►
From YouTube: What's an Open Source Program Office(OSPO) and why would a company want invest in one? #DemoDays
Description
2:05 - Program Start
5:23 - What is an OSPO?
7:21 - What industries benefit from OSPOs?
9:05 - Why would I want to start one?
16:09 - How does an organization start one?
21:26 - How does the OSPO know what policies to create?
27:13 - How to make good decisions with respect to software licenses?
29:21 - Where does the OSPO properly fit on the org chart?
36:46 - What is a good mix of people/roles in the OSPO
40:43 - Is this part-time work?
42:48 - How do I know that the OSPO is being successful?
48:55 - What about OSPO failures? What are the biggest risks?
53:06 - Additional resources
How do you manage the open source projects that your company uses? What's an OSPO, who should be in it and why?
Am I allowed to reuse this code inside a company project? What legal liabilities do we have around this piece of code that came from a partner or open source? These and many more questions are tackled by Natalie Bradley, Regional Director, Professional Services and Zack Koppert, Senior DevOps Engineer and open source project maintainer.
Next on Demo Days: https://www.linkedin.com/events/6793666268064034816/
B
Awesome,
thank
you
so
much
aj,
hello.
Everyone
super
excited
to
be
here
to
talk
to
you
all
about
open
source
program
offices,
so
many
people
reach
out
to
us
at
github
to
talk
about
how
they
should
manage
their
open
source
projects.
What
their
usage
should
look
like.
So
I'm
excited
for
us
to
share
some
guidance
with
you
all
today
and
even
more
excited
to
spend
some
time
with
zach
copert.
He
is
an
open
source.
B
Pro
he's
been
helping
many
large
enterprises
in
public
sector
in
commercial
industries
and
is
going
to
share
all
the
good,
the
bad
and
the
ugly
with
us
today,
a
quick
intro
of
myself
aj
mentioned.
I
am
a
regional
director
for
a
professional
services
organization.
I
help
our
customers
get
the
best
experience
with
github
prior
to
github.
C
B
Awesome
so
you've
got
you've
got
all
the
experience
lined
up
and
we've
got
a
really
good
chat
set
up
here
for
any
of
you
that
are
in
the
audience,
feel
free
to
drop
questions
in
the
chat
on
the
topic
of
ospos
and
we'll
try
to
get
them
answered
for
you
live
today,
we'll
mix
it
up
with
some
real
world
examples,
your
questions
and
zach's
experiences.
So
with
that,
let's
get
down
to
it,
zach,
let's
start
with
the
basics,
just
level
set
for
us
all
here.
What
is
an
open
source
program
office.
C
Okay,
so
an
open
source
programs
office
is
going
to
be
normally
a
team
of
people,
sometimes
only
one
person,
but
normally
a
team
inside
an
organization
and
they're
going
to
be
responsible
for
all
things:
open
source
right,
it's
the
central
hub
for
whether
you're
consuming
inner
source
or
producing
an
open
source
project
or
some
government
customers
of
yours
wants
to
know
what
open
source
is
in
your
projects
and
policies
for
legal.
All
of
that
sort
of
thing.
C
So
in
sort
of
real
world
examples
for
folks
that
don't
have
an
open
source
programs
office,
the
types
of
questions
that
lead
to
creating
it
is
like.
Okay,
a
developer
might
ask:
am
I
allowed
to
reuse
this
code
inside
of
a
company
project,
this
code
being
some
open
source
project,
and
how
did
they
know?
How
did
they
figure
that
out
or
or
some
legal
folks
might
be
wondering
about
a
product
that
recently
released
and
say
well
what
legal
agreements
do
we
have?
What
legal
liabilities
do
we
have
around?
C
You
know
this
this
part
of
the
code
or
this
code,
that
we
didn't
write
or
code
from
a
partnership
or
different
things
like
that,
and
ospos
really
help
navigate
and
traffic
control
all
of
the
different
networks.
In
terms
of
the
you
know,
legal
engineering,
business
leadership
and
decisions
open
source
intersource
when
to
do
which
and
how
to
protect
your
private
code
as
well.
B
It's
awesome,
so
it's
it's
a
group
of
people
that
are
really
defining
that
process
in
your
organization,
so
you
don't
kind
of
have
every
man
for
themselves
or
a
group
over
here
doing
they're
doing
it
one
way
and
somebody
over
here
kind
of
doing
whatever
they
want
a
different
way.
Okay,
so
is
this:
is
this
just
for
tech
companies
like?
Are
they
the
the
leaders
of
this?
Is
this?
Is
this
where
an
awesome
fits
best.
C
Yeah,
so
tech
companies,
I
think,
is
the
obvious.
Yes,
of
course,
most
large
tech
companies
have
open
source
programs,
offices
or
they're
thinking
about
them.
There
have
been
some
surveys
that
have
come
out
from
the
linux
foundation,
just
about
that
increasing
wave
of
tech
companies,
sort
of
growing
and
doing
that,
but
it
is
definitely
not
just
for
tech
companies.
So
if
you
look
at
a
couple
of
different
industries
that
are
traditionally
non-tech
low-tech
or
maybe
they're
increasing
or
investing
in
tech,
you
look
at
banking
and
financial
services.
C
C
B
C
Yeah,
so
what
the
industry
has
found
from
several
different
surveys
is
really
that
there's
this
97
percent
number
right
97
of
companies
are
building
their
their
technology
platforms,
their
products,
their
services
by
using
open
source.
Sometimes
there's
not
a
lot
of
awareness
from
that
from
the
leadership
level,
because
they're,
not
the
ones,
including
this.
You
know
time
zone
library
so
that
things
time
zones
automatically
get
converted
in
your
product
or
service.
C
But
the
developers
know
this.
You
know
they're
the
ones
that
are
including
that
library
and
writing
that
code
and
so
with
90
of
products
or
services
being
built
on
top
of
open
source.
That
means
that,
well,
your
organization
is
likely
very
likely
using
open
source,
and
that
can
be
a
great
reason
to
start
an
open
source
programs
office,
because
you've
already
got
something
to
manage.
C
You've
already
got
legal
liabilities
to
manage
with
the
licenses
and
projects
that
you're
consuming
in
your
products
and
that's
a
great
place
to
start
implementing
policy
and
and
what's
going
to
be
allowed,
are
we
going
to
allow
some
of
these
more
infectious
or
copy
left
license
projects
in
into
our
code,
which
is
a
whole
other
topic
we
might
get
into
later,
but
yeah
it's.
It
doesn't
take
much
if
you're
an
organization
operating
today
and
you're,
even
not
producing
but
using
technology
you're,
probably
using
open
source.
B
So
it's
a
big
deal
for
an
organization,
but
it
probably
is
beneficial
for
those
that
are
just
starting
off
so
they're
not
getting
themselves
into
too
much
trouble
further
down
the
line.
Have
you
do
you
have
any
experience
with
like
require
recommending
one
or
setting
one
up
in
an
organization
in
your
in
your
previous
life.
C
Yes,
yes,
so
I
have
I
have
done
this
before
and
so
I
the
story
goes.
I
was
a
software
manager
at
at
a
previous
company
and
I
was
assigned
to
a
new
team.
So
I
just
switch
onto
a
team.
I've
got
a
bunch
of
you,
know,
senior
developers
and
up
that
are
reporting
to
me.
I'm
trying
to
figure
out.
What's
going
on,
I
have
some
one-on-ones
with
them
and
figure
out
like
hey
what
you're
working
on
catch
me
up
to
speed.
C
You
know
the
new
kid
here
and
so
talking
to
one
of
the
developers.
Actually
it
turns
out
that
they're
working
on
a
an
embedded
web
server,
so
I
don't
know
if
you're
familiar
with
the
web
server
scene
in
open
source,
but
there's
like
a
jillion
of
them.
Some
of
them
are
really
well
established,
really
well
thought
out.
C
You
know
if
you
want
one
with
every
feature
under
the
sun
or
if
you
want
one,
that's
very
minimal
and
can
run
on
small
embedded
systems.
There's
it's
all
available
out
there
right
and
big
communities
around
these,
and
so
I'd
asked.
You
know
this
very
experienced
developer.
Like
hey.
Did
we
consider
you
know
open
source
options
on
this?
You
know
how
long
have
you
been
working
on
this?
C
We
have
high
standards
for
quality
and
anything
that
we
do
here
is
going
to
be
better
than
what's
what's
out
there.
I
just
had
to
let
that
soak
in
like
okay.
This
is
not
my
understanding
of
open
source
and
how
it
works,
nor
my
experience
with
some
projects
out
there
that
are
just
rock
solid,
and
so
you
know
after
having
conversations
about
like
well,
have
you
heard
of
tensorflow
like
it's
a
pretty
big
deal,
it's
pretty
well
established.
Have
you
heard
of
kubernetes?
It's
pretty
well
established.
It's,
you
know
very
high
quality.
C
These
are
open
source
projects
and,
there's
you
know
we're
not
using
code
snippets,
we're
using
whole
projects
so,
like
there's,
some
room
anyways,
there's
a
there's,
a
whole
long
story
about
that
specific
issue,
but
that
was
really
what
led
me
to
saying.
I
think
we
need
to
figure
out
how
to
sort
of
maximize
at
this
organization
our
use
of
open
source
digging
around
that
a
little
bit
more.
I
actually
found
that
there
were
multiple
departments,
you
know.
C
So
you
start
to
see
these
patterns
of
reuse
that
are
opportunities
right
to
reuse
code.
More
often
across
your
product
portfolio.
That's
going
to
help
your
engineering
efficiency
go
up
if
you're
not
rewriting
things
over
and
over
reinventing
the
wheel,
so
to
speak,
and
so
you
know
that
inner
source
opportunity
paired
together
with
the
gosh.
C
I
think
this
is
a
big
opportunity
for
us
to
digitally,
disrupt
ourselves
and
transform
a
little
bit
into
a
newer
company
and
and
be
a
more
efficient
one
in
terms
of
engineering,
and
I
think
this
is
really
going
to
propel
us
forward.
I
put
some
numbers
together
and
talked
about
what
open
source
programs
look
like
out
in
the
wild,
what
other
people
had
already
put
together
and
it
was
pretty
wild
the
response-
I
I
really
didn't-
expect
it,
but
they
they
were
like
okay
go
for
it.
C
I
just
joined
this
new
team
and
then
yeah
now
I
gotta
start
planning
this
other
direction
so
yeah
it
was
a.
It
was
a
lot
of
fun
to
establish
that
and
to
to
help
out
with
that,
so
yeah
that
that's
kind
of
the
story
of
how
I
ended
up
getting
into
this,
and
then
you
know
now,
I'm
in
github
and
and
consulting
with
tons
of
different
open
source
programs
offices
to
to
help
them
get
up
and
running
or
even
just
patch,
certain
areas
they're
struggling
with.
B
That's
awesome
so
for
anyone
listening,
that's
interested
in
this!
You
you've,
you
know,
fought
the
good
fight
right,
brought
it
to
their
attention
kind
of
made
open
their
eyes
to
it
and
realized
it.
But
how
does
how
would
an
organization
actually
go
about
starting
an
ospo
inside
their
company
like
what?
What
are
those
steps
they
need
to
take?
So
somebody
said
yeah.
This
is
what
we
need.
Where
do
you
go
from
there.
C
Yeah,
that's
a
really
good
question.
So,
if
you're
looking
to
start
in
ospo
it's
it
can
be
sort
of
a
complex
topic
because,
just
like
you
can
imagine
that
an
engineering
organization
at
company
a
and
company
b,
they
don't
do
the
same
thing
right.
They
they
do
what
the
company
needs
them
to
do,
and
so
not
to
give
a
non-answer
here.
But
it's
a
little
bit
similar
with
the
open
source
programs
office
right.
If
you
have
more
of
a
compliance,
heavy
industry.
C
Well,
okay,
getting
started
on
your
policy
working
with
your
legal
team
and
really
outlining
what
legal
risk
is
acceptable
and
what
needs
to
be
mitigated
and
and
developer.
Training
for
how
to
use
open
source
might
be
your
your
steps
and
your
focus
and
then,
if
you
let's
say
have
again
another,
maybe
very
large
organization
and
developer
productivity
is
really
unwieldy,
like
you're,
just
not
sure
what
people
are
working
on
and
things
like
that,
then
focusing
on
producing
open
source
and
writing
policies
for
that.
C
Encouraging
folks
to
do
that,
and
maybe
even
applying
a
little
bit
of
pressure
for
folks
to
open
source
can
really
do
some
amazing
things.
Like
clean
up
your
code,
all
of
a
sudden
there's
wonderful
documentation,
because,
oh
my
gosh,
this
is
going
to
be
open
source
or
you
know.
Some
of
the
folks
were
sorry
for
the
analogy,
but
they
know
where
the
the
the
bodies
are.
C
They
know
where
the
skeletons
are
hiding
in
the
code
yeah,
they
those
all
of
a
sudden,
get
cleaned
up,
even
though
maybe
they've
been
sitting
there
for
15
years,
just
at
the
at
the
mere
opportunity
of
open
sourcing
it
so
yeah.
I
would
say
overall,
though,
in
terms
of
like
actually
starting
one
with
it
being
so
different,
the
the
biggest
things
that
I
can
give
you
for
that
should
be
on
your
like
starting
plan.
C
Your
outline
is
clear
goals
right
and
a
clear
definition
of
what
you're
going
to
do
when
I
went
to
my
manager
and
and
and
up
the
chain
to
propose
an
open
source
programs
office.
I
was
very
succinct
in
you
know
just
two
three
sentences
about
what
I
was
going
to
do
and
the
impact
that
it
was
going
to
have.
C
B
C
Yep,
yep
and,
and
hopefully
you're
pitching
that
to
some
leader
that
is
going
to
in
return
kind
of
sponsor
you
and
say.
Yes,
this
is
aligned
with
our
company
goals,
just
aligned
with
our
current
priorities,
and
it
fits
in
our
timeline
here
and
then
they're
going
to
be
the
executive
champion
so
that
later
on
down
the
road,
if
you
weren't
to
have
gotten
that
approval,
you
sorted
to
this
grassroots
thing,
which
is
great,
but
but
you
don't
want
to
leave
it
grassroots
right,
because
then
they
might
look
and
go.
C
Okay,
I'm
looking
at
time
cards
here,
I'm
looking
at
jira,
and
I
see
that
everybody
seems
to
be
spending
about
10
of
their
time
on
these
open
source
collaboration
projects
that
just
aren't
tied
to
revenue
like
cut
that
out.
You
know
because
they
don't
understand
the
the
goals
and
the
benefits
of
doing
it.
They
they
might
see
it
pop
up
as
a
potential.
You
know
leaky
vessel
like
oh
we're,
wasting
time
here
and
so
yeah.
C
You
really
want
to
get
that
sponsorship
and
then
write
out
a
plan
about
how
you're
going
to
achieve
those
goals
right
if,
if
the
legal
risk
is
really
your
concern,
chart
a
course
from
where
you're
at
today
to
creating
a
policy.
That's
going
to
help
you
standardize
on
how
to
use
open
source
and
license
acceptability,
and
all
of
that.
B
So
it
is
a
good
amount
of
meaningful
work,
but
it's
a
good
amount
of
work
to
to
pull
this
together.
Setting
your
or
your
your
goals
that
leadership
support
is
definitely.
It
sounds
like
it's
definitely
a
lot
more
than
just
pat
on
the
back.
Yeah
zach
you've
got
this,
but
actually
something
that
they'll
back
you
and
support
you
and
ensure
that
people
understand
the
messaging
and
then
full
plan.
What
are
we
doing?
Why
are
we
doing
it?
How
are
we
gonna
establish
this
earlier?
B
You
had
mentioned
the
legal
issues,
the
security
issues
we're
talking
about
corporate
ip,
so
I
assume
that
this
group
is
actually
going
to
be
responsible
for
policy
like
actually
create
and
determine
and
decide
like.
This
is
the
process.
These
are
the
policies
we're
following:
how
do
they
know
what
policies
to
create
is
this
like
someone
in
legal
is
going
to
tell
them
like
this
is
what
we
can
and
can't
do,
or
how
does?
How
does
somebody
know
who's
stepping
setting
this
up
out
of
the
great
policy.
C
C
So
policy
should
have
legal
there,
but
it
shouldn't
be
only
illegal.
They
should
have
developers
there
because
they're
going
to
be
the
ones
that
have
to
live
with
the
you
know,
weight
of
developing
under
this
load
right,
but
then
also
business
leadership,
making
those
business
decisions
about
like
okay,
I
feel
comfortable
for
my
product
line,
taking
this
amount
of
risk
or
or
innovating
in
this
area,
but
I
want
to
avoid
that
or
you
know
that
they
need
to
be
in
the
room
too.
So
it
needs
to
be
a
very
diverse
group.
C
That's
coming
up
and
agreeing
or
approving
these
policies
and
then
there's
some
mechanics
about
putting
those
in
place
and
enforcement
as
well
in
terms
of
like
what
the
policies
should
actually
be
again
aligned
with
your
goals,
but
I
guess
one
thing
to
mention
is:
is
that
there
are
other
open
source
programs
offices
out
there.
So,
if
you
were
to,
you,
know,
try
and
do
this
in
your
organization-
you
don't
have
to
do
this
alone,
right
and
and
because
open
source
offices
are
sort
of
into
like
sharing
and
being
open
and
transparent,
which
is
great.
C
So
microsoft
has
a
very
similar
one
opensource.microsoft.com
program
and
they
they
say:
okay,
there's
using
contributing
and
releasing
so
there's
a
couple,
different
ways
and
different
examples
from
some
fairly
large
organizations,
that
sort
of
know
the
tech
thing
and
so
yeah.
I
guess
the
other
part
I'd
say
is
that
you
know
you
can
you
can
look
at
examples
of
what
exists
out
there
in
the
wild?
You
can
look
at
what
your
company
needs,
but
the
third
area
is
to
look
at
really
standards.
C
So
there's
a
thing
called
the
open
chain
project
and
that's
gonna.
That's
an
international
standard
for
open
source,
license
compliance
and
then
there's
also
clearly
defined
dot.
Io
and
that's
gonna,
be
helping
open
source
projects
actually
be
clear
about.
Well
what
license
do
you
use
on
this
project?
How
hard
is
that
to
discover
what
dependencies
does
your
project?
C
Have
you
know,
so
you
can
imagine
that
maybe
I'm
I'm
using
project
a
in
in
my
company,
which
is
open
source
created
by
somebody
else,
and
it
has
let's
say
an
mit
license:
okay,
but
then
that
project
has
a
dependency
in
open
source
that
it
uses.
These
are
called
like
transitive
dependencies
or
sub
dependencies,
and
and
maybe
that
dependency
has
a
different
license-
is
that
license
even
compatible
with
the
license
that
I'm
using
it
under?
C
These
are
all
questions,
for
you
know
your
legal
group
to
figure
out
and
your
your
automation
and
tools
to
really
highlight
for
you
so
and
then
the
last
one
is-
and
I
find
this
one
very
helpful-
it's
called
reuse
api
and
it's
like
if
you're
familiar
with
a
linter
linter
goes
through
the
code
and
you
know,
checks
for
common
errors
and
not
only
compilation
errors
but
style
errors
right,
well.
Reuse.
Api
is
like
a
linter
for
your
projects
to
be
able
to
check
the
reusability
and
a
lot.
C
C
Yeah
so
they're
experts
in
this
field.
I
am
not
one
of
them,
but
my
advice
is
mostly
to
seek
out
those
experts,
whether
it's
consulting
with
a
legal
firm,
that's
familiar
with
open
source,
getting
mentored
by
a
company
who
has
an
ospo
who
is
dealing
with
this,
especially
large.
You
know
ones
like
like
microsoft
or
google
or
somebody
else,
and
the
other
piece
would
be
to
from
a
developer
perspective
and
a
software
manager
perspective,
which
is
what
I
was
in
that
last
role.
C
B
B
Yes,
a
little
bit
now
go
figure,
take
advantage
of
the
open
source
for
that,
so
so
you're,
an
organization
you're
making
like
huge
strides
and
setting
up
your
osmo
right
you've
got
your
goals.
You've
got
your
sponsorship,
you've
got
a
plan
and
you
you've
got
resources
but
like
what
does
that
look
like
in
an
organization
like?
Is
it
on
the
org
chart?
Are
we?
B
C
C
C
Yeah
yeah,
so
I
I
think
going
back
to
kind
of
the
the
core
question
about:
where
does
it
fit
in
the
organization?
I
think
that
done
well,
you
could
put
it
anywhere,
but
done
well
from
what
I've
seen
just
out
in
industry
and
from
what
I've
tried
to
accomplish.
Myself
is
really
it
needs.
It
needs
to
be
on
your
chart.
It
needs
to
be
established.
It
needs
to
not
be
shadow
ops,
it
needs
to
be
in
the
light.
It
needs
to
be
recognized
as
aligned
to
strategy.
C
So
if
one
of
the
big
thing
or
the
big
thing
that
your
organization
is
focusing
on
is
customer
service,
you
know
that's
your
big
push
for
this
year
and
you,
you
know,
need
to
make
great
strides
in
customer
service.
All
the
metrics
are
around
that
your
open
source
programs,
office
proposal
and
your
goals
and
your
metrics
need
to
tie
and
be
aligned
to
customer
service,
which
is
totally
achievable
for
open
source
programs
offices.
C
There's
a
big
customer
impact,
especially
open
sourcing
example
code
on
how
to
use
your
product
or
service,
or
I
could
get
into
a
ton
of
things
there,
but
back
to
the
back
to
the
org
chart
part
like,
but
where,
on
the
order
chart,
does
it
sit
a
lot
of
times
you'll
see
that
open
source
programs
offices
are
a
program's
office
program
management
tends
to.
If
you
have
a
program,
management
org
tends
to
be
a
natural
fit
for
this.
C
So
if
you
have
a
very
well
integrated
I.t
that
has
positive
relationship,
not
adversarial
relationship
with
your
developers
and
your
it
group
is
well
networked
into
the
actual
business
decision
makers.
You
know
so
legal,
your
your
people
who
hold
revenue
as
a
part
of
their
performance
and
and
decision
makers
there
if
they're
well
networked
then
I've
seen
that
work
well.
I've
also
seen
it
sort
of
be
a
silo,
they
implement
things
and
it
can
be
very
heavy-handed
or
never
really
reach
the
developers
like.
C
C
It
leads
to
approvers
and
signatures
and
meetings
and
delays,
and
it
can
not
be
very
fun,
but
it
might
be
the
way
that
it
fits
into
your
organization.
The
way
that
it's
structured
right
now.
So
something
to
consider
the
alternative
that
I've
seen
be
pretty
successful
is
in
ospo
as
a
service,
and
you
know
I'm
not
just
saying
that,
because
infrastructures
as
a
service
and
code
as
a
service
and
everything
is
as
a
service
but
to
think
about
it
as
a
service
model
and
also
as
a
self-service
model.
C
How
do
I
get
a
github
identity
as
a
part
of
this
organization
and
start
contributing
to
my
my
organization's
open
source
projects
being
able
to
just
type
your
name
in?
Have
that
all
done
for
you
and
then
you
know
the
person
just
gets
a
username
and
boom.
They
can
go.
I've
seen
that
work
really
well,
it
helps
with
managing
the
size
of
the
ospo
being
able
to
keep
it
small
and
focused
on
leverage
points.
C
So,
as
you
lean
into
automation,
you're
leveraging
the
fact
that
okay,
we
decided
how
to
do
this
once
and
then
boom.
We
implemented
it
everywhere
and
scaled
it
versus
okay.
There's
some
legal
person
sitting
in
a
basement
somewhere
reviewing
each
and
every
license
on
each
and
every
project,
and
they
start
to
you
know,
lose
their
mind
a
little
bit
because
it's
a
very
repetitive
process
and
it
can
be
automated.
It
can
be
implemented
as
policy,
and
so
then
developers
can
really
self
service.
On
that.
B
That's
awesome
and
that's
your
as
you're
describing
this
I'm
totally
picturing
like
an
org
chart
and
seeing
this
little
box
move
around
in
all
the
places
that
it
can
possibly
go
and
then
eventually
like
in
my
head.
This
box
is
getting
smaller
and
smaller
because
of
the
number
of
people,
but
like
what
is
the?
What
is
the
mix
of
an
osmo
like?
Is
it
is
this
people
that
are
self-volunteering?
B
C
So
the
key
is
to
focus
on
having
a
mix
right
so
you're
shooting
for
diversity
and
for
all
the
benefits
that
that's
going
to
give
you
in
this
area.
Right.
You've
got
diversity
of
backgrounds
if
you've
got
developer,
minded
folks,
legal-minded,
folks,
business-minded
folks,
that's
how
you're
going
to
make
a
decision
that
is
appropriate
for
the
organization
and
considering
and
weighing
all
of
the
different
perspectives
and
factors
right.
C
So
what
do
those
rules
actually
look
like
well
program
management,
like
I
mentioned
already,
is
going
to
be
a
big
one:
who's
running
your
open
source
program
as
a
whole
and
managing
the
measurement
of
it
against
its
own
goals,
as
well
as
maybe
managing
being
a
functional
manager,
manager
of
people
that
are
in
that
org.
Then
there's
community
managers,
so
you
can
imagine
that
you
launch
an
open
source
project.
It's
you
know
maybe
has
some
great
engagement
from
the
community.
C
The
community
has
questions,
they
have
contributions,
they
have
things.
Somebody's
gonna
have
to
review
that
code.
Someone's
gonna
have
to
answer
those
questions
and
and
help
set
up
the
tone
of
the
community.
Too.
Do
people
just
get
to
come
in
on
a
github
issue
and
request
that
something
be
done
by
next
week
or
do?
Is
it
that
they
propose
things?
Is
it
you
know,
do
they
have
to
be?
The
idea
have
to
be
approved
before
they
start
writing
code.
All
of
these
workflows
and
just
community
management
pieces
around
an
open
source
project.
C
You
know
that
is
what
kind
of
a
community
manager
focuses
on,
and
so
it
can
be.
Also
marketing,
though
your
brand
is
out
there
when
you
start
up
an
open
source
thing
and
open
source
projects,
often
if
they're
large
enough
have
sub
brands
of
themself,
so
you
can
think
of
you
know.
Google
and
kubernetes
are
the
first
ones
that
kind
of
come
to
mind.
Well,
it's
not
called
google
net
google
netties.
C
C
C
B
C
Yeah
these
are,
these
are
normally
the
folks
where
this
is
100
of
their
job.
I'm
talking
about
you
know
listing
off
all
of
these
different
roles
as
if
they're
different
people,
sometimes
it's
the
same
person.
You
know
a
person
who
I
worked
with
a
ip
lawyer
who
also
went
to
through
a
cs
degree.
So
he
had
the
developer
perspective
and
the
legal
perspective.
That's
a
great
leverage
point
to
be
able
to
just
hire
one
person
instead
of
two
people
who
are
going
to
argue
a
lot.
C
Maybe
so
he
just
gets
to
argue
in
his
head
instead,
so
yeah,
there's
there's
that,
but
then
there's
also
the
scale
of
your
organization
that
you
need
to
consider
a
startup
would
not
hire
12
people
with
different
back.
You
know
they're
going
to
hire
somebody
who
can
wrap
their
arms
around
it
all,
but
maybe
not
be
an
expert
at
any
one
of
them.
C
So
when
I
started
up
in
ospo
and
proposed
that
I
said,
I
would
like
to
start
it
personally
one
person
and
I
would
like
to
grow
to
three
people
if
I
can
hit
these
targets
in
these
metrics.
I'd
like
to
you
know,
implement
a
policy
I
like
to
do
all
this
stuff
and
then
I'd
like
to
bring
on
two
developers
and
have
them
build
some
automation.
So
we
can
get
our
license.
Compliance
in
place
or
select
a
tool,
and
so
you
you
can
grow.
There's
growth
paths
you
definitely
don't
have
to
like.
C
I
can't
afford
12
people,
so
I
can't
ospo.
That's
that's!
Definitely
not
it.
It
can
be
as
small
as
one
person
I've
seen.
People
do
this
part
time,
but
just
you
know
word
of
caution,
probably
preaching
in
the
choir
here,
but
watch
out
for
burnout
right.
So
how
many
extra
jobs
roles
will
you
take
on
at
your
job
before
you
know,
you're,
on
a
path
to
just
drown
in
in
the
work
and
and
not
be
able
to
take
yourself
out
of
it.
B
And
you
become
that
bottleneck
right
that
everybody
else
is
relying
on
because
you're
doing
so
many
other
things,
but
that,
like
totally
leads
perfectly
into
my
my
next
question,
which
is
like
I've
got
these
dedicated
resources,
whether
it's
one,
whether
it's
12.
How
do
I
know
that
they're
doing
what
we
need
them
to
do
like?
How
do
we
know?
This
is
successful
that
I
should
be
reinvesting
and
getting
to
that
third
person
or
12th.
C
You
know
you're
you're,
measuring
something
and
you're,
seeing
if
you
obtained
it
and
it's
sort
of
your
your
goal
or
your
experiment
you're
going
to
try
it
out,
see
how
you
do
and
and
then
you
have
data
at
the
end
of
it
of
whether
it
worked
or
not.
So
that's
how
you
know
you've
had
success,
but
you.
C
Yes,
but
I
think
in
practice
that's
a
lot
harder
than
it
sounds,
especially
like
folks
get
a
lot
of
stage
fright
in
choosing
a
metric
like
what,
if
it's
the
wrong
metric,
what
if
I'm
you
know
measuring
my
goal,
is
here,
but
I'm
sort
of
measuring
this
tertiary
thing
to
try
and
see
or
gain
insight
into
if
I've
achieved
that
goal?
Oh
well,
if
I
don't
hit
it,
my
performance
is
tied
to
that
and
my
job,
my
salary
is
tied
to
that,
but
also
like
I
didn't.
C
C
C
They
have
a
slack
channel,
that's
open
to
folks
to
be
able
to
jump
into
and
ask
questions
jump
in
there
and
ask
other
folks
what
metrics
are
driving
your
open
source
programs
offices
and
see
what
you
get
or
tell
them
like.
I
might
have
a
couple
of
years
ago
and
say:
well,
I'm
trying
to
drive.
You
know
my
open
source
programs
office
off
of
the
metric
of
dollars
saved
by
using
open
source
instead
of
developing
internally.
C
If
I
had
actually
done
that,
I
might
have
got
some
great
advice
to
save
me
from
a
lot
of
pain
that
I
went
through
of
trying
to
explain.
When
I
tell
the
the
cto
like,
oh
well,
we
saved
we
saved
three
million
dollars
this
year
like
by
using
open
source
instead
of
developing
our
own
solution
and-
and
he
kind
of
goes
like
well
so
where's-
the
three
million
like
which
account
put
that
in
and
I
would
like
to
redirect
that
to
this
other
project.
I
want
to
fund
and
I
go
well.
C
I
don't
have
the
money
right
now,
but
you
know
it's
it's
funny
money
so
to
speak.
So
you
have
to
be
very
careful
about
the
metrics.
You
choose
there's
tons
of
folks
out
there
who
are
experts
in
this
and
especially
if
you're
open
source.
You
know
I
talked
about
the
to
do
group
and
that's
very
open
source
focus,
there's
also
innersourcecommons.org.
C
They
also
have
a
slack
channel
where
you
can
ask
people
who
are
actually
doing
this
in
their
organizations
and
are
experts
at
it
have
done
it.
Maybe
in
many
organizations
like
hey,
how
do
I,
what
what
metrics
are
good
for
measuring
inner
source?
Here's
my
goals
and
I'm
struggling
to
find
you
know
ways
to
do
that
so
reach
out
for
help.
There's
tons
of
places
to
do
that,
and
I
guess
alignment
is
probably
the
other
thing
to
talk
about.
C
C
So
my
metric,
for
that
is
an
output
from
maybe
a
tool
that
we've
built,
or
you
know,
maybe
we're
using
fosa
or
something
else
that
tracks
what
open
source
projects
we
have
and
what
licenses
we
have
and
whether
it
complies
against
our
policy.
Then
you
have
a
report
that
says:
okay
yeah.
We
we
shored
up
all
the
holes.
We
achieved
success
here,
we've
reduced
our
legal
risk
or
you
know.
Maybe
your
goal
is.
I
want
to
make
a
big
splash
in
open
source.
We
have
a
lot
of
things.
C
I
think
would
be
helpful
for
other
folks
to
use
outside
of
our
company,
and
so
I
want
to
open
source
10
projects
by
the
end
of
the
year.
Good
luck
to
you
quantity
is
quality.
There's
a
lot
of
reasons
I
would
say.
Maybe
this
isn't
a
great
goal,
but
just
as
an
example,
you
know
you
can
you
can
measure
that
you
can
say:
okay
well,
github
repos,
that
were
public
at
the
beginning
of
the
year
and
github
repos
that
were
public
at
the
end
of
the
year.
C
And
now,
if
you
want
to
improve
the
quality
of
that
metric,
you
might
say.
Okay,
did
I
build
a
community
around
those?
What
type
of
user
engagement
did?
I
have
on
a
per
project
basis,
so
your
ospo
might
have
its
set
of
like
parent
goals
and
metrics,
but
then
each
open
source
or
inner
source
project
might
have
its
own
as
well.
For
the
specific
note
that
it's
trying
to
hit,
if
that
makes
sense,.
B
Yeah
yeah,
so
I
I
mean
it's
it's
safe
to
say,
and
it's
probably
great
for
everybody
here
to
hear
this-
that
small
wins
are
wins
right.
So
if
you're
just
saying,
I
want
to
make
sure
that,
like
we're,
not
opening
any
legal
holes,
that's
going
to
get
us
in
a
ton
of
trouble.
Maybe
that's
not
tangible,
but
it's
it's
definitely
moving
you
in
the
right
direction.
So
I
would
guess
probably
don't
try
and
set
your
sights
up
here.
B
C
You
know
normally,
I
avoid
talking
about
this
type
of
thing,
but
since
you
asked
so
directly,
I
I
think
that
the
the
common
thing
that
I
see
that
I
find
personally
to
be,
I
would
even
say
extremely
unsuccessful-
is
the
theatricality
of
like
our
previous
example,
of
releasing
12
open
source
projects.
I
think,
when
folks
take
all
their
old
stuff
and
all
of
their
like
this
isn't
core
to
our
business
stuff,
and
they
don't
think
about
whether
it's
actually
useful
for
anyone
else
to
use.
C
They
don't
put
any
effort
into
building
a
community
around
it
or
you
know
anything
like
that
or
they
put
a
policy
in
place
and
the
policy
is
like
use,
good
common
sense
when
reading
legal
licenses
developer,
who
was
never
trained
to
do
this,
those
type
of
things
are
they're
checking
the
box
and
it's
it's
sort
of
like.
This
is
really
a
big
thing
in
innovation
spaces.
C
You're
you're,
you
know
you're
doing
the
play,
but
it's
not
reality
and
that's
where
I
think
ospos
are
unsuccessful.
Now
I
don't
see
that
a
lot.
Thankfully,
I
think
that
the
the
folks
that
are
engaging
in
open
source
and
inner
source
at
a
programs
level,
especially
for
these
large
organizations,
are
some
of
the
most
thoughtful
people
that
I've
just
honestly
ever
met
they're
they're
very
concerned
about
really
doing
this
and
and
they
have
the
right
heart
and
goals
in
mind.
You
know
everybody's
driven.
C
C
I
could
tell
you
that
you
know
if
some
of
the
largest
open
source
projects
that
we
know
of,
like
linux
or
other
things,
if
they
weren't
open
sourced,
would
they
have
the
adoption
that
they
have
by
their
customers.
Well,
no
so,
and
billion
dollar
businesses
have
been
built
around
servicing
and
supporting
those
folks.
So
yeah
there
is
a
great
business
opportunity,
and
so
my
point
is:
is
that
everyone
sort
of
comes
at
it
for
a
different
reason
and
a
different?
C
You
know
imperative
of
why
to
do
this
and
I
think
that
those
are
all
fine
as
long
as
you
accomplish
that
goal.
But
I
think
if
your
goal
is
to
look
like
other
tech
companies
who
are
open,
sourcing
things
and
that's
seen
as
like
gonna
up
your
recruitment
because
you're
you
talk
about
open
source
in
your
application
all
the
time,
but
none
of
it's
really
useful.
That's
it's
just
a
it's
an
act
and
it's
a
shame,
because
it's
a
big
missed
opportunity,
so
yeah.
B
It's
so
much
like
right,
any
endeavor
an
organization
is
going
to
change
the
way
they're
doing
something
start
small
focus
in
on
what's
realistic,
rather
than
trying
to
accomplish
the
world
in
a
small
period
of
time,
so
that
in
a
year
you
can
look
back
and
be
like
we
did.
We
did
these
little
things,
but
they're
going
to
move
us
so
much
further
forward
over
the
next
year,
rather
than
looking
back
and
saying.
Well,
that
was
a
disaster.
B
B
C
Yeah
yeah
so
to
do
group
which
has
case
studies,
the
slack
works
space
and
tons
of
folks
who
are
leading
and
participating
in
ospos,
it's
a
great
community.
You
know,
I
think
it's
hosted
by
the
linux
foundation,
there's
a
certain
way
to
become
a
member,
but
you
don't
have
to
do
that
just
to
jump
into
the
slack
and
learn
from
folks
and
so
yeah.
C
I
would
especially
dive
into
the
case
studies
there
if
you're
thinking
about
like
how
does
this
actually
play
out
at
companies
and
even
companies
that
you've
heard
of
right,
and
so
you
can
help
socialize
the
idea
of
an
oscar
at
your
organization
by
sharing
those
ospo
con,
which
is
a
conference
specifically
about
open
source
programs
offices.
That's
coming
up
in
september.
C
So
I'm
actually
looking
forward
to
that
coming
up
later
this
year
and
then,
if
you
are
a
government
or
a
university,
and
the
to-do
group
sounds
interesting
to
you,
but
you
have
a
little
bit
of
this
peace
in
your
mind
about
like
well,
but
it's
different
for
tech
companies,
it's
different
in
the
private
sector.
How
does
this
work
in
government
in
public
universities
and
things
ospo
plus
plus,
like
actual
plus
sign?
C
That
group
is
is
going
to
be
the
equivalent
of
the
to
do
group,
but
for
the
inner
source
side
of
things
and
often
when
you're
looking
at
how
to
you
know,
improve
your
engineering
efficiency.
Consuming
more
open
source
is
great
and
even
producing
open
source
is
great
because
you
get
contributions
back,
but
a
huge
key
for
a
ton
of
organizations
is
going
to
be
the
ability
to
reuse
and
share
code
inside
of
your
organization's
walls.
So
to
speak
so,
and
that's
really
sort
of
what
I
mean
by
inner
source.
B
Awesome-
and
I
can
imagine
a
lot
of
universities
and
government
and
tech
companies-
probably
do
that
quite
well
or
are
starting
to
get
on
that
bandwagon
as
well
any
see
we
are
we're
almost
up
on
time,
but
do
you
have
any
parting
words?
Anything
you
want
to
reiterate
on
message
to
hamper
down
with.
I
think
one
of
the
big
takeaways
I
got
from
this
discussion
with
you
is
that
there
is
a
lot
of
work
that
does
go
into
it.
B
It
is
a
real
organization
with
real
people
and
and
salaries,
but
it's
a
it's
such
a
positive
investment
for
any
company.
If
we're
we're
looking
at
the
way
of
the
world
and
98
of
software
is
built
on
open
source,
then
you
know
you
want
to
definitely
put
your
your
money
toward
where
it's
going
to
get
you
the
most
value
and
the
least
amount
of
trouble
out
there
in
the
world,
any
any
parting
thoughts
or
anything
from
you.
C
Yeah,
I
think
I'll
just
mention,
I
feel
like
the
one
area-
maybe
I
talked
around
and
that's
when
you're
starting
an
open
source
programs
office
for
a
lot
of
existing
organizations
for
the
a
lot
of
the
ones
that
I've
worked
with.
At
least
this
is
a
big
change.
You
know
when,
when
I
introduced
the
idea
that
we
were
that
we
were
starting
when
we'd
already
got
approval.
C
This
is
not
what
that
is
doing.
It's
actually
protecting
us
from
doing
that.
You
know
and
talking
about
all
of
the
details
thereof,
so
I
think
that
the
the
most
important
thing,
if
I
could
have
you,
take
away
one
thing
from
all
of
the
details
of
why
you
should
start
nospo
is
that
the
key
to
success
with
implementing
an
ospo
is
to
have
an
inspirational
leader
who
can
help
folks
in
your
organization,
bottom
to
top
side
to
side
everywhere
in
the
organization
understand
clearly
what
the
vision
is.
C
Today
we
don't
have
an
ospo
in
three
years
if
we
had
an
ospo
what
what
will
be
different?
Why
will
this
have
made
an
impact?
And
why
is
it
so
important
that
we
do
this
right
now,
and
I
think
that
when
you
have
a
leader
who's
able
to
take
that
baton
and
sort
of
go
around
and
light
the
candles
of
everybody
else?
C
That's
when
you
get
that
beautiful
picture
of
an
organization,
who's,
living
and
breathing
open
source
you've
got
the
diversity
of
perspective
of
all
of
those
different
folks
contributing
to
it,
and
also
one
thing
that
I'll
note
is
that
most
people,
don't
at
least
everyone
I
know-
doesn't
live
forever.
They
don't
stay
at
a
company
forever,
and
so,
when
that
inspirational
leader
removes
themselves
from
the
organization,
if
you
have
lit
all
of
those
candles
out
there,
then
you
have
a
much
better
chance
of
it.
C
Living
the
program
living
on
past
you
and
continuing
to
reap
benefits
and
you've
built
something
inside
a
company.
That's
going
to
deliver
fruit
for
that
company
way
past
your
direct
contribution,
your
time
in
that
company.
So
I
think
that's
a
really
magical
thing
and
and
that's
sort
of
what
I
love
planting
the
seed
in
different
organizations.
For
so
I
love.
B
That
I
love
that
analogy
or
just
that
picture
right,
because
it's
it's
so
much
the
momentum
and
building
it
to
make
it
lasting.
Instead
of
that,
just
like
one
guy,
you
know
running
a
marathon
with
a
torch
and
everybody
sees
it
but
they're
you
know
haven't
gotten
on
the
bandwagon
like
yeah.
Absolutely.
I
totally
agree
with
that.
It's
it's
got
to
be
someone.
That's
gonna
help
others
with
that
momentum
and
keep
it
going
for
the
long
term.
So,
thank
you
so
much.
B
I
think,
we're
about
a
time
to
wrap
it
up.
I
really
appreciate
you
taking
the
time
talking
to
all
of
us
sharing
your
experiences
and
your
genius
and
knowledge
about
all
of
this,
and
I
know
we've
been
talking,
we've
been
talking
at
github,
we've
been
talking
about
open
source
all
month
and
I'm
pretty
sure
you're
you're
involved
in
one
of
the
next
demo
day
sessions.
Can
you
give
us
a
little
sneak
peek
on
what's
coming
up
next.
C
Yeah
yeah,
so
first
thank
you
for
that
you're
far
too
kind
with
the
genius
remarks,
but
yeah.
I'm
definitely
gonna
tell
my
wife
about
that
later,
yeah
in
terms
of
the
sneak
peek
for
for
next
time,
yeah,
like
you
mentioned,
been
focusing
with
demo
days
on
open
source
all
month
and
next
month,
we'll
really
be
transitioning
into
inner
source.
So
these
are
both
two
very
important
components
of
an
ospo,
so
that
fits
into
what
we're
talking
about
today,
but
looking
into
our
journey
ahead.
C
If
you're
going
to
join
us
again
next
week
for
for
the
inner
source
session,
that's
first
we're
going
to
be
talking
about
sort
of
an
inner
source
initiative.
How
do
you
start
an
inner
source
initiative?
What
are
all
the
components
we'll
get?
This
very
high
level
view
of
all
the
different
pieces
involved
and
then
in
following
sessions
throughout
the
month,
we'll
be
diving
into
each
one
of
those
different
parts
to
try
and
give
you
some
tangible
takeaways
and
even
code
to
help
you
manage
an
inner
source
initiative.
C
So
I
am
so
excited
that
I
got
the
opportunity
to
speak
here
today.
This
was
super
cool.
It
was
great
to
go
back
and
forth
with
you
natalie,
but
I
am
equally
excited
for
the
sessions
coming
up.
I
think
they're
going
to
be
really
great,
so
hopefully
we'll
tune
in
for
those
and
ask
questions.
We
love
the
questions.