►
From YouTube: The city guide to open source - GitHub Universe 2019
Description
Presented by Devon Zuegel, Product Manager of the Open Source Economy Team at GitHub
What can cities teach open source about governance, funding, and managing limited resources?
In this talk, we'll take a stroll through von Haussmann's Paris, the subways of NYC and Hong Kong, the streets of Singapore, and modern day San Francisco. We'll look at how the long history of the design, maintenance, economics, and politics of physical infrastructure can inform how we approach our digital infrastructure—and where we'll need to draw a new map for ourselves.
About GitHub Universe:
GitHub Universe is a two-day conference dedicated to the creativity and curiosity of the largest software community in the world. Sessions cover topics from team culture to open source software across industries and technologies.
For more information on GitHub Universe, check the website:
https://githubuniverse.com
A
A
Yesterday
we
announced
that
github
sponsors
which,
as
a
quick
refresher,
is
a
way
to
financially
support
the
open
source
projects
that
you
depend
on
yesterday,
we
announced
that
we
are
now
able
to
support
organizations
to
receive
funding.
So
if
your
projects
like
to
identify
as
a
team
as
a
group
of
people
as
opposed
to
just
an
individual
developer,
you
can
now
receive
funding
as
a
group.
So
we're
really
excited
about
this.
A
One
of
the
things
that
really
kicked
off
my
interest
in
this
space
is
a
good
friend
of
mine.
Named
Nadia
egg
ball
wrote
a
report
in
2016
called
roads
and
bridges,
roads
and
bridges
talks
about
how
there's
all
of
this
unseen
labor
that
open-source
maintainer
--zz
do
that,
most
of
which
goes
unpaid
and
is
on
volunteer
time
and
I'm
somebody
who
is
really
interested
in
cities
and
urban
design
and
urban
development
and
I'm
a
big
nerd
about
land
use
policy.
A
So
it
really
was
very
interesting
for
me
to
see
Nadia
talking
about
open
source
in
terms
of
city
infrastructure.
That
metaphor
really
struck
me
and
pulled
me
in
so
today.
I'm
gonna
combine
my
two
interests:
my
love
of
cities
and
my
love
of
open
source
and
share
a
little
bit
of
the
the
parallels
that
I
see
between
them.
So
what
we'll
do
is
we're
gonna
talk
about
three
parallels
and
one
really
big
difference.
A
So
as
a
quick
rundown,
we're
gonna
talk
about
how
cities
and
open
source
projects
make
decisions,
how
they
secure
funding,
how
they
manage
congestion
and
when
to
define
governance.
So
this
is
all
very
abstract.
I
promise
I'll
go
more
in
more
detail
as
we
go
through
the
talk.
So
to
start,
both
open
source
projects
and
cities
have
to
make
really
difficult
decisions
that
often
affect
a
lot
of
different
stakeholders
who
have
different
goals
and
different
opinions
about
what
should
happen
so
to
start
with
cities.
A
There
are
I
think
it's
really
interesting
to
look
at
this
as
one
primary
dimension
around
how
who
gets
the
decision-making
power?
Is
it
a
singular
person?
Is
it
a
centralized
source
that
gets
to
make
one
call
or
is
it
sort
of
more
of
a
chaotic,
and
you
know
maybe
democratic
thing
where
everyone
gets
to
say
everyone
gets
to
be
part
of
the
decision
and
a
very
interesting
instance
of
that
first
extreme
is
Paris.
So
Paris
is
one
of
the
most
beloved
cities
in
the
world.
A
It
is
really
famous
for
these
beautiful
boulevards
that
cut
through
the
city
and
the
way
that
they
got
there
was
in
medieval
Europe.
Paris
was
previously
this
warren
of
very
small
streets
with
no
sewage
people
had
were
very
sick
all
the
time.
It
was
incredibly
difficult
to
get
across
the
city
because
if
you
can
imagine
those
little
areas
between
the
big
boulevards,
that
was
the
entire
map,
and
so,
if
you
wanted
to
cross
the
city,
you
had
to
go
through
this
maze
of
little
streets
just
to
get
anywhere.
So
during
Napoleon
the
thirds
rain.
A
This
is
not
Napoleon
Bonaparte
but
hit
one
of
his
sons.
Sons
I
think
he
was
an
emperor,
and
so
he
got
to
say
exactly
what
was
going
on
and
he
was
looking.
He
looked
at
his
streets
and
he
looked
at
the
people
living
in
Paris
and
dying
of
cholera,
and
he
said
you
know
what
we're
not
going
to
put
up
with
this
anymore.
So
he
put
a
man
in
charge
named
Baron
von
Houseman
and
Baron
von
Houseman
was
given
basically
complete
reign
over
Paris.
A
He
could
do
whatever
he
wanted,
and
one
could
argue
that
he
had
a
very
good
impact
on
Paris.
They
now
have
sewage,
that's
definitely
a
plus.
They
also
have
these
boulevards
to
go
through
the
city,
but
at
the
same
time
to
create
those
boulevards.
He
just
sliced
through
the
city
with
a
red
line
and
tore
down
entire
neighborhoods.
A
So
I
think
this
is
an
interesting
image,
because
you
can
see
exactly
where
all
there
were
houses
there
were
businesses.
There
were
community
centers
at
every
point
where
those
roads
now
cut
through
and
Vaughn
Houseman
just
decided.
You
know
what
it's
for
the
greater
good.
We
need
to
just
cut
right
through
an
interesting
contrast
is
San
Francisco,
which
we
are
all
familiar
with
here
here
in
San
Francisco
we
have
a
pretty
serious
housing
crisis
that
I
promise.
A
There's
this
thing
called
discretionary
review,
which
effectively
means
that
any
given
person
in
San
Francisco
can
say
you
know
what
I
don't
like
that
building,
let's
not
build
it,
and
this
is
makes
it
really
hard
to
get
anything
done.
So
the
there's
a
very
different
sort
of
story
here
and
with
a
very
different
set
of
trade-offs
and
I.
Think
in
open-source
communities.
You
see
it
very
similar
things
in
open-source
communities
you
have.
A
Some
extremes
like
on
on
Elm
is
a
really
interesting
example
of
a
programming
language
where
there's
a
guy
named
Evan,
who
just
is
really
talented
and
has
really
strong
opinions
about
what
the
project
should
be
and
when
he
found
it
out,
he
very
much
set
the
tone
you
know
this
is
this
is
my
thought
baby
that
I
am
going
to
make
this
my
brainchild
and
do
it
the
way
I
want
and
I
have
this
vision
for
what
Elm
will
be?
If
you
are
with
me,
let's
go.
A
If
you
are
not
with
me,
that's
fine,
but
you
don't
have
to
use
Elm,
but
don't
complain
about
it.
Javascript
is
on
the
opposite
end
of
the
spectrum,
where
JavaScript's
is
much
more
community
driven,
there's
a
committee
that
makes
a
lot
of
decisions
about
what
goes
in
the
formal
language
and
before
even
that
you
can,
you
can
get
some
things
inserted
into
battle
battle.
A
If
you
want
to
have
your
own
little
flavor
of
JavaScript-
and
this
means
that
JavaScript
both
can,
it
can
be
harder
to
make
really
major
decisions
that
affects
everybody,
but
it
can
also
be
easier
to
make
smaller
decisions
and
have
a
little
local
experimentation.
So
to
look
at
these
trade-offs,
I
think
in
elm
evan
has
made
some
really
hard
choices.
He's
deprecated
things
he's
made
certain
parts
of
the
language
not
backwards
compatible
because
he's
like
you
know
what
that
thing.
A
At
the
same
time,
it's
quite
hard
to
get
that
little
experimentation
inside
of
Elm
because
you
have
to
get
through
Evan
and
if
he's
not
excited
about
it
or
if
he's
busy
because
he's
one
dude,
it's
just
not
going
to
happen,
whereas
in
JavaScript
land
you
can
play
with
your
own
little
flavor
of
JavaScript
locally
play
with
it.
Share
it
with
your
friends
and
if
it
catches
on,
then
it
could
get
adopted
into
the
formal
language
over
the
course
of
years.
A
However,
the
history
of
San,
Francisco's
land
use
policy
comes
from
a
very
real
place
that
is
quite
similar
to
the
Vaughan
Houseman
situation,
so
in
in
the
mid-century
in
the
mid
1900s
America
was
not
very
responsive
to
certain
groups.
They
that
we
did
not
really
listen
to
the
certain
communities
and
if
you
look
at
this
highway
map
of
some
of
which
these
highways
were
built,
they
just
cut
right
through
minority
neighborhoods
and
they
were
not
listening
at
all
to
people.
A
Luckily,
before
they
built
all
of
these
highways
and
cut
through
the
V
communities,
San
Franciscans
stood
up
and
said:
you
know
what
we
don't.
We
don't
stand
for
this.
We
want
our
voices
to
be
heard,
and
this
is
way
too
centralized
in
the
planning
department.
We
need
to
include
more
people
in
the
community,
so
this
is
actually
a
lot
of
the
history
for
why
we
ended
up
with
the
housing
policy
that
we
have
where
any
any
neighbor
can
say.
A
Okay,
so
how
to
secure
funding?
This
is
something
that
I
think
open-source
communities
could
learn
a
lot
from
from
certain
cities,
so
in
Sam,
in
in
New,
York
City
there's
a
crumbling
subway
station.
Despite
having
millions
of
trips
every
single
day,
I
think
it's.
Six
million
trips
are
made
on
the
New
York
City
subway
every
day,
and
yet
they
don't
really
take
care
of
this
infrastructure
at
all.
Some
of
the
people
in
this
audience
may
see
some
parallels
with
certain
projects
and
open
source,
or
we
just
don't.
We
don't
invest
quite
enough
by
comparison.
A
Hong
Kong
has
this
beautiful
shining
MTR
station
MTR
is
the
the
company
that
runs
the
the
the
the
public
company
that
runs
the
railroads
in
Hong
Kong,
and
that
contrasts
to
go
back
to
that
slide
is
just
incredible
to
me
when
I,
when
I
was
in
Hong
Kong
about
a
year
ago,
I
was
amazed
by
this
great
train
system
and
the
way
that
Hong
Kong
does.
It
is
also
the
way
that
Singapore
does
it.
A
They
have
this
funding
mechanism
called
value
capture,
and
what
that
does
is
the
the
railroad
company
actually
owns
the
land
around
the
stations,
and
so
when
they
build
a
new
station,
it
increases
the
value
of
land
all
around
it
right.
If
you
could,
you
could
imagine
if
we
were
to
add
a
new
BART
stop
in
some
neighborhood.
That
would
increase
the
value
of
all
of
the
property
there,
because
it's
really.
A
Mozilla
tea
to
the
rest
of
the
metro
area,
so
Hong
Kong
owns
this
land
and
then
they
rent
it
out,
and
so
they
actually
make
money
when
they
create
value,
as
opposed
to
just
bearing
all
of
the
costs
of
that
infrastructure.
Because,
again,
if
you
think
about
Bart,
if
Bart
didn't
exist,
the
entire
metro
region
of
francisco
of
oakland
of
east
bay,
all
this
stuff
would
be
much
much
lower
because
we
wouldn't
all
be
able
to
coordinate
with
each
other
and
collaborate
and
go
visit.
Your
friend
in
oakland.
A
Unfortunately
bart
and
the
new
york
city
subway,
don't
actually
capture
any
of
the
value
that
they
create
their
sort
of
bearing
all
of
the
costs
again
and
giving
all
of
the
benefits
away
to
the
community.
And
so
I
think
that
this
is
most
mostly
a
good
thing
and
we
are
able
to
fund
some
amount
of
infrastructure
this
way.
A
So
how
does
this
relate
back
to
open
source?
You
guys
are
probably
all
sitting
in
the
audience
wondering
why
I
keep
going
on
about
cities,
so
I
think
that
open
source
projects
should
really
start
thinking
about
things
in
these
terms
and
the
people
who
use
those
open
source
projects
in
the
companies
as
well.
A
Right
now,
we
sort
of
see
open
source.
A
lot
of
people
see
open
source
as
a
freebie
that
we
can
just
take
and
we
can
take
and
take.
But
the
problem
is
that
if
we
don't
put
money
back
into
that
system,
it's
gonna
make
it
a
lot
harder
to
take
things
out
of
it
in
the
future
and
the
best
part
of
it
of
this
is
when
you
fund
your
infrastructure,
you
get
a
lot
more
value
out
than
you
ever
put
in.
A
So
I
think
this
is
the
mental
model
that
we
should
have
for
thinking
about
infrastructure,
much
more
than
sort
of
the
traditional
route
which,
frankly,
is
still
born
in
u.s.
infrastructure
systems
that
we
are
not
investing
back
in.
This
is
gonna
take
time.
New,
York,
City
and
San
Francisco
still
haven't
really
figured
this
one
out.
It's
it's.
The
the
value
capture
model
of
Hong,
Kong
and
Singapore
is
still
quite
rare
in
the
world,
but
it's
growing
in
popularity
and
so
I'm
just
trying
to
evangelize
it
everywhere,
because
I
think
it's
the
way.
A
We
should
think
about
all
sorts
of
infrastructure,
whether
it's
physical
or
digital,
okay.
So
a
third
thing
we
can
talk
about
is
how
to
manage
so
in
cities.
Congestion
looks
like
traffic
right,
you
you
get
on
the
road
and
everyone's
blocking
you
and
things
are
moving
slowly.
Street
is
congested.
An
open
source
congestion
looks
like
opening
the
pull
request
or
opening
an
issue
and
seeing
that
there
are
500
more
in
front
of
you
and
that
the
maintainer
is
just
super
busy
and
does
not
have
time
to
get
to
your
issue
or
pull
request.
A
Even
if
it's
a
super
important
one
in
cities,
the
way
that
we
manage
congestion
is
through
something
called
a
congestion
tax
in
Singapore,
again,
I
always
go
back
to
Singapore
as
a
source
of
really
interesting
examples
in
Singapore
they
have
this
congestion
tax.
That
says,
every
time
you
go
on
the
road,
every
single
mile
that
you
go,
you're
gonna
get
charged
a
small
fee,
very,
very,
very
small,
but
it's
just
because
when
you're,
when
you
get
on
the
road
and
you're
complaining
about
traffic,
you
are
also
traffic.
A
It's
interesting,
to
contrast
this,
to
the
way
that
issues,
work
and
github
or
tickets
work
in
JIRA
or
all
sorts
of
things
like
that,
the
or
you
know
message
boards,
where
it's
really
really
cheap.
If
we,
if
we
put
it
in
sort
of
money
terms,
it's
really
cheap
to
open
an
issue
and
create
more
work
for
somebody.
All
you
have
to
do
is
be
like
hey
dude,
this
thing's
broken
and
then
the
maintainer
is
sitting
there
being
like.
Ok,
you
didn't
really
give
me
ways
to
reproduce
it
like
now.
A
I
have
to
walk
you
through
this
whole
process.
It's
frustrating-
and
maybe
maybe,
if
it's
a
feature
request
that
you've
made.
Maybe
you
don't
even
care
really
about
the
feature,
but
now
you've
left
the
maintainer
thinking
all
weekend
about
should
I,
like
should
I
should
I
respond
to
this
guy.
It's
kind
of
a
dumb
feature
but
like
I,
want
to
be
polite
and
you've
created
all
of
this
work
and
there's
this
really
big
imbalance
between
the
amount
of
work
created
and
the
amounts
of
effort
that
was
actually
put
into
opening
it.
A
So
don't
worry,
we're
not
going
to
add
a
congestion
tax
to
issues
and
github,
but
I
think
that
this
is
a
very
interesting
thought
experiment
of
like
what
what
a
congestion
tax
look
like
and
I've
already
started,
seeing
some
open-source
maintainer
playing
with
something
that
looks
a
bit
like
this.
They
say.
A
Yet
again,
most
cities
still
don't
do
this
and
that's
why
it's
so
hard
to
commute
to
work
in
the
Bay
Area.
So
a
little
bit
of
economics
here
to
explain.
What's
going
on
so
our
standard
model
of
fixed
and
marginal
cost?
Is
this
right
you,
if
you're,
creating
a
factory
that
makes
widgets
the
fixed
cost?
Is
the
cost
of
creating
the
factory
and
the
marginal
cost
is
every
time
you
make
a
widget?
How
much
more?
Do
you
pay
right?
A
It's
useful
to
break
this
down
a
little
bit
into
some
subcategories,
so
the
marginal
cost
includes
the
manufacturing
cost,
which
can
be
like
the
materials.
The
cost
of
running
the
machines
there's
the
distribution
cost,
which
is
you
know
you
need
to
get
the
widgets
to
the
people,
and
maybe
you
have
trucks
that
are
driving
them
to
stores
all
over
the
place.
There's
the
servicing
cost.
A
Let's
say
the
widget
I,
don't
know
bites
of
kids
fingers
off
and
now,
like
his
mom,
calls
in
and
says
hey
my
kid
has
a
problem
that
then
you
have
to
pay
for
all
of
the
the
call
centers
that
now
like
service
that
thing
open-source
and
software
people
usually
think
of.
Has
it
having
zero
marginal
cost?
A
They
think
it's
all
fixed
costs,
you
just
do
the
development
upfront
and
then
you're
good
to
go
and
I
think
that
this
is
the
source
of
a
lot
of
the
misconceptions
around
funding
open
source
and
why
we
think
that
we
don't
need
to
fund
it
because
you're
like
well,
the
fixed
cost
is
there
you
just
make
copies
and
yes,
I
would
download
a
car
if
I
could
kind
of
this
is
sort
of
right.
I
think
this
does
point
to
something
that
is
really
beautiful
about
software,
where
most
of
the
investment
does
happen.
A
Upfront
and
then
does
scale
immensely,
but
it
misses
something.
So
the
manufacturing
cost
is
basically
zero.
Copying
bits
is
very
cheap.
Distribution
cost
is
basically
zero.
You
just
go
to
the
github
repo
or
you
know
down
the
tarball
or
whatever,
but
the
servicing
cost
is
not
free.
Every
time
someone
opens
a
bug
report
that
imposes
a
cost
on
the
maintainer
or
the
developer,
the
community
that
is
maintaining
this
work,
and
that
is
not
seen
for
a
very
long
time.
A
Think,
oh,
you
just
wrote
the
code
like
it's
really,
not
that
bad
and
don't
you
like
writing
code,
but
actually
servicing
costs
is
extremely
high,
not
to
mention
dealing
with
the
people,
problems
that
can
come
about
from
the
earlier
problems
that
we
talked
about
around
like
making
decisions
and
deciding
what
the
funding
is.
So
I
would
really
like
people
to
stop
sharing
the
meme
that
software
has
zero
marginal
costs,
because
it's
really
not
true
and
I
think
it
leads
to
a
lot
of
bad
behavior
down
the
line.
A
People
people
understandably
make
like
I
do
think
that
it's
an
understandable
misinterpretation,
but
we
need
to
remember
that
there's
that
servicing
cost
that
really
does
bear
on
people.
So
again,
the
way
we
deal
with
that
cost
that
the
marginal
cost
in
cities
is
that
condemned
congestion,
tax
I
think
would
be
really
cool
to
start
seeing
people
think
about
how
to
price
that
that
cost
that
is
currently
entirely
being
borne
by
maintainer
x',
and
I
want
to
spread
that
around
a
little
bit
and
make
it
so.
A
The
whole
community
bears
it
okay,
so
those
are
three
things
that
I
think
we
can
learn
from
cities
and
I'm
really
excited
to
see
more
people
experimenting
with
this
over
time.
There's
been
really
an
explosion
of
people
trying
things
like
this
and
I've
incepted
these
ideas
and
enough
people
that
some
some
things
have
started
happening,
but
there's
something
that
we
have
very
different
between
physical
infrastructure
and
digital
infrastructure.
A
So
here's
the
place
where
I
want
to
change
gears
a
little
bit
and
say
you
know
we're
gonna
have
to
figure
this
one
out
on
our
own,
and
that
is
when
to
define
governance.
So
let's
say
I
want
to
build
a
dam.
If
I
want
to
build
a
dam,
it
is
really
hard
to
even
get
started.
I
have
to
get
permits.
I
probably
have
to
have
a
master's
degree,
at
least
so.
There's
like
there's
three
years
of
my
life
I
have
to
go
through
environmental
review.
I
have
to
go
talk
to
all
the
communities.
A
I
have
to
raise
a
ton
of
capital.
The
list
goes
on
and
on
before
you're,
even
starting
to
lay
the
first
rounds
of
cement,
and
so
what
this
means
is
that
there's
a
really
high
barrier
to
entry
for
building
something
like
a
dam
or
any
other
physical
infrastructure.
One
of
us
cannot
just
walk
out
there
and
be
like
you
know
what
I
want
to
build
a
new
highway
today.
No,
that's,
not
that's!
Not
an
option.
A
By
contrast,
starting
a
new
open-source
project
is
really
easy.
All
you
have
to
do
is
go
to
github,
create
a
new
repository
and
there
you
go.
You
have
an
open-source
project,
maybe
no
one's
using
it,
and
maybe
it's
totally
useless,
but
you
can
start
and
there's
no
one's
stopping
you.
This
is
one
of
the
most
beautiful
things
about
open-source.
It's
one
of
the
reasons
why
we
get
so
much
innovation,
really
talented
people
who
maybe
society
doesn't
necessarily
respect
or
like
overlooks
a
little
bit.
You
can
just
go
in
and
be
like.
A
A
It
is
also
the
source
of
some
of
the
biggest
challenges
of
open-source
and
what
I
mean
by
that
is
that
open-source
developers
often
start
projects
without
really
having
a
very
long
term
plan
for
what
it'll
be
because
they're
just
hacking
away
on
a
Friday
night.
They
eating
their
pizza
with
their
friends
or
whatever,
maybe
at
a
hackathon,
and
they
don't
necessarily
even
know
what
it'll
be
in
a
week,
let
alone
in
three
years
or
five
years.
So
a
very,
very
often
story
that
I
hear
from
amazing
open-source
maintainer
x'
is
that
they've
started
this
project.
A
What
is
and
is
not
the
community
who
does
get
to
decide
rules
who
doesn't,
if
you
start
receiving
funding
who
gets
to
make?
Who
gets
the
money?
I
can't
tell
you
how
many
open-source
projects
I
know
of
who
are
raising
money,
but
as
soon
as
they
have
the
money,
it
just
sits
in
the
bank
account
because
they're
like
who
do
we
actually
allocate
this
to
this
is
really
hard.
A
A
What
what
structure
should
I
be
putting
in
place
now
that
it's
big
and
I
think
it's
cool,
because
it
means
that
we
get
to
lazy
load
governance.
It
means
that
with
a
dam,
you
have
to
decide
governance
and
make
all
these
decisions
up
front.
Now
we
only
get
to
decide
if
the
project's
actually
working,
but
the
thing
is
when
the
project
is
actually
working.
You
do
need
to
decide
these
questions.
A
You
do
need
to
nail
these
down,
because
otherwise
you
will
be
completely
overwhelmed
with
work
and
with
really
hard
ethical
questions,
even
okay,
so
that
was
sort
of
a
whirlwind
through
my
two
obsessions,
which
are
cities
and
open-source
and
I
would
love
to
talk
about
these
basically
without
any
end
whatsoever.
But
I
also
want
to
leave
you
guys
some
time
to
ask
questions
about
github
sponsors
about
how
github
thinks
about
open
source
about
any
of
the
research
that
I've
done
or
about
cities
to.
A
If,
if
you
are
interested
so
I
will
leave
you
with
just
one
story,
that
I
think
will
I
find
really
helpful,
but
you
might
find
kind
of
gross
in
Japan
as
late
as
I
think
the
1950s
they
didn't
have
sewage
in
most
major
cities.
So
if
you've
ever
been
to
Tokyo
I
mean
imagine
Tokyo
with
no
sewage
system,
not
a
pretty
sight.
A
If
you
look
at
those
those
barrels
they're
sitting
on
the
sidewalk,
you
can
guess
what
those
are
filled
with
and
if
you
go
to
Tokyo
today
it
is
a
sparkling
clean
city.
It
smells
great,
it
doesn't
smell
bad
at
least
and
they've
really
solved
this
problem,
and
so,
while
I
think
every
once
in
a
while
I
get
a
little
down
about
how
little
we
are
funding
open
source
and
how
little
support
maintainer
'z
have.
But
we
are
not
in
this
bad
of
a
situation.
A
We
are
not
literally,
you
know,
I'm
actually
not
going
to
go
there,
but
the
point
is
we
can
make
progress
really
quickly.
It
only
took
about
a
decade
or
two
for
them
to
fully
replace
all
of
this
infrastructure
and
solve
these
problems,
and
they
did
it
and
not
a
very
bump
on
Houseman
style.
They
did
it
without
disrupting
a
ton
of
people's
lives
because
they
had
better
technology
and
they
were
more
thoughtful.
So
I
like
this
picture,
because
it
reminds
me
that
we
can
make
progress
quickly.
A
We
can
do
a
lot
even
when
it
seems
like
a
very
desperate
situation
and
so
I'm
really
excited
to
see
more
experiments,
encourage
more
experiments
and
and
try
new
things
out
inside
of
the
github
product
too.
So,
if
you
guys
have
any
ideas,
please
let
me
know
I
think
oh,
this
is
this:
is
their
new
sewage
system.
It
works
a
lot
better
if
you
have
any
ideas
whatsoever.
A
A
B
A
A
So
it
really
is
what
got
me
into
it
all
in
the
first
place,
and
actually
how
I
met
NAT,
the
the
CEO
of
github
I'm,
very
involved
in
housing
policy
in
San
Francisco,
and
he
is
too,
and
so
we
didn't
even
talk
about
software
for
the
first
like
year
that
we
knew
each
other.
I
didn't
really
know
that
he
had
any
technical
background.
We
were
just
talking
about
cities
and
like
yelling,
about
city
planning,
meetings
and
things
like
that,
and
it
really
inspired
the
whole
project
actually
so
I.
A
This
is
a
little
bit
abstract,
but
I
think
it's.
It's
useful,
still,
I'm
a
really
big
believer
that
coordination
problems
are
some
of
the
biggest
problems
in
the
world
and
a
coordination
problem
is,
if
you
guys,
are
familiar
with
prisoners,
dilemmas
or
tragedies
of
the
Commons.
This
idea
of
you
know
two
people
or
multiple
people
who,
when
they
act
in
their
own
best
self-interest,
they
do
things
that
are
worse
for
all
of
them.
I
think
those
are
some
of
the
worst
problems
in
the
world.
A
A
prime
example
of
this
being
global
warming
right
where
everyone
doesn't
want
the
planet
to
melt.
I
think
we're
all
on
the
same
page
with
that,
but
at
the
same
time,
every
time
you
get
in
the
shower,
you
just
want
to
leave
it
on
a
little
bit
longer
or
you
know
you
want
to
buy
that
new
thing
that
like
comes
in
a
big
plastic,
wrap
or
you
know
it's
a
little
bit
more
efficient
to
take
an
uber
than
to
bike.
So
I'm
gonna
do
that.
A
But
if
we
all
do
that,
then
it
adds
up
and
it
screws
us
all.
So
I
think
the
way
that
we
are
not
really
funding
open-source
at
the
moment
is
the
same
kind
of
problem
where
we
would
all
be
better
off
if
our
digital
infrastructure
was
great
and
we
could
all
use
it
without
a
hitch,
and
it
was
well
maintained
over
the
long
term.
But
no
one
really
wants
to
be
the
person
who
actually
foots
the
bill,
and
so
I
was
originally
spending
a
lot
of
my
time.
A
As
a
software
engineer,
myself,
I
understand
how
this
stuff
works.
I
know
the
people
involved.
I
can
do
things
here
and
then,
when
not
became
the
CEO
I
sent
him
a
very
long
email
being
like
you
guys
should
do
something
about
open
source
and
then
the
he
emailed
me
back
saying
you
should
do
something
about
open
source,
come
try
and
github.
So
there's
there's
a
very,
very
strong
thread
between
those
and
in
my
brain.
I
cannot
separate
them
at
all.
D
A
We
get
it
as
a
really
strong
product
philosophy
of
creating
a
canvas
for
developers
to
do
things
on,
and
we
don't
really
want
to
define
what
people
do
as
long
as
it's
within
the
bounds
of
our
respectful
community,
and
so
we
don't
want
to
pick
winners
and
we
we
invite
everyone
to
get
sponsored
through
through
getting
sponsors,
I
mean
I
can
say
like.
If
we
were
picking
winners,
we
would
not
have
chosen
certain
projects
that
we
just
didn't
know
about.
So
it's
really
important
for
us
to
leave
it
as
open
as
possible
to
everybody.
E
Hey
I'm,
a
big
fan
of
yours,
I,
think
that
this
is
a
fantastic
talk
and
I'm
super
grateful
for
the
sponsors
project.
I'd
like
to
know
what
you
think.
As
far
as
the
role
that
github
has
in
in
the
funding
and
sustainability
of
open
source
like
to
what
extent
do
they
feel
it's
a
responsibility
and
an
obligation
versus
just
something
that
we
should
a.
A
Responsibility
and
obligation
on
the
part
of
github
or
so
get
it
would
be
nothing
without
open
source
developers
are
the
core
of
the
github
community,
and
it's
why
people
come
to
github
I.
Think
actually,
an
interesting
and
interesting
metaphor
here
is
like
twitch
or
YouTube.
So
twitch
and
YouTube
have
a
very
small
portion
of
the
user
bases
that
are
streamers
or
you
know,
video
creators,
but
if
they
didn't
upload
content
like
no
one
would
want
to
come
to
that
community
at
all.
I
think!
A
That's
where
the
parallels
end,
there's
other
things
that
are
not
similar,
but
I.
Think
once
you
frame
it
as
that,
where
it's
like
maintained
errs,
are
what
make
github
what
it
is,
then
it's
pretty
it's
just
obvious
good
business
sense
forget
him
to
be
doing
this,
so
that's
one
thing,
I
think
like
it's,
it's
just
out
of,
like
almost
economic
self-interest.
It's
a
great
idea
to
do
this
and
I
think
it
was
I'm
very
excited
for
github
to
be
taking
this
on.
A
For
that
reason-
and
it's
also
one
of
the
reasons
why
I
thought
github
was
a
good
place,
because
it's
not
just
a
question
of
sort
of
charity
or
just
out
of
self
out
of
being
like
good
good
company,
but
it's
actually
something
that
is
sustained
for
the
long
term,
because
it's
really
necessary
for
the
long-term
health
of
the
whole
ecosystem.
On
the
other
hand,
I'm
doing
this
because
I
think
that
github
also
has
sort
of
an
obligation.
A
Github
again
would
be
nothing,
and
so
it's
very
important
to
give
back
to
that
community,
because
they've
done
so
much
for
for
github
and
for
all
of
the
companies
that
use
software.
That
I
think
it's
a
big
thing,
but
that
said,
I
don't
think
it
hubs,
the
only
one
that
has
the
obligation
I.
Think
every
company
that
uses
open-source
has
an
obligation
here.
Every
developer,
who
uses
open-source
and
so
I'm
I,
think
we're
starting
to
see
a
cultural
change
where
people
are
starting
to
recognize
this.
But
it's
it's.
F
Hi,
as
you
extend
the
digital
infrastructure,
metaphor
and
kind
of
looking
at
city
infrastructure,
the
way
that
you
were
looking
at
it
there's
there's
a
border
in
there
somewhere,
where
it
stops
being
city
infrastructure
and
starts
being
the
plumbing
in
your
house.
And
when
we
look
at
the
way
companies
tend
to
fund
open
source
projects.
That
line
is
very
blurry
and
I
and
I
wanted
to
hear
kind
of
how
you
think
about
that
particular
aspect
of
the
problem.
A
A
This
is
one
of
the
places
where
I
think
it
creates
a
lot
of
the
tension
that
we
start
seeing
in
issues.
We've
probably
all
seen
those
flame
war
issue,
threads
that
happen
and
I
think
one
of
the
sources
of
tension
here
is
that
companies,
understandably
once
open
source
to
be
as
useful
as
possible
to
them,
but
the
maintainer
say
he
dude.
This
is
my
road
map
and
also
I'm,
not
gonna.
Just
go.
A
Do
some
boring
integration
work
for
you
for
free,
that's
not
going
to
happen,
and
so
there's
this
mismatch
of
expectations,
because
companies
think
oh
wait.
Why
doesn't
the
maintainer
want
this
to
be
as
useful
to
everyone
as
possible
and
the
maintainer
thinks?
Why
does
this
company
feel
entitled
to
my
time
both
very
understandable
reactions,
and
so
one
of
the
things
that
I
think
can
smooth
that
over
is
if
the
company
says
hey,
we
will
pay
you
to
do
this.
We
are
excited
to
use
the
software.
We
just
need
it
to
be
tweaks
in
this
way.
A
We
need,
you
know
a
wrapper
around
it
to
make
it
more
useful.
Here's
some
cash
to
make
that
happen,
and
so
that
can
align
the
incentives
where
you
know
the
the
maintainer
is
saying:
okay,
you
know
I'm
happy
to
do
some
of
this
custom
work.
It's
not
really
what
I'm
personally
most
excited
to
do
and
the
core
of
the
project,
but
it's
a
way
to
make
it
a
little
bit
more
useful
to
more
people.
Maybe
it'll
also
be
useful
to
the
rest
of
the
community.
Some
people
will,
you
know,
have
to
have
this
particular
integration.
A
I'll
make
that
happen
so
I
think
money
can
sort
of
make
that
simpler.
I
do
want
to
note
that
it
can
be.
It
can
also
create
risks
where
it
could
potentially
sort
of
change
the
road
map,
and
it's
not
just
about
changing
just
a
wrapper,
but
maybe
it
changes
some
part
of
the
core
library
and
I
think
that's
something
that
we
have
to
be
very
careful
of.
So
this
is
one
of
the
reasons
we've
been
rolling
out.
A
Github
sponsors
very
slowly
is
because,
while
I
believe
very
strongly
that
we
should
be
funding
open
source,
we
cannot
make
this
change
it's
too
quickly
because,
like
we
learned
in
Paris
it
ruins
people's
lives.
No
I
mean
the.
If
you,
if
you
completely
change
the
paradigm
of
the
way
something
is
done,
it
can
create
a
lot
of
problems,
even
if
the
old
paradigm
was
kind
of
broken.
A
G
Hey
thanks,
so
you
mentioned
in
the
talk
about
governance,
and
definitely
when
you
get
the
money
coming
into
a
project,
you
have
the
governance
question:
how
to
spend
it.
Yeah
does
does
github,
have
plans
to
provide
support
or
training
or
materials
to
help
new
projects
or
projects
that
are
growing,
adopt
best
practices
or
help
them
find
the
model
that
works
for
them.
That's.
A
A
good
question
so
right
now
the
answer
is
just
me:
I,
it's
not
really
a
formal
thing,
but
anytime
I
maintainer
wants
to
talk
about
that
stuff.
I
am
more
than
happy
to
talk
about
it
and
give
input
and
I'm
I
really
have
a
cool
job,
because
I
get
to
talk
to
all
of
these
different
people
and
sort
of
share
the
best
practices
across
them.
So,
while
I
don't
think
I
have
I,
don't
personally
have
all
the
answers
that
I've
created
myself
I've
like
collected
a
lot
of
really
good
answers
from
other
people.
A
That's
not
scalable,
though.
Unfortunately
I
don't
have
a
cloning
machine.
Not
yet
someone
hopefully
builds
an
open-source
version
of
that
at
some
point
soon,
but
I
think
that's
something
that
we
could
do
better
at
I.
Also
think
that
best
practices
are
not
very
well
defined.
Right
now,
often
my
advice
comes
in
the
form
of
here's
a
menu,
and
this
is
what
I
think
the
consequences
are
of
different
options.
A
But
I
don't
really
know
here
you
go,
but
as
we're
learning
more
through
this
program
and
as
we
as
we
sort
of
collect
a
set
of
options,
I
think
best
practices
will
continue
to
sort
of
rise
to
the
top
and
I'm
excited
to
see
that
happen.
So
long
story
short.
We
don't
have
anything
like
that.
Right
now,
I
think
it's
something
we
can
and
should
build
in
the
future,
but
I
want
to
hold
off
until
we
have
better
answers,
because
the
what's
worse
than
no
advice
is
bad
advice,
so
we
want
to.