►
From YouTube: Argo Contributor Experience Office Hour 15th July 2021
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
Good
morning,
good
evening,
one
more
time,
everyone,
let's
start
from
sharing
my
screen.
If
I'm
not
mistaken
today,
we
have
ux
centric
meeting
we're
supposed
to
discuss
ui
changes
before
we
start
and
since
I
always
forget
about
changing
github
discussion
moderator,
I
want
to
do
it
right
away
and
basically
thanks
feeling
for
answering
questions
last
week
and
basically
the
next
one
is
jonathan,
any
okay,
unboxing
good.
Thank
you
all
right,
so
we
won't
forget
to
do
it
this
time,
and
I
noticed
that
we
didn't
have
much.
We
added
two
topics.
One
was
added
by.
A
I
guess,
regina
and
next
one
I
added.
I
think
it's
going
to
be
meaty
to
think
you
know
they
just
basically,
I
I
I'm
proposing
to
brainstorm
about
a
big
change.
We
made
on
details
page
and
we
kind
of
redesigned,
filter,
alex
collins,
did
it
yeah
and
just
wanted
to
get
feedback
and
see
if
anyone
can
suggest
any
improvements.
C
Hey
so
this
I
did
add
this,
but
it
was
samya
who
wanted
to
bring
it
up
so
samia.
If
you
want
to
discuss
this
at
all.
D
Me,
oh
sorry,
it
was
you
my
dad,
thank
you
regina.
So
I
worked
on
this
original
issue
about
this
layout
of
two
icons.
If
you
can
scroll
down
a
little
bit
so
you
see,
one
is
external
link,
the
screenshot,
sorry,
okay,
this
one
yeah,
so
you
see
two
icons
listed
there.
One
is
external
link
and
the
other
one
is
put
on
list.
D
D
A
And
is
it?
Am
I
clear
that
the
problem
is
that
basically,
when
you'll
click
here,
he
doesn't
really?
If
there
are
multiple,
he
doesn't
know
which
one
is
going
to
be
opened
or
you,
you
kind
of
concerned,
about
this
type
of
drop
down
in
general,.
D
I'm
thinking,
maybe
a
better
icon,
just
for
only
one
icon
which
can
open
the
drop
down
list,
so
the
user
can
choose
which
link
they
want
to
go,
but
I
don't
know.
Maybe
someone
has
better
idea.
I
did
google
like
icon
for
multiple
links.
D
A
I
want
to
bring
up
one
more
issue
related
to
the
same
icon.
I
just
know
that
from
experience,
so
since
I'm
sharing,
I
will
just
show
you,
we
have
the
page
and
we
have
refresh
button
here
and
basically
I
noticed
that
users
do
not
understand
what
this
thing
is,
and
no
one
knows
that
we
have
a
drop
down.
This
has
this
hard
difference
yeah,
so
I'm
not
100
sure,
but
maybe
it's
same
type
of
problem.
Basically,
this
icon
is
not
self-explanatory.
It's
not
clear
enough.
It's
a
drop
down!
Yeah.
F
A
To
me,
external
urls
are
important,
but
this
is
super
important
and
it's
kind
of
it's
a
thing
that
we
really
want
users
to
be
aware
of,
and
they
don't
know
so
maybe
I'm
glad
you
brought
it
up.
So
maybe
we
should
just
you
know,
create
a
new
ticket
to
redesign
drop
down,
because
this
pattern
doesn't
seem
to
work.
C
Yeah
we
could,
I
don't
think
any
of
the
you
I
and
ux
people
are
here
like
from
red
hat,
like
serena
or
beau
or
bridget,
but
we
could
reach
out
to
them
and
see
if
they
have
any
suggestions
about
that.
G
A
E
okay,
basically
refresh
is
maybe
a
better
word,
is
reconcile.
Refresh
is
just
you
know,
control
it.
A
C
A
I
am
100
agree,
it's
not
imaginary
problem.
It's
like
if
you
search
for
hard,
refresh
and
github,
you
will
see
it's
like
the
same
discussion
happens.
All
you
know
very,
very
frequently.
People
asking
how
do
I
get
rid
of
the
cache
answer
is
use
hard
refresh
and
no
one
like
everyone
doesn't
have
no
clue
like
what
are
you
talking
about
and
then
usually
we
attach
a
screenshot
and
they're.
Just
like
oh.
F
A
G
D
Yes,
just
one
little
suggestion:
if,
if
we
cannot
find
a
solution
for
both,
I
mean
for
the
drop
down,
we
can
just
simply
disable
the
external
link
icon
so
that
the
user
will
just
click
on
the
drop
down
and
choose
one
link.
A
A
D
D
G
To
show
hey,
hey,
alex
on
that
note,
since
we're
looking
at
refresh
and
hard
refresh,
doesn't
sync
also
have
a
similar
level
of
caching
right
now,
where
we
cache
in
advance
what
the
commit
id
out
there
is,
so
that
we
don't
necessarily
look
at
that
commit
id,
and
then
we
also
check
whether
we
have
the
information
from
that
commit
id.
So
technically,
do
you
foresee
even
sync,
going
through
a
similar
exercise,
yeah.
A
Actually
refresh
and
think
so,
a
refresh
is
basically
it's
a
com.
You
know,
operation
of
comparing
of
two
states
and
seeing
right,
comparing
plus
keeps
each
other
applying
that
that's
yeah,
I
see,
does
does
things
it
uses
exact
same
cache.
So
sync
when
you
click
it
and
okay,
maybe
maybe
I
see
where
you're
going
so
technically
user
could
have
a
choice
to
basically
we
could
have
sync
and
hudson.
It
would
be
a
terrible
name,
but
we,
you
know
user,
could
potentially
choose
to
think
and
avoid
using
cached
manifests
yeah.
A
We
use
cached
option
because
if
you
throw
away
cash
and
try
to
redeem,
then
sync
might
do
something
else
like
not
what
you
are
expecting
to
to
do
to.
G
A
G
A
G
I
think
sounds
good.
I
think
it's
probably
worth
to
note
these
down
in
the
ticket
so
that
you
know
if
you
design
and
if
and
if
there's
a
chance
to
indicate
that
to
the
user,
what
is
being
cached
and
uncashed,
then
that's
fine,
but
then
I
think
we
do
not
plan
on
exposing
a
lot
of
the
cluster
state
cache
to
the
user
anyway.
So,
yes,
I
think
probably.
A
A
Simply
because
we
we
know
that
it's
impossible
to
like
git,
don't
give
us
push
notifications
unless
you
can
and
that's
what
I
need
to
give
some
way
for
user
to
you
know
to
control
it
yeah.
If
we
don't
have
the
button,
then
user
would
have
to
either
wait
like
two
or
three
minutes
or
rely
on
the
hook.
That's
that
was
my
motivation
to
even
create
that
button.
H
Yeah,
I
would
say
it
couldn't
hurt
to
include
what
you
just
talked
about
in
the
docs.
E
H
As
definitions
or
related
to
the
particular
actions
in
the
particular
section,
I
don't
think
I've
seen
it
in
there.
Maybe
it
is
maybe
it
was
added
more
recently,
and
I
know
we
have
like
general
definitions
of
what
syncar
and
what
refresh
are,
but
it
couldn't
hurt
to
include
a
bit
more
information
on
the
actual
behavior
of
those,
because
I
suspect
that
that
is
something
that
folks
would
find
interesting
or
useful.
H
A
A
A
So
basically
we
had
a
it's
been.
It
was
clear
that
we
had
a
problem
with
filter.
Basically,
filter
grow,
the
filter
of
resources
on
application,
details
page
it
used
to
be
a
small
drop
down
and
we
added
you
know
more
and
more
options
of
what
user
can
filter
into
the
drop
down
and
it
just
became
difficult
to
use.
A
You
know
it's
deployed
to
demo
instance
of
fargo
cd
and
we
agreed
that
we
work
on
it
and
basically,
we
you
know,
made
the
change,
but
we
want
to
keep
improving
it
and
I
wanted
to
you
know:
just
discuss
random
ideas,
see
what
people
I
what
people
think
I
had
some
my
own
ideas
to
improve
it,
but
I
basically
don't
really
trust
myself,
so
I
wanted
to
hear
from
as
many
people
as
possible
yeah.
So
this
is
how
the
filter
looks
like
right.
A
A
Didn't
really
belong
here,
so
it
was
moved
pretty
much
here,
so
we
have
a
button
show
height
and
then,
if
user
click
show,
we
show
bunch
of
options
that
user
can
use
to.
You
know
to
play
with
filter
options,
and
now
it's
much
more
feature
which
kind
of
it's
possible
to
sync
by
you
know,
by
status.
By
help,
I
think
status,
health
status
kinds
are
visible
via
that
namespace.
So,
basically,
if
you
have
more
than
one
namespace
you
can
choose,
this
is
the
empty
name.
A
Space
is
a
bug,
it's
a
final
bug
which
have
to
be
fixed
and
basically,
we
added
more
room
for
other
options.
I'm
pretty
sure
we
can
come
up
with.
You
know
way
more
other
options
and
yes
and-
and
I
can
start
complaining
about-
you
know
things
that
I
don't
like
this
honors
and
owned
on
owners
and
owned.
It's
kind
of
it's
tricky,
but
I
kind
of
like
it.
It's
an
advanced
filter
to
me.
So,
basically
you
can
say
I
want
to
be
everything
owned
by
stateful
set
and
I
think
it
works
this
way.
A
A
Yeah,
it's
I
mean
the
discussion
is
this
meeting
today
is
supposed
to
you
know
we
just
have
people
who
care
about
ui
and
want
to
give
feedback
about
ui,
and
I
think
it's
the
biggest
change
in
the
you
know
one
of
the
biggest
changes
we
made
and
yeah.
That's.
Why.
I
Yeah
I
mean
yeah
I'd
like
to
hear
people's
feedback.
I've
based
this
on
how
I
want
to
use
it
for
large,
apps
and
a
large
app
might
contain.
You
know
lots
of
different
objects
that
are
kind
of
like
loosely
related
to
it,
so
you
might
have
two
deployments
in
it
that
do
two
different
things
and
you
might
want
to
just
look
at
one.
So
it's
kind
of
orientated
around
just
looking
at
you
know,
subsets
of
apps.
I
You
don't
have
the
label
filter
on
this
one.
Do
you.
I
Yeah
there
are,
there
are
a
couple
of
kind
of
incomplete
features
in
this,
and
I
kind
of
left
it
as
three
things.
One
is
that
the
way
that
it
lays
out
on
the
screen,
I
think
remington
can
see
this.
E
I
That
it
doesn't
really
make
very
efficient
use
of
space
here.
Does
it
the
kinds
column
could
be
like
twice
as
wide
and
half
as
deep,
and
it
would
be
more
efficiently
laid
out
on
it?
That's
one.
A
I
A
A
You
know
a
different
developers
would
work
on
backhand
changes,
so
it's
basically
now
we
have.
We
have
enough
time
to
think
about
improving
it
and
you
know
and
polish
it
and
so
on,
and
I
I
wanted
to
collect
ideas
and
then
transform
them
into
tickets.
So
one
I
agree:
it's
not
the
most
efficient
way
to
use
space,
because
you
know,
I
think
it
maybe
looks
better
on
a
small
application
that
has
couple
resources,
but
this
particular
application
demonstrates
that
it's
not
efficient.
I
J
I
Knows
better
so,
for
obviously
I
mean
just
just
going
through
the
sync
status
same
as
usual
health
center
same
as
usual
kinds,
you
know
you're
just
just
the
same,
but
you've
just
got
more
space
to
do
with
it
name
spaces
if
you've
got
multiple
namespaces
mixed
within
the
same
app,
and
you
only
want
to
look
at
one
of
them.
So
in
our
use
case,
that's
argo
events
and
ongo
workflows
or
in
different
name
space.
This
allows
you
to
just
look
at
the
argo
event
stuff
the
created
within
is
intense.
I
I
didn't
want
that.
I
actually
wanted.
I
want
to
see
what
what's
been
recently
synced
in
my
app
what's
changed
recently,
but
that
data
isn't
available.
Who's
created
within
is
kind
of
a
a
proxy
for
that,
and
there's
also
labels
in
the
second
pl.
That
requires
back-end
changes
again,
that's
not
perfect,
because
it
only
has
the
labels
of
live
resources,
not
target
resources,
and
I
actually
kind
of
really
want
to
filter
on
the
second
one
rather
than
the
third
one.
But
it's
pretty
good
enough
for
a
lot
of
use
cases.
I
would.
A
Right,
yeah
and
I
think,
okay,
it's
going
to
be
recorded,
hopefully
we'll
catch.
The
suggestions
I
I
also
I
thought
about,
I
feel
like
ownership
created
within
it's
kind
of
you
know
filters
that
you
use
less
frequently
because
you
know
her
status
status
is
like
you,
you
basically
filter
wait.
I
I'm.
I
still
need.
I
Sorry
I
didn't
get,
I
didn't
get
ownership,
so
I
wanted
to
come
back
because
the
ownership
one
is
a
bit
of
a
weird
one.
It's
completely
all
the
other
ones
are
like
and
filters
synced
and
healthy,
and
it's
a
pod
and
its
name.
Spaces
are
go
cd
and
it
was
created
in
the
last
five
minutes,
but
ownership
is
not
the
same
as
that.
Ownership
is
more
like
if
it
was
included
by
any
of
the
other
filters.
I
So
say
I
filtered
out,
and
I
ended
up
with
just
one
pod
at
the
end
of
this
ownership
basis,
also
so
show
any
resources
that
are
either
owned
or
own
that
resource.
I
So
if
it
was
a
if
I
want
to
find
a
particular
pod
and
then
look
at
the
deployment,
I
can
filter
the
to
the
pod
with
the
other
filters
and
then
click
owners,
and
then
it
will
show
me
the
replica
set
and
the
deployment
or
whatever
or
whatever
it
is
I
mean
you
can
show.
I
mean
this
is
a
good
example,
just
click
on
pods
and
then
click
on
and
there
you
can
see.
A
J
I
can
you
try
something
so
I
can
so
uncheck
pod
but
check
replica
set.
Yes,
okay,
scroll
down,
you
don't.
J
A
I
I
don't
know
how
useful
this
will
be,
because
I
think
what
I
would
probably
do
is
use
like
an
app
filter,
because
everything
always
has
like
a
label
on
it
of
app
equals.
I
don't
know
workflow
controller
and
you
know
the
replica
set.
Has
the
label
and
the
pod
has
the
label
as
well?
So
if
you
had
labels,
you
just
use
the
labels,
and
actually
this
I
don't
know
how
useful
this
owner
would
be.
I.
I
A
J
J
Yeah,
I
agree
with
that.
I
think
yeah
I
think
parent
children
is,
is
maybe
more
clear
what
the
it's
gonna
do
and
then
and
then
the
fact
that
it
acts
a
little
bit
differently
some
way
to
visually
separate.
It
would
be.
I
Yeah,
so
you
kind
of
confirmed
what
I
suspected,
I
think,
and
also
like,
using
the
terms
owned
and
known
as
this
kind
of
technically
current
correct,
but
not
very
clear.
I
don't,
as
I
say,
I
don't
know,
which
term
I'm
meant
to
be
using
but
yeah,
maybe
some
visual
differentiation
of
that
particular
because
it's
not
even
a
filter.
Is
it
yeah.
I
I
don't
I
don't
know
how
useful
it
would
be,
but
also
I
I
I
don't
want
it
no.
J
This
is
useful
because
often
I
do
the
same
exact
thing.
It's
like.
I
show
me
my
deployment,
but
then
I
ultimately
I
really
want
to
see
the
pods
and
then
so
then
I
probably
do
usually
what
I
do
is
like
click
pods
with
deployments,
and
then
maybe
I
want
to
see
replicas
too,
but
I
can
save
a
click
by
just
saying
deployments,
plus
children.
I
I
wanted
I
mean
this
is
a
question
for
other
people
as
well
like
that
created
without
a
kind
of
you
said
age,
but
I
kind
of
want
like
I
want
to
know
what's
been
recently.
Synced
is
actually
the
use
case
I
wanted
for
this
one,
but
there
there
isn't
that
data
isn't
available.
As
far
as
I
know
I
mean
you
can
correct
me
if
I'm
wrong,
but
I
think
that
would
be
a
cool
feature
and
if
you
scroll
up
alex.
A
I
Yeah
one
thing
that
might
be
nice
in
the
user
interface
again,
this
is
kind
of
I
was
just
doing
something
that
was
easy.
To
be
honest,
everything
about
this
is
easy
because
I
had
a
limit.
You
know
I
had
two
days
to
work
on
this
is
these
replica
sets
could
be
kind
of
you
know
like
a
pile
of
playing
cards
like
a
deck
of
playing
cards,.
I
At
that,
from
the
top
you
know
say,
I
had
10
playing
cards.
This
is
kind
of
like
10
playing
cards
laid
out
on
the
table,
but
because
they're
all
replica
sets,
I
feel
like
I
could
visually
stack
all
these
rectal
cassettes
with.
You
know
showing
the
overlay
that
there
are
multiple
replicas
and
I
kind
of
want
to
do
this
in
data
flow.
Actually,
when
there
are
multiple
replicas
to
kind
of
lay
them
out,
so
you
can
see
multiple
copies,
but
only
the
most
pertinent
one
is
on
the
top
of
this
deck.
I
I
Set,
but
also
partly
beyond
my
available
time
to
do
it.
I
J
Yeah,
I
think
we've
talked
about
this
before,
but
I
think
right
now,
everything's
a
generic
tree
thing,
yeah,
there's
any
special
casing
for
anything
in
that
tree,
but
I
think
everyone
wants
something
like
that.
I
have
ideas
for
the
parent
and
children,
though
another
idea,
so
so
really,
this
is
a.
This
filter
is
affecting
kinds
right
like
really
you're,
including
more
kinds.
Oh
you're,
not
so
to
kind
of
ex
help.
Explain
this.
I
think
parent
and
children
should
be
somehow
incorporated
into
the
kinds
box
as
a
separate
visualization.
J
So
you
have
you,
can
you
can
cherry
pick
specific
things
in
the
clients,
but
you
can
also
include
the
kinds,
children
or
or
you're,
basically
including
more
of
the
kinds
implicitly.
Basically,
you
think
merch,
these
two
yeah,
but
with,
like
you,
know
a
separate
section
for
include
parents
and
children
yeah.
I.
A
I
know
that
basically,
labels
filter,
that
alex
was
going.
He
wanted
to
edit
here
as
well.
It
would
also
benefit
from
ownership.
It's
like
you
can
say.
I
want
to
find
all
children
of
component
labeled
this
abc,
so
it
would
be
kind
of
same
as
kinds.
It's
like
you
know.
You
either
use
it
with
labels,
all
these
kinds,
always
both
so
yeah.
Basically,
now
I
would
merge
it
into
kinds,
but
once
we
add
labels
panel,
then
it
will
be
unclear.
A
J
But
it
is
the
the
the
the
act
of
checking
that
and
unchecking
it
does
affect,
is
related
to
kinds.
Is
it
not
like
it's
related
to
everything?
It's
like.
Basically,
all
these
filters.
A
Plus
owners
of
everything
filter
by
these
guys,
you
can
say:
let's
say
if
you
have
two
namespaces
here,
you
can
say
I
just
want
to
see
everything
that
own
okay
and
name
space
is
a
bad
example
yeah.
You
can
say
I
want
to
see
everything
that
owns
things
created
one
minute
ago,
kind
of
and
it
yeah
right.
So
it's
not
kind.
You
filter
you
just
first,
you
narrow
down
to
creation,
you
know
by
by
age,
and
then
then
you
expand
using
this
guide.
So
it's
kind
of
this
is
how
I
understand
it.
J
J
All
right,
I
I
take
back
what
I
said
so
I
think
we
can
keep
it
separate.
A
What
like
anyone?
I'm
happy
yeah,
let's
put
it
this
way,
anyone
wants
but
old,
filter
back.
K
Well,
this
isn't
this.
This
isn't
probably
the
feedback
you're
looking
for,
I
just
kept
on
thinking.
I
kind
of
wish
that
filters
button
was
up
at
the
top
with
all
the
other
buttons,
because
I
worry
about
it,
getting
lost
as
I'm
scrolling
around.
That's
that's
my
only
comment
about
it.
I
think
it's
thank
you
so.
I
It
was
originally
there
and
it
isn't
there
in
version
two
zero,
but
I
basically
harmonized
this
with
the
application.
I
call
this
the
resource
filter
and
it's
meant
to
be
kind
of
visually
harmonized
with
the
application
filter,
and
the
thing
I
noticed
is
the
hide
button
is
quite
far
away
from
the
the
I
don't
know
what
you'd
call
the
filters
with
a
little
filter
symbol.
I
J
Actually,
there's
a
point
about:
if
we
scroll
down,
do
we
we
have
to
scroll
back
up
for
fill
additional
filtering.
Is
that
yes,
yes,
no.
A
F
K
I
I
like
the
idea
of
a
panel
because
you
can
also
have
or
or
a
button
that's
persistent,
because
it
can
tell
you
that
filters
are
in
force,
because
if
I
set
one-
and
I
don't
know
if
it's
persistent
but
like
if
I
set
one-
and
I
forgot
about
it-
then
suddenly
I'm
like
oh,
I
can't
find
this
resource.
K
I
can't
find
this
resource
and
then
oh,
I
forgot
to
scroll
up,
there's
like
filters
up
here,
and
I
forgot
about
that
so
having
like
the
filters
persistently
visible,
whether
that's
because
we
move
it
up
or
it's
in
a
panel
or
maybe
it's
a
persistent
float
that
can
open
and
minimize
as
you
scroll
or
something
but
having
it
be.
Persistent,
I
think,
is
important,
especially
if
filters
are
in
place
if
you've
set
filters.
I
think
it's
especially
important.
A
A
Basically,
if
we
just
take
filters-
and
you
know
render
them
vertically
and
you
know,
screen
would
be
kind
of
separated
and
let's
say,
if
we
let
user,
you
know,
hide
and
show
the
parts.
So
in
this
case,
user
would
see
filters
all
the
time
and
could
just
you
know,
scroll
up
and
down
this
area
independently.
K
I
mean
that's
kind
of
why
I
was
thinking
like
if,
if
we
had
the
filter
button
up
next
to
the
refresh
right
or
up
in
that
area,
if
I
click
it,
I
would
kind
of
imagine
that
it
would
expand
a
panel
below
it
that
showed
all
the
filters
and
then,
if
I
retract
it
it
shows
you
know,
maybe
it's
highlighted
orange
and
shows
a
number
that
basically
indicates
that
filters
are
active,
so
I
can
always
see
that's
like
kind
of
lit
up
while
it
while
a
filter
is
set,
and
then,
if
I
click
it,
it
expands
back
out
that
panel.
A
K
Yeah,
probably
between
the
row
of
buttons
and
like
the
app
health
row,
it
would
open
between
those
and
show
me
all
the
filters
available
and
maybe
that
maybe
there's
a
scroll
within
there,
because
you
don't
want
it
to
completely
overtake
the
screen
or
something.
But
then
you
set
your
filters
and
they're
applied
automatically
they're
applied
and
then,
if
you
click
the
filter
button
again
it
collapses.
K
J
That's
kind
of
a
radical
idea,
it's
so
so
you
know
that's
that
argo
cd
on
the
left,
the
basically
the
the
root
node
on
this
tree.
Yes,
this
is
only
ever
like
one
item,
so
we've
basically
used
up.
You
know
probably
two
300
pixels
of
space
column,
I'm
I'm
proposing.
J
We
actually
just
get
rid
of
the
root,
node
and
and
maybe
think
about
leveraging
the
left
space,
because
I
just
visited
the
yelp
page
like
see
how
do
they
handle
filters
and
they
actually
reserve
the
whole
left
side
of
the
screen
for
filtering,
and
one
thought
I
just
had
was
like
what,
if
we
do
something
similar
like-
and
this
is
always
kind
of
something
I've
been
thinking
about-
is
like
what's
the
point
of
just
reserving
the
whole
life
yeah,
I
I.
A
K
K
A
A
You
know
the
rotation
is
kind
of
it.
It
would
kind
of
keep
it
here,
but
still
it
would
save
a
little
space.
It
would
be
like
yeah.
We
definitely
can
make
this
guy
might
take
significantly
less
white
space
yeah,
and
then
I
like
that
option
because
I
feel
like
I
want
to
have.
I
I
think
it's
super
important
to
be
able
to
filter
quickly,
like
I
hate
to
click
back
and
forth
here
with
the
previous
filter.
A
It
was
not
perfect.
This
is
not
perfectly
there
because
now
I
have
to
scroll
up
and
down
and
if
we
just
move
these
guys
to
the
left
like
yelp,
does
it
then
it
seems
to
be.
You
know
a
good
solution.
We
have.
We
have
space
here.
Filters
will
be
available
all
the
time
and
users
don't.
Maybe
we
don't
maybe
even
need
to
hide
these
guys
because
we
already
have
white
space.
So
why
don't
just
use
it
fill
it
with
the
useful
information.
C
Can
I
make
one
more
suggestion
on
this
panel?
Can
you
have
the
ability
to
hide
each
like
expand
and
collapse
like
each
section,
so
that,
like
it's,
not
like
the
entire
thing
like
if
you're
really
not
caring
about
one
of
them,
you
can
just.
I.
J
I
So
just
so,
I
think
those
two
should
be
harmonized
together,
but
again,
it's
kind
of
the
way
that
the
application
page
is
laid
out
is
different
to
how
this
page
is
laid
out
and
again
I
just
had
a
limited
amount
of
time.
I
didn't
want
to
spend
yeah
a
lot
of
it,
given
that
you
know
it'll
take
me
longer
to
get
headway
in
that
area
versus.
I
Can
I
just
like
bounce
a
couple
of
ideas
off
people
as
well:
functionality
related,
rather
than
visually
related,
so
going
to
the
the
age
created
within
filter
my
I
this
is
there,
because
that
data
was
readily
available
and
acted
as
a
proxy
for
recently
syncs,
but
I'm
actually
skeptical.
That
recently,
sync
would
be
useful
because,
typically
you
you
end
up
syncing
up
every
single
resource.
Don't
you
because
you
don't
quite
know:
if
actually
you
need
to
do
that
or
not
yeah.
I
And
if
you
don't
do
it,
you
might
end
up
with
a
slightly
you
know
you
might
end
up
the
resource,
it
didn't
get
sync
that
needs
to
be
synced
because
it
shows
as
being
synced
in
the
in
the
thing,
but
actually
it's
not
synced
underlying
or
it's
slightly
different.
Some
default
value
is
slightly
different
and
in
that
case
every
single
resource
within
the
application
would
be
recently
synced,
but
but
but
some
would
not
would
be
recently
created,
for
example,
a
pod
that
was
created
because
a
new
replica
set
was
created
and
a
rollout
occurred.
I
K
A
A
J
J
Machine
consumable
form
like
you're
relying
on
a
string
output
keeps
here
to
apply.
I
This
is
like
a
general
kubernetes
issue.
Isn't
it
like?
I
want
any
resource
I
want
to
know
who
owns
the
resource
like
which
human
being
is
to
blame.
I
think
I
mentioned
this
before
who's
to
blame
for
this
resource,
and
when
did
they
make
the
mistake
that
they
made
okay,
I'm
kind
of
like
catastrophizing
them?
I
J
J
Also,
it's
also
not
always
accurate.
Sometimes
if
you
have
a
like,
if
you
apply
something
but
there's
a
diff,
but
the
diff
doesn't
actually
affect
anything
and
keeps
ctl
will
still
say
it's
configured
and
it's
not
something
you
can
rely
on
as
a
real
diff.
J
E
I
They
just
did
some
kind
of
tech
debt
thing
where
they
reformatted
some
white
space
on
a
file
and
really
the
you.
You
know
the
the
person
who
create
you
know
designed
that
resource
or
design
the
application
is
the
person
you
want
to
speak
to
I'm
just
riffing.
Now
I'm
going
to
stop
remington
remington
who's,
not
here,
we've
just
committed
him
to
fix
this
is
that
right,
yeah.
I
A
But
I'm
kind
of
I'm
at
least
I'm
really
happy.
We
had
the
discussion
because
you
know
now
it
seems
obvious
how
we
can
improve
it.
At
least
you
know
if
we
simply
make
it
consistent
with
that,
it
would
be
much
better
yeah
and
we
have
we've
got
idea
of
basically
buying
space
for
the
filter
if
we
just
make
these
guys
smaller.
So.
I
Have
you
ever
thought
about
making
this
circular
with
the
application
in
the
middle
and
the
resources
in
a
kind
of
circle
around
it.
A
J
Too,
I
guess
iconic,
I
think
it's
it's
gonna,
be
super
disruptive
too.
I
I
Just
like
reduce
the
amount
of
duplicate
noise.
I
don't
know
how
you
need
some
algorithm,
because
how
do
you
know
that
two
replica
sets
belong
to
a
deploy
or
sorry?
How
do
you
know
that
some
resource,
pods
and
replica
sets
are
obvious
ones,
but
there
might
be
other
ones
that
I
don't
know
about
your
custom
resources.
How
do
you
know
that
they're,
actually
just
like
versions
of
the
same
resource
yeah,
I
guess
you've
got
the
revision.
Haven't
you.
I
I
A
A
I
yeah
so
not
just
too
late.
Maybe
we
can.
I
think,
we've
got
some
good
ideas
and
we'll
try
to
put
them
into
the
ticket
and
and
then
remington
will
be
happy.
He'll
have
a
lot
of
things
to
do.
We
also,
typically,
we
kind
of
we
keep
forgetting
to
triage
features
and
10
minutes
is
not
enough
again
to
to
go
for
it,
because
maybe
we
can
use
10
minutes
to
figure
out
how
we
finally
can
do
this
by
aging.
A
I
I
I
have
a
feeling
that
this
meeting
won't
it's
just
not
long
enough
to
create
teachers.
If
we
try
now,
I
think
we
will
handle
maybe
four
like
because
at
least
two
minutes
needed
to
discuss
every
feature
to
say
yes
or
no
yeah.
What?
What
do
you
think
like?
A
B
C
I
had
added
a
feature
that
I
made
a
pr
for
about
a
status
bar,
but
it's
an
open
pr
right
now.
No
it
it's
okay
it.
If
this
is
more
important,
that's
fine
people
can
just
look
at
one
and
say
yes
or
no.
H
Yeah
so
yeah,
I
just
thought
I'd,
give
a
brief
update
on
application
set.
Second
release:
zero:
two
zero.
I
think
in
the
last
meeting
you
all
discussed
the
next
argo
cd
release.
I
wasn't
able
to
attend
that
meeting.
I
think
I
had
a
dentist
appointment
or
something,
but
in
any
case
yeah
we'll
be
releasing
application
set
new
release,
zero
two
zero
alongside
the
next
dargo
cd
release,
targeting
ideally
the
same
day
and
just
as
a
quick
status
update.
H
All
the
functional
prs
are
in
for
zero
two
zero.
Basically,
everything
that
we
wanted
to
do
in
this
milestone.
That's
left
is
some
doc
updates
I'll
be
putting
out
official
release
candidate
builds
fairly
shortly,
but
in
the
meantime,
folks
are
free
to
check
out
the
development
fields
which
are
up
to
date
with
the
master
branch.
One
quick
thing:
if
anyone
on
this
call
contributed
to
the
application
said:
zero.
Two
zero
release
check
out
issue
291
in
the
application
set
repo
to
make
sure
that
you're
on
the
list
of
contributors.
H
I
want
to
make
sure
that
I
shout
out
everybody
that
had
a
hand
in
the
zero
two
zero
release
and,
if
folks
are
interested
in
status
updates
as
the
release
progresses
as
the
release
process
progresses,
you
can
check
the
app
set
slack
channel
on
cncf
slack.
That's
where
I'll
be
posting
status
updates
and
that's
it
for
me.
J
What's
the
whether
the
highlights
of
the
release
just
say.
H
E
A
C
C
So
this
is
what
it
would
look
like
if
I
had
a
bunch
of
these
apps
with
all
these
statuses,
the
I'm
looking
a
lot
for
feedback
on,
for
instance,
this
gray
color.
If
this
was
the
whole
status
bar
like
it
would
be
if
it
was
just
if
it
was
using
the
actual
data,
it
would
kind
of
just
look
broken
like
I'm,
not
a
huge
fan
of
that.
C
So
if
there's
a
better
color
or
a
different
kind
of
grouping,
remington-
and
I
had
discussed
grouping
these
two
together
like
missing
an
unknown,
but
if
there
is
a
better
grouping
or
any
suggestions
you
have
just
on
this,
please
let
me
know:
there's
an
open
pr
for
it
right.
A
Okay,
one
do
you
think,
is
it
possible
to
move
that
line
next
to
items
per
page,
basically
just
to
save
use
that
white
space
right?
Oh.
C
C
I
can
move
that.
What
happens
when
you
click
on
the
bar?
Nothing
because
we
discussed
having
this
functionality
where
you
could
click
on
it,
but
those
filters
are
already
here
if
you
wanted
that,
so.
C
C
C
Okay,
I
can
do
some
exploration
on.
You
know
how
challenging
that
would
be.
Maybe
in
a
second
pr.
We
could
add
that
if
we
want
this
sooner
but
I'll
look
into
that.
C
A
Like
if
someone
click
on
this,
I'm
kind
of
I'm
trying
to
see
what
purpose
it
serves,
it's
like
it
saves
user
from
going
back
and
forth
to
to
this
status.
Page
kind
of
you
know
what
page
I'm
talking
about
right
on
the
right
top
corner.
We
have
this
graph,
which
kind
of
have
the
same
information,
but
it
has
more.
It
has
kind
of
richer
information,
oh
wait,
no
it
and
it
has
yeah.
A
It
would
be
similar
to
that
and
basically
user
no
longer
have
to
go
to
that
page
only
if
he
wants
details,
so
maybe
if
click
would
simply
redirect
user
here
you
know,
like,
let's
say,
you're
working
with
a
bunch
of
applications,
and
then
you
run
sync
and
then
two
of
100
degraded
and
basically
and
then
you
can
click
on
the
bar.
Actually
I
take
it
back
you,
you
would
click
on
that
bar
just
to
know,
okay,
what
exactly
integrated!
A
I
think
it's
both
that
the
toolbar
and
this
page
are
just
a
summary
and
second
kind
of
give.
You
summary,
plus
some
numbers
that
bar
gives
you
like,
just
a
feeling
that
you
know
yeah
in
general.
Everything
is
good
every
you
know
something
is
bad,
but
if
you
want
details,
you
really
want
filters.
So,
okay
and
basically
I'm
going
to
stop
talking.
Okay,
I
will
just
filter
if
you.
A
E
Yeah,
actually
I
just
had
one
point
I
just
wanted
to
know
regarding
the
migration
of
health
checks
from
argo
cd
to
github's
engine.
I
saw
an
issue
in
github's
engine
that
that
that's
gonna
be
the
plan,
so
just
wanted
to
know.
When
are
we
planning
to
do
that?
I.
A
A
Another
one
is
to
basically
let
people
move
basically
use
extensions
mechanism
to
distribute
how
to
call
it
to
to
package
extensions
and
basically,
in
this
case
we
would
discourage
people
to
contribute
built-in
extensions
into
adversity
and
maybe
encourage
them
to
maintain
some
kind
of
registry,
and
I
wanted
to
point
to
jason
like
what
do
you
think
about
it,
because
you
was
kind
of
you
kind
of
losing
the
extensions
so.
J
So
actually,
what's
the
so
the
the
the
issue
is
that
we
have
the
lua
scripts
in
the
argo
cd
repo,
but
they
probably
should
exist
in
the
git
ops
engine.
Yes,
yes,.
A
And
the
reason
it
was
not
moved
there
first
because
of
parker
like
we
didn't
want
to
force
potential
github
singing
users
to
use
packer,
and
it's
no
longer
a
problem.
Now
we
can
just
use,
go
link,
you
know
embedding,
but
the
question
is
like:
do
we
want
to
do
it
at
all
or
not
because
we've
got
another
way.
You
know
we
will
have
another
way
soon.
Actually.
J
Yeah
yeah,
my
question
was,
as
was
going
to
be,
I
mean
it
sounds
like
the
right
thing
to
do
is
to
move
to
the
guillotine,
but
is
there
any
one
pressing
for
this
like.
J
Yeah
I
mean
I,
unless
someone's
pushing
for
it.
I
mean
I
wouldn't
do
anything
but.
A
I
mean
the
idea
to
do
it
was
to
attract
people.
Basically,
if
you
you
know
have
it
adds
more
value
to
github's
engine.
Basically,
you
can
say
it's
not
just
going
to
sync.
You
will
also
get
for
free,
okay,
so
yeah,
but
I
agree
it's
not
the
you
know.
We
have
more
important
things
to
do.
It's
you
know
it's
not
to
help
cd,
it's
to
help
github's
engine
get
more
users,
yeah.
J
Yeah
I
mean
I,
I
think
I
I
don't
have
strong
opinions.
I
don't
see
why
both
approaches
aren't
compatible,
meaning
we
could
still
have
the
built-ins
in
the
abstention
and
then,
but
we
still
have
this
upcoming
way
where
resource
health
checks
are
are
easier
to
to
manage
in
in
argo
city.
A
Not
to
do
it
because
what
happened
like
recently,
we
learned
that
we
were
approving
health
checks
for
clgs
left
and
right.
It
felt
like
very
safe,
but
then
we
had
several
incidents
that
existing
cargo
cd
users.
A
So
and
maybe
yeah,
we
should
not
do
it.
I
want
to
say
I
like
extensions
more
because
this
way
you
can,
you
have
to
yeah.
If
you
make
this
experience
good
like.
If
it
will
be
really
easy
to
enable
extensions,
then
it
seems
like
a
perfect
solution,
like
user
still
get
same
experience
as
built-in
that
have
very
easy
way
to
opt-outs,
like
yeah.
J
Yeah
yeah,
I
I
think
we
should
kind
of
hold
off
on
it
until
we
have,
unless
someone's
really
pushing
for
it
and
then
and
focus
on
the
extensions
they're.
Improving
that
experience.
G
B
A
Right,
we
already
used
the
whole
hour
and
five
minutes
on
top,
so
I
think
we
have
to
stop
and
maybe
like
maybe
we
can
seek
up
offline.
I
know
that
you
know
shobik,
maybe
usually
have
good
ideas
about
processes.
I
do
not.
I
keep
trying
to
think
about
better
way
to
have
triaging,
you
know
featured
aging
process.
Maybe
we
can
take
it
offline
and
basically.