►
From YouTube: Jenkins UX SIG Meeting 9 Nov 2022
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
Welcome
everyone:
this
is
the
Jenkins
user
experience
special
interest
group.
It's
the
9th
of
November
2022
topics
on
the
agenda
that
I
see
user
feature,
Flags,
vaudec,
ux,
Futures,
talk,
Tim
and
yang
web
components
in
the
Jenkins.
You
act
by
Thorson
Security
reviews
for
ux
poll
requests,
request,
vadek,
open,
ux,
regressions,
possible
review
and
hacktoberfest
results.
If
we
want
to
discuss
them
and
if
time
allows
what
other
topics
Yan
do
you
and
Tim
need
a
a
segment
for
any
recent
things
that
need
to
be
highlighted.
A
Great
all
right
so
Tim
had
warned
that
he's
only
available
for
the
first
30
minutes,
so
we'd
like
to
prioritize
things
to
be
sure
that
those
topics
that
need
hot
discussion
come
first,
so
I
propose.
We
move
this
one
down
so
that
we've
got
web
components
and
user
feature
flags
as
the
top
two
top
two
for
discussion.
C
A
A
C
C
A
D
A
C
A
A
D
That
should
be
good,
so
normally
you
can
see
the
stuff
I
do
not
see
what
you
are
seeing.
So
I
expect
it
to
be
working,
so
you
can
see
here
a
regular
user
to
be
sure
it's
still
connected
and
at
the
bottom
we
have
some
feature
flag
that
are
proposed
only
for
that
user,
meaning
it's
not
a
system
property
that
you
have
to
enable
as
an
admin.
It's
something
that
you
can
enable
or
disable.
As
a
user.
D
That's
really
the
key
point
for
the
uis
back
there,
because
it's
not
something
that
should
have
an
impact
on
the
system,
but
only
on
your
interface.
Your
experience,
username
example
there,
with
the
demo
flag,
if
I,
enable
it
and
save
I,
can
use
it
to
do
things
in
term
of
UI,
for
example,
in
that
part
and
I
will
use
this,
inform
you,
but
the
code
that
is
necessary
for
that.
You
have
a
Jelly
Time
with
a
bit
of
Zoom.
D
You
have
a
new
Jelly
tag,
especially
for
that
you
enter
the
class
of
your
flag,
and
then
you
can
use
the
value
of
your
flag
to
disable
or
enable
the
feature
that
you
want
to.
The
ID
with
this
kind
of
approach
is
to
let
the
UI
team
working
at
their
Pace
at
their
progress
in
terms
of
evolution
of
the
design.
The
different
feature
that
you
want
to
introduce
without
having
to
wait
for
a
full
consumption
for
the
full
community,
which
perhaps
will
never
come
in
a
sense.
But
with
that
you
have
a
way
to.
E
D
The
development
of
the
future
and
to
focus
one
to
request
to
start
another
pull
request
to
go
on
on
the
different
steps.
Instead
of
waiting
too
long,
typically
waiting
for
security
agreement,
this
kind
of
thing,
as
well
as
long
as
it's
sufficiently
isolated,
it
could
be
also
a
good
way
to
reduce
the
potential
impact
there.
D
D
Maintaining
something
that
is
not
used
at
the
end.
Typically,
what
was
done
with
the
previous
new
Adder
that
was
removed
completely
at
some
point,
because
it
was
perhaps
not
used,
and
this
kind
of
thing
should
be
more
I
would
say,
backed
with
data.
That
could
be
interesting
at
that
point.
Instead
of
just
getting
some
feedback
like
the
one
that
are
very
negative
or
very
positive.
I
would
like
to
see
more
what
is
really
the
adoption
from
the
community
for
these
different
things.
D
A
D
Flag
itself
is
just
a
way
to
retrieve
and
to
circulate
and
deserialize
the
value
inside
the
user
property,
so
the
use
of
property,
it's
mainly
the
user,
experimental
flux
property
that
is
keeping
a
map
of
the
different
flags.
If
you
have
done
nothing
from
your
side,
that
property
will
be
empty.
If
you
enable
10
flag,
you
will
get
a
map
with
10
item
and
the
different
flags
retrieving
the
value
doing
their
stuff.
With
that,
let's.
A
D
Example
that
one
it's
a
flag
encrypted
one
just
for
the
demo
and
you
can
put
the
the
ID
of
the
flag
so
that
you
see
which
one
is
inside
Uma.
What
is
the
default
value
I'm
just
using
Boolean
flag
at
the
moment,
but
you
can
Implement
some
for
integral
float
any
form
and
data
that
you
want
to
store
and
you
can
see,
for
example,
digital
description
in.
E
D
The
name
and
the
the
description
you
can
see
here
for
Boolean
three
difference
type
default,
meaning
you
follow
what
is
coming
from
the
code
enable
or
disable
if
you
want
to
force
and
due
to
the
question
I
got
related
to
the
three
state
there.
For
my
point
of
view,
the
default
is
especially
interesting
in
terms
of
telemetry
so
that
we
know
if
someone
wanted
to
get
that
before
or
wanted
to
hold
back
after
this
kind
of
thing
or
to
just
do
nothing.
For
instance,
it's
more
like
the
blank
vote.
If
you
want
internal
information.
A
G
So
that
actually
leads
us
to
to
the
next
topic
so
I'm,
not
basically
it's
like
I'm,
not
the
the
one
that
actually
consumes
too
much
index
jellies
right,
okay,
but
I,
see
it
useful
in
in
the
current
setup.
You
know
so
that
you
actually
can
have
a
quick
integration.
That
would
be.
The
question:
is
there
plan,
for
example,
to
abstract
that,
for
example,
to
use
some?
D
At
this
point,
I
do
not
expect
to
expand
the
system
outside,
especially
as
a
system
flag.
It's
not
a
feature
Factor,
so
it's
really
a
pure
user
effect.
Typically,
what
you
can
think
about
feature
flag
in
general,
it's
more
something
that
you
want
to
have
to
enable
or
disable
the
feature
across
your
different
customer
or
user
engineer.
In
this
case,
it's
for
a
single
instance
to
enable
a
better
view.
If
you
want
in
terms
of
design.
So
it's
a
bit
different.
We
can
have
something
evolve
over
time
for
sure.
D
D
A
G
A
G
So
so,
basically,
actually
having
seen
the
the
the
session
on
on
on
on
the
devops
today,
I
actually
need
to
maybe
rethink
the
whole
integration
path
that
that
I
actually
think
about
so
to
to
make
a
long
story
short.
So
I'm,
a
dedicated
front-end
developer
and
for
me
having
to
use
jelly,
is
something
that's
in
a
modern
components.
Web
application,
it's
not
traditional,
let's
say
there's
no
one
else,
learning
jelly
because
he
wants
to
ride
a
front-end
call
right.
G
It's
like
jelly
is
something
exclusively
born
in
Jenkins
and
and
actually
you
know
used
there.
It's
like
correct
me
if
you're
wrong,
there's,
maybe
two
other
all
three
other
different
applications.
But
basically
the
point
is
why
not
trying
to
not
Implement
our
own
component
but
reusing
something
as
a
first
step
like
standard
right.
G
So,
basically,
right
now,
for
example,
I
I
took
the
liberty
to
create
a
small
PR
or
a
branch
basically
to
to
rewrite
our
check
boxes,
implementation
and
instead
of
using
our
default
CSS
and
everything
I
encapsulated
everything
in
using
a
web
component
provided
by
by
Google
as
a
mvac
check
boxes
and
it's
it's
based
on
material
UI.
So
basically,
you
have
baked
in
there
all
the
usability
and
the
CSS
and
everything
so
there's
no
import,
rather
than
only
yeah
a
small
import
into
into
npm.
B
A
G
So,
basically,
that
that's
the
diff
right
and-
and
you
see
there-
there's
a
just
a
small
change
actually
in
the
underlying
code,
but
basically
I
you
see,
don't
you
don't
see
me
changing
any
CSS,
you
don't
see
changing
anything
else,
then
they
had
some
VA
but
to
make
the
default
magic
work.
But
basically
this
is
the
the
checkbox
that
we
know
until
now.
G
Right,
but
based
on
material
is
in
a
web
component,
and
my
my
point
in
in
using
web
component
is
I
want
to
reach
to
some
kind
of
standard
in
in
the
usage
of
our
implementation,
of
those
special
components.
Right,
chachelli
and
I
would
actually
rather
dedicate
those
things
to,
for
example,
a
web
component.
Why
a
web
component
yeah?
G
Basically,
we
don't
need
to
actually
I
didn't,
have
to
change
any
CSS,
because
if
I
would
edit
just
material
CSS
and
then
implement
or
try
to
implement
the
whole
thing
myself
right,
I
would
run
not
only
into
the
implementation
problem.
I
need
time
to
do
that
and
further
that
I
actually
need
to
have
to
adapt
the
core
much
much
more.
To
allow
me
to
do
that
so
here.
G
Basically,
this
would
be
one
way
to
quickly
update
to
something
and
I
took
right
now:
the
material
Library
as
an
example,
because
in
Cloud
piece
we
are
right
now
striving
to
support
material,
UI
right
and
and
use
that
as
as
a
base
a
baseline
for
our
development
and
then
the
the
example
based
on
on
material
UI.
But
it
could
be
any
other
given
web
component.
So
for
me,
web
component
is:
is
it's
a
way
to
actually
have
like
a
nice
Baseline
in
in
in
what
we
have
as
a
component
and
the
Second
Step?
G
It's
a
split
off
of
backend
and
front
end,
so
Shelley
in
itself
shouldn't
be
used
anymore
because
it's
it's
a
server-side
rendering
technology
and
we
should
actually
strive
for
to
actually
have
like
a
two-tier
applications,
so
basically
classic
Jenkins
responses
in
in
Json,
and
then
we
actually
can
create
a
customized
front-end
actually
for
it
challenging.
There
is
basically
extension
point
right
and
and
the
way
that
we
right
now
actually
integrate
and
use
Jenkins
and
those
things,
but
yeah
that
those
actually
I
need
to
synchronize
with
I.
G
G
A
So
Tim
or
John
any
any
insights
there
I
assume
Torsten
when
you
said
that
you
need
to
review
what
was
presented,
that
that
may
shape
and
refine
and
alter
this.
This
idea
that
you've
got
is
the
general
idea
of
web
components
likely
to
to
continue
still
that
that's
the
idea,
you're
thinking
of
is
web
components
as
a
as
a
key
part.
The
yeah.
G
Go
ahead,
so
so
so
what
components
for
myself
is
is
basically
just
an
API
contract
right.
You
can
Implement
web
component
in
react
or
you
can
Implement
your
component
with
JavaScript
pure
always
lit,
or
there's
million
different
ways
and
Frameworks
where
you
can.
Actually
the
outcome
is
a
web
component.
What
komban
basically
means
that
I
have
like
a
formal
API
between
me
and
my
front
end.
All
right.
That
would
mean
I
could
create
a
back
end.
That
actually
looks
the
same
with
the
with
the
front.
G
G
You
want
yeah,
no,
it's
like
in
in
in
the
challenge
right
now,
it's
like
as
I
understand
what
what
what
this
was
showing
us.
What
is
coming
next
right,
it's
not
based
anymore
on
on
layout
jelly
and
that
your
whole
jelly
thing,
or
maybe
it
is
but
you're
using
in
the
end,
react
as
I
as
I
say.
As
you
wrote
in
in
the
guitar
chat.
C
C
Next
I,
don't
they
have
I,
think
yeah
web
components,
maybe
a
good
thing
to
trial
out
a
bit
more
if
they're
being
tried
a
bit
in
the
Jenkins
project
and
they
would
allow
you
to,
you
know,
write
and
more
modern
front-end
technology
for
the
components
suddenly
say:
I'm
not
sold
on
material
design.
I
think
that
would
be
a
big
change
and
I.
Don't
really
think
it's
the
best
fit.
C
E
Yeah
I
would
Echo
what
Tim
said
about
bringing
plugins
along
for
the
ride.
I
think
if
we,
if
we
start
a
concerted
effort
to
migrate
the
entire
ecosystem
to
a
new
framework,
I
think
that
would
be
great
and
also
a
lot
of
work,
but
having
only
part
of
the
ecosystem
in
one
Paradigm
and
the
rest
in
a
different
Paradigm,
leads
to
more
complexity
and
makes
maintenance
more
difficult.
So
I'd
rather
stay
in
one
Paradigm
overall
than
to
introduce
multiple
paradigms.
A
G
What
what
actually,
what
is
Will
somebody
explains
me
right
now.
For
example,
there
is
already
a
movement
by
getting
let
by
Gavin
right
and
Jenkins
IO
to
actually
Expose
and
use
web
component
right.
So
basically,
for
me,
it
would
be
actually
interesting
where
he
actually
is
using
them
and
where
we
can
see
the
usage
for
it
right.
G
So
it
is
actually
to
to
try
to
actually
see
because
I
I
feel
there
is
a
need
to
actually
enhance
the
the
whole
UI
right
and
I
actually
had
been
one
of
those
to
actually
roll
blue
ocean
back
in
the
day
and
I
actually
always
said.
We
need
to
split
front
and
from
backhand
in
the
end.
We
didn't
do
that
far
enough
right.
So
what
right
now
is
seeing
is
the
big
problem
of
blue
ocean
right.
E
The
the
tables
to
dips
migration
could
be
one
example
of
how
to
do
that
in
a
compatible
way
where
prior
to
introducing
divs
plugins
were
prepared
to
support
both
tables
and
divs
during
a
transition
period,
and
at
that
point,
once
enough,
plugins
have
been
prepared.
We
could
make
the
change
in
core
to
switch
to
divs
and
then
eventually,
once
those
plugins
adjusted
their
Baseline
to
a
newer
core
version
that
used
divs,
then
the
tables
code
could
be
deleted
and
that's
a
general
pattern
that
we
use
whenever
we
make
an
API
change
that
affects
plugins.
A
A
A
Okay,
next
next
topic,
then,
was
the
ux
Futures
talk
this
one
Jan
I
just
wanted
to
highlight
that
Jan
and
Tim
are
presenting
this
talk.
Torsten
already
alluded
to
it.
It's
been
presented
once
in
Europe
time
zone.
Today
it
will
become
available
later
today,
as
well.
In
the
U.S
time
zone,
go
to
the
devops
world
site.
B
A
G
Just
just
one
quick
question
so
so,
basically,
you
know
outside
Cloud
peace
has
their
their
own
plugins
and
and
sometimes
their
hook
in
in
predefined
Dome.
That
we
actually
know
exist
right
and
and,
for
example,
the
header
we
have
a
customer
header
so
seeing
the
future.
For
example,
what
would
it
make
for
our
plugins
to
be
comfortable
against?
You
know
the
next
version
of
Jenkins.
So
basically,
as
we
talk
the
effort
on
actually,
you
know
opting
in
opting
out
of
something
like
that.
B
Yeah,
this
is
still
lots
of
very
early
days
to
kind
of
those
sorts
of
things.
Really
it's
a
kind
of
implementation
of
well
how
we'll
Implement
such
a
design
is
still
to
be
kind
of
answered,
but
I,
don't
it
should
be
just
as
customizable
as
what
we
have
now.
So
if
we
have
Cloud
B
to
kind
of
looking
at
Saturn,
I
mean
they'll
still
be
supported
in
the
future.
B
G
B
B
A
Okay,
next
next
topic,
then
was
open.
Ux,
regressions
and
I
have
to
stop
sharing
my
screen
briefly,
because
anytime,
I,
open
jira
in
a
public
environment,
I'm
an
administrator-
and
there
are
things
I-
can
see.
That
may
not
be
good
to
show
so
just
a
minute
I'm
going
to
stop
sharing
briefly
open
that
and
see
that
it's
okay.
A
A
A
A
Instead
of
using
a
link
that
I
became
conditioned
to
with
my
oh
pick
that
to
refresh
the
page
so
Jan
I
think
it's
intentional
I
know
it's
intentional
and
and
I
I,
like
Tim's
observation
that
really
we
probably
shouldn't
go
back
on
that,
because
it's
been
reviewed
by
four
people
who
are
very
skilled
in
that
area
and
just
go
forward
any
comments
you
want
to
give
on
that.
One
I.
D
G
D
We
remove
that
feature.
Some
time
ago
there
was
a
link
with
auto
refresh
that
was
there
for
10
years.
It
took
I
think
two
years
for
the
pull
request
from
Daniel
to
be
murdered
after
three
or
four
attempts.
So
if
we
can
avoid
this
kind
of
internal
video,
it
will
be
better
for
the
sanity
of
some
people.
Yeah.
A
G
So
is
it,
what
is
is
there
is:
is
there
in
an
AI
rule
to
not
have
like
a
refresh
button
on
a
page,
because
I
think
Marx
use
case
is
valid
right
and
you
may
want
to
expose?
Actually,
you
know
just
a
button
floating
somewhere
in
the
menu.
For
example,
it's
like
refresh
the
the
page
somebody
not
finds
the
refresh
button
on
the
browser.
You
know
what
I
mean.
A
Yeah,
at
least
for
me,
I
think
I
I'm,
not
sure
that
the
refresh
button,
because
the
refresh
button
will
naturally
be
in
a
different
location
and
those
years
of
physical
conditioning
that
I
developed
to
click.
To
a
specific.
This
is
really
embarrassing
to
say
it.
This
way
right.
It
sounds
like
I'm,
a
monkey
or
I'm
I'm,
a
pavlovian
dog,
but
the
reality
is
I'm
I'm
sort
of
conditioned
to
look
for
it
in
a
certain
place.
A
So
so
toasted
I'm
hesitant
to
lobby
for
a
refresh
button
just
because
I'm
not
sure
where
it
would
go
on
the
on
the
layout
and
if
it
goes
there
I
would
expect
Somebody,
Like
Jesse
Glick
would
remind
me
Mark.
You
should
be
submitting
bug
reports
for
things
that
are
not
updating
themselves
with
Ajax
automatically
and
and
I
have
to
shake
my
head
and
say
Jesse's
right,
I
should
submit
bug
reports
for
those
things
and
I
haven't.
A
G
E
Nine
five,
eight
seven
so
I
think
these
are
the
two
most
reported
from
users.
Both
of
these
have
been
reported
on
the
change
log
voting
page
and
the
first
one
is
I.
Think
a
one-line
fix,
there's
a
variable
that
appears
to
have
been
renamed
by
accident
to
a
non-existent
variable,
which
is
turns
out
to
be
null
if
it
doesn't
exist
in
jelly.
So
it
causes
a
null
pointer
exception.
H
E
E
The
pipeline
console
this
is
another
one.
The
users
have
been
voting
that
has
come
up
on
the
change
logs,
so
people
seem
to
not
like
the
fact
that
tailing
the
log
creates
a
lot
of
bouncing
in
recent
releases
by
tailing.
The
log
I
just
mean
scrolling
to
the
bottom
and
watching
build
live,
which
is
I.
E
Think
a
lot
of
us
have
done
this
when
we're
nervously
waiting
for
test
results
to
come
in
and
we're
just
tailing
the
end
of
the
log
file,
and
so
I've
identified
the
line
of
code
that
causes
this
regression.
It's
it's
one
line
so
again,
I
think
this
would
be
another
one
that
users
would
be
happy
to
see
fixed.
A
B
A
Okay,
and-
and
that's
that's-
that's
relative
to
that's
why
this
is
caused
by
link-
should
have
told
me
something
there
it
that
was
this
this
this
issue,
this
ticket
is
the
one
that
shows
the
transition
from
a
top
level
bar
to
Across
the
left
hand
side
and
that
change.
Yeah,
I
I
really
like
that
change,
but
it's
using
smooth
scroll
to
get
to
navigate
around
yeah
I,
see.
B
E
A
A
All
right
so
last
topic,
I
had
then
was
hacktoberfest
results.
Jean-Mark
Mason
is
prepping
a
blog
post.
If
there's
anything,
you
would
like
to
be
sure
is
included
in
that
blog
post
of
places
where
you're
aware
someone
from
the
hacktoberfest
contributing
community
did
something
that
really
helped
the
Jenkins
project.
We
would
love
to
put
that
story
in
there
I'm
aware
of
the
content,
security
policy,
improvements,
vadek
and
the
security
team
created
16
tickets
to
to
encourage
people
to
submit
content
security
policy.
Improvements
for
Jenkins,
plugins
and
12
of
those
tickets
were
resolved.