►
From YouTube: OpenShift Commons Briefing: New Marketplace and Catalog UX for OpenShift 4 Serena Nichols, Red Hat
Description
The User Experience Design Team gave a preview of the marketplace and catalog experience coming in OpenShift 4.0. If you’re looking to get a preview of what’s to come or learn how you can team with the OpenShift UX team to help shape the new experience, please watch this! We are looking to identify customers interested in working with our team to validate use cases around multi cluster management, operator subscriptions and the service catalog. Interested? Fill out this survey!
Speaker: Serena Chechile Nichols (Red Hat)
A
Some
of
the
ideas
and
concepts
here
so
I'm
gonna,
let
Serena
introduce
the
topic,
but
there
will
be
a
survey
that
we'll
share
in
the
chat
that
we're
going
to
ask
everybody
to
fill
out
and
give
us
your
feedback
on
and
we'll
be
sharing
that
with
other
folks
as
well
afterwards.
So
please
be
sure
and
fill
out
the
survey
and
and
help
us
make
the
marketplace
and
catalog
even
better,
though
serena
with
that.
Take
it
away.
Okay,.
B
Great
thanks,
I
just
want
to
quickly
go
over
the
agenda
so,
but
we're
gonna
start
with
is
Rob
is
going
to
discuss
that
a
high-level
the
operator
framework
and
how
that
integrates
into
the
marketplace.
Experience
I'm
not
gonna,
go
over
how
that
Marketplace
experience
affects
the
catalog
experience
and
how
the
catalog
is
now
different
in
upcoming
releases
than
it
was
in
the
three
dot
X
experience
and
then
really
what
our
main
focus
is
like
Diane
mentioned.
B
C
Right,
thank
you.
Everybody,
so
I
want
to
give
a
brief
introduction
to
the
concept
of
an
operator.
This
is
really
at
the
core
of
everything
that
we're
building
around
the
marketplace
in
the
catalog
and
at
the
end
of
the
day,
an
operator
if
you
haven't
heard
this
term
before
is
packaging
your
application.
Actually,
it
can
be
automated,
like
a
cloud
service,
something
that
you
would
expect
to
deploy
or
consume
from
Amazon
or
Google,
or
a
juror
or
any
of
these
cloud
providers.
C
If
you
think
about
something
that
when
you
get
like
a
database
or
a
queue
or
a
machine-learning
service
from
one
of
these
cloud
providers,
there's
a
bunch
of
software
running
behind
the
scenes
there,
that
is
doing
a
ton
of
stuff
and
provisions
all
the
infrastructure.
You
know
these
are
things-
are
all
behind
the
scenes.
You
don't
ever
get
to
see.
They're
scaling
things
out,
they're,
doing
failure,
detection
and
recovery.
They
might
be
doing
backup
all
that
type
of
stuff
is
using
a
number
of
primitives
that
are
actually
available
in
kubernetes.
C
You've
got
your
deployments
and
your
load
bouncers
secrets
in
config
maps.
You
can
stitch
together.
Some
really
smart
services,
and
this
is
what
we
consider
an
operator
now.
The
great
thing
about
that
is
you
give
this
cloud
like
automation,
but
because
you're,
just
using
kubernetes
primitives
at
the
end
of
the
day,
you're
not
tied
to
any
single
infrastructure
provider,
you
can
run
anywhere
that
you
can
install
an
open
ship
cluster.
For
example,
you
can
have
one
of
these
databases,
these
queues,
learning
anything
that
you
can
think
under
the
Sun.
C
C
We
have
an
operator
framework
and
it
helps
you
build
this
type
of
software.
You
know
it's
a
new
type
of
software
and
we
want
to
make
it
really
easy
to
get
started.
So
we
have
an
operator
SDK.
This
is
on
the
build
side
of
things,
so
we
hook
up
all
the
magic
that
you
need
to
talk
to
kubernetes.
There's
all
these
thing.
You
need
to
like
watch
the
API
and
set
up
a
bunch
of
scaffolding.
So
the
idea
with
that
is
that
you
can
come
with
the
business
logic
of
your
application.
C
If
you're
a
software
vendor
like
MongoDB,
you
know
exactly
how
to
you
know,
deploy
and
manage
MongoDB,
and
so
you
can
bring
that
business
logic
to
our
SDK
and
quickly
get
started
with
building
an
offering.
Now,
once
you
have
the
operator
built,
you
know,
you're,
not
just
gonna
run.
You
know
one
or
two
tops
at
this.
Ideally
you're
gonna
be
running.
You
know
hundreds
of
these
things
across
a
cluster
as
they're
managing
different
dependencies
for
different
teams,
and
that's
where
our
lifecycle
manager
comes
in.
C
This
is
a
tool
to
install
and
manage
the
lifecycle
of
those
operators
that
are
installed
in
the
cluster,
as
well
as
it
update
them,
and
this
is
key
to
getting
new
bug,
fixes
and
new
features
down
from
the
you
know,
the
folks
that
really
understand
these
operators,
the
MongoDB
ease
the
Redis,
is
and
then,
if
you're,
building
these
internally
as
a
team
inside
the
organization
and
being
able
to
update
those
is
how
they
get
better
over
time.
And
that's
really
key
to
that
cloud.
C
Like
experience,
a
service
that
is
hosted
for
you
on
a
cloud
you
know,
is
constantly
being
improved
and
operators
at
the
same
way
and
then
last
there
is
operator
metering,
and
this
is
a
Prometheus
based
system
that
allows
for
usage
reporting
of
operators
in
their
operations
clustered.
So
if
you
had
a
node
failure
and
a
number
of
operators
needed
to
you
know
recover
from
that
and
rebalance
data
or
do
some
other
operation,
they
can
start
reporting
those.
C
You
have
a
little
bit
more
insight
of
what's
going
on,
and
this
can
help
you
as
the
admin
of
a
cluster.
It
can
also
help
you,
as
a
software
vendor,
that's
building
an
operator
seeing
how
it's
used
at
scale
and
then
seeing
you
know
possibly
how
any
events
kind
of
cascade
through
the
system,
and
so
all
three
of
these
make
up
the
operator
framework.
We
have
a
github
org,
that's
linked
here.
C
If
you
want
to
check
out
that
and
some
other
other
tools
that
are
there,
and
so
these
are
gonna
form
the
basis
of
what
we're
going
to
talk
about
for
the
rest
of
the
session
next
and
so
the
magical
piece
comes
in,
which
is
the
application
marketplace,
and
this
is
a
place
to
connect
operators
together
out
to
your
OpenShift
clusters.
So
you
know
pretty
simple
steps.
C
You
know
folks
build
operators,
the
SDK
when
they're,
ready
and
stable,
they
will
get
published
to
the
marketplace
and
then
the
marketplace
can
push
down
both
the
install
of
those
operators
onto
a
number
of
your
clusters,
running
the
lifecycle
manager,
as
well
as
push
down
a
stream
of
updates
over
time
to
those
operators,
so
that
you're
getting
those
latest
bits
from
the
the
software
vendor
that
you've
chosen
and
then
when
you're
gonna
use
these
on
the
cluster.
So
a
question
about
how
does
this
interact
with
the
catalog
we're
going
to
go
over
that?
C
So
there's
two
different
personas
that
we're
gonna
talk
about
today,
there's
the
cluster
admins
and
then
the
developers
and
we
think
of
them
as
separate
there's
also,
you
know
some
use
cases
where
those
are
blended,
depending
on
the
team
that
you
have,
and
we
want
to
make
sure
that
you
know
the
installation
of
operators
and
updating
them
is
different
than
folks
that,
just
you
know
one
a
database.
They
want
a
queue
they
want.
You
know
whatever
the
software
is
that's
being
provided
by
that
operator.
They
want
a
monitoring,
stacks,
etc.
C
So
new
in
openshift
will
be
an
operator
section.
You
can
see
this
in
the
nav
on
the
left,
and
this
is
where
the
cluster
admin
will
go
to
kind
of
do
everything
we
just
talked
about
manage
the
lifecycle
of
operators,
see
what
versions
are
installed,
see
what
they're
actually
managing
for
the
teams
inside
of
the
cluster
and
then
do
some
access
control
of
that
total.
A
C
So
here's
what
that
marketplace
looks
like,
and
some
of
the
other
nav
items
in
that
operator
section
really
quick
from
top
to
bottom,
just
and
then
a
band
plane
would
soon.
These
are
at
a
high
level.
So
you
know
the
marketplace.
This
is
kind
of
your
shopping
experience.
This
is
where
you're
discovering
the
operators
that
might
be
available
to
you
right
underneath
that
is
the
cluster
services.
These
are.
The
operators
are
actually
installed
on
your
cluster
and
available
for
teams
to
use,
and
then
you've
got
some
items
that
are
a
little
bit
under
the
hood.
C
These
are
there's
a
concept
of
a
catalog,
so
you
can
hook
up
to
you,
know
our
marketplace
or
if
you
wanted
to
make
your
own
grouping
operators,
you
can
do
that.
Then
the
subscriptions
are
what
powered
the
over-the-air
updates,
like
I,
talked
about
getting
new
versions
of
these
operators
updated
and
installed
on
the
cluster,
and
some
other
objects
that
the
immunes
doll
plan
is
the
actual
kind
of
audit
log
of
what
versions
are
installed
and
what's
being
updated,
currently
about
typing.
C
So
looking
at
the
actual
marketplace
itself
looks
pretty
familiar.
You
know
it's
a
very
modern
web-based
marketplace,
so
you've
got
a
number
of
filters
on
the
left-hand
side.
So
if
you
have
an
engineer
that
needs
some
distributed
storage
for
an
application,
you
can
go
see
what's
available
at
the
storage
section,
you
know
find
more
information
about
the
features
of
that
operator,
who's,
publishing
and
some
of
the
versions
that
are
out
there
that
type
of
thing,
and
so
let's
do
new.
C
For
example,
let's
just
dig
into
one
of
these,
so
if
I
click
on
one
of
these
items
on
the
next
slide,
you'll
see
what
our
details
screen
looks
like,
and
this
is
a
little
bunch
of
information
about
the
Prometheus
operator,
some
information
about
it
on
the
right
and
of
what
does
it
provide?
Why
would
you
want
to
use
this
who's
producing
it?
I've
been
a
bunch
of
interesting
metadata
on
the
left.
C
The
number
of
the
versions
that
are
available-
it's
been
certified
by
Red
Hat,
which
is
an
important
thing,
also
happens
to
be
produced
by
Red
Hat
and
it's
got
a
container
health
index.
So
this
is
the
grade
of
what
containers
it's
using
and
whether
it
gets
updated.
You
know
known
CBE's
and
that
type
of
thing
and
then
some
information
about
the
code
and
its
version.
Now,
if
you
look
through
all
this
information,
you
decide
that
yeah
it's
certified
by
right
now,
it
looks
like
I
trusted
I'm
ready
to
subscribe
this
installed
on
my
cluster.
C
You
would
click
that
button
and
then
quickly
choose
the
update
channel.
This
is
a
little
bit
of
an
interaction
since
that
you
don't
want
to
install
a
specific
version.
You
actually
want
to
install
a
stream
of
updates
over
time,
and
so
this
channel
is
kind
of
the
version
that
you
like
very
similar,
like
how
the
Chrome
browser
is
updated.
That's
all
the
same,
and
then
the
strategy
for
updating
it.
C
C
Would
you
want
your
end-users
to
be
able
to
kind
of
request
an
operator
view
all
the
operators
that
are
out
there
in
the
world
and
say
hey
that
that
my
sequel
operator
looks
really
great
I,
don't
want
to
be
managing
my
sequel
or
you
know
at
least
at
a
high
level.
I,
don't
I
want
to
be
doing
it,
but
I
don't
want
to.
You
know,
configure
the
read
replicas
and
the
backups
and
all
that,
and
so
that's
one
thing
and
then
second
is
inside
of
your
different
teams.
C
You
know
obviously
there's
always
different
needs
and
skill
sets
size
of
teams,
the
goals
that
they
have
and
so
we're
curious.
If
you
want
to
limit
the
use
of
operators
to
specific
teams,
you
might
want
to
let
one
team
run
MongoDB,
but
not
the
other,
for
example,
I
suppose
in
chat.
Let
me
catch
up
on
that
really
quick.
They.
C
So
the
first
question
is:
why
is
this
humanized
marketplace
if
it's
just
for
operators,
so
operators
are
kubernetes
natives,
so
they're
they're
taking
code
and
capabilities
that
kubernetes
can
do
just
like
you
know,
setting
up
service
discovery
and
all
these
things
that
you
would
have
to
reinvent
yourself,
its
harnessing
those
to
run
these
really
sophisticated
application
stacks.
So
it's
you
know
it's.
This
is
not
an
open
shift
marketplace
that
appears
in
open
shift,
but
all
these
operators
work
against
any
kubernetes
version,
or
you
know
arranger
versions,
depending
on
what
API
is
they
use.
C
And
then
it
looks
like
there
was
another
more
of
a
comment
about
this.
All
kinds
of
depends
on
how
licensed
products
are
going
to
work
in
this,
and
that
is
a
great
point.
One
of
the
benefits
of
our
metering
framework
is
that
it
will
unlock
a
number
of
use
cases
for
if
you
wanted
to
do
like
on-demand
billing-
and
you
know,
usage
based
stuff.
C
B
Great
thanks
very
much
so
the
catalog
experience
is
more
focused
on
on
the
developer
for
the
most
part.
So
our
catalog
at
this
point
where
we
have
it,
is
underneath
the
home
navigation
section
and
the
the
use
cases
that
the
developer
visits
the
catalog
to
discover
and
provision
services
for
use
with
their
applications
right.
B
So
in
OpenShift
3x
we
had
a
number
of
things
that
were
available
in
the
catalog
and
what
we're
doing
in
Florida
in
four
is
just
that
the
catalog
is
going
to
be
enhanced
to
also
include
operator
back
services,
so
the
things
that
were
the
the
content
that
was
available
before
things
that
were
backed
by
the
service
broker,
that
stuff
is
still
going
to
be
in
there.
It's
just
that.
We're
gonna
have
it's
going
to
be
enhanced
with
things
that
our
operator
backed.
B
B
So
you'd
then
select
one
of
them
and
then
you
would
say:
okay,
I
want
set
that
to
be
added
to
my
project
and
here's
all
the
details
associated
with
it.
So
the
use
cases
or
the
flow
is
very
similar
in
what
we
had
in
OpenShift
3,
where
you're
still
going
through
a
guided
flow
except
we're
not
doing
a
wizard-based
flow.
We're
still
stepping
you
through
that
guided
flow.
B
Allow
you
to
create
something
and
then
once
it's
created,
we
have
additional
questions
actually
on
the
experience
you
know,
for
instance,
once
you
once
the
services
provision,
where
what
would
you
expect
to
see
and
where
would
you
expect
to
go?
Would
you
expect
to
still
be
sitting
on
top
of
the
catalogue?
Would
you
expect
to
be
pushed
out
to
the
project
overview,
for
instance,
and
then
there's
other
questions
as
far
as
discoverability?
Where
would
you
expect
that
quote-unquote
catalog
to
live?
B
Does
it
make
sense
to
live
in
the
home
navigation
section
and
even
does
browsing
versus
via
categories
work
as
expected
or
filtering
work
as
expected,
because
the
experience
is
different
than
what
we
had
in
3
and
we're
assuming
that
this
is
an
enhanced
experience,
but
we
do
want
to
make
sure
that
we
can
do
some
testing
and
invalidation
before
we
get
this.
You
know
too
far
through
developments,
so
I
think
I
will
open
it
up
to
more
questions
here
or
look
at
the
chat.
B
So
when
I
talked
to
my
customers,
I
spent
a
lot
of
time
on
labels
for
features
it's
hard
with
all
the
hype
from
the
industry.
In
general,
this
comes
across
to
me
as
don't
bother
deploying
to
kubernetes
openshift.
If
you
don't
have
an
operator
and
have
had
that
conversation
multiple
times,
I
think
Rob,
that's
probably
better
answered
by
you.
C
C
I
think
there's
like
operators
the
only
thing
going
in
the
future,
but
we're
excited
to
help
ease
some
of
the
operational
burden,
for
you
know,
building
and
running
these
like
kind
of
infrastructure
building
blocks,
so
maybe
not
necessarily
the
code
that
you
are
deploying
day-to-day,
like
your
JavaScript
app
but
running
databases
and
queues
and
smarter
services
for
machine
learning.
A
good
example
of
this
is
in
the
next
version
of
OpenShift.
There
is
going
to
be
a
node
tuning
operator.
This
is
something
run
by
the
platform.
That's
looking
at
the
workloads
that
you're
running
and
tuning.
C
You
know
the
different
parameters
and
the
kernel
or
on
the
host
to
make
those
better
kind
of
just
seamlessly,
for
you
so
I
think
that's
a
good
example
of
how
operators
can
really
do
a
lot
of
automation
for
you
behind
the
scenes
and
then
at
a
very
high
level.
You
know
you're
saying
yes,
I
need
to
back
it
up
at
this.
B
Yeah
I'd
also
go
back
to
this
comment
to
right,
where
the
catalogs
would
be
enhanced
to
also
include
operator
back
services,
so
we're
not
removing
what
we
had
before
I
think
one
of
the
other
things
that
just
to
note
is
that
we're
I'm,
showing
the
navigation
like
the
screenshots
that
we
do
have
of
them
of
the
navigation
here.
This
stuff
is
all
going
to
be
our
back.
B
So
whether
or
not
the
operator
section
is
going
to
be
available
for
developers,
I
think
that's
one
of
the
one
of
the
questions
we
actually
have
right
is:
do
we
want
to
expose
that
kind
of
terminology,
and
do
we
need
to
to
the
end
user?
Is
that
something
that
should
kind
of
be
more
hidden
to
the
developer
and
just
seen
from
the
administrators
perspective.
C
D
D
They
seem
like
two
very
different
concepts
to
me
and
when
I
go
talk
to
a
customer,
you
know
they
see
the
GUI
like.
Oh,
this
is
awesome.
I
want
to
start
on
boring
my
applications.
Okay,
great,
you
know,
let's
on
board
XYZ
application,
they
start
going
to
a
containerization
process.
You
know
it's
unusual
for
someone
to
say:
well
we're
not
going
to
get
any
value
of
this.
This
is
a
direct
quote.
I've
had
with
multiple
customers.
B
Just
gonna,
say
I
think,
that's
honestly,
I
do
think.
That's
why
we're
trying
to
get
more
customer
feedback
upfront
this
time
right
where
this
is
a
big
change
for
the
next
release.
Not
only
is
it
a
big
change
for
the
marketplace
and
and
the
catalog,
but
the
the
user
experience
itself
is
much
different
than
what
we
had
in
previous
3x
releases
right.
B
So
the
more
discussions
we
can
have
like
this,
whether
it's
online
in
this
meeting
or
offline
with
UX
and
p.m.
would
be
greatly
appreciated,
so
that
we
can
make
sure
that
we
understand
what
the
concerns
are
and
make
any
updates
as
needed
to
the
experience
before
we
get
before
we
get
the
release
out.
D
Happy
to
happy
to
do
that,
I
guess.
My
other
question
would
be
as
an
IFV
along
the
same
lines.
Yeah
I've
got
I've
got
containers
right
now
in
the
catalog.
You
know.
Am
I
included
in
this
marketplace
by
having
a
container
in
the
catalog,
you
know,
do
I
have
to
build
an
operator,
or
you
know
my
system
to
be
included
in
this
new
marketplace.
Eh.
What's
the
what's
the
path
of
getting
into
this
marketplace
as
an
ISP.
C
Yeah,
so
we're
you
know,
have
an
EAP
program
kicking
off
here
very
shortly,
so
the
way
that
we
progress
with
some
of
the
operator
partners
is,
we
had
a
small
group
of
ones
that
we're
testing
out
in
311,
that's
Revis,
MongoDB,
dynaTrace,
Couchbase
and
TD
and
Prometheus,
and
then
amq
streams,
which
is
a
concur
product
from
Red
Hat,
and
so
you
can
try.
It
goes
out
in
311
today,
and
so
in
is
be.
C
A
If
you're,
coming
to
kook
on
there's
a
face-to-face
kubernetes
operator
workshop
that
I'm
hosting
with
Rob
and
unbunch
of
other
folks,
it's
hands
on
the
last
day
of
coop
con
and
the
folks
who
are
doing
that
Tony
Campbell
and
that
Dorn
are
also
doing
a
whole
bunch
of
virtual
sessions
that
were
signing
people
up
to
and
we're
just
starting
rolling
all
this
out.
So
I
apologize.
If
you
didn't
get
the
get
any
outreach
yet
because
it
hasn't
been
sent.
A
But
what
we're
trying
to
do
is
make
sure
that
everyone
who's
already
in
the
programs
that
we
have
get
into
the
EAP
and
get
the
resources
they
need
to
make
operators
and
or
to
get
them
into
the
system.
Here.
The
way
it's
set
up
now,
if
that
makes
sense
and
Rob,
if
you
want
to
add
any
more
to
that
flavor.
C
Yeah
so,
like
Dianne,
said
basically
having
a
certified
container
or
set
of
containers
is
the
first
step
and
then
what
we're
going
to
do
is
use
that
same
process
to
certify
the
operators.
It's
kind
of
a
you
know,
plus
plus
certification,
and
then
you
know
your
operator
can
use
all
those
certified
bits
under
the
hood,
such
that
everything
you
know
is
being
scanned
and
vetted
by
Red
Hat
and
then
made
available
the
customers
kind
of
the
universe.
D
C
The
experience
that
we're
looking
to
build
here
is
that
system
updates
so
being
able
to.
You
know,
constantly
improve
these
applications,
and
so
an
operator
is
a
really
great
way
to
actually
have
additional
logic
in
there,
especially
when
you're
running,
like
a
stateful
service,
for
how
to
upgrade
that
and
how
to
actually
manage
it
versus
just
like.
You
know,
dumping
out
a
container
onto
the
cluster,
which
is
you
know,
totally
great
and
using
them
like
just
of
deployment.
But.
A
Arena,
if
you
can
put
that
survey
link
into
the
chat
now,
what
we're
gonna
ask
is
everybody
who,
who
has
any
feedback
and
mark
I,
will
definitely
reach
out
to
you
right
away
and
try
and
and
make
sure
that
we
get
your
feedback
in
and
also
to
do,
some
more
education.
There
is
a
lot
of
content
around
operators
around
working
with
helm
and
ensel,
and
building
operators
there's
also
an
operator's
sig
that
meets
the
third
Friday
of
every
month.
A
So
we
just
really
haven't
we've
just
starting
to
educate
that
the
universe
on
this
and
as
I
mentioned
I,
think
in
the
chat
earlier
at
KU
Khan,
there
will
be
a
hands-on
half-day
session.
That's
free!
That's
kubernetes
operator
framework
workshop
on
the
fried
the
end,
the
last
friday
of
the
seattle
one
and
we're
doing
some
virtual
ones
as
well,
so
there's
lots
of
opportunities
to
get
educated
on
operators,
but
I
think
what
this
is
raising
is
some
of
the
conceptual
bits
that
need
to
be
better
articulated
and
documented
and
rolled
out.
A
B
C
C
There
is
a
number
of
ways
that
operators
can
say
how
they
want
to
be
installed
and
the
main
one
that
we're
finding
is
for
it
to
be
installed
for
all
users
of
the
cluster,
so
you're
an
admin
and
you
install
it-
and
you
say:
hey-
make
this
available
to
everybody.
This
is
most
of
the
ISPs
out
there
and
then
there'll,
be
other
models
of
sign
up.
I
want
these
specific
projects
to
be
able
to
access
this
operator
as
well.
C
A
Thank
you
very
much
for
taking
the
time
to
do
this
and
I
will
socialize
this
and
hopefully
we'll
get
some
feedback
and
then
we'll
have
Sarina
and
Rob
back
with
when
we
get
closer
to
release
time
and
and
hopefully
get
our
documentation
up
to
snuff,
to
clarify
all
of
these
different
aspects
of
Service,
Catalog
operators
and
multiple
places,
but
it
is
it
is,
it
is
new
conceptually
and
that's
why
we're
kind
of
starting
to
try
and
validate
it
and
run
it
by
everybody,
meaning
you
want
to
add
any
final
words.
Yeah.
B
I,
just
thanks
very
much
I
think
there's
a
number
of
questions
that
were
brought
up,
which
are
great
and
just
as
you
notice
that
this
last
slide
it's
like.
We
have
a
lot
of
other
use
cases.
We
want
to
start
talking
about
as
well,
which
once
we've
got
the
current
things
to
address.
We've
got
a
lot
more
to
go
further
places
to
go
so,
as
we
mentioned,
feel
free
to
reach
out
with
that
survey
and
we'd
love
to
hear
your
feedback
thanks
for
joining
us
today
and.