►
From YouTube: JupyterLab Team Meeting - 17 May 2023
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/170
A
Hello
and
welcome
to
the
May
17th
Jupiter
lab
weekly
call.
Today
we
have
12
people
on
the
call.
It's
sad.
Let's
hope
more
come
next
week
or
maybe
more
come
late,
that's
cool
too.
A
So
last
week
was
Jupiter
con.
We
didn't
have
this
call
and
some
people
on
the
call
were
there.
So
maybe
that's
something
I'll
talk
about,
but
aside
from
that,
I
think
there's
a
lot
to
talk
about
this
week.
So
why
don't
we
get
started
with
Jason
w.
B
Hey
good
morning
good
evening,
everyone
so
I,
unfortunately,
couldn't
make
it
to
Jupiter
con,
but
I
did
see
yesterday
or
on
Monday
that
Jupiter
lab
400
is
out.
Congratulations
to
everyone.
B
I
started
drafting
a
blog
post
based
heavily
on
the
400
change
log
I
linked
a
draft
there
that
should
hopefully
be
World
visible,
so
you
guys
can
all
make
comments
or
edits
or
something
there's
also
a
team
Compass
issue
that
I
think
Jeremy
opened
I'd
like
to
get
this
like
out
on
the
public
blog
soon.
So
then
we
can
blast
it
out
on
social
media.
We
can
start
a
thread
on
discourse
because
when
I
was
doing
triage
yesterday,
I
can
now
start
asking
hey
Jupiter
lab
400
is
out.
B
Do
you
still
see
this
on
4-0?
So
then
we
can
kind
of
identify
what's
supported
and
what
isn't
I
also
made
a
couple.
Other
changes,
but
thoughts
about
like
promotion
for
4-0.
C
I
think
the
one
thing
that
would
be
great
to
get
would
be
a
final
release
of
the
Jupiter
collaboration
out
first,
so
that
there
is
RTC
you
know
available
for
those
who
like
to
try
lab
four
and
I
think
it
should
be
there
very
soon
that
maybe
next
week
so
I
guess
the
blog
post
could
go
out,
maybe
earlier
next
week,
but
maybe
one
two
weeks
yeah
some
that
sounds
reasonable.
D
I
love
the
outline
of
the
draft
and
I,
don't
that
it
includes
pictures
which
were
added
since
3.0
was
released
and
not
only
features
which
are
the
new
message.
3.6
and
I
think
this
is
also
important
to
update
our
general
documentation
and
maybe
course
referenced
our
potential
kind
documentation
all
the
ideas
that
we
collect.
Indeed,
look
for,
strap,
which
is
change
lock.
It's
really
missing
some
organization
and
and
pretty
screenshots
with
the
new
features
foreign.
B
Yeah
I
mean
the
blog
post
is
meant
to
be
more
promotional
than
comprehensive,
like
I
didn't
put
in
screenshots.
Yet
someone
shared
the
the
PowerPoint
slides
from
Jupiter
con
and
the
way
that
Medium
lays
out
posts
doesn't
really
give
us
a
great
deal
of
flexibility,
and
so
we
we
need
to
be
kind
of.
If
we
want
to
include
lots
of
images,
we
may
need
to
cut
down
on
the
amount
of
text.
We
include
and
defer
to
the
changelog
for
a
lot
of
the
more
meaty
topics.
D
A
Thanks
for
working
on
the
blog
post,
yeah
I
think
as
we're
getting
reactions
and
some
issues
come
in
to
4.0
we're
getting
some
feedback
about
what
a
large
scale
upgrade
might
look
like
for
people.
So
if
we
see
issues
come
in
that
seem
like
low
hanging
through
high
value
and
we
can
publish
patches
and
stuff
the
more
of
those
are
integral
to
what
you
get
when
you
pip
install
Jupiter
lab
when
the
blog
post
goes
out,
you
know
the
softer
aligning.
A
A
Sure
yeah,
to
be
honest,
that
sort
of
tone
is
really
dispiriting
and
I
could
have
imagined
Us.
Closing
that
issue
written
nicely
ourselves
or
something.
A
No,
no,
no
I
wasn't
enraged
myself,
but
you
know
their
their
posts.
You
can
read
it
yourself.
It's
not
yeah,
anyway,
the
more
those
we
get
in
to
patch
releases.
By
the
time
the
blog
post
is
published
and
the
fewer
people
will
have
reactions
like
that.
So
that'll
that'll
be
good
for
us
too,
like
we've
effectively
done
a
soft
launch
here
right,
even
even
at
Jupiter
con.
It
seemed
a
bit
optimistic
to
predict
that
you
know.
A
E
Issue
I
think
we
also
should.
E
Really
cool,
otherwise,
you
know
other
people.
B
Might
see
the
same
at
triage,
I
tried
to
respond
to
the
the
poster
with
kindness,
rather
than
tone,
policing,
right
away
and
and
I
also
tried
to
reflect
it
by
saying
we
apologize
for
the
regression
and
we
welcome
contributions
to
reinstate,
to
reinstate
it
and
I
got
a
little
heart
reaction
from
the
the
original
filer
of
the
issue.
So
Mike
is
volunteering
to
be
the
tone
policeman
thanks.
B
Mike
I
agree
that
the
the
posts
some
somewhat
cast
aspersions
on
us,
like
we
somehow
were
being
negative
towards
people
like
him
who
buy
the
cool
60
keyboards,
it's
100,
keyboards
or
so
20
20.
E
A
B
I
I
think
I
think
that
the
person
yeah.
B
A
Mike,
thank
you
for
volunteering
to
say
something
on
the
issue
and
I
hesitated
to
say
something
because
right
now
we
don't
have
yeah
a
an
active
code
of
conduct
incident
response
committee,
which
means
all
of
those
fall
back
on
the
Jupiter
executive,
Council
and
so
I
wasn't
sure
if
I
was
waiting
in.
A
As
you
know,
a
member
of
the
Jupiter
lab
Community
reading
this
issue
and
trying
to
gently
correct
someone
or
as
something
more
official,
so
I
I
am
grateful
for
well
for
your
level
head
and
your
your
your
Kiln
with
kindness
approach,
Jason,
but
also
Mike.
Thank
you
for
wanting
to
to
help
steer
this
in
a
better
Direction
and.
A
Someone
reading
that
thread
actually
finds
it.
You
know
bad
enough
that
they
want
to
file
a
code
of
conflict
report
and
that's
okay
that
but,
but
you
know
not,
everything
has
to
be
if
we
can
solve
it
great
cool.
But
let's
not
talk
about
this
more
because
we
have
so
many
other
things.
If
that's
okay,.
B
A
B
A
Cool
okay,
so
Jeremy!
You
are.
C
Yeah,
so
since
the
follow
or
final
is
out,
I
open
a
meta
issue
to
track
some
follow-up
items
after
the
release,
there
might
be
more
issues
as
we
discussions
before
that
are
coming.
There
will
be
great
topics
in
the
point
release
in
coming
days.
Another
thing
was
that
the
default
Branch
was
renamed
to
main.
So
it's
been,
you
know
waiting
for
a
very
long
time,
because,
with
the
reason
we
update
the
default
branch,
sometimes
the
change
log
is
not
correctly
generated.
C
So
that's
why
we
were
waiting
for
like
the
final
photo
or
release.
So
now
it's
done
so,
if
you're
using
Forks,
don't
forget
to
update
your
branch
yeah
and
and
that
then
it's
about
notebook
7.
So
if
anyone
is
interested
in
working
on
on
that,
you
know
in
the
end
it's
just
like
working
on
Twitter
apps.
So
if
you're
interested
I
hope
it's
super
helpful,
so
we
can
get
a
RC
or
final
release.
D
E
C
The
coming
weeks,
there
is
a
seven
almost
done
on
the
repo
I
think
we'll
be
looking
into
identifying
some
reason.
Blocker
issues
to
know
what
kind
of
issues
are
really
important
to
fix
before
the
final
release,
I
posted
a
couple
of
them
in
the
notes
that
would
be
great
to
get
in
before
the
first
rc,
but
there
is
more
to
triage.
C
So
if
you
have
time-
and
you
want
to
also
have
a
look
at
them
and
see
what
to
do
about
it,
some
of
them
are,
you
know,
bug
fixes
or
updating
the
menu,
so
they
match
better
with
the
classic
notebook
I
mean
anything
would
be
super
helpful,
so
yeah
I
think
that's!
That's
all
I
had
yeah.
A
Maria
emphasize
some
of
that,
if
you
work
on
Jupiter
lab
or
have
worked
on
Duplo
and
have
some
idea
what
it's
like,
that's
what
working
on
Jupiter
notebook
is
like
now,
and
so
you
don't
the
the
barrier
to
entry.
If
you
are
working
on
some
of
the
things
in
this
ecosystem
already
is
basically
a
constant.
You
don't
have
to
learn
new
things,
but
this
is
a
project
that
a
lot
of
people
use,
and
you
know
if
you
want
to
have
your
name
on
it
in
the
changelog,
that's
cool!
A
You
should
do
it
yeah
thanks
for
bringing
it
up
Jeremy
cool!
Now
we
are
up
to
Mike.
D
Okay,
so
my
first
question
is
whether
we
want
to
release
the
movements
and
after
Jupiter
about
4.1
or
should
notebook
7
will
be
based
on
opens
hero,
and
the
next
question
is
very
closely
connected
to
that.
D
C
And
I
think
it
would
all
depend
on
when
we
would
like
to
make
a
photo.
One
I
was
hoping
that
we
could
get
RC
for
the
notebook
7
before
the
end
of
the
month,
which
would
be
like
in
about
two
weeks.
So
I
don't
know
if
getting
the
follow
one
for
that
would
be
realistic
within
that
Tech
frame.
So
I
guess
it's
fine.
C
If
we
base
notebook
7
on
4.0.x,
which
will
include
fixes
rendered
in
4.0
and
then
the
second
question
I
think
usually,
if
I
remember
correctly,
for
3.0
I
think
there
were
like
a
couple
of
days
or
maybe
one
two
weeks
before
the
the
main
branch
switched
to
to
follow
the
one.
C
But
we
could
dig
into
the
issues
and
see
how
we
did
it
last
time,
but
I
I
guess
usually
the
the
process
is
to
wait
a
little
bit
to
for
issues
to
come
in
and
then
fix
this
to
be
addressed
before
starting
working
on
the
new
features
yeah.
Maybe
some
other
people
have
opinions
on
that.
E
Yes,
it's
more
of
a
question
so
for
the
notebook
7
release
cycle
after
it's
come
out.
If
Jupiter
lab
does
so
when
we're
on
4.1,
if
triple
lab
does
like
a
4.2
release.
E
Notebook
is
a
separate,
build
similar
to
a
lab
app
right,
so
it
would
need
to
also
do
a
minor
release
version
following
a
Drupal
Miner
race
version,
click,
update
it's
package
versions,
or
is
it
like
a
dependency
that
can
be
pulled
in
dynamically?
How
how
those
like
diversions
and
the
fixes
flow
Downstream.
C
So
there
is
the
python
version
which
can
be
pulled
in
independently
for
crypto
app,
so
you
could
have
a
newer
to
grab
with
a
maybe
order
in
the
book
Seven
but
I
think
for
Notebook
7.
We
could
also
batch
features
or
manual
releases.
For
example,
if
there
is
4.1
and
4.2
for
chapter
lab,
there
could
be
on
your
7.1
for
Notebook
server.
That
includes
everything
from
those
two
releases.
I.
Think
yeah,
like
you
said,
it's
separate,
store,
I.
Think
it's
up
to
the
seven
to
decide
when
to
make
reduces.
C
C
Requires
maybe
a
little
more
work.
I
was
thinking
packages,
but
it
sounds
doable.
E
C
Yeah
but
there's
a
good
point:
yeah
I'm
thinking
trying
to
think
of
something
that
could
break
notebook.
Seven
when
lab
is
updated,
Maybe
I,
don't
know,
maybe
the
3.6,
for
example,
the
release
of
lab
could
have
been
an
issue
with
notebook
7.
C
D
You
know
my
question
was
motivated
by
one
of
the
pull
requests,
adding
capitro,
which
is
something
based
on
sorting
notebooks
first
and
that's
currently
tagged
for
all
X
and
if
that's
a
single
request,
which
adds
a
thing
I
think
it's
admissible
to
merge
that,
in
into
a
patch
release
slightly
bending
the
rules
of
sampler
requests,
which
would
be
required
by
notebook,
7.
D
I
I'll
be
slightly
more
towards
releasing
4.1
and
basic
notebook.
7
Final
on
that.
But
I
understand
that
this
doesn't
seem
to
be
the
case.
C
C
I
think
it's
really
up
to
us
and
I
mean
yeah
again
depending
on
how
far
is
a
photo
one.
Yes
for
that
particular
one,
but
that
particular
PR
I
guess
we
could
maybe
make
exceptions
an
exception
for
that.
One.
D
So
my
next
point
is
about
so
this
is
different
from
extension
manager,
extension,
disables
and
or
extensions,
or
allows,
to
instance,
them
plugin
manager,
exposures
or
plugins
with
the
descriptions
and
allows
to
enable
all
disabled
individual
plugins,
which
creates
a
new
routes
for
users
to
modify
how
the
environment
looks
and
which
features
includes
or
excludes.
But
it
also
moves
the
balance
of
power
in
a
way.
D
It
is
previously
if
an
assistant
administrator,
disabled,
some
pictures
to
just
download
and
terminal
and
use
the
kernel
which
doesn't
expose
a
way
to
run
commands
on
the
server
side.
D
This
system
administration
always
have
been
assured
to
have
a
control
and
to
mate,
to
have
have
made
life
of
the
user
more
difficult
when
it
comes
to
like
downloading
things
which
they
might
maybe
not
are
not
permitted
to
download
and
plugin
manager.
Adding
this
exposing
this
to
the
content
doesn't
change
the
situation
when
it
comes
to
security,
because
we
never
give
priorities
that
we
are,
in
fact
exposing
API,
which,
for
a
user
Advanced
enough,
would
allow
to
circumvent
any
restrictions
of
this
kind.
D
But
it
now
exposes
that
in
the
user
interface,
which
is
great
because
features
which
are
heavy
and
not
used
can
be
disabled,
but
then
the
conversation
that
we
had
with
Brad
led
me
to
sketch
a
new
command
line
interface,
option
which
is
locking
of
plugins,
so
that
will
enable
server
administrators
to
lock
the
plugin
from
disable
coordinating
by
RBC
user
interface.
D
Nick
is
saying
that
it
sounds
like
a
third
party
extension
in
chat
and
that
there
is
like
just
for
awareness
and
comments
in
the
pull
requests.
One
of
the
questions
is
how
to
make
it
annoying
enough,
so
that
users
who
are
not
knowledgeable
enough,
don't
disable
things
which
they
shouldn't.
So
there
is
a
disclaimer
checkbox.
There
is
also
a
app
dialog
box
ensuring
that
users
do
not
disable
plugins,
which
are
dependent
by
other
plugins
and
similar
restrictions
between
certain
and
important
I.
Guess
it's
just.
D
D
And
this
is
just
to
see
whether
anyone
has
objections.
Okay,
this
particular
shortcut,
it's
the
same
as
in
vs
code,
and
it's
only
active
one
search
box
is
open.
Yeah
just
wanted
to
see
and
avoid
any
breaking
changes
and
users
who
are
not
having
liquid
shortcuts.
F
Yeah
I
don't
have
an
opinion
on
shortcuts,
but
I've
seen
a
couple
of
issues
and
one
of
them
we
create
we
discussed
today
about
like
shortcuts.
So
I,
don't
know
you
have
to
be
careful
with
those
I
guess:
I've
seen
a
couple
of
fishes
that
we
removed
shortcuts
for
a
control,
Pancho
release.
So
I
don't
know
it's
like
minefields.
You
got
to
be
careful.
A
D
That's
a
very
good
question
and
that's
what
I
would
like
to
hear
from
my
knowledgeable
reviewer,
because
it
duplicates
some
of
the
interfaces
which
are
used
in
lumino,
but
they
are
currently
private,
but
in
Lumina
they
are
private
for
a
good
reason,
because
they
also
include
functions
for
enabling
and
disabling
and
exposing
these
Metals
could
potentially
lead
to
more
consistent
state.
So
we
may
want
to
only
expose
a
subset
of
these
interfaces
and
we
could
do
that
in
luminal.
D
If
that's
a
better
place
or
we
could
decide
that
Lumina
doesn't
offer
that
itself.
But
then
part
of
the
implementation
is
strictly
dependent
on
the
bootstrapping
process
in
Jupiter
lab,
because
nominal
doesn't
have
a
concept
of
an
extension
and
if
we
want
to
get
a
benefit
of
being
able
to
disable
extension,
plugins
and
extensions.
D
Even
if
those
do
not
follow
the
naming
convention,
which
is
something
that
practical
summations.
When
we
were
discussing
the
issue
that
led
to
this
PR,
we
would
actually
make
some
changes
which
are
included
in
this
VR
on
the
lab
site
too.
D
There
is
a
plug
added
in
this
pull
request
which
locks
all
plugins,
and
this
was
the
way
I
implemented
it.
It's
not
necessarily
the
best
way
that
could
be
implemented
as
a
separate
server
explanation,
which
is
shipped
with
Jupiter
lock,
but
can
be
disabled
as
you
suggest
or
or
there
is
a
Blog,
a
plug
CLI
plug,
so
that's
may
be
appropriate
and
what
you
think.
E
Yeah,
so
for
if
you're,
an
admin
and
you're
saying
like
all
the
users
should
never
be
able
to
adjust,
which
extensions
are
installed
or
enabled
just
removing
the
endpoints
is
the
safest
one
right,
because
you're
sure
that
it,
even
if
there
was
a
vulnerability,
they
can't
use
it
because
the
API
endpoint
isn't
there
right,
so
that
that
would
clearly
be
preferable.
But
so
the
lock
is,
of
course,
a
good
second
choice.
That's
because
it
wasn't
clear
to
me
whether
the
lock
was
only
for
extensions
or
for
plugins
or
for
both.
D
So
there
is,
this
is
an
interesting
topic.
There
is
no
differentiation
between
extension
and
plugin
in
the
command
line,
interface
and
actually
part
of
the
doc
strings
refers
to
the
mass
packages,
and
this
is
Legacy
because
when
it
was
first
implemented
well,
first
back-end
doesn't
know
about
plugins,
but
there
were
requests
to
make
it
aware.
D
So
initially
it
was
implemented
as
a
kind
of
regular
expression
and
then
in
parts
of
the
code,
like
disabling
extensions,
looks
at
Splits
the
string
by
column
and
if
the
first
part
matches
it
disables
entire
extension,
and
if
there
is
a
column,
it's
true.
It
passes
that
in
page
to
the
front
end
and
then
problems
is
responsible
for
disabled
individual
plugins.
E
E
But
yeah
it's
since
it's
just
the
JavaScript
right.
It
doesn't.
D
It's
never
bullets,
proof.
G
Yeah
I
mean
part
of
that
too,
is
you
know:
we've
got
stuff
like
like
notebook
now
and
it's
absolutely
possible
to
create
extensions
that
work
in
Notebook,
but
they
will
throw
a
bunch
of
console
errors
because
they
will
not
have
resolved
tokens,
and
so,
if
this
mechanism,
actually,
if
we
had
a
declarative
way
outside
of
it,
you
know,
if
you
look
at
index.js.,
mustaches
or
I,
don't
know
handlebars
whatever
the
hell.
It
is.
If
it
actually
didn't
hit
those
code
points
to
those
code.
G
Putting
it
in
the
lab
is
fine,
because
that's
that's
where
it's
got
to
go,
but
thinking
about
how
this
works
with
notebook
and
and
how
the
interplay
of
now
multiple
applications
at
different
levels
of
complexity,
I
mean
notebook,
is
actually
four
or
five
different
applications
right.
It's
got
different,
you
know,
pages
to
it
or
whatever
and
there's
things
that
just
won't
work
in
one
of
in
some
of
them.
They
will
not
load
yeah,
okay,
so
anyhow,
it
would
be.
G
G
Yeah
I
mean
I.
Think
it's
not.
You
know
that
that
that's
a
longer
conversation
you
know
on
what
the
goals
of
of
five
would
be.
Obviously
we
need
to
get
four.
Oh.
You
know
a
couple
a
couple
of
four
rows
out
before
thinking
about
that,
but
that
does
sound
like
a
solid
workshoppy
kind
of
thing.
You
know
what
what
is
it
actually
going
to
do?
G
You
know
get
the
Trove
classifier
start
cutting
Alphas,
you
know
whatever
just
like
that
would
probably
I
think
40.
We
started
the
alpha
brand.
We
started
the
4-0
branch,
basically
immediately
after
three
right,
like
it
was
very
fairly
early
on
that
we
started
planning
for
the
next
release.
Just
took.
D
A
while
I
I
just
wanted
to
add
that
what
makes
is
suggesting
it
would
be
useful
to
the
overarching
effort,
and
so
this
PR
is
motivated
to
allow
users
to
disable
some
things,
not
only
to
customize
how
it
looks,
but
also
for
the
path
of
those
reasons,
and,
of
course,
page
load
time
would
suffer
if
we
are
still
including
these
assets.
Some
of
them
are
not
included
because
webpack
is
is
pretty
smart,
but
some
of
them
will
still
be
loaded
and
yeah
that
that
requires
actually
a
larger,
refactor
and
I.
A
We
don't
have
anything
else
on
the
agenda.
I
always
some
seconds
to
see
if
anyone
wants
to
add
something
to
the
agenda,
but
if
not
I
will
stop
the
recording.