
modern
motherboards.
So
there
there's
there's
nothing
else
like
it.
B
A
project
called
alder
lake
and
we're
seeing
that.
So
now,
commissioner,
like
european
commission
staff,
is
asking,
can
you
ever
give
a
presentation
about
this?
Okay
dude
about
so
it's
it's,
it's
really
about
as
a
group
to
get
the
kind
of
attention,
but
it's
it's
lots
of
little
folks,
lots
of
little
really
put
lions
to
to
to
to
gather
together
and
and
create
this
next
generation
internet
thing,
which
is
well
a
bigger
sum,
or
you
can
say
it's
part
of
this.
So
it's
we
know
it's
been
vetted.
B
B
How
are
we
going
to
do
it?
So
we
fund
some
alternatives
that
prove
that
you
don't
need
to
do
it
their
way,
but
you
can
do
it
another
way
that
is
privacy.
Preserving
gdr
compliant
actually
has
hard
guarantees.
Obviously,
after
that
it
becomes
a
lobby
on
saying.
Well,
we
we
know
for
a
fact
that
it's
that
we
can't
do
it.
We
know
for
a
fact
that
we
don't
need
facebook
to
do
social
networking.
D
So,
b,
before
going
to
a
question
by
crunches
that
I
think
fits
well
into
this
as
well.
But
do
you
think
so
annette
is
basically
between
the
politicians?
Let's
say
who
have
their
perspective
and
the
developers
who
a
lot
of
individual
tools
and
and
helps
to
orient
and
understand
this
whole
ecosystem
and
make
stories
that
that
are
coherent,
that
integrates
different
tools
and
different
approaches?
Do
you
think
that
governments
should
have
themselves,
let's
say
an
entity
or
a
ministry
or
something
similar
to
a
net
or
is?
B
We
it
the
fact
that
we're
detached
from
everybody-
and
we
can
just
do
whatever
we
think-
is
good
for
society.
I
think
it's
a
great
benefit.
What
what
should
be
happening?
I
think
there
should
be
a
counterpart
at
these.
Well
there's
something
called
the
open
source
program
offices,
which
is
which
in
many
governments
are
now
starting
to
to
appear.
B
I
think
they
should
be
ground
making
is
something
different,
and
this
also
has
a
global
aspect.
So
so
you
can't
be
too
nationalistic
about
who
does
what?
But
you
want
to
have
a
collaboration
across
well
across
the
globe,
and
so
I
think
the
fact
that
we
can
operate.
We,
I
think
we
have
projects
in
something
like
70
countries
or
so
concurrently.
B
Maybe
we'd
have
something
more
high
security,
something
we
can
control
ourselves.
So
maybe
we
should
so
it's
it's.
I
I
guess
to
inspire
government
to
say
well
we're
not
taking
proprietary
close
crap
anymore,
and
if
I
mean
our
hope
is
to
go
into
the
real
economy,
we
can
seed
all
these
projects,
but
we're
like
a
a
a
a
a
hot
house
once
once
the
they
research
maturity.
B
We
want
these
projects
to
be
part
of
the
real
economy
where,
where
budgets
are
thousands
of
times
bigger,
I
mean
if
the
the
government
spends
tens
of
millions
on
on
identity
management
and
we
spend
well
projects
of
50
000
euros.
Obviously,
at
some
point
in
time
we
have
to
be
better
than
what
is
available
once
we
reach
a
stage
where,
where
it's
comparative
and
from
there
on
it
should
hopefully
just
take
off
the
goal
is
not
to
to
create
a.
I
mean,
I
think,
accepting
the
cost
of
of
id.
B
Browsers
are
are,
are
the
thing
that
we
use
every
day
all
day
same
as
mobile
operating
systems,
and
if
we
can't
trust
them,
because
we
got
them
for
free
with
a
with
an
advertisement
so
to
say-
and
the
advertisement
keeps
keeps
an
eye
on
us
all
day
long,
we
should
just
say
well
we're
going
to
pay
for
it
to
our
nose
anyway.
So
why
not
just
pay
directly
and
we
control
it
rather
than
to
wait
for
somebody
else
to
pick
up
the
bill.
B
So
my
hope
is
that
we
can
actually
get
into
the
real
economy,
and
that's
also,
I
guess
why
why
this
this
talk
inspired
to
give
this
talk.
If
all
of
research
has
the
same
beneficial
conditions
where
people
can
just
jump
in
and
say
that's
useful
I'll
take
it
all
I
need
to
do
is
copy
the
next
nyx
flake
and
and
have
the
component
working
and
and
be
able
to
tinker
with
it
and
all
of
and
everything
is
open
source,
which
is
also
unique
to
this
program.
B
But
there's
no
reason
for
it
not
to
we.
We
have
public
money
so
so
enforcing
public
code
everywhere
is
just
a
scratch
of
the
pen
away
so
to
say:
yeah,
it's
it's
a
system
we
have
to
as
a
system
go
the
the
the
towards
this
road
of
digital
commons.
I
think,
but
yeah,
I'm
hopeful
that
it
can
be
right
and.
D
B
One
comment
is
that
I
don't
think
we
should
be
in
between
the
politicians
and
projects.
I
think
we're
standing
next
to
them,
but
but
we're
in
elnett
is
a
tiny
organization,
we're
just
a
handful
of
people
that
that
tried
to
try
to
make
everything
move
in
all
the
right
places,
but
we
need
the
digital
rights
community,
like
the
eddies
of
europe
and
the
effs,
and
we
work
with
them
from
time
to
time,
but
we
should
be
doing
more
of
that
and
also
people
themselves.
I
mean
this.
B
D
B
Yes,
we
do
because,
basically,
the
the
the
the
proper
company
with
proper
products
may
not
have
the
money.
Basically,
if,
if
you
ask
venture
capital
to
put
money
up
front
and
then
extract
it
later,
with
with
an
incremental
benefit
and
you're
saying
we're
going
to
buy
enough
of
these
devices
in
the
public
sector
alone,
even
in
a
single
company,
people
are
buying
sometimes
millions
worth
of
of
of
these
kinds
of
security
devices.
B
So
if
we
borrow
the
money
so
to
say
to
to
an
organization
like
nitro
key
saying,
you
don't
need
to
invest
from
your
pockets,
you
could
just
well.
We
take
away
the
risk
we
pay
for
for
further
development,
and
when
it's
done,
you
can
start
producing
it
and
any
well.
The
benefit
for
us
is
that
anybody
else
can
produce
it
as
well,
so
they
have
to
be.
B
They
have
to
have
a
a
an
excellent
reputation
in
delivery,
excellent
operations,
because
otherwise
one
of
their
employees
can
just
step
out
and
say
I'll,
be
no
trokey
and
and
and
and
make
a
better
product.
So
so
the
competitiveness
of
open
source
is
the
guarantee
that
it's
it's
worthwhile
for
society
and
if
we
accept
the
cost
that
that
we
that
we're
going
to
pay
for
it
anyway,
because
the
public
sector
is
one
of
the
biggest
consumers
of
every
technology
that
we
can
possibly
think
of.
Then
we'll
have
having
our
well.
B
You
can't
refuse,
because
there's
no
alternative
or
you
have
an
offer
where
an
organization
like
nitro
key
is
just
the
best
around,
so
they
get
the
money
and
as
long
as
they
stay
the
best
around
they
get
the
money
if,
if
all
of
them
like
in
the
case
of,
for
instance,
next
cloud
versus
own
cloud,
if
the
entire
team
leaves
the
company,
the
contracts
may
still
be
in
place,
but
people
will
move
their
development
on
the
outside
and
I
think
that's
a
healthy
thing.
It
it
well.
B
D
Okay
yeah,
so
I'm
just
asking
one
more
follow-up
question
that
I
had,
which
may
sound
a
little
bit
naive,
but
sometimes
free
and
open.
So
software
and
funding
it
is
associated
with
some
kind
of
altruistic
approach
and
so
to
make
it
clear,
you're
not
against
people
becoming
rich
with
with
their
software.
B
No,
it's
it's
it,
but,
but
I
mean
there's:
there's
there's
a
boundary
where
obscenely
rich.
I
don't
think
people
should
have
a
billion
times
more
than
somebody
else.
So
so,
and
I
think
that
currently,
given
that
we
have
that
that
that
position
right
now,
if,
if
you
manage
to
do
that
with
the
approach
that
we
have
you
could
you
could
you
can
have,
I
mean
there's
lots
of
folks
that
made
lots
of
money,
doing
open
source
software
free
and
open
source
software.
B
B
B
B
A
B
A
B
So
so
I
don't
think
we
have
a
proper
solution
for
that.
So
I
mean,
if
you
know
where
we're
coming
from
is,
is
from
the
barren
land
of
doing
nothing
and
and
and
perhaps
somebody
crafting
some
debian
packages
that
will
may
or
may
not
work
at
some
point
in
time.
Taking
the
next
step
is
is
obviously
going
towards
next
packaging
delving
into
the
specific
issues.
So
it's
it's
it's.
There
are
so
many
edge
cases
that
we
still
have
to
consider.
B
We
care
about
the
topic
at
hand
and
we
want
to
invest
in
it
and
please
tell
us
what
to
do
to
fund.
Please
tell
us
where
to
invest
because
we
yeah
you're
on
the
inside.
You
see
all
the
edge
cases,
you
see
all
the
problems
and
we're
just
happy
if
somebody
solves
them
and
that's
how
it
works,
people
notice
a
problem.
They
come
to
us,
ask
us
for
money,
we
give
them
money
and
they
solve
the
problem
and
everybody's
happy.
B
Well,
it's
it's
called
europe
and
it's
actually
just
go
and
live
anywhere
where
you
want,
including
on
the
countryside
and
basically
deliver
the
technology
set
through
this.
I
don't
one
of
one
of
the
we
I
was
discussing
this
with
a
colleague
earlier
today.
One
of
the
one
of
the
problems
that
you
traditionally
have
is
we've
been
discussing.
For
instance,
having
like
an
ngi
channel
like,
like
you,
have
the
the
main
knicks
package:
don't
have
an
ngi
channel,
but
we
really
want
people
to
be
part
of
the
mainstream.
B
We
really
want
people
to
be
in
in
the
silicon
world
to
talk
to
the
silicon
people
and
in
the
real-time
messaging
world
to
talk
to
the
real-time
messaging
people,
because,
ultimately-
and
then
we
want
to
have
like
this
shared
backing
of
of
of
of
of
the
the
distribution
and
packaging
and
quality
assurance,
but
at
the
same
time
we
want
yeah.
We
want
the
communities
to
be
effective
and
we
well,
I
I
I
think,
there's
so
many
people
working
on
this
type
of
technology.
B
It's
just
that
this
is
a
really
effective
sparta,
with
lots
of
heavy
hitters
in
a
very
sm.
In
a
very
crowded
space
but
yeah,
we
want
them
not
in
a
war
mode,
but
we
want
them
to
be
taking
over
society
step
by
step,
so
to
say
so
so
I
I
I
I
I
thank
you
as
one
of
the
organizers
of
last
year's
nick's
camp.
I
think
matthew
has
as
a
good
idea
gather
people
anywhere
it
can
be
hacker
camps
can
be.
B
A
A
A
So
that
was
yet
an
other,
very
amazing
lecture.
We
went
a
little
bit
over
time,
but
that's
okay,
very
interesting
questions
and
discussions
had
to
be
done
so
once
again,
thanks
to
the
people
that
made
this
possible,
the
european
commission
and
I'll
net
the
nexus
foundation
tweak
and
for
more
questions,
and
if
you
guys
want
to
discuss
a
little
bit
more
about
this
lecture
and
previous
lecture
for
that
matter,
head
over
to
the
matrix
room.
A
Where
you
can
ask
questions
have
discussions
and
it's
it's
a
nice
place
to
be,
and
then
yes,
the
next
lecture
will
be
next
week
same
place
same
time,
like
always
by
conrad.
If
I'm
pronouncing
that
right
about
reproducible
science
with
nixon
geeks
and
I
hope
you're
looking
forward
to
that,
I
know
I
am,
and
I
will
see
you
then
goodbye.