►
From YouTube: JupyterLab Team Meeting - 7 June 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
B
C
Oh
yeah
sure
so,
thanks
to
everyone
who
helped
with
the
Jupiter
lab
4.0
blog
post
that
was
published
earlier
this
week,
I
know
we
tried
to
have
it
published
by
project
Jupiter,
but
I
have
the
byline
that
wasn't
intentional,
but
I'm
grateful
for
everyone's
support
for
that.
So
we
also
tweeted
about
it.
So
please
boost
that
up,
get
more
exposure
for
that.
We've
already
gotten
a
lot
of
like
engagement
on
that
and
yeah
thanks
everyone
for
your
your
support.
There
I,
don't
think
I
have
anything
else
on
my
agenda.
A
Okay,
I
wanted
to
talk
about
two
things.
One
is
I
added
a
team
Compass
issue
about
hosting
this
call,
so
I
was
ill
for
the
last
two
calls
and
I
asked
the
mailing
list
for
the
Jupiter
lab
Council
for
help,
but
really
I
would
like
it.
If
we
had
a
group
of
hosts
I.
Think
I
can't
remember
how
long
it's
been
that
I've
been
the
default
host
on
this
call.
But
it
wasn't
always
this
way
and
it
doesn't
always
have
to
remain
this
way.
A
D
A
So
so,
okay,
so
so
that's
that's
our
resource
to
help
someone
who
isn't
exactly
sure
what
hosting
this
call
requires.
But
it's
not
a
heavyweight
commitment,
I'm
sure,
you've
noticed
and
yeah
I
mean
my
intuition
is
to
initially
ask
that
you
grow
up
Council,
but
really,
if
there's
someone
who
is
interested
in
doing
this
and
they're
not
on
the
console,
it's
a
great
way
to
get
involved.
A
That's
fine,
actually,
I,
don't
think
that
should
be
a
barrier,
but
if
you
are
interested
in
occasionally
doing
this
or
often
doing
this
or
every
single
time
doing
this
any
of
any
combination
of
wanting
to
host
the
score.
Please
chime
in
on
that
issue,
and
maybe
then
a
group
of
us
can
meet
and
talk
about
how
to
either
create.
B
E
A
E
C
Having
a
like
a
rotation
of
like
two
or
three
people
who
would
host
it,
I
feel,
like
that's,
usually
a
better
mechanism
than
just
like
putting
out
a
call
for
volunteers,
because,
as
you've
experienced
that
huge,
that's
while
well-intentioned,
usually
leads
to
one
person
hosting
it
by
default.
Most
of
the
time
so
having
like
a
rotation
might
be
good.
A
Yeah
so,
okay,
two
things
there
one
is
the
call
that
I'm
putting
out
for
volunteers
is
who
wants
to
be
in
a
group
that
we
do
that
with
not
ad
hoc
just
bringing
up.
So
basically,
who
are
those
you
know,
X
number
of
people
and
that's
right
so
project
Jupiter
uses
a
shared
one,
password
group
account,
and
so
anyone
who
wants
to
do
this
we
can
give
them
access
to
the
zoom
account
to
do
it.
A
The
only
weirdness
about
that
is
that
sometimes,
because
we
are
located
in
multiple
places,
the
zoom
account
wants
to
do
two-factor,
authentication
and
that
two
Factor
authentication
is
with
the
project
in
Jupiter.
Gmail
account
yep,
so
I
am
not
sure
how
we
get
around
that
yet.
But
that
is
a
solvable
problem
and
if
we
don't
know
how
to
solve
it,
we
can
ask
maybe
somebody
from
the
security
group
for
help,
but
I,
don't
think
that's
a
barrier
because.
C
My
one
password
access
includes
access
to
the
Gmail
account,
so
I
was
able
to
log
in
then
hop
over
to
Gmail
get
the
code
enter
the
code,
it
adds
complexity,
but
it
did
work
for
me,
but
you
know
it's:
it's
increasing
the
surface
area
of
the
one
password,
which
is
a
potential
security
concern.
I
know.
E
A
A
C
A
B
A
E
F
So
I
just
wanted
to
to
ask
about
the
follow-ups
following
the
release
of
Jupiter
app
and
four,
and
thanks
for
the
community
for
doing
the
release,
and
so
can
start
by
the
bottom.
So
the
language
package
they've
been
released.
They
also
contains
the
the
string
from
notebook
master,
so
that's
mean
also.
They
contain
the
internet,
internationalization
for
the
coming
notebook
V7
just
to
let
no
people
that
are
working
on
that
project,
the
example
the
extension
examples
most
of
them
are
like
almost
all
of
them
are-
are
done.
F
The
challenging
one
was
the
collaborative
document.
Example:
I
have
updated
the
code,
it's
working,
but
I'm
fighting
with
the
integration
test
for
it.
Basically
in
the
integration
test,
the
action
are
done
too
fast
for
the
collaboration
backhand.
So
I
have
issue
that
I'm
not
seeing
when
I'm
doing
things
manually.
Hopefully,
I
will
figure
it
out
quickly.
F
They've
been
updated
on
Condor
Force
tunic,
but
I
have
no
idea
so
Nick
on
the
chat
asked
if
the
language
pack
has
been
updated
on
condo
Forge
I
did
it
manually
because,
indeed
I
don't
know
why
the
bot
never
pingers
know
when
there
is
a
new
version
paper
for
that
one.
So
we
need
every
time
to
to
do
it
manually,
but
I
did
it.
It
may
have
been
like
yesterday
on
Monday
or
something
like
that
for
counterfeiture.
F
Where
was
I
so
the
example
are
almost
done.
We
have
a
trouble,
that's
we're
still
fighting
with
the
Jupiter
that
demo
project
so
for
people
that
are
unaware
of
it
for
the
binder
that
we
have
on
renmese
and
then
lots
of
pricing
documentation.
We
don't
use
the
the
main
branch
from
the
Jupiter
project
directly,
but
we
use
another
repository
to
get
a
more
stable
images
and
avoid
building
too
often,
unfortunately,
between
the
trouble
of
binder
and
the
trouble
of
solving
the
new
environment.
F
F
D
Even
if
we
will
update
the
demo
repo,
it
still
wouldn't
be
stable
right,
because
binder
doesn't
have
capacity
to
save
to
Cache
images.
So
it
would
still
attempt
to
build
everything
and
it
fails
like
it.
85
percent
of
the
time.
I
think
we
need
to
look
into
like
binder
light
or
or
something
like
that.
F
Yeah,
if
people
have
time
for
that
I'm
happy
to
if
people
wants
to
to
take
that
for
now,
I'm,
just
like
taking
the
the
shorter
path
and
I
agree
like
it's
a
trouble
with
binder,
so
hopefully
we'll
get
more
stability
on
my
inner
side,
soon,
Alexandria,
and
so
the
two
latest
are
about
styling.
We
have
the
project,
that's
Jupiter
lab
pigments.
It's
about
bringing
the
styling
from
Jupiter
lab
compatible
with
pigments
I've.
Seen
that
Rosio
has
a
PR
open.
I,
don't
know!
Do
you
have
time
to
finish
it?
B
F
And
also
we
have
a
follow-up
action
that
is
needed
on
NB
convert.
We
need
to
update
the
style
in
there,
and
people
at
constack
will
do
it
so
just
to
to
mention
it
there,
and
hopefully
it's
not
gonna
break
too
much
ring.
One
of
the
trouble
with
nbconvert
is
that
in
Jupiter
lab
we've
changed
the
structure
of
the
notebook
done
by
using
a
table.
So
let's
see
how
it
goes.
F
Yeah
and
if
somebody
sees
other
projects
I'll
put
aside
all
the
extension
by
purpose,
but
since
because
there
are
another
kind
of
business,
but
if
people
have
other
projects
like
that
that
need
to
update
that
are
not
listed
there,
yeah
don't
hesitate
to
comment
either
on
the
chat
or
directly
on
the
Note,
because
yeah,
we
should
be
sure
that
we
have.
We
have
all
those
those
Repository
insight
and
that
we
we
do
what
is
needed
on
them.
F
F
Unfortunately,
that
package
depends
on
lumino
and
because
we've
been
to
lumino2
in
that
the
people
that
are
doing
development
with
Jupiter
Lab
3,
they
get
that
version,
and
so
they
get
a
crazy
errors
and
complete
it's
complex
for
them
to
to
get
a
working,
Dev,
environment
and
so
there's
been
a
fixed
pushed
and
and
yeah
skip
sleep
check
and
we
are
all
good
yeah
but
yeah.
Officially,
it's
a
bad
practice.
F
Okay
and
then
my
last
point
is
now
that
I'm
a
dad,
my
bandwidth
is
a
bit
smaller,
so
I'd
decided
to
remove
the
performance
monthly
meeting
from
the
Jupiter
calendar
because
it
has
been
like
the
last
two
meetings
were
either
cancel
or
there
is.
There
were
on
nobody,
so
I
remove
it
from
the
calendar
because,
for
example,
normally
there
is
one
after
this
one
but
I'm
not
able
to
to
us
that
that
call
so
I
prefer
to
be
honest
to
the
old
community
and
I,
remove
it
from
the
calendar.
A
C
We
briefly
tried
this
out
there.
There
was
this
person
who
was
like
opening
a
lot
of
pull
requests
with
the
proof
of
concept,
saying
hey,
we're
we're
uffizi.
We
have
this
open
source
project
that
we
sell
as
a
service
and
we
make
it
available
to
open
source
projects
for
free.
You
know
clearly
kind
of
a
self-serving
promotional
thing
and
we
tried
it
out
and
in
my
observation,
ufizi
was
super
fast
way
faster
than
binder,
because
it
spun
up
I
think
it
spun
up
an
instance.
C
Every
time
someone
opened
a
pull
request,
as
opposed
to
binder,
which
spun
up
an
instance
when
you
requested
it,
the
major
downsides
of
opeat
city
were
one
the
instance
was
shared,
so
the
you
know
if
I
open
a
pull
request,
and
then
you
Frederick
mess
around
with
the
instance
and
then
I
open
the
instance.
I
see
your
your
stuff,
so
it's
it's
potentially
problematic
in
that
way,
and
the
other
is
that
no
one
really
stepped
forward
to
like
maintain
the
uffizi
instance.
C
F
Well,
one
thing
I
want
to
try,
which
is
also
related
to
the
CI
being
slow
is
I
would
like
to
use
a
to
try
to
set
up
the
CI
on
docker
and
that
could
help
also
to
to
use
other
platform
like
ufitsi
by
having
directly
a
Docker
image.
That's
ready
and
publish
from
more
that
could
also
have
not
saying
anything
like
the
the
travel
for
me.
If
it's
he,
especially
as
you
said,
because
it's
sharing
the
environment,
it
doesn't
seem
suitable
for
linking
the
equivalent
of
the
Jupiter
lab
demo.
B
G
Yeah
I
mean
it
was
pretty
I
mean
I
use
it
quite
often
actually
for
to
work
on
lab
and
it
works
pretty
well,
but
you
need
to
sign
in
that's
the
only
issue,
it's
a
much
harder
for
anyone
to
to
try
something,
but
if
you
already
signed
in
then
you
can
review
PR's
pretty
well
with
this,
because
you
just
click
on
the
button
and
you
get
redirected
to
to
get
Bud.
You
get
the
server
you
do
you
check
your
thing
and
then
you
just
edit
the
stop
automatically.
So
that's
one
way
for
sure.
C
B
A
B
A
F
Yeah
I
try
to
find
time
yeah,
especially
because
one
of
the
like
I,
think
the
main
bottleneck
is
aside
from
the
Macaw
stuff,
because
the
limitation
desire
is
the
integration
test
and
the
solution
for
the
integration
test
is
actually.
You
can
split
the
old
test
in
multiple
jobs.
That's
how
they
do
it
like,
even
in
the
official
playwright
package,
but
the
trouble
is
like
yeah
if
you
don't
use
Docker
if
for
E,
if
you
split
the
old
test
like
in
10
10
batches
and
for
the
10
batches,
you
need
to
build
the
image
again.
Yeah.