►
From YouTube: Platform Engineering: Then and Now - Open Source Friday
Description
This Open Source Friday, @ramiyengar and I will be discussing Cloud Foundry's role in the evolution of Platform Engineering. Join us TODAY at 1pm ET on twitch.tv/github !
RSVP: https://gh.io/osf-cloudfoundry
A
Hello,
everyone
welcome
to
open
source
Friday.
My
name
is
Abby
I'm
filling
in
for
Roselle
this
week.
She
has
the
flu
and
strep
throat
so
wishing
Roselle
some
some
speedy
recovery,
but
glad
to
see
you
all
here
and
it's
great
to
see
people
yeah,
saying
hi
in
the
chat
so
hi
to
Alice
area.
So
this
is
also
my
first
time
running
a
stream
using
stream
yard.
A
So
if
I
do
something
really
funky,
please
be
gracious
but
good
to
see
you
all
yeah,
say
hi
in
the
chat
say
where
you're
from
oh,
we
have
Columbia.
That's
fun,
highlight
that
and
I'm
here
today
with
ram
Iron
Guard,
so
I'm
Rama
Rama
pass
it
to
you
to
introduce
yourself.
A
While
you
figure
that
out
yeah
I
still
don't
hear
anything
I
wonder
if
that's
my
end
nope
we
had
several
people
from
Colombia.
This
is
exciting,
so
hey
Carlos,
oh
Kenya,
that's
great,
hey,
Felix
great!
To
have
you
here.
I'm
calling
from
Canada
I
should
have
said
that
earlier
I'm
based
in
Toronto,
ROM
I
still
don't
hear
you
can
everyone
else
hear
Rob
hello
from
Turkey.
A
Nigeria,
we'll
give
you
a
moment:
I'll
just
have
fun
here,
with
the
chat
results
coming
in
hey
someone
from
La
welcome.
A
Someone
is
also
here
from
Canada,
yes,
I'm,
thanks
for
calling
in
from
Canada,
you
have
Kenya
Morocco
Zimbabwe.
This
is
truly
a
global
open
source
Friday
today,
Indiana
welcome
the
Netherlands.
A
A
Sweden
Peru
no
sound
from
right,
yeah!
Okay,
you
can't,
but
you
can
hear
me
I'm,
not
the
one
doing
stream
yard
wrong.
Yeah
Mexico
has
so
many
Dallas.
Oh
it
won.
Oh
it
Scrolls
too
fast
I
couldn't
highlight
it
there.
We
go
hey
Toronto
I'm
in
the
beaches
in
Toronto
every
dish.
If
you
could
say
where
in
Toronto
are
you,
we
should
have
like
a
Toronto
open
source,
Friday
sometime
I,
think
that
would
be
fun.
A
We
can
all
call
in
from
the
same
spot,
Maybe,
Romania,
Cuba,
India,
oh
I'm,
very
excited
that
so
many
people
are
saying
where
they're
coming
from
a
lot
of
Cuba
in
a
row,
oh
and
I,
think
ROM
is
going
to
rejoin
us
record,
Ivory,
Coast
and
again
I
apologize.
This
was
my
my
first
time
hosting
streamer
and
I
probably
should
have
made
sure
his
audio
Works
before
we
clipped
the
broadcast.
A
Oh
UAE
I
used
to
live
in
Saudi
when
I
was
a
kid.
So
that's
fine
yeah
try
to
thank
you,
Roselle
we're
almost
leaving
he
has
left
and
hopefully
he
will
return
soon
wait
from
the
hometown
of
Santa.
The
Santa
is
that
it's
that's
not
the
North
Pole.
Is
that
or
is
the
Santa
something
else
Buenos
Aires
nice?
Oh
Ghana,
welcome,
Manuel!
A
A
B
I
am
chief
evangelist
at
Cloud
Foundry
Cloud,
Foundry
Foundation
is
a
non-profit
tasked
with
governance
around
the
open
source,
Cloud
Foundry
project
and
a
whole
bunch
of
associated
open
source
projects,
so
psyched
to
be
here
on
open
source.
Friday
I
think
it's
definitely
a
great
stage
for
cloud
Foundry
and
Cloud
Foundry
related
Technologies,
but
also
big
thanks
for
having
me
here.
B
Yeah
so
a
little
bit
of
history
about
Cloud
Foundry
itself.
The
project
has
been
around
for
a
little
over
10
years
now,
which
is
like
a
million
years
in
the
software
world.
B
Cloud
Foundry
has
been
very
successful
in
the
VM
space
about
five
or
maybe
six
years
ago.
They
also
started
to
grow
into
containers
and
container
orchestration,
but
obviously
we
know
who
the
winner
in
those
races
were,
so
it's
seen
a
lot
less
adoption
in
terms
of
people
using
it
for
the
purposes
of
containers
and
container
orchestration.
But
more
recently,
we've
also
launched
a
project
which
is
technically
our
third
attempt
at
putting
the
cloud
Foundry
experience
over
kubernetes
and
I'm
going
to
Showcase
that
project
a
little
bit
today.
B
So
it's
about
creating
this
platform
experience
over
kubernetes
and
providing
the
same
kind
of
developer
experience
and
polish
and
self-service,
and
all
of
those
beautiful
things
that
cloud
Foundry
is
known
for.
A
B
I
think
the
the
key
is
in
what
you
just
mentioned.
Right
kubernetes
can
be
a
little
complex
and
daunting
for
any
team
to
start
making
use
of
now
in
terms
of
pure
marketing,
like
everybody
wants
to
be
using
kubernetes
right,
I
mean
the
the
Google.
Does
it
and
all
the
big
names
that
I
know
Uber?
Does
it
and
Airbnb
does
it
and
Spotify?
Does
it
and
wow
it?
You
know
we
should
all
be
doing
it.
So
you
know
pure
marketing.
B
Speak
aside,
if
you,
if
there
are
software
engineering
teams
out
there
who
are
using
containers
at
scale,
then
you
know
Cloud
Foundry
is
I
mean
sorry.
Kubernetes
is
a
bit
of
a
no-brainer
right.
You,
you
want
something:
that's
keeping
your
containers
afloat
and
making
sure
that
stuff
doesn't
go
down
and
just
the
nature
of
the
problem
itself.
That
kubernetes
is
solving
is
extremely
complex,
and
so
you
know
kubernetes
as
a
technology
has
to
be
complex
too
and
there's
there's
nobody
there's
nothing.
B
People
can
do
about
it
right
and
then
there's
also
the
notion
of
15
ways
to
solving
a
given
problem,
and
you
know
no
one
person
is
right.
No
one
team
is
right:
no
one
technology
can
be
a
silver
bullet
and
an
answer
to
everything,
and
so
the
cncf
landscape,
as
we've
seen,
has
exploded
in
terms
of
you
know
all
these
different
technology
pieces
that
some
of
which
solve
the
same
problem,
some
of
it's
all
different
problems.
And
what
have
you
right?
B
So
you
can
have
a
handful
of
projects
for
like
cicd
stuff
and
then
you
can
have
like
a
handful
of
different
projects
that
do
networking.
You
can
have
like
a
few
projects
that
do
Ingress
and
you
can
have
a
few
problem,
a
few
projects
that
do
load
balancing
well
and
so
there's
all
these
different
projects.
B
There's
you
know
a
community
that
has
like
a
decades
worth
of
experience
in
solving
this
golden
path
to
production
in
solving
it
at
scale,
keeping
it
secure,
keeping
it
active,
keeping
it
under
development
all
the
time
and
because
we
have
a
very
active
maintainer
Community
who
have
a
lot
of
Mind
share
in
this
developer
experience
over
complex
infrastructure
space.
B
You
know
it
was
very
natural
to
apply
all
of
these
solutions
that
we've
learned
over
the
years
to
kubernetes
today
and
if
there's
something
new
tomorrow,
then
there's
probably
going
to
be
a
cloud
Foundry
for
technology
X
in
future
too,
but
that's
really
what
it
is.
That's.
A
B
Correct
I
mean
it's:
it's
fully
open
source,
so
the
added
Advantage
is,
you
can
take
this
Cloud
Foundry
query
fee
as
it's
being
called,
you
can
take
kurifi.
You
can
swap
out,
let's
say
the
build
workflow
with
something
that
you
already
use.
You
can
swap
out
the
deploy
workflow
to
you
know
something
that
you
might
already
be
using
and
so
Cloud
one
of
the
downsides
of
cloud
Foundry
that
we
often
hear
is
it's
way
too
opinionated,
and
so
this
is
an
attempt
at
sort
of
you
know
solving
that
as
well.
B
So
you
have
a
very
modular
piece
of
technology
that
you
can
pick
and
choose
and
you
can
just
take
it
as
you
want
to
follow
the
opinion
of
the
cloud
Foundry,
Community
or,
like
I
said
you
know,
pick
and
choose
and
do
something
bespoke
with
it.
That
suits
your
organization.
So
that's
that's
really.
The
goal
here.
A
B
I'm
going
to
try
so
the
definition
of
cloud
native
is
about
taking
technology
that
we
have
today
that
works,
but
does
not
necessarily
work
in
a
simple
scalable
Manner
and
transform
that
into
an
experience.
That
is
that
that
works
on
the
cloud
that
is
scalable
on
demand
and.
B
It
comes
with
like
a
whole
raft
of
I
want
to
say
prerequisites,
but
it
comes
with
like
a
whole
raft
of
attributes
associated
with
it,
which
is
it's
open
source.
It
starts
with
source
code
every
time,
irrespective
of
what
is
the
tool
that
you're
trying
to
make
Cloud
native
and
it
starts
with
being
able
to
containerize
and
deploy
that
technology
as
a
workload
on
kubernetes.
So,
for
example,
if
you
were
doing
cic,
let's
say
CD
right,
continuous
delivery
of
your
applications
and
then
now
you
want
to
do
Cloud
native
CD.
B
You
basically
Avail
all
the
advantages
that
kubernetes
has
for
application
workloads
and
you
make
that
available
to
tooling
as
well.
You
make
that
available
to
security
as
well.
You
make
that
available
to
load
balancing
as
well
right.
So
you,
you,
basically
expand
this
notion
of
there
being
a
resilient
scalable
way
to
run
workloads,
and
you
apply
that
to
all
of
the
other,
tooling
and
other
things.
There
is
and
so
basically
get
everything
to
run
on
an
orchestrated
container,
runtime
yeah.
A
B
B
Let's
see
so
this
is
going
to
be
terminal
Windows,
just
the
way
your
developer
audience
likes,
I
guess
so,
let's
see
I'm
going
to
start
off
with
this
code
base,
it's
the
corifi
code
base
that
I
just
checked
out.
Let
me
quickly
showed
this
on
GitHub.
B
B
Right
so
the
bulk
of
you
know
the
action
at
Cloud
Foundry
happens
on
GitHub,
so
you
know
a
big
heartfelt
thanks
from
the
community
for
doing
what
you
folks
do
and
being
the
best
at
it.
So
this
is
the
home
for
kurifi.
It
is.
It
is
where
all
the
action
ends.
So
anybody
who
wants
to
get
started
I,
think
you
know
the
the
GitHub
page
is
like
a
great
place
to
start
and
the
readme
has
various
links
to
what
are
relevant
and
again
just
a
bit
of
History
lesson
here.
B
This
I
mentioned
that
this
was
like
the
third
attempt
at
getting
Cloud
Foundry
to
work
on
kubernetes,
so
like
the
first
attempt
at
doing
that
back
in
the
day
was
in
the
form
of
two
or
three
projects
known
as
quarks
and
fissile
and
Irene,
and
all
these
things.
B
So
if
you
really
go
back
into
the
history
books
about
Cloud,
Foundry
you'll
probably
run
into
these
names,
but
those
were
like
the
first
attempts
at
getting
Cloud
foundries
orchestrator
just
replaced
with
kubernetes
and
then
making
sure
that
all
of
the
tooling
around
just
operated
normally,
but
just
that
final
step
of
containerizing.
The
workload
and
placing
it
in
the
kubernetes
runtime
was
done
by
a
custom
component
back
then,
and
so
obviously
there
were
issues
with
that.
B
You
know
kubernetes,
as
a
technology
was
maturing
and
developing
its
own
set
of
you
know,
Tools
around
what
cloud
Foundry
was
already
doing,
which
is
placing
a
workload
and
then
building
the
artifact
itself,
and
then
you
know
putting
load
balances
around
it.
B
Configuring
Ingress
and
all
of
that,
so
it
was
becoming
very
difficult
to
support
Cloud
Foundry
components
for
the
kubernetes
core,
and
so
the
next
attempt
at
doing
this
was
replacing
a
lot
of
the
I
I,
don't
want
to
say
clunky
and
piss
off
a
lot
of
community
members,
but
a
lot
of
the
clunky
Cloud
Foundry
based
tooling,
with
more
Cloud
native
and
kubernetes
Native,
tooling
right,
and
so
we
had
the
second
iteration
of
this.
B
But
then
it
also
carried
forward
like
a
historic
problem,
which
is
you
can't
really
get
a
cloud
Foundry
going
where
you
know
you
come
back
from
work.
You
just
kick
off
a
project.
You
know
you
make
a
few
changes,
you
deploy
it.
You
test
it
and
stuff
like
that.
B
So
Cloud
Foundry
was
never
capable
of
doing
this
lightweight
hobby
programmer
sort
of
workflow,
and
this
was
important
because
adoption
for
open
source
technology
and
adoption
for
developer
first
tools
happens
locally
first
and
then
you
know
gradually
builds
to
other
installations
and
things
like
that,
and
so
it
was.
It
was
something
that
the
cloud
Foundry
Community
felt
that
they
were
missing
on
historically
and
even
with
this
new
technology.
B
So
keeping
these
things
in
mind
and
keeping
the
fact
that
cloud
Foundry
really
needed
to
be
just
an
API
shim,
as
opposed
to
something
very
complex
and
difficult
to
you
know,
manage
around
kubernetes
and
you
had
a
very
simple
lightweight:
kubernetes
API
and
then
all
you
needed
was
another
simple
lightweight,
Cloud
Foundry
API,
and
they
could
do
you
know
all
of
the
communication
through
these
apis,
and
so
that
is
the
vision
with
which
the
current
version
has
started,
and
so
some
of
that
is
documented
here
for
those
that
are
interested
in
reading
this.
B
It's
an
open
document
feel
free
to
go
through
that,
and
you
know,
we've
also
managed
to
document
some
of
the
architectural
decisions
we've
made.
So
a
lot
of
this
is
cids
that
are
running
for
different
for
accomplishing
like
different
workflows
and
then
there's
other
parts.
B
You
know
in
terms
of
what
are
what
happens
in
like
traditional
Cloud
Foundry
environments,
that
we
are
transposing
to
Cloud
Foundry
on
kubernetes
and
what
we
cannot
do
as
well.
So
those
are
all
documented
well
as
well,
but
yeah.
That's
that's
a
good
point
to
get
started
with
Cloud
Foundry
with
query
fee,
and
you
know
this
is
basically
a
clone
of
that
now
I'm
not
going
to
go
into
the
installation
itself.
B
But
what
I
have
here
is
a
kind
cluster
that
is
already
running
a
cloud
Foundry
locally
for
me.
Right
and
I
am
also
going
to
open
up
canines,
which
will
connect
to
like
the
local
cluster
and
can
show
us
a
few
things.
B
So
you
can
see
the
different
name,
spaces
that
are
created
here
and
the
different
tools
that
are
basically
a
part
of
Cloud
Foundry
on
kubernetes
or
the
query
fee
architecture
right.
So
we
make
use
of
contour
and
we
make
use
of
like
Bridle
packs,
which
are
a
couple
of
the
cncf
projects.
B
In
addition
to
like
a
whole
bunch
of
things
and
let's
see
I
I-
think
the
important
part
here
is
the
deployment
experience
itself.
Okay,
so
I
am
going
to
use
the
magic
CF
push
command,
which
is
basically
what
pushes
a
code
base
onto
the
remote
instance
and
in
this
case
the
remote
instance.
Let's
see,
I
can
do
CF
forget
right
and
it
will
tell
me
what
my
endpoint
is
and
what
my
organ
space
is.
So
orgs
and
spaces
are
how
corifi
or
Cloud
Foundry
does
multi-tenancy
on
any
remote
cluster.
B
So
we
have
a
version
of
multi-tenancy,
that's
capable
of
doing
accomplishing
true,
you
know
user
Separation
on
kubernetes
now,
obviously,
as
an
admin,
I
have
access
to
all
of
these
namespaces
and
clusters.
But
the
idea
is,
you
know
you
give
your
users
enough
access
just
so
that
they
know
what
you
know
they're
capable
of
running.
So
this
one
command
is
a
demonstration
of
all
of
the
apps
that
I
have
access
to
so
right.
B
B
I
need
to
say
that
this
is
the
path
so
I'm
missing
that,
and
so
this
idea
with
Cloud
Foundry
is
that
you
pass
along
like
a
single
command
and
Cloud
Foundry
is
able
to
deploy
that
for
you
now.
What
I'm
pushing
here
is
the
node.js
app,
but
what
I
also
want
folks
to
notice
is
there's
nothing
that
I
specified.
That's
not
JS
specific
right.
B
So
all
I'm
saying
is
you
know,
here's
the
code
go
and
apply
this,
and
this
was
like
the
big
marketing
thing
around
Cloud
Foundry,
where
there
was
a
haiku
about
Cloud
Foundry,
which
basically
said
here's
my
code
run
it
in
the
cloud
for
me:
I,
don't
care
how
and
the
the
that
single
command
experience
is
what
demonstrates
that
capability.
B
So
one
of
the
other
things
that
you
know
I
should
highlight
is
this
project
called
paqueto
build
packs,
so
build
packs
again
were
a
thing
that
started
in
started.
Along
with,
when
you
know,
Cloud
Foundry
Heroku
back
in
the
day
were
a
big
deal
and
I
guess
continue
to
be
a
big
deal
and
I.
Think
most
folks
will
agree
with
me
when
I
say
Heroku
is
like
the
gold
standard
of
developer
experience.
B
As
far
as
you
know,
deployment
goes,
it's
definitely
being
replaced
these
days,
there's
a
lot
more
that
we
can
use,
but
not
in
like
an
Apples
to
Apples
manner.
You
know
I
love
the
new
GitHub
actions.
Experience
I'll,
you
know,
love
the
experience
that
Helm
and
terraform
and
all
of
these
other
tools
are
bringing
into.
But
you
know
not
a
day
goes
by
when
I
don't
think.
B
Oh
wow,
this
was,
you
know,
already
a
thing
with
Heroku
back
in
the
day,
but
it's
basically
that
experience
where
all
you
need
to
do
is
specify
like
a
single
command
and
then
it
returns
with
like
a
URL
that
you
can
make
use
of
to
access
your
apps,
and
so
this
endpoint
here
is
basically
the
root
that
has
been
created
for
the
app
now
because
of
the
use
of
build
packs.
B
This
whole
experience
of
you
know
you
don't
have
to
necessarily
specify
the
language
and
the
tooling
and
the
framework
and
the
the
cloud
Foundry
controller
is
able
to
figure
out
that
there
is
a
node.js
app
thanks
to
you
know
some
of
the
hooks
that
are
there.
The
build
pack
has
a
life
cycle
stage
known
as
a
detect,
and
then
you
know
it's
built
to
select
the
appropriate
sort
of
build
pack
for
building
this,
and
so
it's
it's
able
to
extract
environment
variables.
B
It's
able
to
extract
the
right
kind
of
you
know,
install
commands
and
the
build
commands
and
also
the
start
command,
and
so
it
specifies
what
it
has
been
using.
There's
obviously,
ways
to
you
know
override
all
of
this
and
add
additional
stuff,
and
things
like
that.
But
the
general
idea
here
is,
you
know
even
saying
defaults
are
good
enough
for
the
the
workflow
to
succeed,
and
so
you
can
make
use
of
this
URL
excuse
like
the
security
thing,
because
we
don't
have
like
certificates
configured,
but
that's
basically
it
you
know.
B
So
you
you
just
specify,
what's
the
what's
the
app
that
you
want
to
run
from
the
path
and
it's
now
deployed
on,
you
know
that
kubernetes
cluster,
that
we
were
seeing.
B
Pretty
yeah-
and
you
can
you
can
do
this
for
like
a
bunch
of
other
languages
and
things
as
well,
so
I
have
let's
say
a
Go
app
here
and
I'm
just
going
to
make
use
of
this
sample.
So
you
can
see
that
it
has
like.
You
know,
a
very
straightforward
set
of
files.
It
does
nothing
fancy,
but
then,
if
I
say,
CF
push
go
sample.
B
It's
going
to,
you
know,
go
through
the
exact
same
process,
so
basically
packages.
You
know
the
source
pushes
it
to
the
cloud
controller.
The
cloud
controller
will
then
invoke
the
right
kind
of
build
pack,
so
it
can
detect
that
you
know
there's
a
go
dot,
mod,
that's
doing
the
dependency
management,
and
so
let
let
me
pick
that
particular
build
pack
and
then
create
like
the
workload
for
it,
and
you
know
it's
going
to
if
there
are.
B
If
you
do
repeated
sort
of
CF
pushes
for
like
the
same
stuff,
then
it's
going
to
do
a
lot
of
caching.
That
makes
builds
faster,
but
you
know
this
is
this
is
basically
the
exact
same
experience,
and
then
you
know
the
what
I
would
like
to
highlight
once
again
is
it
did
not?
They
really
specify
that
this
is
a
go
application
at
all,
so
it
was
just
CF
push.
You
know
app
name.
A
A
Different
cases,
it's
awesome
that
you
do
have
a
couple
questions
actually
before
that
there's
a
fun
comment.
Just
Cloud
native
is
like
having
a
magical
computer
that
can
do
anything.
I
thought
this
was
a
fun
description
of
clothing,
but
it's
not
just
one
computer.
It's
like
having
lots
of
computers
that
can
work
together
as
a
team.
So
thanks
ubar
for
that
nice.
Nice
comment.
I.
A
B
A
There
was
another
comment
here:
this
new
version
version
hopefully
runs
on
k3s.
Any
comment
on
that.
A
B
I
will
definitely
remember
to
test
it
and
post
on
like
cloudform
resources.
So
if
you
want
to
follow
us,
there,
I'll
I'll
definitely
make
sure
to
post
something
there
because
I
know
it's
tested
on
kind
clusters.
I've
done
mini
cube
in
the
past,
for
like
bigger
Cloud
providers,
we
regularly
test
Azure,
gcp
and
AWS
kubernetes
services
I'll
make
sure
to
give
it
a
spin
on
k3s
and
let
you
know
thanks
for
the
question.
A
B
B
B
Yeah
corifi
is
a
part
of
VMware
tanzu
and
it's
under
the
name.
Application
adapter.
B
B
B
Well,
interoperability
has
been
a
strong
point
of
the
cloud
Foundry
Fort.
We
have
demonstrated
openshift,
Integrations
and
ways
to
work
with
openshift
in
the
past,
and
I
should
imagine
that
it
is
possible,
but
I
can't
comment
on
the
current
state
how
it
stands
because
I
haven't
seen
anything
in
terms
of
testing
this
or
having
tried
it
out
myself,
but
I
should
imagine
it
would.
B
Thanks
Abby
I
mean
what
I
what
I
have
tested
this
on
in
the
past
is
using
GitHub
actions.
B
So
there's
like
an
older
version
of
this
that
Iran
and
then
you
know
this
has
been
tested
in
the
past
too.
So
one
of
the
things
that
something
like
this
does
very
well
is
the
interface
provides
a
simple
way
to
deploy
from
any
tool
of
your
choice.
So
we've
we've
done
GitHub
actions.
We've
done
Jenkins
like
just
the
regular
classic
Jenkins
experience.
B
You
can
recycle
all
of
these
existing
workflows
that
you
have
with
the
CF
for
kids
or
the
query
fee
interface
right,
so
Cloud
Foundry
on
kubernetes
is,
you
know
just
an
end
point,
and
so
all
you
need
to
do
is
basically
be
authenticated
and
to
that
so,
like,
like
I,
said,
you
basically
figure
out
this
end
point
and
then,
as
as
long
as
you're
logged
in
and
you
have
the
right
kind
of
spaces
and
other
things
configured,
you
can.
A
Nice
yeah
and
I'd
love
to
see
we're
gonna
demo,
like
the
GitHub
actions
with
this.
A
B
Yeah,
so
there's
there's
two
pieces
of
this
that
we
actually
configured
right,
so
I
was
very
excited
when
GitHub
container
Registries
came
out
and
that's
what
prompted
this
demo
more
than
anything,
so
I
made
use
of
the
GitHub
container
registry
to
store
all
of
the
images
that
are
exported
at
the
end
of
the
build
process.
And
basically
you
know
the
the
there
is
a
CF
action
task
that
you
can
make
use
of
and
it
basically
you
know,
deploys
to
ACF
endpoint
of
choice.
So
here
you
know
you
configure.
B
Like
the
end
point,
let
me
make
this
slightly
bigger.
B
Yeah-
and
you
know
you
make
use
of
this
job
where
you
create
the
end
point
like
I
mentioned,
provide
the
username
and
the
password
along
with
like
the
org
and
the
space
that
you
know,
the
the
GitHub
action
has
to
use
and
push
is
going
to
be
like
the
default
deployment
command.
So
you
can
configure
all
of
that
information
in
like
an
entry
point
script
and
that
basically
gets
called
every
time.
B
B
It
was
he
was
pretty
nifty,
and
so
the
idea
that
you
know
you
you
can
make
use
of
any
remote
endpoint
and
just
you
know,
deploy
to
that
simply
was
was
just
too
enticing
and
and,
like
I
said,
the
big.
The
big
part
was
the
container
Registries
that
also
came
with
GitHub.
So
I
was
very,
very
curious
to
make
use
of
that
yeah.
That's.
A
Great
yeah
that
seems
so
smooth
just
like
every
time
you
push
or
like
make
a
release,
then
it'll
just
deploy
yeah,
that's
awesome
cool!
So
it's
great
that
carifi
is
open
source.
Do
you
have
any
thoughts
on
how
we
can
make
open
source
more
sustainable
or
how
Cloud
Foundry
is
able
to
offer
coryphia's
open
source.
B
I
mean
open
source
has
been
the
cloud
Foundry
ethos
from
the
start,
I,
don't
think,
there's
a
point
in
time
that
I
can
go
back
to
when
I.
Don't
remember
eating
you
know,
driven
by
the
community
and
being
very
open
and
Democratic
in
that
way.
B
So
there's
definitely
you
know
challenges,
though,
being
a
multi-company
open
source
project,
you
sort
of
have
to
embrace
the
ambiguity
of
just
one,
but
so
many
organizations
around
you
and
I
think
I
think
that
is
like
the
biggest
challenge
and
there
is
in
terms
of
being
able
to
sustain
this
across.
B
You
know
changing
outlooks
and
you
know
evolving
purposes
and
and
different
end
goals
for
all
of
these
different
companies
right
so,
but
I
think
the
community
has
always
been
very
strong,
very
close-knit
and
you
know
coming
together
not
just
in
terms
of
technical
contributions,
but
also
like
documentation
and
Outreach
and
evangelism,
and
all
of
these
functions
has,
you
know,
been
a
typical
challenge
with
the
community,
but
you
know
we
we're
working
through
those
there's.
B
There's
I
think
a
few
great
initiatives
that
have
happened
over
the
years
that
has
made
things
simpler
and
more
sustainable.
As
as
far
as
open
source,
Community
communities
go,
which
is
I
mean
a
very
small
example
of
this-
is
marking
issues
as
good
first
issues
right.
So
when
you
invite
folks
to
the
community
and
then
they
go
through
your
issues
and
your
backlog
and
then
there's
like
a
good
first
issue
there.
B
It
you
know,
invites
people
to
pick
it
up
and
start
contributing
and
start
making
inroads
right,
and
this
whole
era
of
kovid
that
we've
had
where
you
know,
people
have
been
forced
to
collaborate,
remote
and
not
see
each
other's
faces,
but
make
great
engineering
decisions.
B
So
I
think
I
think
I
think
the
the
open
source
Community
has
been
doing
that
for
years
before
kovid,
and
so
the
this
this
whole
remote
were
forced
sense
of
remote
working
has
been
I,
think
a
an
unforeseen
and
unplanned,
but
you
know
rather
good
outcome
of
having
gone
through
the
pandemic.
You
know
right,
but
but
I
don't
want
to
boast
about
the
pandemic
too
much
for
obvious
reasons,
but
I
think
I
think
this
one
side
effect
has
been
something
good
for
the
tech
community
at
large.
B
Where
we're
now
able
to
you
know
think
in
terms
of
how
can
we
all
contribute
being
remote
and.
A
B
A
A
little
bit
more
Symphony
making
all
the
decisions.
This
way,
you
have
a
little
bit
more
accountability,
but
it
is
tough,
getting
everyone's
aligned
and
like
understanding
what
everyone's
goals
are.
So
Andrew
actually
had
a
question
here
related
to
that
amazing,
insightful,
multiple
org
challenges,
how
best
deal
with
multi-org
and
best
practices.
Any
thoughts
on
that.
B
I
think
that
single
question
itself
like
deserves
an
an
episode
of
its
own
and
there's.
You
know
a
lot
of
big
and
small
open
source
organizations
that
we
can
look
to
right
like
just
look
at
the
cncf.
B
Their
growth
has
been
tremendous,
and
just
the
sheer
number
of
projects
that
you
know
keep
knocking
at
their
doorsteps,
so
to
say,
is,
is
very
difficult
to
keep
up
with
and
so
and
but
they've
all
you
know
focused
on
providing
this
amazing
platform
and
so
I
think
if
communities
approach
this
from
not
just
a
technical
standpoint
like
educating
people
about
projects
is
just
as
important
as
you
know,
doing
technical
contributions
and
I
don't
know
making
a
code
base
10
more
performant
in
a
quarter
right,
so
making
sure
that
you
know
stuff
is
documented
in
road
maps
are
updated
and
I
think
you
know
I
I,
don't
want
to
keep
harking
back
to
all
of
the
GitHub
features
that
that
already
Provide
support
for
this.
B
But
you
know
it's
all
right:
there,
projects
and
road
maps
and
Milestones
and
tags
and
releases,
and
all
of
it
is
available
natively
in
GitHub,
Version
Control,
and
so
you
know
the
the
tools
are
there
in
order
for
companies
to
make
these
multi-org
open
source
projects
a
transparent
and
I
think
you
know
an
active
pursuit
of
this
transparency
and
keeping
in
mind
the
goals
of
not
just
building
projects,
but
also
I
mean
building
technical
competency,
but
also
making
sure
that
the
non-technical
parts
of
it
are
checked
off
like
make
a
YouTube.
B
Video
of
you
know
how
this
works
and
write
a
small
blog
post
now
and
then
about
you,
know
what
you're
working
on
and,
if
you'd
like,
to
see
more
contributions
in
a
particular
area
and
just
share
information
with
the
larger
Community
about
you
know
what
problems
you
solved
this
month
or
this
week
or
this
quarter
or
you
know
just
tiny
things
like
that
I
think
go
a
long
way
in
terms
of
showcasing
how
unified
you
are
in
terms
of
the
core
vision
of
the
project,
and
you
know
I
think.
B
Ultimately,
it
all
boils
down
to
identifying
the
right
people
within
these
orgs
to
you
know,
come
forward
and
take
leadership
roles
and
and
be
active
contributors,
and
you
know
maintain
that
sense
of
community
to
a
point
where,
even
if
they
switch
roles
or
move
companies
or
don't
necessarily
do
the
same
line
or
they're
not
in
the
same
line
of
work
anymore,
they
do
have
you
know.
A
Coming
up
soon,
I
have
another
question
yeah
and
he
plans
to
contribute
Griffey
to
the
cmcf.
B
At
the
moment,
it's
governed
by
the
cloud
Foundry
Foundation
much
in
the
same
way
as
any
cncf
project
would
be.
The
goal
definitely
is
to
be
part
of
that
cncf
landscape.
In
the
past
quadrant
there's
there's,
you
know
a
lot
of
emptiness
there
right
now.
B
So
the
cloud
Foundry
Foundation,
just
in
terms
of
background,
is
a
directed
fund
of
the
Linux
Foundation.
Much
like
the
cncf
is
also
a
part
of
the
Linux
foundation.
So
both
of
these
foundations
are,
you
know,
managed
by
the
same
parent
org
and
have
the
same
kind
of
Outlook
in
terms
of
how
we
manage
and
keep
the
project
alive
and
things
like
that.
B
But
if
you
know
the
powers
to
be
decide
that
kurifi
probably
has
a
better
home
there
and
you
know,
depending
on
the
decisions
made
by
our
governing
board
and
their
governing
board-
and
you
know
all
of
the
different
tocs
and
other
committees
that
we
have,
then
it
could
happen.
But
there
are
no
plans
at
the
moment
of
doing
anything
like
that.
Yeah.
B
Like
I
said,
our
GitHub
repo
is,
is
the
perfect
starting
point,
so
you
know
GitHub
Cloud
Foundry,
query
fee
will
get
you
there.
B
All
of
our
backlog
is
marked
with
like
good
first
issues,
where
people
can
jump
in
every
little
part
of
cloud
Foundry,
whether
it's
you
know
query
fee
related
or
otherwise
is-
is
governed
by
working
groups,
much
like
other
open
source
communities.
So
you
can
hop
on
to
like
calls
that
they
have
and
everything
is.
You
know
well
documented
in
terms
of
the
decisions
that
people
are
making
and
they're
all
linked
off
of
our
repos.
So
you
know
to
repost
the
place.
A
Nice
sweet
so
yeah
go
check
out
the
curfew,
repo
I'll
post-
that
in
the
chat
again
so
if
you're
in
I
think
twitch
or
YouTube
You'll
see
this
chat
but
we'll
yeah,
we'll
post
it
somewhere
for
everyone
to
find
one
cool,
any
other
questions
from
the
audience.
So
you
can.
If
this
is
your
last
call
for
questions,
audience
we're
wrapping
up
on
time
soon,
but
in
the
meantime
I
think
I'll
ask
a
fun
question:
if
that's
okay,
we'll.
B
B
I
I,
don't
think,
that's
a
fair
question
to
ask
somebody
who
lives
in
like
the
Indian
subcontinent
because
there's
like
so
much
to
choose
from
right,
but
if
I
really
had
to
make
a
choice.
There's
a
there's,
a
South
Indian
dish
called
idli,
which
is
basically
like
tiny
cakes
made
of
like
rice
and
lentils.
And
it's
like
a
very
comfortable
food
for.
A
A
A
Now,
what's
your
favorite
Beyonce
song.
B
A
That's
fine
cool
any!
Are
you
googling
your
fiance
song
right.
A
A
B
Feel
I
feel
like
a
geriatric
Millennial
today.
B
I
think
I
think
there's
one
with
like
Jay-Z
Crazy,
In
Love
or
something
oh.
B
B
So
one
of
the
old
adages
goes
when
in
a
gold
rush,
try
and
sell
Spades,
and
so
you
know
everybody
is
rushing
towards
kubernetes
right
now
and
there's
probably
like
you
know,
a
ton
of
folks
who
can
benefit
from
like
a
great
platform
experience
and
so
I'm,
happy
I'm
able
to
you,
know,
work
in
that
space
and
it's
been
rewarding
and
you
know
I
invite
anybody
interested
in
like
contributing
to
open
source
to
also
check
out
this.
A
B
B
Made
it
thanks
Abby
thanks
for
having
me
I,
think
umizelle
was
the
one
who
invited
me
to
this.
So
big
thanks
to
her
too,
but
yeah.
A
I
hope
you
get
well
soon
so
chat.
Everyone
in
the
chat,
please
wish
results.
Some
speedy
recovery,
I'm
sure
she'll
appreciate
it
well.
Well,
thanks
everyone,
I'm
gonna,
close
us
off,
but
see
you
next
time
and
Roselle
will
be
back
here
unless
she's
still
sick.
Then
results
should
be
back
next
week
thanks.
Everyone
thanks.