►
From YouTube: Kubernetes SIG Apps 20180723
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
Alright
folks
welcome
to
kubernetes
cig
apps
for
July
23rd,
2018
and
I
will
share
in
the
chat
window.
Today's
meeting
minutes
and
agenda
are
at
that
location,
so
the
one
quick
announcement
I'll
say
the
home
project,
which
has
moved,
which
was
under
cig
apps
and
moved
to
be
a
top-level
CN
CF
project.
The
repos
have
now
moved
under
the
helm
org,
so
anybody
who
is
looking
for
them
under
the
kubernetes
org
or
in
the
kubernetes
helm
org.
A
A
We
do
have
some
coming
for
future
sessions,
but
we
didn't
have
one
for
today,
and
so
with
that
we
have
looks
like
a
single
discussion
topic
and
then
open
discussion
afterwards,
and
so,
if
there's
something
y'all
would
like
to
talk
about,
please
feel
free
to
add
it
to
the
discussion
here
and
with
that
Josh
was
at
you
who
added
the
chart.
Repo
management
UI
chart
Museum,
which
has
to
do
with
home
conversation
yeah.
B
So
yeah,
basically
for
those
of
you
on
the
call
who
are
not
familiar
chart
museum
is
a
way
to
it's.
A
project
for
hosting
your
own
helm,
cart
repository.
So
it's
kind
of
become
part
of
the
helm
project.
That's
on
the
it's
in
the
helm,
github
org
and
a
number
of
people
have
been
asking
me
if
there's
a
UI
for
this-
and
this
is
not
to
be
confused
with
the
monocular
project,
which
is
what
Cube
Apps
is
based
on.
Her
cube.
B
Apps
hub
is
based
on
this
would
be
about
hosting
your
own
shard
repo
and
then
monocular
is
more
about
discovering
charts
across
multiple
across
multiple
sharp
repos.
So
I
put
a
link
in
the
in
the
doc
there's
there's
basically
a
list
of
requirements
I
put
on,
but
more
or
less
it's
the
ability
to
list
all
the
charts
in
your
repo
download
them
upload
them
and
delete
them
all
through
a
UI
and
then
there's
some
there's
some
other
desired
functionality.
But
the
chart
museum
project
is
not
mature
enough
to
handle
some
of
the
use
cases.
B
So,
if
you're
interested
in
kind
of
having
a
user
interface
for
managing
your
own
chart
repo
based
on
chart
museum,
please
comment
on
this
issue
and
as
well
of
if
you're
interested
in
developing
it
I'm
not
sure
that
I
myself
could
actually
put
a
lot
of
time
into
this
and
I'm
not
the
most
up-to-date
on
the
newest
and
latest
and
greatest
JavaScript
stuff.
So
any
help
is
appreciated.
That's
pretty
much
all
I
got
if
there's
any
questions.
A
A
A
C
C
Yeah,
so
we're
actually
talking
about
that
a
bit
at
Red
Hat
prior
to
it
merging.
There
is
a
concern
from
some
of
the
teams
regarding
having
the
icons
explicitly
referenced,
something
that
is
online
and
then
having
we'd
like
to
have
the
ability
to
support,
essentially
loading
those
icons
offline
for
clusters
that
are
air-gapped,
so
I'm,
currently
working
on
a
PR
to
extend
that
where
you
will
be
able
to
define
the
icon
in
line
either
as
binary
or
as
text.
C
If
you
have
SVG,
also
working
on
being
able
to
reference
a
config
map,
if
you
say
have
a
very
large
image
or
you
have
a
very
large
SVG
that
you
don't
want
to
be
there
in
the
app
CRD
manifest.
So
that
should
be
done,
and
hopefully,
today
or
sometime
this
week.
I'll
have
that
up
to
share
with
you
guys
perfect.
A
D
A
A
As
far
as
offline,
you
know,
because
lots
of
folks
are
gonna-
want
to
proxy
that
image.
You
know
always
think
of
the
way
that
github
does
stuff.
If
you
embed
a
link
to
an
image,
if
I
remember
right,
they've
got
a
domain
like
github
User
Content,
something
like
that,
or
they
will
actually
run
it
through,
and
then
they
serve
it
after
making
sure
that
it's
not
something
nefarious.
A
There
are
situations
where
you'd
want
to
do
something
like
that.
The
question-
and
we
can
hash
it
out.
There
is,
should
that
be
part
of
the
spec
or
an
implementation
detail
for
folks
behind
it,
who
maybe
do
pre-processing
on
the
spec
before
it
goes
into
a
cluster.
I,
don't
know
the
answer
to
it,
but
that
is
one
of
the
things
that
we
had
talked
about
before
is
where
does
this
live?
Something
like
this
when
you,
when
you
have
that
situation
like
at
github
or
I?
A
Imagine
Google
Cloud
if
they
were
gonna
use
this
or
any
of
the
other
clouds,
and
they
were
gonna
use
those
icons,
they're
gonna
store
it
locally,
cached
locally,
so
it
goes
faster,
not
reach
out
over
the
web
and
Trust
URL
doesn't
go
away.
You
know,
make
sure
it's
secure
and
somebody's,
not
embedding
something
nefarious
in
it
stuff.
You
know,
there's
there's
lots
of
reasons
to
do
it.
Where
is
that
right?
You
know
extension
point,
that's
something!
A
This
was
based
on
the
web
apps
pack
and
they
have
a
purpose
field
for
the
purpose,
for
an
image
rather
than
just
you
know,
size,
information
and
type.
They
have
a
purpose,
and
we
have
been
talking
about
bringing
that
in
because
there
might
be
a
difference
between
an
icon
versus
a
banner
versus
something
else,
and
we
might
look
at
introducing
that
idea
as
well,
just
so
as
you're
thinking
about
this.
That's
another
concept
sure
use
a
PR.
C
Yeah,
just
so,
you
guys
have
some
reference.
A
lot
of
this
came
out
of
items
and
discussion
points
with
the
operator
lifecycle
manager,
which
I
no
I'll
see
if
I
could
get
a
link
into
the
chat
here,
but
that's
something
that
a
lot
of
folks
are
RedHat
are
currently
working
on
as
well
to
essentially
bring
part
of
that
is
bringing
essentially
a
catalogue
of
operators
to
your
cluster,
which
are
then
madan
managed
by
the
operator
or
lifecycle
manager.
Okay,.
C
A
A
In
addition
to
regular
URLs
to
store
the
data,
if
you
think
there's
something
else
beyond
that,
that
would
be
useful,
I'd
love
to
know
the
reasoning
behind
it
whatever
it
might
be.
I
understand
that
that
you
know
this
does
end
up
usually
mean
base64
encoded,
which
is
a
larger
data
set
than
just
the
image
in
other
formats.
E
D
A
Yeah
and
there's
a
number
of
ways
we
could
look
at
this
right
like
if
it's
a
URL
structure.
Do
you
do
something
like
config
map
colon,
slash,
slash
the
name
of
it,
something
like
that
or
or
just
some
other.
You
know
defined
URL
that
works
for
this,
or
you
can
key
off
of
that
first
part
of
the
URL
to
know
what
to
do
with
it.
There's
things
like
that.
A
We
could
do
because,
when
I
look
at
git
colon,
slash,
slash,
I,
probably
more
want
HTTP,
but
you
know
it,
and
then
some
of
this
is
even
gonna
have
to
do
with.
How
do
you
reference
it
to
view
it
right?
Am
I
pulling
it
in
and
proxying
it
through
to
view
it
and
that's
just
my
origin
source,
because
with
the
config
map
and
somebody
viewing
it
in
a
browser
somewhere
like
me
at
home
versus
some
other
place,
how
does
that
config
map
get
turned
into
a
URL
that
can
now
be
sent
through
the
browser?
A
E
I
think
all
of
this
in
it's
all,
going
to
be
fronted
by
tooling
right,
there's
some
sort
of
tooling
that's
going
to
be
looking
through
your
application
instances
and
deciding
how
to
render
them.
So
the
tooling
becomes
responsible
for
deciding
whether
it's
going
to
resolve
the
config
map
reference
or
render
the
data
URI
or
pass
through
the
URI
or
how
that's
going
to
be
handled.
Isn't
there's
another
layer
here
that
comes
into
play
before
the
end
user.
D
I'm
just
wondering
if
the
kubernetes
project
has
an
idea
of
like
a
standard,
URL
format
scheme
for
referring
to
cluster
resources,
because
that
seems
like
the
kind
of
thing
that
eventually
lots
of
projects
are
going
to
need,
and
it
makes
more
sense
for
them
to
define
it
than
for
everybody
to
define
their
own.
So.
D
E
D
A
That
can
be
done
as
well,
which
is
all
tooling
on
top
of
it,
and
so
this
is
when
the
poll
request
comes
in.
This
is
one
of
those
things
will
probably
poke
at
is
to
say
what
should
be
in
here
versus.
What's
the
responsibility
of
tooling.
On
top
of
it,
because
so
part
of
our
I,
so
one
of
the
things
we
had
with
all
of
us-
and
please
this
was
our
thinking
so
feedback
on
it
is
always
welcome-
was
to
make
this
as
light
as
possible.
So
it
conveyed
enough
information
to
enable
implementers
on
top.
A
We
don't
want
to
go
too
over-the-top
and
what's
defined,
so
is
so
that
it
that
it
infringes
on
the
implementation
details.
We
want
to
give
the
implementers
enough
that
they
can
go,
do
what
they
need
to
do
or
then
try
to
force
an
implementation
right,
because
that's
that's
a
kubernetes
thing
here.
We
want
to
be
more
more
platform
in
this
aspect
and
enable
the
ecosystem,
and
so
how
much
of
this
is
worth
being
in
the
SPAC
versus
being
up
to
implementers
to
do
all
kinds
of
wonderful
things
with.
A
A
D
A
All
right
folks,
well
with
with
that
in
mind,
if
there's
nothing
else
today,
we'll
give
you
a
bunch
of
time
back
and
have
a
wonderful
week,
we'll
see
you
online
there's
some
nice
conferences
going
on
I
hope
some
folks
are
enjoying
those
or
get
a
chance
just
to
stream
anything
that's
streaming
thanks!
Everyone
have
a
wonderful
week.