►
From YouTube: JupyterLab Team Meeting - October 19, 2022
Description
A meeting to share and discuss features, ideas, issues, and pull requests in JupyterLab and other Jupyter frontends. This meeting is open to anyone and everyone.
Join future calls via the Jupyter community calendar: https://docs.jupyter.org/en/latest/community/content-community.html#jupyter-community-meetings
Notes for upcoming meetings can be found on the agenda: https://hackmd.io/Y7fBMQPSQ1C08SDGI-fwtg
Past notes can be found on the JupyterLab team compass: https://github.com/jupyterlab/team-compass/issues/152
A
Hello
and
welcome
to
the
October
19th
weekly
Jupiter
lab
call.
Today
we
have
12
people
on
the
call
and
if
you
just
joined,
please
find
a
link
to
the
agenda
in
the
chat
and
please
sign
in
and
if
there's
something
you'd
like
to
talk
about.
A
Please
add
a
bullet
point
and
if
there's
something
that
will
require
more
than
let's
say
five
minutes,
please
add
it
to
the
additional
discussion
section
at
the
bottom.
Given
that
we
have
a
fairly
small
cohort
today
and
given
that
the
agenda
looks
light
I,
don't
imagine
we're
going
to
run
into
this
problem.
So
if
there's
something
you've
been
holding
off
on
showing
people
now's
both
a
good
time
and
a
bad
time,
it's
a
bad
time,
because
the
call
is
lightly
attended.
A
A
A
Is
welcome
to
the
publishing
steps
there
is
a
and
it
is
all
Jupiter
release
or
based
these
days
and
it's
a
minor
version
release.
So
hopefully
it
is
not
too
big
a
deal.
Oh
hey!
There's.
C
B
So
I
can
do
that.
Then
I
don't
have
the
bandwidth
to
like
thoroughly
test
it
or
anything.
But
I
have
the
bandwidth
to
like
walk
through
the
releaser
and
make
sure
it
gets
published.
A
Okay,
great
is
there
anything
that
you
think
you
need
to
know
beforehand,
I
think
the
release.
A
Awesome
yeah,
so
it's
not
really
a
different
process.
You'll
just
be
picking
you'll
just
be
doing
a
minor
instead
of
patch.
So
fair
enough
great!
Thank
you!
So
much
Alex!
That's
awesome!
A
Okay!
Let's
wait
for
Isabella
to
join
us
later
on
the
call
and
let's
go
to
Alex.
B
Yeah,
this
is
more
just
an
FYI
I.
Finally,
have
the
bandwidth
again
to
jump
back
into
what
I've
dubbed
launcher
2.0,
which
we
last
discussed
in
March,
so
kind
of
an
FY
to
the
community,
I'm
jumping
back
into
that.
If
anybody
has
like
strong
opinions
on
UI
or
ux
or
features
that
they
want
to
see
in
a
updated
redesigned,
modularized
launcher.
B
Please
put
them
on
that
issue
that
I've
linked
so
I'm,
aware
of
them
as
I
begin
work,
and
if
somebody
really
wants
to
help
me
on
it,
they
can
ping
me
on
on
that
issue
or
on
getter
and
I
can
see
how
I
can
break
it
up
so,
but
that
is
something
I
would
really
like
to
get
done
this
calendar
year,
if
possible,
we'll
see
how
my
bandwidth
continues
to
give
me
space
for
that.
B
A
B
I'm
targeting
it
targeting
as
a
non-blocker
for
4.0,
if
I
can
get
it
done
by
4.0,
I
would
love
to
get
it
in,
but
I
am
intending
to
write
it
in
a
way
that
it
is
backwards
compatible.
So
it
is
essentially
an
opt-in
beta
feature,
okay.
Well,
so
it
could
technically
go
out
in
4.1
with
the
way
I
plan
on
writing.
It.
B
Plan
on
extending
them
adding
adding
essentially
overloading
the
API.
So
if
you
call
the
old
apis,
it
will
just
add
it
will
add
functionality
close
to
or
similar
to
what
is
currently
on
the
launcher
to
the
new
launcher.
If
you
have
the
new
launcher
turned
on,
if
you
don't
have
the
new
launcher
turned
on,
it
will
still
just
show
the
old
launcher.
B
How
I'm
intending
to
write
it
specifically
because
I
don't
want
to
I,
don't
know
if
I'll
have
the
time
to
get
it
in
by
4.0
and
I.
Don't
even
know
if
it's
something
we
want
to
just
completely
overhaul
in
4.0
and
I'd,
rather
have
that
flexibility
of
essentially
deprecating
the
old
launcher
for
a
version.
A
Sure
so
I
guess
there's
two
different
ways
of
thinking
about
this.
One
is
the
strictly
semantic
versioning
of
the
API,
which
I
think
we
are
committed
to
the
other
is
it
could
be
a
radically
different
experience,
but
if
the
apis
are
Backward,
Compatible
you're
still
good
to
release
that
as
a
4.1
feature,
so
yeah
cool
okay,
thanks
for
coming
back,
though
that's
great
yeah
any
questions
comments
on
this
I
know
people
have
opinions
about
launcher.
C
Because
I
want
to
go
in
there
and
give
some
feedback
what's
in
scope,
is
it
just
about
the
extensibility
looking
at
that
GitHub
issue,
or
are
you
kind
of
like
seeking
high
value
ideas
for
the
launcher
in
general,.
B
Both
my
initial
goal
is
to
Simply:
rework
it
so
it's
more
extendable,
so
you
can
do
more
than
just
add
those
cards
like,
so
you
can
essentially
make
it
kind
of
like
iOS
or
Android,
where
you
don't
just
have
to
have
an
icon.
That
opens
up
an
app,
but
it
can
do
more
than
that
like
so
you
could
add
a
widget
or
something
to
it.
B
So
that's
going
to
be
kind
of
my
starting
point
is
just
getting
the
same
functionality,
but
with
a
more
versatile
way
of
adding
things
and
then
thinking
about
adding
new
features
to
it.
After
that
kind
of
basis
is
written
and.
B
Or
in
this
case,
I
was
thinking
of
like
widgets
for
iOS,
but
I
have
not
yet
figured
out.
How
what's
the
word
like
what
level
the
different
extendable
objects
inside
the
launcher
is
going
to
be?
Are
they
going
to
be
actual
Jupiter
lab
widgets?
Are
they
going
to
be
widgetlets?
Are
they
gonna,
like
I,
haven't
figured
that
part
out
yet.
C
A
You
could
introduce
a
concept
the
way
we
introduce
the
concept
of
the
main
area
widget,
you
could
introduce
the
concept
of
a
you,
know,
launcher
card
or
launcher
widget
or
something
like
that.
Where
the
default
one
accepts
the
same
arguments
that
the
current
launcher
does
and
sort
of
under
the
covers
gets
instantiated.
It
looks
pretty
much
the
same,
but
a
user
or
not
user.
A
But
an
extension
author
could
choose
to
implement
one
that
is
more
sophisticated
and
as
long
as
it
is
inheriting
from
launcher
widget
it
shows
up
and
I
don't
know,
has
an
animation
inside
or
something
cool
or
or
whatever,
whatever
you're
sort
of
imagining
I
think
introducing
a
thing
that
others
inherit
and
can
extend.
My
might
be
one
Avenue
to
take
here,
but
that's.
B
Really
close
to
the
idea
that
I
had
in
my
head,
but
specifically
making
an
extendable
cost
for
it
is
a
solid
Next,
Step
yeah,
no
I
I
like
that
using
an
extendable
class,
specifically
I'll,
probably
launch
chop
of
that
point.
A
So
Gabriel
points
out
that
animations
have
side
effects
that
we
should
consider
yeah,
but
how
else
are
we
going
to
make
Jupiter
lab
irritating
to
use?
Is
the
question
under
construction
everywhere
right
cool?
Any
other
questions
comments
about
this
I
think
this
is
exciting.
F
This
could
be
a
segue
into
Isabella's
thing,
I.
Think
if
I
understand
what
she
wrote,
let
me
check
look
at
that
real
quickly,
yeah
yeah,
so.
F
She's
been
tackling
the
accessibility
issue
of
Reflow,
it's
related
to
zoom
or
the
Jupiter
lab
UI
and
there's
a
lot
in
there,
but
like
one
of
the
things
that
we
we've
been
looking
at
is
like
having
a
button
somewhere
in
the
user
interface
that
pulls
up
the
command
palette.
I
think
we
want
to
try
to
work
on
putting
more
and
more
stuff
explode,
I
think
there's
a
number
of
commands
currently
that
are
not
in
the
command
palette,
and
we
want
to
get
as
many
as
we
can
into
there
and
then
also
just
surface
surf.
F
I
don't
know
if
this
is
related
at
all
to
the
launcher,
but
I
thought
that,
just
when
I
hear
the
launcher
and
I
hear
like
like
making
it
do
more
things,
I
thought.
Oh,
this.
Those
things
are
maybe
kind
of
sort
of
tangentially
related.
A
There
could
be
an
area
in
the
launcher
or
a
button
in
the
launch,
or
something
like
that.
That's
true.
We
used
to
have
a
sidebar
button
that
brought
up
the
command
palette
when
the
command
palette
was
a
sidebar
item.
A
B
Could
could
be
a
pretty
low
hitting
fruit
to
have
the
First
new
launcher,
widget
be
putting
the
command
palette
like
a
command
palette
button
or
like
bar,
or
something
at
the
top
of
the
launcher?
That
just
allows
you
to
click.
It
open
right
from
the
launcher,
so
it.
A
That
that's
true
on
one
model
but
on
the
other,
you
know
and
you'll
you'll
I,
think
think
about
this
more
deeply
than
the
rest
of
us,
but
that's
pretty
precious
real
estate
too,
because
if
this
is
sort
of
first
view
of
what
you're
going
to
do
inside
this
application,
especially
if
you're
new
to
the
application,
it
probably
ought
to
lead
you
into
creating
a
notebook
right.
I
mean
that
probably,
is
the
first
experience
you
should
have
inside
lab
if
you've
never
used
it
before,
but
I'm,
not
exactly
sure,
yeah
I'm,
not
exactly
sure.
A
Where
else
a
button
for
command
pal
could
go.
William
says
in
the
chat
in
Jupiter
classic
there
was
a
command
power
button
right
in
the
middle
of
the
button
bar
it's
very
discoverable
and
Mike
says
in
the
chat.
There
was
an
idea
of
showing
press
X
to
open
command
palette
on
the
monitor,
yeah,
that's
similar
to
what
you're
saying,
Alex
and
Ely
is
saying:
isn't
there
already
a
button
that
pulls
up
the
command
palette
in
lab
yeah?
That's
right,
it
used
to
be
in
the
sidebar,
but
that
got
changed
to
a
modal.
A
So,
right
now
each
extension
author
has
to
make
a
decision
about
which
commands
go
in
the
menus
which
menu
they
go
into,
which
commands
have
keyboard
shortcuts,
what
the
shortcuts
are
and
which
commands
get
added
to
the
palette
and
what
category
they
fall
under
and
it's
I
wouldn't
be
surprised
at
all.
If
that
needs
an
audit
just
for
the
core
commands
to
see.
If
we
truly
have
the
commands
that
we
want
in
the
palette
and
I,
don't
really
even
know
what
the
criteria
are
to
like.
How
do
we
know
which
commands
belong
there?
F
I,
do
think
that
if
you
look,
if
you
look
at
vs
code
and
you
look
at
their
docs,
they
think
of
the
command
pal,
and
this
is
the
way
we've
been
thinking
about
it
as
well.
If
they
look
at
the
command
palette
as
an
accessibility
feature,
so
they
try
to
surface
as
much
functionality
as
they
possibly
can
into
the
command
palette,
and
so
I
think
that
I,
like
that
way
of
thinking
about
it,
I,
don't
know.
That's
all
I
can
say
at
the
moment.
B
Yeah
I
I
agree
I
like
that
thinking
about
it,
but
I
also
believe
that
we
have
not
been
thinking
about
that
that
way
in
Jupiter
lab
I.
If
we
the
yeah
what
Darian
said
about
doing
an
audit,
we
really
need
to
audit.
What's
in
the
palette,
because
I
remember
the
last
time
I
opened,
the
palette
I
was
like.
B
Why
are
half
these
things
in
here
and
why
are
so
many
things
not
in
here
I
think
that
when
it
was
in
the
sidebar
I
liked
how
it
was
how
it
was
when
it
was
in
the
sidebar
I
feel
like
when
we
moved
it
out
of
the
sidebar.
We
should
have
reevaluated
what
was
in
there
for
the
new
usage
and
I.
Don't
think
we
did
I
think
we
just
kind
of
moved
it
and
left
the
same
list
of
items.
A
Yeah
we
didn't
modify
the
content
of
it.
We
just
modified
the
user
experience
of
it,
so
I'm,
looking
at
vs
code
and
I
see
oh
their
first
item
under
view.
In
the
view,
menu
is
also
command
palette
and
it
has
a
keyboard
shortcut,
but
I
don't
see
a
button
in
the
interface
to
launch
the
command
palette
there,
either
and
I.
A
Think
it's
probably
because
since
theirs
also
is
a
mobile
and
not
a
sidebar
thing
and
the
main
place
where
they
have
buttons
is
either
the
status
bar
or
the
sidebar
they've
chosen,
not
to
put
one
there
but
I'm,
not
I,
don't
know
I
mean
maybe
the
the
difference
between
vs
code's
menu
system
and
ours.
A
In
addition
to
being
native,
is
at
least
on
a
Mac
and
I
I,
don't
know
if
it's
cross-platform
it
has
the
built-in
search
in
The
Help
menu,
which
always
helps
me
because
right
now,
I
didn't
know
that
it
was
going
to
be
under
the
view.
Menu
like
it
is
in
lab,
I
searched
for
it
and
discovered
that
it's
there
that's
a
feature
that
seems
both
hard
to
implement
and
super
useful.
A
F
Just
I
wasn't
quite
sure
the
last
point
you
made
Darian.
You
said
something
I
like
something
got
lost
and
oh.
A
So
on
Mac
apps
The
Help
menu.
The
first
item
is
always
a
search
bar
if
author
of
the
application
has
implemented
it
I
think,
and
so
that's
how
I
discovered
where
the
command
palette
item
was
in
vs
codes,
menu
system
and
I've
always
wished.
We
had
that
feature
in
Jupiter
lab,
because
you
can
launch
the
palette
and
search
for
stuff
and
that's
really
great,
but
people
who
are
used
to
using
Macs
probably
look
under
help
hoping
to
find
such
a
thing.
A
D
A
D
I
got
it
Mike
and
I
just
shared
a
an
issue
that
I
opened
in
the
last
week
to
add
a
Mac
style,
help,
menu
and
I
think
this
was
Brian
Granger's
suggestion
that
we
opened
one
to
add
just
that
functionality,
the
the
Mac
help
menu
will
also
visually
highlight,
and
you
can
see
a
screenshot
in
the
issue
where
the
item
is
yeah.
B
D
A
C
A
A
I
think
and
I
don't
want
to
speak
for
Alex.
So
as
soon
as
I
finish
this
please
Alex,
chime
in,
but
I
think
that
no
new
Concepts
that
don't
currently
exist
in
lab,
like
environment,
doesn't
exist
as
a
com
as
a
real
Concept
in
lab
can
be
introduced
in
this
in
this
pass
of
a
new
launcher.
But
maybe
Alex
has
more
ambitious
plans.
I,
don't
know
what
are
you
thinking
else?
No.
B
I,
using
the
terms
that
you
were
using
this
would
be
a
facelift.
The
reality
here
is
taking
the
apis
from
an
API
level.
The
idea
is
to
make
it
easier
to
add
more
complex
things
to
that
launcher
than
just
run
this
command.
Here's
an
image
and
a
name
for
it
and
create
a
tile,
but
to
allow
users
to
using
the
term
daring
use
to
create
a
launcher
widget
and
that
launcher
widget
should
theoretically
be
able
to
do
lots
of
things
so
maybe
down
the
line.
B
B
Yeah
I
I'm,
theoretically
not
planning
on
adding
very
many
if
Amy
at
all
new
launcher,
widgets
Beyond,
what's
currently
of
what
it
currently
does
like
the
obviously
the
first
launcher,
which
I'm
going
to
create
is
like
command
tile
like
here's.
What
we
have
now
and
past.
That
is
very
much
follow-up
and
up
to
the
discretion
of
people
who
want
to
implement
new
things.
E
A
Yeah
so
there's
also
the
the
behavior
of
the
launcher,
where,
if
there's
nothing
on
the
screen,
it
opens
up
if
you're
on
the
screen
and
there's
a
launcher
and
you
use
it
to
launch
something,
it
remains
open.
But
then,
if
you
close
it
and
you
have
something
else,
it
doesn't
come
back,
and
so
some
of
those
behaviors
are
intentional,
but
they
may
benefit
from
a
revisit
as
well.
Is
that
within
the
scope
of
what
you're
thinking
about.
B
That
is
that
stuff
is
not
in
scope
of
what
I'm
working
on
I'd
say.
It's
parallel,
okay,
really
cool
same
package,
but
if
people
want
to
work
on
that,
I
could
say
that
we
wouldn't
even
step
on
each
other's
toes
doing
those
in
parallel.
A
B
And
I've
recently
touched
those
things
when
we
did
a
bunch
of
work
with
the
with
the
toolbar
changes,
the
tab
bar
changes
and
then
also
when
we
tinkered
with
the
file
browser
and
how
it
related
to
launcher.
We
I
forget
I,
think
the
launch
in
the
file
browser.
We
switched
the
dependency
order
on
those
recently
and
then
we
also
added
the
plus
button
to
the
tab
bar
that
now
opens
the
launcher
and
when
we
did
both
those
features,
we
tinkered
with
a
lot
of
the
stuff.
A
Got
it
I
think
also
there's
a
PR
that
is
approved,
but
not
yet
merged
by
Nikola,
with
a
change
to
the
behavior
of
the
simple
mode
where
it
does
not
load
the
content
of
your
workspace.
In
the
background
and
one
of
the
things
that
he
does,
there
also
touches
this
launch
of
behavior.
Although
I
don't
I,
don't
remember
exactly
what
I
think
it
was
maybe
to
not
create
a
launcher
in
the
background
or
something
I'm,
not
I'm,
not
100
sure,
but
yeah
this
this.
A
This
piece
of
it
is
sort
of,
as
you
say,
parallel
to
the
content
of
the
launcher,
so
cool.
Okay,
great,
why
don't
we
move
on
to
Gabriel.
F
I'm
muting,
okay,
I,
am
doing
what
is
valid
normally
does
on
this
meeting
and
letting
you
know
that
there
is
an
accessibility
meeting
15
minutes
after
the
hour
after
this
meeting
and
are
welcome
to
join.
A
Great,
thank
you
very
much.
I
think
I
might
miss
that
because
of
the
location
and
event
I
am
in,
but
I
am
going
to
miss
it
and
feel
bad
about
it.
F
I
noticed
that
there
is
a
suspiciously
similar
looking
background
on
both
your
video
and
on
the
Jason
grouts.
Today,.
A
A
A
About
that
all
right
moving
on
is
this
I'm
suffering
from
jet
lag?
Yeah
no
I
mean
he
did
he.
He
he's
from
eight
time
zones
away.
No
one
can
blame.
Him
is
Isabella
on
the
call.
Currently,
no
I
don't
think
she
is.
Okay
is
someone
else
who
has
been
working
on
accessibility,
comfortable
discussing
this
point.
F
I
I,
just
I,
think
Isabella
will
be
in
a
much
better
position
to
go
over
those
things
over
this
issue.
In
this
comment
because
she
wrote
it
herself
and
has
been
really.
F
A
Then
I
I
will
do
that
thing
that
I've
been
soliciting
multiple
times,
which
is
if
anyone
on
this
call
is
interested
in
hosting
this
call
and
have
been
holding
back.
Please
stop
holding
back.
If
there's
anything
I
can
say
or
do
that
will
make
it
easier
for
you
to
host
this
call,
I
will
say,
and
do
that
thing.
A
Oh
I
recorded
that
and
that's
a
pretty
strong
claim
within
reason.
Yeah.
A
Not
just
within
reason
within
reason
and
within
my
capabilities.
Those
are
the
caveats,
but
no
really
I
mean
I
think
it
would
be
great.
A
Isabella
has
volunteered
to
do
some
of
them
and
I
appreciate
it,
but
you
know
there's
plenty
of
calls
where
there
is
no
one
who
has
volunteered
to
do
it
and
I
am
happy
to
continue
doing
them.
I,
don't
I,
don't
have
a
problem
doing
it,
but
I
also
think
it's
nice
to
hear
other
people's
voices
so
yeah
consider
that
it's
a
pinned
issue
on
the
Jupiter
lab
team
Compass.
It's
very
easy
to
find,
doesn't
get
that
much
traffic,
but
you
know
send
some
traffic
that
way.
A
Please
Okay
cool,
so
I
guess
before
Isabella
joins,
it
seems
like
we've
had
some
no
I'm
afraid
I
cannot.
Babysit
sorry
I'm
like
well
depends
on
how
many
calls
you
take
actually
so
we
that
that's
negotiable,
but
there's
some
conversation
in
the
chat
still
going
about
the
the
command
palette.
Discoverability.
So
William
pointed
out
that
having
the
button
in
a
discoverable
place
has
been
good,
both
in
classic
and
in
cocalc
and
yeah
I.
I
guess:
do
you
collect
metrics?
G
I
don't
collect
metrics,
but
I
can
just
tell
from
you
know,
support
feedback
that
people
do
regularly
find
the
palette
and
you
know
explore
it.
So
it's
extremely
discoverable
and
also
if
people
are
used
to
Jupiter
classic
it's
just
exactly
the
same
spot
as
interpreter
classic
and
plus
people
also
look
at
the
little
drop
down
for
cell
versus
markdown,
and
it's
right
next
to
that.
So
I
don't
know,
but
it
definitely
should
be
about
like
the
palette
should
be
very
much
about
Jupiter
notebooks.
G
If
it's
you
know
in
the
button
bar
for
Jupiter
notebooks,
whereas
in
Jupiter
lab
the
palette
that
you're
talking
about
might
be
much
more
General
than
just
Jupiter
notebooks.
So
there's
like
a
subtlety
there
I
think
between
the
two
potential
locations,
it'd
be
nice.
If
you
could
open
it
up
without
a
notebook,
open,
yeah.
E
G
A
A
That
is
much
more
about
notebook,
simply
because
it's
a
smaller
application
with
fewer
commands,
I,
don't
I,
don't
know
yeah,
as
Jason
said,
having
to
have
a
notebook
open
to
find
that
button
makes
sense
in
a
UI
where
you're
only
looking
at
notebooks,
really
in
a
UI
where
you're
looking
at
a
launcher,
when
you
first
open
it
or
a
code
console
or
a
text,
editor
Etc,
we
have
other
real
estate
that
it
could
potentially
go
in
I'm,
not
sure
put
a
button
on
the
menu
bar
right
next
to
the
help
menu
yeah
there
could
be
a
button
on
the
menu
bar
next
to
the
help
menu.
A
Help
menu
because
the
command
palette
itself
is
searchable,
except
it's
not
truly
analogous,
because
the
menu
system
has
different
commands
in
it
than
the
command
palette.
That's
not
one-to-one
exactly.
G
A
All
right
is
there
anything
anyone
else
would
like
to
discuss.
Oh
Mike
cool,
you
are
up.
E
So
I
created
a
pull
request,
which
we
moved
about
15
unused
styles,
some
of
those
were
unused
for
like
six
years
and
we
haven't
done
a
great
job
with
removing
features.
Removing
the
Styles
as
well
so
I
opened
a
proposal,
pull
requests
with
a
function
which
would
allow
you
to
check
for
unused
styles
on
runtime
in
kalata.
E
This
is,
of
course,
prone
to
generating
false
positives
and
false
negatives
for
various
reasons,
but
there
is
some
reasonable
defaults
in
the
track,
which
would
hopefully
be
useful,
and
if
we
think
that
this
is
a
good
approach
that
could
be
also
reused
by
extension,
Outdoors,
we
use
karata
to
test
the
extensions
and
hopefully
reduce
the
number
of
CSS
rules
which
we
have
and
to
not
use
largely.
E
This
is
relevant
to
the
performance
effort,
because
the
recalculation
of
styles
is
taking
a
very
long
time
for
large
notebooks
and
workspaces,
and
if
anyone
has
thoughts
about
this,
I
think
that
maybe
this
is
not
the
best
approach.
Maybe
there
is
a
better
way.
Please
leave
feedback
on
the
pull
request
and
Next
Step
that
I'm
thinking
of
is
trying
to
enforce
the
convention.
How
the
CSS
rules
should
use
classes
specific
classes
as
possible
to
reduce
the
style
matching
time
and
increase
performance.
A
E
That
the
first
sentence
has
a
link
to
another
Quest
I'm,
just
linking
that.
A
E
Just
just
to
clarify
it's:
it's
not
so
clever,
because
when
you
run
that
on
the
top
level
of
Jupiter
lab
up,
there
are
many
components
which
are
not
in
the
dome
currently
or
for
some
reason.
I
have
do
not
have
like
the
most
active
style
class,
so
you
will
have
lots
of
false
positives
and
I
I
ended
up
with
a
list
of
300
Styles,
which
I
just
went
for
mostly
one
by
one,
and
what
is
it
possible
that
if
this
is
not
used
anymore?
E
So
that
was
a
little
bit
of
manual
effort,
but
it's
much
easier
to
add
kind
of
regression
checks
when
limiting
those
to
one
components,
one
at
a
time.
So
the
the
other
pull
request
adds
this
regression
checks
to
the
file
browser.
So
we
can
see.
We
can
be
sure
that
in
the
future
we
are
not
adding
Styles,
which
are
not
used
or
removing
components
without
removing
Styles.
So
that
would
be
a
gradual
effort
where
we
can
decide
on
component
by
component
basis,
whether
that's
West,
closing
or
not.
I,.
A
A
E
Just
one
more
thing
to
do
the
lap
fresh
install
of
the
latest
version
has
about
8
000
styles,
and
only
a
small
fraction
of
those
all
right
come
from
our
Styles.
Most
of
those
are
like
in
in
three
point
x,
that
is
blueprint
in
the
latest.
This
isn't
awesome,
and
this
is
the
toastify.
C
E
E
H
H
Styles,
even
in
my
limited
CSS
Adventures,
so
I'd
be
very
happy
to
see
something
with
this
yeah.
A
G
It
might
be
look,
it
might
be
worth
comparing
it's
a
very
different
approach,
I
think.
But
it's
it's.
For
example
a
among
other
things,
a
webpack
plugins,
so
it
just
analyzes
your
source
code
when
doing
the
build.
So
it's
more
like
tree
shaking
your
CSS
rather
than
not
shipping
to
you
not
putting
CSS
in
your
source
code
at
all
that
you
don't
need,
so
it
solves
a
different
problem
but
might
be
relevant
for
comparison.
H
Hello
thanks
for
making
it
till
I
got
here
sorry
about
conflicting
calls.
First
thing:
I
need
to
remember
to
say:
Jupiter
Community
College
next
week,
all
are
welcome
I've
linked
to
the
Discord
and
discourse
in
the
agenda,
and
that
should
have
all
the
info.
You
need
I,
think
I
have
the
hack
MD
link
corrected
I.
They
must
have
changed
their
permissions
thing
because
I
keep
getting
on
like
everything,
I'm
doing
people
mentioning
anyway.
Let
me
know
if
there's
issues
with
that
hack,
md2
and
I
am
happy
to
fix
it.
H
Two
weeks
in
a
row
would
be
very
funny,
but
yeah
that'll
be
exciting,
so
hope
to
see
you
there
I
would
say,
feel
free
to
be
Halloween,
inspired
and
dress
up
if
you
wish,
but
I
have
no
super
exciting
plans.
Yet
to
probably
the
more
important
thing
there
has
been
an
issue.
That's
been
open
for
perhaps
an
embarrassing
to
me
amount
of
time
about
Jupiter
lab
in
kind
of
high
browser,
Zoom
situations.
H
This
is
an
accessibility
concern
for
those
of
you
that
don't
know
we
need
to
be
able
to
zoom
to
400
Zoom
according
to
Wiki
standards
and
I.
Believe,
Firefox
and
chrome
actually
support
up
to
500
percent.
Now,
if
you
open,
Jupiter,
lab
and
start
zooming
in
that
high
you're
gonna
have
a
fun
time,
but
maybe
not
the
most
productive
time
with
things
so
trying
to
come
up
with
some
solutions
around
what
a
better
experience
for
that
would
look
like
because
right
when
you
start
zooming
in
that
much,
we
have
limited
real
estate.
H
I,
don't
know
if
we
were
ever
at
the
point
of
additional
discussion.
So
the
first
thing
I
wanted
to
do
is
kind
of
flag
that
issue
for
you
all
and
request
some
feedback,
because
we're
gonna
have
to
do
something
pretty
opinionated
to
be
honest,
to
make
it
fit
at
that
size
and
I
know
that
can
be
controversial,
I'm
happy
to
walk
through
stuff
I.
Just
don't
want
to
take
more
time.
If
there's
other
discussions,
we
need
to
have
first.
H
Okay,
cool
well
I'm,
going
to
share
my
screen,
then,
on
this
issue
and
yes,
I,
agree:
Lane,
there's
a
lot
of
use
cases
where
I
would
just
like
Zoom
to
be
better
for
for
all
kinds
of
reasons,
but
I
wanted
to
give
some
feedback
on
like
where
the
threshold
was
coming
from.
H
So
right
now,
I
have
it
on
this
comment:
I
link
to
there's
kind
of
two
main
paths
that
I
see
they're
like
diverging
paths:
I
have
a
personal
feeling
we'll
end
up
somewhere
in
the
middle
kind
of
compromising,
but
either
we're
going
to
emphasize
the
document
right
so
like
this
is
in
the
main
document
area
or
we're
going
to
emphasize
the
rest
of
the
UI
surrounding
it
right,
like
all
the
stuff
that
isn't
the
document,
which
is
kind
of
what
we
currently
do
in
the
sense
that,
like
the
sidebar,
will
overtake
right
if
you're
going
to
zoom
in
so
with
this.
H
My
opinion
needed
to
take
is
when
you
start
zooming
in
you
actually
lose
the
sidebars
more
things
get
collapsed
into
the
command
palette.
You
can
have
the
command
palette
as
a
button,
which
I
know
is
something
that
has
been
requested
in
passing
before,
because
it's
only
kind
of
buried
in
a
main
menu
right
now
or
with
the
shortcut
so
you're
able
to
navigate
to
all
those
things
through
there.
The
UI
is
not
lost.
You
just
aren't
seeing
it
like
it's
not
taking
up
screen
real
estate.
H
The
main
problem
I
have
to
solve
with
that,
which
is
what
was
playing
out
and
I
wasn't
talking
about,
was
in
most
cases,
I
think
it
would
make
sense
to
have
kind
of
sidebar
items
that
are
not
dependent
on
the
document
area.
Take
up
kind
of
like
the
main
documentary
it
currently
does
for
the
settings
editor
or
the
launcher
or
technically
UI.
H
That
act
like
a
document
and
kind
of
mimicking
that
behavior,
but
for
things
like
the
table
of
contents
that
are
document
dependent,
we'd
have
to
come
up
with
another
kind
of
overlapping
solution,
because
it
does
rely
on
whatever
document
you
have
focused
and
then
the
other
approach
is
much
closer
to.
What
we
have
would
be
right.
Everything
just
start
zooming.
We
just
make
it
so
that
the
sidebars
are
going
to
intentionally
take
up
the
whole
Space
when
we
start
working
with
them
instead
of
kind
of
unintentionally
right
now.
H
They
just
push
things
out
of
the
way
and
aren't
optimized
within
the
space
like
there's
a
lot
of
kind
of
Maximum,
with
things
that
when
the
text
starts
to
scale
but
the
box
that
the
text
in
is
not
does
not
work
quite
right,
so
yeah,
so
those
are
kind
of
the
two
approaches
I'm
trying
not
to
throw
too
much
at
you
at
once.
There's
a
lot
to
work
out
within
both
of
those
approaches.
To
be
honest
with
you,
so
let's
see
what
are
we
seeing?
H
Oh
searching
for
the
scroll
issue,
yeah
yeah,
so
the
thing
I
will
say
that's
also.
One
of
my
concerns
is
I.
Think,
just
because
you're
zooming
in
doesn't
mean
you
should
be
getting
like
drastically
different
ux
right,
you're
still
in
Jupiter
lab.
That's
kind
of
one
of
my
fears
with
this
document
focused
where
we
lose
some
of
those
main
UI
areas
visually
and
just
have
them
packed
away
somewhere
else,
but
at
the
same
time,
maybe
this
is
controversial.
H
I
do
think
that
the
document
is
probably
the
main
focus,
I
kind
of
like
that.
It
allows
like
this
is
like
what
am
I
trying
to
say.
This
is
in
a
lot
of
other
information
dents
and
like
kind
of
document-centric
uis.
What
I'm,
seeing
a
lot
when
they
go
to
mobile
or
go
to
other
small
screens,
is
that
they
kind
of
put
the
document
first
and
put
the
UI
kind
of
the
side.
I
think
it
supports,
like
that.
You
can
read
and
not
need
to.
H
Edit
is
kind
of
the
motivation
right,
because
you
could
just
be
opening
this
to
read
very
zoomed
in
you.
Don't
actually
have
to
be
editing
it
and,
if
you're
not
editing,
all
these
things
are
kind
of
in
the
way.
At
the
same
time,
is
that
too
divorced
from
the
rest
of
the
Jupiter
lab
ux,
so
I'm
talking
a
lot
I'm
just
trying
to
throw
out
the
ideas
that
I've
been
thinking
of,
but
I
would
love
to
be
quiet
and
listen
to
you
now.
A
So
in
I
think
a
not
fully
thought
out
attempt
to
think
about
mobile
experience
before
Jupiter
Lab
3
I
added
this
attribute
on
the
application
class.
That
was
the
mode
whether
it
was
desktop
or
mobile,
and
the
idea
was
when
you
go
to
mobile
you're,
basically
using
simple
mode
that
ended
up
being
a
bit
too
simplistic,
but
the
premise
behind
it
was
exactly
what
you're
saying
and
given
that
many
people's
primary
experience
is
gonna,
be
Jupiter
notebooks
UI,
given
that
many
people's
primary
experience
in
lab
is
notebooks.
H
A
Yeah
I
think
it
was
closing
the
sidebars
and
removing
the
status
bar,
maybe,
but
it
probably
like
we
probably
can't
afford
to
get
rid
of
the
top
area
and
only
have
the
menus.
We
probably
you
know
this
and
even
the
menus
themselves.
I,
don't
yeah
I'm,
not
sure.
There's
definitely
thought
that
needs
to
go
into
that.
I
just
meant
the
actual
API,
for
it
was
probably
wrong
having
a
mode.
A
That's
only
two
choices
and
sitting
on
the
top
level
applications,
probably
not
right
and
I'm
afraid
of
removing
it
because
Jupiter
front
end
is
now
used
by
a
bunch
of
other
things,
but
we're
not
really
using
it
that
way.
Anymore.
Really,
I
know
other
people
have
thoughts
about
this.
So
please,
chime
in.
H
C
Yeah,
this
is
just
like
a
demo
because,
like
you
were
asking
what
should
we
prioritize?
And
it's
like
it's
pretty
clear
that
you
know
the
notebook
has
no
space
to
breathe
and
this
isn't,
like
you
know,
240
and
the
only
time
I
would
be
doing.
This
is
because
I
want
to
share
content
in
the
notebook
with
someone
else
like
during
a
demo.
C
One
thing
that
immediately
comes
to
mind
is,
like
you
know
these
pandas
tables,
like
that's
the
number
one
output,
so
I
would
say,
and
they're
they're
not
like
resized
in
any
way
the
column
widths
and
then
a
major
problem
that
I
haven't
brought
up
because
it's
kind
of
nitpicky
is
you
in
output
cells.
H
C
H
A
I
think
this
does
show,
though,
so,
for
example,
the
the
counter
prompt
being
on
the
left
side,
probably
stops
making
sense
when
you're
so
zoomed
in
it
probably
makes
better
sense
to
use
vertical
space
instead
of
horizontal
space
there
and
the
margins
disappearing,
like
you,
don't
need
the
the
Shadow,
the
Box
Shadow
and
the
margins
on
the
document
when
you
have
such
little
space
to
work
with
and
there's
probably
a
lot,
because
there's
only
three
minutes
left
on
the
call
I'm
going
to
stop
recording.