►
From YouTube: JupyterLab Team Meeting - 8 February 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
February
8th,
weekly,
Jupiter
lab
call,
and
today
we
have
16
people
on
the
call
and
we're
doing
the
normal
agenda
driven
call,
but
then,
at
the
end
of
the
call,
we're
going
to
pick
up
our
final
couple.
Triaging
sessions
for
Jupiter
lab
four
to
make
sure
all
of
the
issues
are
correctly
categorized.
So
why
don't
we
get
started
first
person
on
the
agenda
today
is
Jason
w.
B
Sure,
good
morning,
everyone
I'm
Jason
Weil,
so
400
release
plan
follow-up.
There
was
a
bit
of
ambiguity
last
week
where
Mike
and
I
both
published
separate
proposals
for
the
40
stuff
so
and
then
I
I
proceeded
with
mine,
unaware
of
Mike,
so
I
apologize
for
that
I've
been
following
up
with
Mike
directly
about
this.
B
What
I
did
last
week
was
move
all
unassigned
issues
that
were
in
the
400
Milestone
to
410,
and
then
a
few
people,
notably
Mike
and
Frederick,
moved
issues
that
belonged
in
4-0
back
from
4010
to
400
I'm,
not
going
to
be
doing
that
going
forward.
That
was
just
a
one-off
thing
to
move
like
issues
that
were
in
the
40
Milestone.
That
may
not
have
belonged
there
to
a
future
version
going
forward.
B
The
idea
is
that
for
enhancements,
those
should
be
ready
by
the
end
of
this
month,
like
we
want
to
get
new
features,
Frozen
in
good
shape
by
the
end
of
this
month,
for
bugs
for
new
issue
for
newly
filed,
bugs
we're
going
to
be
taking
a
look
at
the
weekly
triaging,
which
is
Wednesday
at
10,
A.M
Pacific,
and
consider
whether
these
are
going
to
go
in
for
4-0.
They
don't
have
to
have
an
assignee.
That
was
an
area
of
confusion.
B
After
some
conversation,
new
issues
in
40
don't
need
an
assignee,
but
we
do
need
to
work
quickly
to
find
assignees
find
owners
for
these
we're
going
to
reach
a
point
at
some
point
this
spring,
where
either
Somebody's
gotta
fix
a
40
bug,
or
we
have
to
accept
that
it's
going
to
go
to
a
future
version,
I'm
curious
about
what
other
folks
have
had
experience
with.
But,
like
that's
been
my
experience
working
in
releases
in
other
types
of
software.
B
A
I
think
there
might
not
be
a
generic
answer
if
the
bug
is
the
kind
of
bug
that
is
a
show
stopper.
We
need
to
have
individuals
looking
at
it
make
that
call.
So
we
wanted
to
release
Jupiter
lab
one
because
all
the
features
existed,
but
we
had
a
bug
that
it
just
rendered
so
slowly
for
some
notebooks
and
Firefox
that
we
thought
it
was
unacceptably
slow
and
that
became
a
showstopper,
so
I
think
there
isn't
like
a
generic
answer
here.
A
There
might
be
some
bugs
that
push,
but
then
again
we
have
a
really
really
hard
Jupiter
con
deadline
here.
So
it
might
instead
just
light
a
bunch
of
fires
under
a
whole
bunch
of
people,
but
whatever
it
is,
some
bugs
might
really
need
frantic
attention
quickly
or
they
can
be
pointed
back.
I
don't
know.
Does
anyone
else
have
a
thought.
C
B
Right
I
mean
I've
I've,
never
worked
on
a
piece
of
software
that
shipped
with
zero
bugs.
Eventually,
you
reach
a
point
where
something
has
to
be
like
a
known
issue.
Somebody
pointed
out,
there's
also
a
4-0x
milestone,
which
I've
seen
before
in
my
experience
where,
if
something
came
in-
and
we
should
do
it
in
like
a
401
or
402
release,
but
it
shouldn't
block
the
4-0
someone
pointed
out,
there's
very
few
items
that
are
in
40x
might
be
worth
taking
a
look
and
see
if
they
truly
belong
in
400,
but
that's
another
option.
D
Yeah,
just
just
to
reiterate
what
Darian
said.
My
sense
is
that,
as
we
get
closer,
human
judgment
on
a
per
issue
basis
becomes
critically
important
and
it's
important
that
we
we
have
consensus
along
the
council
or
the
the
part
of
the
subset
of
the
council.
That's
that's
doing
that
activity,
and
so
we
I
I
think
it
makes
sense
to
do
some
broad
passes.
D
But
we
can't
ever
rule
out
the
possibility
for
human
judgment
because
yeah
we
we
don't-
and
this
is
what's
a
makes
open,
source
fun
and
exciting.
Is
that
even
though
we
set
aspirational
dates
for
releases,
we
don't
really
do
date
based
releases
and
it's
entirely
possible
that
a
month
or
two
from
now
we
would
say
Orlando's
not
ready,
we're
delaying
until
after
Jupiter
to
come
and
I'm
not
I'm
not
advocating
for
that.
But
certainly
that
has
happened
plenty
making
a
face
just
historically
with
Jupiter
lab
and
many
other
Jupiter
sub
projects.
D
E
B
Plug
for
if
you're
interested
in
triaging
new
issues,
I
always
welcome
additional
informed
feedback
at
the
weekly
triage
meeting,
which
is
Thursday
at
10,
A.M
Pacific.
A
Great
thanks,
Jason
next
up
is
me.
A
This
will
be
short,
so
I
sent
an
a
message
to
the
Jupiter
lab
Council
mailing
list,
asking
for
volunteers
from
the
council
to
represent
the
official
Jupiter
lab
sub
project
and
give
a
talk
on
Jupiter
lab
4
at
Jupiter
con,
and
you
know,
I
was
a
bit
too
I
I
set
up
to
four
people,
but
actually
it's
only
a
half
hour
call
so
I
think
two
is
probably
the
optimal
number
here
and
one
person,
volunteered
I,
think
maybe
one
more
person,
and
that
would
be
perfect.
A
So
if
you
are
on
the
council
got
that
message,
please
respond
going
forward.
We're
going
to
try
to
more
formally
have
sub-project
representation
in
jupitercon,
but
as
the
new
governance
is
very
new
and
as
this
Jupiter
Khan
is
imminent,
we're
not
we're
not
doing
all
of
the
sub
projects,
presumably
this
year,
because
it
would
just
be
too
much
to
organize.
However,
we
really
really
really
ought
to
have
a
Jupiter
lab
talk
and
I'm
glad
we've
got
at
least
one
volunteer
so
far.
A
That
does
remind
me,
though,
it's
been
a
while,
since
we
nominated
anyone
to
the
Jupiter
lab
Council.
So
if
you're
on
the
council-
and
you
know
someone
who
wants
to
be
nominal
nominated,
that
was
a
good
time
we
should
get
back
on
that.
Please
be
inclusive
and
expansive
in
your
view
of
Who
belongs
on
the
council.
People
who
participate,
who
you've
seen
working
on
this
for
a
little
while
they
probably
ought
to
have
a
say.
A
Yes,
so
in
the
in
the
mailing
list,
one
person
says
I
nominate
so
and
so
and
a
second
person
says
I
second,
the
nomination
and
we
can
try
the
the
The
Forum
based
School
GitHub
private
repo
voting,
but
really
it's
probably
okay
to
do
as
we've
done
with
most
of
the
others,
which
is
have
the
votes
be
within
that
same
thread
in
the
message
group.
That's
fine!
A
We
don't
publish
these
particular
vote
results
because
they're
about
people,
but
obviously
if
a
person
is
nominated
and
within
a
week
gets
a
number
of
votes
to
pass,
which
is
oh
wait,
it's
not
a
number
of
votes.
It
passes
it.
It's
it's
just
that
the
number
of
votes
coming
in
the
majority
of
them
have
to
be
in
favor.
I
believe
is
that
right.
A
Thank
you
Jason,
so
yeah.
Basically,
it
takes
about
a
week
and
a
day
or
two,
and
we
only
should
do
one
at
a
time
because
the
next
nominee
should
be
someone
was
also
considered
by
every
member
of
the
council.
So
we
don't
want
like
three
nominations
outstanding:
we
do
them
sequentially,
yeah,
I,
guess,
that's
it
so
yeah!
Please
do
those
things,
and
my
third
point
is
I
will
not
be
here
two
weeks
from
today.
A
So
somebody
please,
would
you
host
this
call
if
you
are
happy
to
just
put
your
name
in
the
bullet
point
on
the
agenda
and
with
that
I'm
happy
to
pass
it
on
to
Brian.
D
Thank
you
for
letting
me
go
real
quick
I'll
keep
it
short.
I
saw
that
I,
I,
think
Frederick
and
some
others
maybe
have
been
working
on
a
privacy
policy
for
Jupiter
lab
I
saw
the
back
Port
PR
here.
D
I
got
pinged
unsolicited
feedback
by
Jupiter,
lab
user,
who
was
expressed
concerned
about
this,
and
their
feedback
was
basically
it
was.
It
was
vague
and
unclear
about.
F
D
But
in
their
minds
we
were
acknowledging
if
you
just
download
like
if
you
happen
to
be
using
Jupiter
lab,
even
if
someone
else
installed
it.
These
third
parties
may
be
collecting
data
from
you
and
so
I
I
think
it
would
be
helpful
to
connect
the
dots
between
what
user
behaviors
trigger
those
third
parties
to
trigger.
So,
for
example,
say
if
you
pip,
install
Jupiter
lab
that
that
will
collect
data
on
that.
D
A
Is
the
is
the
linked
issue
appropriate
Brian,
the
one?
That's
in
the
chat,
I.
E
Oud
no,
no,
like
as
I,
did
another
issue
that
was
open
earlier
today
for
or
yesterday,
but
basically
the
user
also
was
referencing
to
a
version
of
the
privacy
policy
that
was
during
a
PR.
So
not
a
new
official
thing
same
thing,
so
I
don't
know
if
it's
the
same
people
that
contacted
you
Brian,
but
probably
as
you
propose
some
announcements,
it
could
be
good
to
to
create
a
new
issue
and
and
discussed
what
to
add.
A
Cool
okay,
great!
Thank
you
thanks,
Mark,
okay,
next
person
on
the
agenda
is
Andrew.
C
Okay,
so
okay,
everyone
I,
am
having
a
local
test,
runs
failing,
and
this
was
reproduced
on
other
Mac
osm1
machine,
so
this
happens
to
a
Boston
head
of
Master
and
on
the
latest
Jupiter
lab
version.
It's
connected
to
Canvas,
Pixma
dependency,
I'm.
Looking
to
see
if
this
affects
only
me
or
us
rather
or
is
this-
is
like
a
more
a
global
problem.
Maybe
it's
specific
to
Apple
silicon
machines
too.
So
there
is
an
issue
I
created
13938.
C
If
you
have
any
advice
on
how
to
fix
this,
or
if
you
can
reproduce
this,
and
also
have
this
problem,
please
let
me
know,
particularly
you
can
just
post
in
the
tissue.
So
I
will
also
link
this
issue
in
the
chat
and
that's
that's
it
for
me.
E
Hey
Andrew,
as
I
commented
on
the
issue,
one
of
the
travel
is
canvas
because
canvas
is
a
npm
package,
that's
a
binary
one,
and
so
the
M1
processor
and
the
following
one
are
a
weird
architecture.
Let's
say
currently.
E
To
like
it's
a
recent
architecture,
let's
call
it
like
this,
so
it
may
be
tough
to
figure
out
how
to
how
to
find
a
work
around.
B
C
Oh,
so
that's
one
of
the
reasons
why
I'm
calling
this
out
I
want
to
understand.
If
this
is
my
my
particular
problem
of
my
machine
or
a
greater
problems
that
affect
subsection
of
Egyptian
population,
so
yeah,
we
can
talk
about
this
later
with
Json,
but
if
anyone
else
can
reproduce
this,
please
Post
in
the
thread.
That
would
be
useful.
Thank
you.
E
F
We
used
a
like
two
thousand
dollars
of
the
donations
to
Sage
to
buy
an
M1
Max
studio
with
32
gigs
of
RAM
and
I've,
given
a
bunch
of
accounts
to
Sage
developers
and
I'm
happy.
If
anybody
writes
me,
I
can
give
you
an
account.
F
It's
just
SSH
access,
but
you
know
it's
a
handy
way
to
get
access
to
a
good,
M1,
Max
processor.
So
just
email
me
I'll,
put
my
email
in
the
chat.
Thank
you,
but
we're
using
it
for
lots
of
little
build
stuff
and
whatever.
C
William
I'll
does
that
if
I
won't
be
able
to.
E
So
we
finally
have
released
Jupiter
that
360
oh
yeah,
and
we
have
already
at
361
so
for
people
that
did
not
follow
up
the
reason
for
2361,
so
Darian
is
worth
to
provide
in
the
service
package.
The
the
API
took
over
the
new
event
system
on
the
new
event
API
from
Jupiter
server
V2,
and
it
used
async
iterable
that
are
available
since
quite
sometimes
in
most
of
the
web
browser
that
that
were
not
available
in
the
type
Target
that
we
were
using
to
produce
the
JavaScript
assets.
E
So
at
the
time
we
thought
it
was
good
enough,
so
we
decided
to
upgrade
the
Target
and
ship
it
like
this.
It
brought
issue,
in
extension
that
were
bumping
Jupiter
lab
because
they
were
also
having
the
same
old
Target
for
the
JavaScript
asset.
So
I
did
a
PR.
That's
like
reverting
back
the
target
to
to
not
have
changed.
That's
why
there
is
a
361
and
I
added
in
the
documentation,
some
information
in
the
immigration
guide,
or
you
can
fix
the
issue
you
can
see
in
your
extension.
E
E
So
last
week
we
discussed
the
where
to
place
the
codes
that
we
will
be
extracting
for
a
written
collaboration
and
from
the
Jupiter
core
to
the
X
Jupiter
server
away
doc.
There's
no
Jupiter
lab
Jupiter
RTC,
so
it
is
move
under
or
organization,
but
Jeremy
points
out
that
there
are
actually
already
another
repository
called
simply
RTC
in
the
Jupiter
lab
organization
that
I
forget
totally
about
at
the
time.
We
discussed
that,
and
so
the
question
now
is
that
what
do
we
want
to
do?
E
I
probably
should
open
a
new
issue
on
the
team
Compass
to
discuss
that,
but
basically
the
the
the
repository
the
first
one,
just
RTC
is
an
order.
Experiment
that
was
I
think
mainly
led
by
a
sold
in
Eric.
A
train
Auto
merge
for
a
live
collaboration
and
so
yeah.
We
probably
should
clarify
the
situation
for
to
avoid
confusion
for
users
or
developers
interesting
to
see
what
we
are
doing
for
RTC.
G
Yeah
I
remember
opening
an
issue
about
archiving
The,
repo
I,
think
it's
like
one
year
old
or
even
more,
but
maybe
it
makes
more
sense
to
to
just
merge
them
as
Mike
suggest
on
the
chat,
but
that
is
the
yeah.
I
was
just
thinking
that
if
someone
comes
to
the
org
and
search
for
RTC
and
find
you
know
two
research
that
they
might
be
confused,
you
know
where
the
work
is
happening.
A
Merging
them
sounds
like
a
good
idea.
Does
anyone
disagree
with
that.
E
A
Cool
great
okay,
Enter
me
you're
up.
G
Yeah,
it's
just
a
quick
update
that
there
is
a
new
notebook,
7
priorities
and,
most
of
the
time
once
there
is
a
Jupiter
app
for
queries.
There
is
a
notebook,
seven
previous
shopping
afterwards.
So
it's
just
a
reminder
that
if
you,
if
you
would
like
to
test
it
on
the
you
know,
check
things
whether
your
extension
works
with
notebook
server,
then
please
do
either
locally
by
installing
the
notebook
package
or
on
binder
I
maintain
a
gist
that
I
posted
in
the
node.
G
So
it's
a
bit
easier
to
to
try
it,
and
there
is
I
press
also
a
link
to
a
request
that
I
opened
two
weeks
ago,
and
it
came
up
several
times
like
when
talking
to
people
using
YouTube
app
for
like
teaching
in
general,
and
they
sometimes
they
use
the
code
console
and
today,
I'm
like
having
the
prompt
at
the
bottom
of
the
record
console
is
a
bit
awkward.
G
Sometimes
when
they,
you
know
they
cast
their
screen
on
a
bigger
screen
in
the
classroom,
and
there
is
an
issue
in
the
lab
report.
I
think
is
seven
years
old.
Also
also
that
suggests
you
know
making
that
configurable.
So
I
started
something.
I
should
also
be
possible
to
try
it
on
binder.
Otherwise
you
have
a
screencast
in
the
in
the
top
command
and
one
suggestion
was
to
put
the
input
prompt
at
the
top
and
then
reverse
the
order
of
the
history.
G
So
the
current
state
of
this
PR
does
this,
but
maybe
you
want
to
tweak
some
of
the
things
or
at
least
make
that
configurable.
So
so
it's
out
there
if
you're,
interested
and
I
think
that's
it.
The
last
point
was
about
the
yeah,
the
RTC
repo,
that
we
already
discussed
before.
A
A
One
thing
that
would
be
kind
of
cool
and
we're
pretty
far
from
is
something
like
the
configuration
of
like
toolbars,
where
you
can
drag
stuff
on
and
off
it'd
be
cool.
If
you
could
take
the
two,
you
know
components
basically
or
the
three
components
with
toolbar
being
one
of
them
just
drag
them
into
an
order
and
lock
it
or
something,
but
we're.
D
A
H
Sorry
yeah,
it
says,
building
meeting
after
this,
if
anybody
wants
to
join,
starts
15
minutes
after
the
hour
agendas
in
hack,
MD
put
anything
in
there.
You
want.
H
If
you've
got
any
questions
about
accessibility,
we're
happy
to
spend
some
time
answering
those
too
so
yeah.
Thank
you.
A
H
So
he's
been
so
react,
Json
schema,
form,
five
just
came
out
and
that
upgrades
to
ajv8.
So
this
is
going
into
starting
some
of
those
upgrades
and
moving
to
the
new
ajb
eight,
which
has
some
like
really
nice
validation
features
a
little
bit
faster
and
stuff.
So
that's
the
best
I
know
from
there.
H
I
can
look,
I,
speak
Nick,
okay,
all
right
so
ajv8
has
a
safe
validation
mode,
which
this
I
think
like
it's
something
to
the
effect
like
the
types
are
pre-compiled
or
something
and.
H
Just
are
not
running
unsafe
code
at
any
point,
but
you
can
so
this
avoids
some
unsafeness
there's
some
security
and
then
the
goal
would
be
to
make
it
easy.
For
extension,
authors
to
just
extend
schema
form,
which
would
then
mean
you'd,
write
your
type
interface
and
by
writing
your
type
interface.
You
would
get
your
UI
as
a
result,
so
it
would
really
help
standardize
a
lot
of
the
basically
the
UI
would
be
built
off
of
schema.
H
It
brings
some
standardization
there
and
hopefully,
other
folks
will
be
able
to
use
that
I
know
he's
put
it
in
some
of
the
Dead
in
a
few
Death
Beds
projects
as
a
test
and
been
successful.
There.
E
That
is
to
say,
instead
of
a
ring
as
a
public
API
the
react
component
from
schema
form.
It
may
be
interesting
to
expose
a
widget
so
that,
for
example,
if
there
is
a
new
major
version
of
that
Library,
we
can
update
it
within
the
3x
cycle,
for
instance,
so
it
should
make
a
or
life
easier
for
deciding
to
move
forward.
A
A
Okay,
so
do
you
guys
think
I
should
stop
the
recording
for
the
triage
or
keep
recording
I
can't
remember
what
we
did
last
time.
B
I
think
we
did
the
triage
during
the
recording
okay,
but
we
also
really
went
like
over
our
our
time
limit
so
and
I
also
see
no
additional
discussion
scheduled
today.
Does
anybody
have
anything
that
they
wanted
to
talk
about
off
the
Record.
A
Okay,
how
about
this
good
point
Tony
in
the
chat
show
of
hands
for
continuing
the
recording
until
the
hour
is
over.