►
From YouTube: JupyterLab Team Meeting - 1 March 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
March
1st
weekly
Jupiter
lab
call,
and
today
we
have
20
people
on
the
call
and
if
you
haven't
yet,
please
find
the
agenda
in
the
chat.
Please
sign
in
please
add
a
bullet
point.
A
If
there's
something
you'd
like
to
discuss
and
as
a
reminder,
whatever
time
we
have
left
will
probably
be
doing
some
triage
with
outstanding
issues
and
trying
to
figure
out
where
we
are
with
the
4.0
issues,
because,
as
you'll
recall
yesterday
was
the
feature
freeze
date
that
doesn't
apply
to
bugs,
but
it
does
apply
it
to
new
features.
A
B
A
Not
okay,
does
anyone
know
about
this
subject
that
he's
written
about
or
should
I
just
read
what
it
says.
A
Okay
sounds
like
no
one
does
so
he's
linked
to
a
program
called
the
mlh
Major
League
hacking
program
and
apparently
they're
looking
for
Jupiter
lab
programmers
to
support
students
working
on
the
Jupiter
lab
git
extension,
and
it
doesn't
look
like
a
huge
time
commitment.
But
you
know
it's
real.
So
if
anyone
is
interested
in
mentoring,
some
students
working
on
a
pretty
important
High
use,
I
visibility
extension
get
in
touch
with
Fred,
please
yeah,
I,
guess
I,
don't
have
any
additional
context,
so
I'll
leave
it
at
that.
C
Are
they
based
out
of
any
specific
location
like
Seattle
or
New
York
or.
A
Is
it
purely
remote?
I
have
no
idea
like
I
I,
don't
know
anything
beyond
what
Fred
wrote,
but
hopefully
their
page
will
have
answers
for
you.
Suggestion.
A
That's
what
happened
with
that's
what
happened
with
the
intern
program
at
Cal
Poly
for
a
while.
C
D
Mlh
fellowship
says
fully
remote,
but
I
don't
know
if
that's
what
he's
referring
to
but
David,
it's
probably
best
to
follow
up
with
Frederick
directly
about
that.
A
I
think
that's
probably
the
best
and
if
you
don't
have
his
email,
just
ping
me
I
do
all.
A
Okay
sure
thing
all
right,
so
it
looks
like
Jeremy's
got
a
binder
link
up
and
running.
Why
don't
we
all
up
into
that
and
at
the
same
time,
Fernando
can
fill
us
in
on
the
RTC
issues
he's
been
seeing.
E
Yeah
no,
it
was
mostly
just
a
quick
update
to
thank
David
and
Carlos.
We,
we
did
jump
into
a
breakout
room
at
the
end
of
last
meeting
and
we
were
able
to
kind
of
track
down
the
well.
We
they.
So
we
all
logged
into
one
of
my
campus
servers
and
and
they
were
able
to
understand
what
was
going
on
Carlos,
if
you,
if
you
want
to.
E
If
you
want
to
say
more,
it
was
mostly
just
to
update
people
that
you've
now
merged
108
into
into
the
the
main
branch
and
and
that
this
will
be
we'll
be
yeah.
E
Well,
I,
don't
I,
don't
know
if
he's
here
today
and
but
Carlos
is
here
today
the
for
people
who
may
be
using
it
right
now.
Until
all
of
this
settles
down
I
guess
the
the
high
order
bit
is
don't
keep
more
than
one
browser.
Tab
open
across
server
restarts.
There's
an
interesting
mechanism
with
local
storage,
where
information
Pops
from
one
tap
to
the
other
for
local
storage
then
makes
its
way
down
to
the
server.
E
I
haven't
followed
all
the
details
in
108,
but
it's
mostly
a
thank
you
kind
of
good
news.
Work
in
progress
and
I
did
not
turn
it
off
on
our
servers.
I
just
gave
my
students
a
big
warning
on
this.
So
thanks
to
David
and
Carlos's
rapid
response,
we
were
able
to
continue
the
experiment
for
now,
which
I'm
very
happy
to
do,
because
we're
going
to
get
if.
E
G
G
Basically,
ygs
has
so
the
white
provider
has
an
option
to
sync
different
peers
through
the
broadcast
channel
of
the
browser.
So,
basically,
if
you
have
to
tabs
open
it
syncs
both
documents
through
the
broadcast
channel
or
through
the
local
restaurants,
and
what
was
happening
is
that,
basically,
when
so
Jupiter
had
automatically,
if
you
configure
it
after
X
minute
of
inactivity,
the
server
session
shuts
down
and
when
you
try
to
you
need
to
launch
a
new
one
and
it
opens
a
new
tab.
Basically,
you
had
the
old
cap
open.
G
It
will
sync
the
old
document
with
the
new
tab
and
we'll
duplicate
the
the
content,
because
of
touch
on
those
documents
are
different
and
that's
what
was
happening.
The
problem
is
that,
apart
from
the
crosstab
communication,
also
when
you
open
the
second
tab
with
a
new
server
session,
it
updates
the
cook
in
your
browser.
G
So
the
first
step
also
updates,
authenticated
and
is
able
to
communicate
with
the
server
through
the
workshop
that
and
a
single
document.
So
we
needed
to
add
the
document
sessions
or
a
small
implementation
of
the
document
sessions
to
be
able
to
fix
it.
It
wasn't
only
to
deactivate
the
crosstab
communication
Yas
and
that's
all.
G
A
E
G
Yeah
yeah,
let
me
open
it
up
with
you
on
YouTube
for
that
yeah.
If
we
think
it's
probably
in
in
Jupiter
lab
court,
we
will
work
on
that
is.
G
18
Jupiter
collaboration
for
the
what
Drive,
URL
thingy
between
the
collaboration,
Repository
I,
prefer
to
keep
it
there.
So
I
will
it's
more
visible
and
I
will
teach
that
is
probably
important
and
the
other
the
whole
base
team
I
think
might
open
it.
An
issue
ready
for
that
in
Jupiter,
lab.
A
Okay:
okay,
great
any
other.
E
I
I
also
added
a
link
there
to
Min
added
to
the
Jupiter
Hub
docs
a
little
tutorial
on
how
to
set
up
collaborative
group
accounts
so
that
so
that
we
begin
documenting
how,
if
you're,
using
to
perhaps
host
your
live
instances,
and
you
want
to
set
up
shared
Unix
accounts
that
map
to
that
map
to
group
usage
under
Jupiter
Hub
users.
How
to
set
that
up.
E
E
E
A
A
Oh,
so,
okay,
that
that
sort
of
I
I
was
just
speaking
with
the
muon,
but
what
I
was
saying
was:
please
sink
the
state
of
the
agenda
minutes
with
what
content
you
want
in
the
hack
MD
for
now,
I,
don't
think
we
should
have
the
canonical
agenda
be
sitting
in
this
notebook
because
it
seems
like
we
have
some
issues
yeah.
So
if
you
have
updated
the
agenda
inside
the
first
cell
of
this
notebook,
please
also
update
the
hack,
MD
and
yeah.
Something
weird
happened
to
the
scroll
bar.
A
A
We
could
probably
spend
a
lot
of
time
here
playing
with
this,
but
we
do
have
other
things
on
the
agenda.
Are
there
any
sort
of
critical
things
to
discuss
with
regard
to
RTC
that
anybody
wants
to
talk
about,
or
should
we
move
on.
A
Okay,
let's
move
on.
Thank
you
Jeremy
for
bringing
up
the
button
anyway.
Thank
you,
Carlos
for
the
update
and
thanks
Fernando
for
dog
fooding.
This
also.
H
Muted
and
I
didn't
know.
A
A
A
H
If
you,
if
you
fire
up
the
the
development
binder
from
4.0.0
I,
guess
it's
tracking
on
Master
control
Z
does
not
work.
The
browser.
F
A
B
A
Because
because
there
is
one,
so
it
was
already
the
case
that
cell
level
undo
was
performing
unintuitively.
Where
say
you
have
two
cells,
you
type
in
cell
one
ABC,
you
type
in
cell
2
d
e
f,
you
go
back
to
cell
one
and
type
ghi.
A
Then
you
do
undo
ghi
goes
away,
then
you
do
undo
again
and
the
def
in
cell
2
goes
when
I
instead
of
the
ABC
in
cell
one,
because
it
was
doing
document
wide,
but
people
expect,
when
they're
in
a
Cell
for
the
undo
to
only
manipulate
that
cell,
not
something
off
screen.
So
that's
part,
that's
what's
being
fixed
in
this
PR
that
Carlos
linked,
but
hopefully,
if
there's
a
broader
undo
issue,
it's
manifest
in
this
PR
as
well.
C
Yeah
so
I've
had
this
math
Jax
PR
out
open
for
a
while.
Let
me
just
paste
the
link
in
the
chat,
so
everybody
can
see
it
and
it's
been
open
for
quite
some
time
and
I
was
wondering
so
basically
what
this
does
is
it
upgrades
the
Jupiter
lab
math
tracks
package
from
Matchbox
2
to
mathjax3
and
we're
also
dropping
the
version
number
from
the
package
name
just
because
it
doesn't
really
like
pers
like
otherwise.
C
We
would
have
to
repeat
this
process
for
every
major
version
of
Mac
Jax
that
comes
out,
and
one
of
the
recurring
discussions
that
was
happening
is
that
people,
especially
Frederick
I,
wish
Frederick
were
on
the
call
right
now,
but
he's
not.
C
One
of
the
concerns
is
that
the
bundles
at
the
package
size
for
math
Jacks,
is
20
megabytes
and
one
of
the
the
way
that
I've
resolved
it
is
I'm
using
a
dynamic
import.
So
basically,
it's
only
loaded
when
you
try
to
render
math
Jacks
like
if
you're
trying
to
render
a
latex
in
a
notebook,
so
I
was
wondering.
Is
this:
are
there
any
other
concerns
with
using
Dynamic
Imports
like
I,
like
I?
C
Don't
know
if
they're,
because
Frederick
seems
to
be
really
strongly
advocating
for
like
using
a
script
file
like
basically
loading
the
script
from
a
CDN
when
the
extension
is
being
loaded
and
I.
Just
don't
really
see
how
that's
I
I,
don't
really
see
how
that's
any
better,
because
that's
kind
of
the
same
right
like
you're
loading
it
dynamically.
But
now
we
have
this
external
dependency
on
a
third-party
CDN
for
law,
Tech
rendering
to
work
which
I
just
don't
really
see
as
practical
and
so
in
summary,
I
I
have
two
questions.
One.
C
Is
there
anything
wrong
with
Dynamic
Imports
like?
Is
there
like
a
hidden
concern
here
that
Frederick
is
not
voicing
or
and
two
what
are
the
minimum?
What
is
the
minimal
set
of
next
steps?
That's
required
to
get
this
PR
merged,
like
I'm
sure,
like
I'm
sure
that
people
have
a
lot
of
questions
and
concerns,
but
the
fact
of
the
matter
is,
at
the
end
of
the
day,
I
didn't
write.
C
This
extension
at
all
I
just
took
upon
the
responsibility
of
trying
to
migrate
it
to
Jupiter
lab
4.,
so
yeah
just
open
to
feedback
here.
A
Just
a
question
about
Brett's
suggestion
here
so
I
know
that
in
the
screenshot
he's
showing
a
CDN
script
type,
but
can
that
seem
method
of
having
a
script
type
injected,
be
referring
to
locally
served
content
like
is
he
specifically
asking
for
the
CDN,
or
is
he
specifically
asking
for
this
method
of
converting
version
2
to
version
three.
C
A
Guess
is
that
because
the
mass
tax
people
have
written
an
upgrade
path,
I'm
I'm,
imagining
that's
where
this
comes
from,
but
I'm,
not
speaking
from
any
real
experience,
I
feel
like
Mike
and
Nick,
who
have
some
comments
in
the
chat
and
also
William
might
have
some
input
here.
So
I'll
stop
speaking.
H
H
H
Size
yeah
I
mean
so
the
other
thing
would
be
like
we
got
to
bring
it,
but,
for
example,
I
got
a
couple
lab
extensions
that
have
to
ship
their
own
math
Jacks
because
they
do
stuff
to
it.
That's
kind
of
what
you're
stuck
with
we
have
Jupiter
server
math
jacks,
which
is
a
single
point
of
failure.
So
there's
not
as
much
of
a
concern
but
I
mean.
Is
it?
Is
it
not
a
much
of
a
concern
if
you're
a
lab
extension,
author
or
you're?
H
A
scientist
that
needs
one
package
to
be
turned
on
is
my
my
concern
about
this
particular
like
yeah,
okay,
so,
okay,
so
that's
that's,
not
real
yeah
I
would
I
would
almost
see
it
if
we're
worried
about
the
package.
Also
the
package
size.
A
lot
of
that
is
fonts
that.
H
So
you
just
ship
one
and
leave
people
the
ability
to
change
their
math,
Jacks
URL,
but
frankly,
I
feel
like
delegating
this
to
Jupiter
server,
math
jacks
and
getting
a
3-0
compatible
version
of
that
is.
Is
your
jam
I,
don't
know
it's
very
difficult.
C
Yeah
I,
just
to
be
frank,
like
this
PR
basically
removes
as
page
config
options
from
the
front
end
and
I
didn't
know
that
there
was
a
use
case
for
having
a
configurable
math
tracks
URL.
It
seemed
that
people,
the
the
people
who
reviewed
my
PR,
seem
to
be
okay
with
this
I.
Didn't
really
think
of
that
concern.
But
like
is
there
a
use
case
for
like
what
speci,
what
specific
use
case
does
a
configurable
math
tracks
URL
address.
A
C
I
Yeah
so
when
you
say
having
Jupiter
server
server,
the
the
main
point
is
that
you
want
to
keep
the
server
wherever
the
assets
are
coming
from
on
the
server
you
want
to
keep
them
in
sync,
with
whatever
lab
extension
version.
You
have
right
if.
C
I
I
Just
just
to
clarify
and
and
the
and
the
config
is,
of
course,
what
is
useful
for
us.
It's
different
ways
of
configuration,
font
and
and
like
a
differential
attack,
features
that
are
supported
right.
D
I
So
having
the
contractor
at
least
the
metrics
too,
confident
that
I
know
about
makes
sense,
there
are
things
that
are
only
configuring,
the
the
the
runtime
and
some
things
to
configure
which
extensions
are
loaded
and
how
much
Network
traffic
you
generate
from
that
right.
I
I
C
I
It
exposed
those
on
the
the
API.
I
C
I
That's
good,
but
the
main
point
I
think
is
that
if,
if
this
is
somebody's
mentioning
that
the
the
package
size
concerned,
I'm,
not
sure,
if
I
is
it
like
the
npm
package,
people
care
about
or
is
it
the
assets
that
people
care
about,
because
I
agree
that
whatever
solution
we
do,
should
it
makes
make
sure
that
the
different
parts
of
match
X3
are
in
kind
of
different
chunks
yeah,
so
that,
depending
on
what
config
you
use,
you
don't
load
in
all
of
Mac
decks.
C
Yeah,
so
the
so
just
for
context
that
mathjax
full
package
it
is,
it
includes
like
declaration
files
too,
which
obviously
don't
get
included
in
the
JavaScript
bundle.
So
the
20
megabyte
like
numbered
you
see
on
npm
is
not
is.
It
is
not
reflective
of
the
actual
contribution
like
the
actual
Network
load.
It
creates
actually.
B
I
Yeah,
so
so
for
the
old
one,
what
we
What's
Done
in
the
notebook
rep
on
the
carried
over
into
the
Jupiter
match,
actually
published
that
it
basically
goes
through
the
package
when
you're
doing
the
build
step.
It
goes
through
the
assets
of
match,
X
and
strips
out
stuff
that
we
know
that
the
Jupiter
doesn't
need
yeah.
C
F
F
J
Thing
that
I,
would
that
I
would
add,
is
just
that
our
images
are
our
Docker
Jupiter
server.
Images
are
currently
about
five
gigabytes,
and
that
starts
to
get
into
load
times.
You
know
people
swap
ends.
J
C
Right,
but
this
is
a
dynamically
imported
dependency,
so
it's
not
going
to
be
included
like
it's
not
going
to
be.
The
client
is
not
going
to
try
to
load
this
as
soon
as
Jupiter
lab
the
Jupiter
lab.
Slash
screen
shows
up
right,
like
it's
only
going
to
load
it
when
necessary.
So
whenever
it's
trying
to
render
a
block
of
law
Tech,
it's
going
to
first
like
check.
If
hey,
have
we
loaded
the
match
X
package
into
the
client?
Yet
if
not
make
a
request
to
like
Jupiter
server
to
load
these
assets
in
and
then.
J
No,
this
is
swap
in
time
for
the
Jupiter
server
onto
the
onto
the
cloud
cluster
right.
You
know,
I
mean
wait,
The
Hub
comes
up,
you
say:
Okay
I
want
a
an
XYZ
environment,
you
you
click
on
it
and
then
it's
it
it
and
then
the
next
thing
that
happens
is
the
Jupiter
server
spun
up
and
and
there's
download
time
from
registry
to
to
spin
up
the
Jupiter
server.
So
after
a
while,
the
image
gets
big
enough.
That
starts
to
get
noticeable
again.
J
C
J
Can
tell
you
a
fair
amount
of
it
is
that
we
offer
both
English
and
we
offer
a
number
of
languages,
and
the
translation
packages
are
pretty
large.
C
C
So
did
I
guess
it
does
increase
yeah
because
it
needs
to
install
the
JavaScript
asset
to
build
the
end
yeah.
So
it
would.
It
would
actually,
in
this
case,
increase
the
docker
image
build
size
by
20,
megabytes
and
ice
and
the
time
it
takes
to
download
that,
but
also
be
reflected
in
additional
Docker
build
time,
but
relative
to
5.4
gigabytes.
I'm,
not
sure.
If
that
is.
J
J
C
A
So
it
might
be
worthwhile
to
Ping
the
people
on
this
call
who
were
interested
in
the
subject
and
also
Fred
on
the
issue
and
either
see
if
you
can
have
a
resolution
to
the
questions
that
came
up
in
within
the
pr
or,
if
not,
you
know,
maybe
like
a
half
hour
call
or
something
with
as
many
people
who
are
interested
and
able
to
go
to
try
to
get
it
out
the
door,
because
it's.
C
C
Yeah
I'd
say
we
spin
this
off
into
a
separate
call.
I
really
do
want
to
let
everybody
else
present
as.
C
Yeah
all
right,
my
last
Point
so
update
on
the
Jupiter
AI
extension.
We
do
plan
on
making
a
first
release
soon,
but
we
do
need
alignment
from
the
broader
community
on
what
we
would
name
this
package,
because
obviously
we
don't
want
to
clobber
and
fight
over
npm
package
names.
So
a
few
suggestions
one
would
just
be
to
Simply
name.
It
really
simple
and
like
Kebab
case
like
basically
Jupiter
Dash,
AI,
Jupiter,
Dash,
AI
Dolly.
C
C
If
somebody
like
beat
us
to
it
and
then
do
Jupiter,
Dash,
AI,
GPT
and
then
just
publish
that
there's,
it
would
be
much
more
difficult
to
resolve
that
issue
right
and
so
the
Alternatives
would
be
to
scope
it
under
the
Jupiter
or
Jupiter
lab
or
npm
organizations,
or
a
third
alternative
would
be
to
scope
it
under
a
new
dedicated
npm
organization
called
like
Jupiter
Dash,
AI
I.
Just
so
I
was
just
thinking
like
what
are
the
next
steps
we
can
do
to
like
finalize
discussion
on
this.
A
C
A
I
think
this
might
have
a
fairly
uncontroversial
answer,
so
the
proposal
was
that
this
project
goes
into
Jupiter
incubator
and
is
sponsored
by
the
Jupiter
laptop
project
and
when
it's
ready
it
graduates
to
the
Jupiter
lab
org,
so
presumably
the
packages
it
publishes
probably
should
be
under
the
Jupiter
lab
scope
and
npm
the
way
other
packages
that
we
publish
like
Jupiter
lab
get.
You
know
that
kind
of
thing.
A
Does
anyone
disagree
with
that
interpretation
to
me
that
seems
sort
of
not
controversial,
but
then
again
you
know.
Maybe
that's
not
correct.
Does
anyone
disagree
with
that.
C
So
is
there
a
precedent
for
allowing
incubator
programs
to
still
use
the,
and
the
Jupiter
lab
npmorg
like?
Is
that
we
don't?
Okay
with
everybody.
A
The
incubator
is
being
sort
of
re
rethought
in
light
of
the
new
governance.
Incubator
didn't
have
the
the
notion
of
a
sub-project
sponsoring
incubated
packages
until
recently,
so
okay,
so
Mike
has
a
suggestion
of
using
something
like
Jupiter
lab
experimental
instead
of
Jupiter
lab
so
yeah
I.
Guess,
because
it's
not
obvious
that
everyone
does
agree.
Maybe
you
should
open
a
team
Compass
issue
we'll
have
the
conversation
there
and
resolve
one
name
to
publish.
C
I
would
just
final
comment
on
this
before
I
create
the
team
Compass
issue.
I
would
like
it
so
I
would
like
to
not
have
to
migrate
the
packages
after
Jupiter
AI
graduates
from
the
incubator
program,
just
just
out
of
like
purely
yes
purely
out
of
practical
concern
like
that
is
that
does
break
a
lot
of
things
and
releaser.
It
involves
like
a
lot
of
additional
effort
and
ambiguity,
so
yeah.
A
Powerful
case
there,
but
yeah
I
mean
Mike
asks
how
many
packages
are
there
right.
C
I
A
H
Hello
I
was
writing
a
thing
the
other
day
where
we
were
talking
about
the
structure
or
packages
on
what
for
and
notebook
7
should
ship,
and
you.
H
Out
some
mermaid
and
put
it
in
the
issue
and
I
was
like
you
know
what
we
just
need
this.
We
need
this
in
Jupiter
lab.
It
works
in
hedge
dock,
the
thing
underlying
the
thing
that
we're
doing
our
stuff
with,
because
our
product
doesn't
work
and
it
works
on
gitlab
and
works
on
GitHub
with
all
the
same
syntax.
H
So
let's
just
do
that,
so
this
PR
does
that
it
also
updates
marked
to
4.2.12,
just
because
you
know,
if
you're
working
on
it
and
it
lazily
loads,
marked
and
lazily
loads,
this
mermaid
stuff-
and
you
can
draw
really
cool
stuff
with
it.
So
it
got
in.
H
You
know
right
before
the
freeze
and
someone
put
a
number
on
it,
but
if
you're
interested
please
go
try
out
PR
on
the
binder
or
whatever
you
want
to
do,
and
I
would
really
like
to
say
that
we,
you
know
caught
up
with
where
gitlab
was
in
2015
and
caught
up
where,
where
GitHub
was
last
year,
with
our
ability
to
document
things
visually,
the
only
caveat
is
that
it's
big
we
can
strip
out
one
megabyte
of
it,
because
it's
behind
an
experimental
flag
that
only
works
on
one
of
its
10
diagrams,
but
but
it
is
really
good
at
what
it
does
like
it
can
actually
draw
yeah.
H
Oh
it's
all
dynamic.
As
hell
brother
man,
I'm
I'm,
like
the
patron
saint
of
dynamic,
Imports
yeah,
you
go
look
on
those
get
planes
yeah,
it's
it's
actually
taking
like
70k
off
the
the
bottom
line
from
the
the
moon
animation,
starting.
J
G
H
Pretty
much
that
went
all
also
adds
a
cash
for
syntax,
highlighting
because
that
creates
Dom
every
time,
apparently
like
it,
creates
extra
Dom
that
then
gets
destroyed,
which
is
weird,
so
you
know
try
to
avoid
that,
but
yeah
like
it
handled.
H
You
know
it
can
draw
all
of
the
dependencies
from
all
of
the
package
jsons
in
like
a
second
or
two
in
the
entire
Jubilee
lab
repo.
So
you
know
it's
not
a
thing
you're
going
to
use
all
the
time,
but
it
can
do
the
thing
that's
about
it.
E
E
E
H
E
H
E
Okay,
I've
I've,
seen
I've
seen
specifically
I
was
asking
specifically
about
Rowan
I've,
seen
him
to
be
extremely
collaborative
and
careful.
I
I'd
appreciate
understanding
what
they're
referring
to,
because
that
response
was
very
unexpected
from
you
and
so
I'd
appreciate
understanding
where
that
came
from.
H
I'm
I'm
just
out
here
trying
to
help
keep
y'all's
packages
running
and
I
get
a
lot
of.
That's,
not
the
executable
books
way
and
it's
you
know
it's
just
tiresome
and
I,
mostly
disengage.
Aside
from
keeping
your
packages
shipped.
H
So
yeah,
that's
deep
personal
stuff
doesn't
matter,
we
can
be
on
gfm,
that's
compatible
with
gitlab
and
GitHub,
and
all
these
guys,
with
this
PR
and
I,
think
it'd
be
a
nice
addition
to
perform
and
we
got
it
in
before
the
deadline.
H
A
He
posted
about
and
updated
Jupiter
lab
desktop
version.
3.6,
Dot,
1-2
I,
don't
know
what
Dash
two
means.
It's.
A
Cool
and
so
let's
see
what
to
say,
support
for
Server,
launch,
arcs,
customization,
support
for
adding
server
environment
variables
and
a
fix
for
some
session
launch
failures
caused
by
which
is
using
iframes.
A
So
I,
don't
know
if
you
folks
are
using
Google
app
desktop,
but
I
think
it's
amazing
that
it
exists,
because
there
are
tons
of
users
who
are
afraid
of
the
command
line,
but
not
afraid
of
Jupiter,
and
so
I
think
this
is
a
really
good
piece
of
software
and
if
you
are
curious,
please
play
with
it.
I'm
sure
I
might
would
appreciate
more
more
eyes
on
it.
B
Hey
everyone,
so
just
two
shirt
like
asks:
first,
is
I'm:
inviting
contributors
to
participate
in
discussion
about
settings,
editor
styling
in
particular,
about
accordions
and
need
for
them,
because
I'm
about
to
start
work
on
potentially
removing
them
and
rolling
the
design
back
to
what
it
was
like.
You
know,
in
Json
editor
without
accordions,
you
can
proceed
to
the
link
in
the
hike
MD
or
in
the
chat,
yeah
and
participate
in
this
discussion.
B
So
the
last
commentary
link
to
and
the
one
before
it
in
particular
and
another
one-
is
I'm
seeking
advice
if
it's
possible
and
if
it's
possible,
how
to
do
it,
how
to
detect
the
debugger
is
active
from
shell.js.
It
would
help
me
to
fix
the
bug
with
W
who's
coming
to
be
not
working
in
succession.
So
it's
a
second
link
I
sent
to
the
chat
and
also
on
hack
MD,
so
yeah.
If
would
be
grateful
for
any
participation
and
any
contributions,
that's
it
in
my
end.
Thank
you.
A
For
your
last
question
it
if
there
isn't
a
trivial
way
to
discover
this,
what
you
could
do
is
modify
the
command
that
enables
the
debugger
and
disables
it.
To
also
add
a
data
attribute
to
The
Notebook,
but
keep
in
mind
it's
it's
subtle
right.
If
you
have
four
notebooks
open
and
one
of
them
in
the
background
has
debugger
activated.
Is
that
what
you
need
to
know
or
do
you
only
need
to
know
if
the
active
one
has
it
or
what?
A
A
Oh
and
yeah
so
I
think
there's
probably
some
feedback.
That's
going
to
go
into
your
PR
about
the
settings
editor
folks
on
this
call.
If
you
have
comments,
please
make
it
there
so
that
Andrea
can
see
them
and
respond.