►
From YouTube: Jupyter Community Call - January 26, 2021
Description
Recording from the Jupyter Community Call in January 2021.
The notes from this call can be found here: https://jupyter.readthedocs.io/en/latest/community/community-call-notes/2021-january.html
Read more about these calls in Discourse: https://discourse.jupyter.org/t/jupyter-community-calls/668
A
Wow
there
we
go
hello,
hello,
welcome
and
let
me
know
if
you
need
anything
if
you
need
the
agenda
links
again
in
the
chat,
I'm
happy
to
paste
them
cool,
so
hi,
happy
new
year,
everyone
and
welcome.
I'm
gonna
share
my
screen
real
quick.
If
I
can
figure
out
how
I'm
doing
that
is
this,
the
biggest
community
called
we've
had
so
far.
I
think
it's
the
most
people
and
the
most
full
agenda
actually
yeah,
I'm
a
little
worried
about
getting
through
it
all,
but
good
problems
to
have.
B
A
Yeah,
hello,
hello,
why
can't
I
do
both
of
these
at
once?
Okay,
I'm
obviously
still
figuring
out
my
share
screen
situation.
However,
let's
just
yeah
here
we
go
always
the
problem.
You
think
I
would
have
figured
it
out
by
now.
There
we
go
cool
great,
so
reminder
this
is
january.
26Th,
welcome!
Welcome
to
our
first
jupiter
community
call
of
2021.
A
woohoo,
I'm
isabella
for
those
of
you
that
may
not
have
met
me,
I'm
a
designer!
I
like
these
calls,
because
I
think
it's
worth
celebrating
what
we're
doing
around
here
and
having
people
show
off
what
they
do.
So
thank
you
for
taking
the
time
to
be
here
and
welcome.
I'm
really
glad
to
see
so
many
people
and
so
much
enthusiasm
on
the
agenda.
It's
very
exciting.
A
There
are
two
things
I
want
to
mention
really
quick,
though
just
so
everyone
knows,
community
calls
are
recorded,
so
whatever
that
means
to
you
what
you
can
and
can't
share
just
keep
that
in
mind.
I'd
also
like
to
say,
we
do
follow
a
code
of
conduct,
because
this
is
a
jupiter
community
event.
You
can
find
this
at
any
time
for
any
case
that
you
need
it
and
keep
in
mind.
A
This
applies
to
me
as
well,
so
we're
all
here
on
the
same
terms,
and
that's
all
I
need
to
do
with
that,
but
those
are
just
my
reminders,
so
I
don't
forget
to
tell
you
anything
important
now.
I
also
want
you
to
know.
If
you
haven't
been
here
before,
what's
gonna
happen,
is
we
have
the
agenda?
A
I'm
gonna
just
go
down
those
items
kind
of
in
order,
and
since
though
we
have
a
lot
of
people
today,
I
first
wanted
to
ask:
is
there
anyone
who,
like
would
be
okay
if
we
run
out
of
time
presenting
partially
next
month
that
we
can
move
to
the
end?
And
if
not,
that's,
okay,
I'll
just
watch
the
time
really
closely,
but
just
an
offer.
A
C
In
fact,
I
I
had
some
changes
to
making
the
presentation,
so
I
would
be,
I
would
be
all
right
doing
it
next
month.
Okay,
to
give
me
some
time.
A
A
A
Great,
thank
you
all
so
much
for
your
flexibility
too.
I
did
not
anticipate
this
many
agenda
items
I
may
have
to
change
the
template
so
that
we
control
it.
But
thank
you
thank
you,
and
in
that
case
I'm
gonna
start
going
down
the
agenda.
The
first
things
that
we
have
are
like
short
reports,
announcements,
kind
of
celebrations
and
the
first
person
we
have
there
is
I'm
thinking
frederic.
I'm
probably
not
saying
that
right.
I'm
sorry.
E
E
So
the
idea
right
now
is
just
an
offer
for
people
that
have
hard
time
who
don't
have
so
much
free
time
to
manage
their
extension
for
jupiter
lab
to
put
it
in
organization,
and
I
will
start
to
to
prepare
stuff
for
having
a
clearer
policy,
because
there
is
also
some
discussion
that
needs
to
be
done.
Should
we
be
part
of
officially
or
not
of
the
of
the
jupiter
organization?
E
So
but
that's
what
will
be
in
the
future,
but
right
now
there
is
already
the
organization
on
github
and
there
have
been
a
couple
of
extensions
already
available
and
moving
that
organization,
and
hopefully
more
portugal.
A
A
That
was
a
lot
of
people
working
very
hard,
and
I
want
to
acknowledge
that
so
yay
congrats,
I
also
wanted
to
say
I've,
seen
a
resurgence
in
like
worrying
about
and
working
on,
accessibility
concerns,
particularly
in
the
jupiter
lab
community,
because
that's
where
I
spend
most
of
my
time-
and
I
wanted
to
thank
the
small
group
of
people
who
are
working
on
that
and
also
since
you're
all
here
invite
you
to
our
meetings.
We
meet
every
other
week,
but
we
happen
to
align
this
week.
A
F
Oh
yeah,
I
just
wanted
to
say
that
the
plotly
community
is
getting
up
on
jupyter
lab
3.0.
F
Pre-Building
their
extensions,
it's
been
a
pain
to
you
know,
get
those
extensions
installed
in
the
past.
They've
had
to
do
a
lot
of
troubleshooting
and
yeah.
They
just
are
they're
getting
more
open
to
you
know.
They're
moving
away
from
you
can
only
run
charts
in
the
in
their
cloud,
and
things
like
that.
So
it's
just
really
good
to
see
their
support
of
the
community.
A
Okay,
then,
thank
you
so
much
we're
moving
on
to
the
agenda
items
yet.
The
first
thing
I
noted
was-
I
saw
ross
and
julian
had
signed
up
very
early
on
the
agenda,
but
I
didn't
see
anything
to
present
and
that's
fine.
I
just
wanted
to
check.
I
don't
even
know
if
they're
yeah,
hey
yeah.
Oh,
yes,
okay!
Thank
you.
Do
you
have
something
that
you
want
to
present
or
were
you
just
signing
up.
G
Not
preemptively
present
not
this
time
anyway,
so
I'm
joining
from
refinitive.
So
as
a
product
manager,
we
have
a
jupiter
notebook
due
to
lab
implementation
that
delivers
quite
a
few
financial
calculations,
so
a
bit
like
bloomberg,
effectively
data
aggregator
and
data
distributor
with
some
analytics,
and
we
just
like
to
get
more
involved.
Basically.
So
I
think
this
is
probably
our
first
step
of
dipping
our
toe
in
the
water
figuring
out.
What's
going
on
and
we're
contributing,
I
think
in
the
first
instance
the
japanese
translation.
A
A
H
Oh,
I
do
get
to
talk
correctly,
yeah,
so
I'll
even
turn.
My
pictures
on
hi
sitting
here-
it's
lovely
in
atlanta
january,
lovely
right.
You
know
no
ice
storms,
so
we
have
two
things
that
I
threw
up
on
there.
One.
H
Three
dot
x
line
supports
jupiter
lab
three
works,
great,
less
filling
all
that
stuff
we've
done
a
couple
releases
now
the
conform
stuff
is
pretty
solid.
You
can
like
just
ask
for
r
and
all
that
other
stuff,
it's
fine.
We
got
eight
big
features
that
you
know
either
refine
or
complement
stuff,
that's
already
in
the
jupiter
lab
experience,
but
we
add
additional
things
like
linsting
and
all
that
kind
of
stuff.
So
big
focus.
This
is
on
the
on
the
epinema
or
eponymous.
H
I
don't
even
know
jupiter
components,
but
we
support
lots
of
other
servers
as
well
like
latex
and
docker,
and
all
this
stuff,
and
then
we've
started
the
jet
process
to
get
that
created
as
an
official
jupiter
sub
project.
There's
a
lot
of
work
to
do.
There's
a
lot
of
work
to
do.
There's
a
lot
going
on
the
language
server
protocol,
but
we
want
those
features
that
will
be
fun.
H
H
Down
and
these
changes
are
going
to
be
landing
in
3.1
they're
ready
to
test,
I
put
them
under
test
and
didn't
find
anything
super
broken.
So
that
was
exciting.
But
if
you
have
an
extension
that
does
do
a
lot
of
heavy
stuff
in
the
notebook
trying
it
out
with
2.3.0
rc0
will
help
them
find
the
problems
and
we'll
break
that's
it.
A
A
Okay,
in
that
case,
I'm
looking
through
the
list-
and
I
think
everyone
else
on
the
list
today-
are
people
that
said
that
they
could
go
next
month,
but
it
doesn't
seem
like
we
need
that
right
now,
so
I'm
just
going
to
return
to
the
top
of
the
list
and
go
back
to.
Is
it
lloyd?
Sorry?
But
it
is
your
turn
if
you
are
ready
and
if
not
no,
you
said
you
wanted
to
go
next
month,
because
you're
preparing.
K
F
A
E
E
E
Deploy
those
kind
of
extensions,
even
if
they
are
replacing
the
core
extension
so
basically,
first
you
can,
you
can
directly
publish
on
conda
or
pip.
So
for
the
other
extension
I
have
done
this
cell
toolbar,
so
you
can
directly
just
look
and
install
them.
So
that's
that's
really.
I
think
a
very
nice
feature
for
user
to
to
not
have
to
to
think
about
two
packaged
one
for
the
front
end
one
for
the
the
back
end
is
everything
will
be
in
pip
or
contest.
So
that's
really
very
nice.
E
So
I
will
let
it
download
all
the
stuff-
and
I
will
speak
about
what's
nice
also
about
jupiter
up
tree,
it's
so
the
ability
of
being
able
to
disable
directly
extension.
So
this
is
the
package
json
of
the
alternative
launcher
and
now
it's
possible
directly
in
the
extension
to
to
say
that.
Okay,
if
you
install
this
extension,
you
need
to
disable
that
launcher
from
the
the
core
jupiter
lab,
and
this
is
done
through
configuration
and
for
the
end
user.
It
doesn't
need
to
do
any
additional
commands.
E
So
in
the,
if
I
look
at
the
readme
of
the
package,
so
basically
for
jupiter,
jupiter
3,
the
the
user
just
need
to
type
that
and
it
will
work
and
if
you
compare
for
gp,
let's
tear
up
two
and
then
it
it
was
another
story.
I
mean
you
need
to
to
use
a
specific
installation
process
for
the
front
end
and
you
need,
after
that,
to
disable
the
back
of
the
the
core
extension.
E
So
that's
what
I
think
the
user
work
through
to
install
extension,
very
nice
and
I'm
very
glad
for
all
the
work
that
the
core
developer
and
for
that.
So
thank
you
thanks
a
lot
and
the
other
one
I
want
to
speak
quickly
is
also.
It
brings
back
lots
of
discussion
that
was
done
at
the
beginning.
I
E
Jupiter
lab
is
the
the
the
idea
of
having
some
some
kind
of
cell
toolbar
in
jupiter
lab.
So
this
is
the
first
implementation.
So
if
you
look
at
the
shift,
what
you
will
see
is
that,
on
the
selected
cell,
you
get
some
icon
for
transforming
this
file
from
marlon
to
code,
to
delete
them
and
also
you
get
the
ability
to
tag
to
a
directory
tag
and
to
visualize
directly
the
tags
inside
the
notebook,
so
that
will
make
the
the
life
of
the
user
much
more
easy.
I
E
Tag
so
because
lots
of
people
know
about
parameters
from
paper
mile,
but
if
you
have
some
process
in
which
you
have
a
keyword
to
hide
tags
and
stuff
like
that,
you
can
also
add
them
and
that
make
the
life
of
your
user
also
much
much
better.
But
I
have
to
say
that
now
my
user
and
my
company
are
using
it
since
two
months
almost,
and
so
there
is
definitely
still
some
ux
problem
like
currently.
The
position
of
the
cell
is
changing
because
of
the
appearance
of
that
cells
toolbar.
I
E
E
E
So
this
is
the
kind
of
thing
here
you
can
see
so
by
default
there
is
only
parameters
as
default
default
tag,
but
you
can
add
more
of
them,
so
there
is
some
interactor
to
access
or
indirectly
yourselves
and
that
we
are
people
that
don't
know
so
much
about
the
shortcut,
the
keyboard
shortcut
that
are
available
and
that
I
think
helped
people
to
to
use
jupiter
lab,
especially
the
the
beginner.
A
C
A
E
What
do
in
this
case
is
just
because
the
I'm
replacing.
I
Something
if.
E
C
A
E
M
This
is,
this
is
damian
very.
The
thing
is,
the
launcher
is
really
nice
as
well,
but
I'm
really
interested
in
the
attacks
in
the
context
of
the
price
slideshow
story,
I
can't
foresee
people
using
price.
I
mean
they
should
be
the
price
level
extension
I'm
working
on
right
now
that
could
potentially
use
that
to
actually
select
and
make
those
cells
in
the
specific
slide
they
want.
So
that
is
great
I
will
I
will.
I
will
use
that.
Thank
you.
E
A
A
N
Spaghetti
okay,
so
this
is
a
project
I
started
like
a
month
or
two
ago,
and
maybe
just
before
I
present
what
I
I
just
want
to
present.
What
is
the
intention
of
this
extension
so
usually
when
I,
when
I
work
with
notebooks,
I'm
myself
a
scientist?
N
What
I
usually
do
is
I
create
some
global
imports
and
then
I
load
some
data
and
then
I
modify
the
data
and
then
I
plot
the
data,
and
then
I
modify
the
data
again
in
a
different
way,
and
then
I
plot
it
again,
and
I
do
that
over
and
over
again-
and
eventually
I
at
some
point,
I
end
up
with
trying
to
make,
for
example,
the
plots
of
the
ends
modification
of
the
data
and
the
amp
modification,
and
it
very
quickly
becomes
quite
well
quite
untidy.
N
So
the
way
I
usually
try
to
fix
that
is
through
using
headings
so
whatever
is
in
the
same
heading
should
be
executed
at
the
same
time.
But,
of
course
it
doesn't
really
work
very
well,
because
if
you
want
to
make
the
plots
of
the,
for
example,
the
first
and
the
end
iteration
of
your
modification
of
the
data,
you
don't
really
know
where
to
put
that
parts.
So
this
is
what
my
notebooks
look
like,
but
this
is
not
what
I
would
my.
N
I
would
like
my
notebook
to
look
like
what
I
would
like
them
to
look
like
is
to
initially
have
some
global
imports.
This
is
my
globals
and
then
load
some
data
and
then
have
like
different
branches
going
out
of
my
data
with
different
modifications
and
then
different
branches
going
out
of
my
data
to
make,
for
example,
to
make
some
plots.
N
So
so
the
way
I
think
about
my
notebook
is
more
in
terms
of
graph
than
in
terms
of
just
a
linear
list
of
cells
and
just
to
give
you
a
an
idea
of
what
it
looks
like
one
of
my
notebook
looks
like
this
and
it's
absolutely
unmaintainable
and
I
have
no
idea
which
cell
should
go
first,
which
cell
should
go
at
the
end,
and
this
is
the
problem
I
often
find
myself
in
and
then
I
I
realized
that
well.
N
Juvederm
lab
is
really
awesome,
because
you
can
very
quickly
create
extensions,
so
I
decided
to
create
my
own
extension
instead
of
complaining
about
the
developers.
Let's
try
to
do
something
about
it.
So
what
I
the
the
idea
I
had
was
to
well
to
think
about
processing
my
data
and
representing
it
in
terms
of
graph,
while
leveraging
all
the
power
of
jupiter
notebooks,
especially
having
rich
outputs
and
the
ability
to
interact
with
with
python
or
whatever
kernel
you
have,
and
so
to
do
that
I
created
the
ib
spaghetti
projects.
N
N
And
then
you
just
connect
the
different
cells
all
together
and
if
you
use
typing,
with
python
3.6
and
over,
you
can
also
well
you're
only
allowed
to
connect
sockets
that
are
that
share
the
same
types
together,
so
that
it
really
helps
you
to
connect
the
stuff
that
goes
together
all
together
and
I
can
try
to
do
a
demo,
but
I'm
usually
cursed
with
demos.
So
it
may
not
work
at
all
all
right.
N
So
I
start
with
some
globals
and
then
I
define
some
some
functions
and
whatever
I
have
registered
node,
it's
a
function
that
is
going
to
become
a
node
in
the
graph
and
there
are
just
python
functions,
so
you
could
execute
them
without
the
name
of
my
package
at
all,
and
if
you
load
that
in
the
extension
you
okay,
so
I
just
do
it,
it
asks
me
for
a
kernel
and
then
it
starts
it.
Fires
up
the
kernel
to
execute
the
globals
yeah
there
we
go
and
then
well
here
it's
it's
a
demo.
N
So
there's
something
that's
that
is
already
in
the
units
and
if
you
click
on,
for
example,
load
data
sets
you
get
the
source
code
of
your
data
sets
and
oh
that's,
a
glitch
so
on
the
left
side
is
what's.
What
is
the
code
of
the
function
on?
The
right
side
is
how
it's
going
to
be
injected
and
executed
in
your
kernel,
and
then
you
can
pipe
at
the
output
of
this
load.
N
Dataset
function,
which
is
in
here
for
export
to
here
a
function,
is
called
protection
plots
and
if
I
click
on
run
graph,
it's
going
to
execute
my
graph
from
left
to
right
and
take
into
account
the
dependencies.
So
whenever
I
change
load
data
sets
all
the
things
that
are
downstream
are
going
to
be
marked
as
being
dirty
and
have
to
be
re-executed,
and
if
I
modify
something
downstream,
it
doesn't
change
anything
upstream.
So
if
I
run
it,
I've
got
a
blue
thing.
N
That
tells
me
oh
it's
running
and
when
it's
green
it
is
done,
and
so
here
you
see,
some
outputs
from
whatever
function
is
being
executed
and,
for
example,
this
one.
It's
a
projection
from
astrophysical
data
sets
because
I'm
doing
astrophysics
and
the
nice
thing
is
that
if
you
well,
you
can
add
nodes.
For
example,
I
can
add.
Let's
say
I
don't
want
to
project
along
the
in
this
case
along
the
x-axis.
N
I
can
ask
it
to
project
it
on
the
y-axis,
and
I
just
connect
that's
here
and
if
you
look
at
the
code
that
is
going
to
be
executed,
no
axis
is
equal
to
y
and
if
I
run
it,
I
get
a
different
prediction.
So,
instead
of
projecting
on
the
whatever
axis
it
was
now
I'm
projecting
along
the
x-axis,
the
y-axis,
sorry,
yeah
and
well.
I
guess
that's
more
or
less
all
the
things
that
I
wanted
to
say.
N
So
the
the
idea
is
to
try
to
really
like
all
the
things
that
I'm
showing
here
on
screen
are
just
pure
jupiter
lab
widgets,
except
except
for
this
graph,
of
course,
because
what
their
photograph
and
I'm
still
very
much
in
an
authoritarian
face.
So
any
any.
H
N
No,
but
I
I
that
that's
one
thing
I
would
like
to
add
support
with,
for
I
don't
know
why
it
doesn't
work,
but
I
don't
know
how
it's
supposed
to
work.
So
it's
internally,
just
a
it's
just
a
code
cell
which
I
imported
from
jupiter
lab.
So
I
was
expecting
it
to
work,
but
it
does
not
I'll
have
to
investigate.
Why.
K
N
H
H
O
F
An
idea
that
comes
to
mind
I'm
reading
a
book
on
desk
right
now-
and
you
know
this
is
obviously
a
direct
acyclic
graph.
It
would
be
cool
to
construct
desk
workflows
this
way,
and
you
know
the
one
the
nice
thing
about
a
dag
is
that
there's
only
one
way
to
traverse
it,
so
this
kind
of
gets
so
it's
something
to
satisfy
the
critics
that
say
jupiter
is
like
non-linear
execution
like
you
can
define
things
and
that
aren't
used
or
things
like
that.
N
Yeah,
definitely
because
what
once
you
represent
your
workflow
as
dag,
you
can
imagine
paralyzed
it's
in
using
desk
or
leveraging
other
other.
N
Yeah-
and
I
I
just
forgot
to
mention
I
just
sent-
I
just
sent
the
the
link
to
the
repo
on
on
the
chats.
I
would
be
really
keen
to
have
some
some
kind
of
feedback
if
you,
if
you
want
to
raise
an
issue
about
something
that
you
would
think,
would
be
great
I'd,
be
I'd,
be
willing
to
well
well
have
that
included.
Somehow,
if
I
find
the
time
to
do
it
in
part
of
my
job.
A
A
A
D
Yes,
give
me
just
a
second,
and
I
will
make
sure
I
can
share
my
screen
here.
D
All
right,
so
I
want
to
quickly
demo
a
are
people
shooting
my
screen
just
started.
Gene
pattern,
notebook
sure
I
just
want
to
quickly
demo
a
jupiter
hub
extension
configuration
that
we've
been
working
on
called
notebook
projects,
and
there
are
a
couple
driving
use
cases
for
this,
but
one
of
the
primary
ones
is
that
we
want
our
users
to
be
to
be
able
to
load
up
a
jupiter
environment
pre-configured
with
whatever
data
science
or
bioinformatics
tools
they
need.
D
Furthermore,
we
want
them
to
be
able
to
install
their
own
packages
and
customize
the
environment
to
suit
their
methods.
We
also
want
them
to
be
able
to
do
this
for
multiple
environments
or
workflows
without
having
to
manage
conflicting
dependencies
or
other
technical
minutia.
Our
users
are
primarily
scientists,
so
we
want
them
to
be
able
to
focus
on
the
science
and
we
want
this
in
a
user-friendly
interface.
D
Okay,
that's
a
lot
of
things,
but
I
should
note
that
we've
set
up
our
hub
for
bioinformatics,
but
it's
generalizable
to
most
domains,
whatever
you're
working
in
so
you
sh.
What
you
should
be
seeing
here
is
my
login
screen.
This
is
just
the
jupiter
hub.
Login
page
with
a
theme
put
on
top
of
it.
D
We've
got
our
own
authenticator,
but
when
we
log
in
we'll
be
brought
to
what
we're
calling
the
notebook
projects
page
and
it's
going
to
be
the
core
of
what
I'll
be
showing
off
here
so
on
this
page,
you
can
view
your
existing
projects
search
through
them,
create
a
new
one
or
do
their
management
tasks
and
to
launch
one.
I
just
click
on
it.
D
You
can
see
that
will
start
up
a
jupiter
instance
here.
So
every
notebook
project
is
essentially
its
own
jupiter
server
running
any
docker
container,
each
with
its
own
environment,
including
locally
installed
dependencies
and
other
components.
So
you
can
see
some
but
there's
some
extensions
here
behind
the
scenes.
Each
project
essentially
consists
of
three
things:
one:
a
working
directory
for
your
notebooks
and
other
files,
and
this
includes
your
jupyter
prof
file
and
user
installed
packages.
D
Two
this
directory
gets
mounted
into
a
docker
container
chosen
from
among
a
white
list
of
available
images,
and
the
image
is
chosen
when
you
first
create
the
project
and
three
kind
of
switching
back
here.
Each
project
has
some
metadata,
such
as
a
name,
a
description,
a
tags
and
the
docker
image.
That's
chosen,
and
you
can
see
some
of
that
here.
You
can
also
edit
any
of
it
after
creation.
By
going
to
this
gear
menu
here
and
selecting
edit
and
you
can
you
can
change
all
that
there
you
can
also
the
gear
menu.
D
Also
lets
you
start
stop
or
delete
a
project
so
to
implement
this
we're
using
bits
and
pieces
developed
by
the
jupiter
community.
Each
project
corresponds
to
a
server
using
jupyter
hub's
named
server
api,
and
when
you
launch
a
project,
we're
using
swarm,
spawner
to
spin
up
spin
down
and
otherwise
manage
docker
containers,
the
products
page.
D
Will
make
use
of
these
sources
vpn
handle
re-rep
and
some
other
management
tasks,
and
if
you
know
what
I
should
note
that
what
you're
seeing
here
is
a
work
in
progress.
We're
going
live
with
this
in
a
few
weeks
in
our
workspace
in
the
future.
We're
planning
to
extend
this
functionality,
namely
for
us
notebook
projects,
are
going
to
become
what
we're
going
to
use
as
the
unit
of
notebook,
publishing
and
sharing.
D
You
know
where
they
get
a
notebook
and
then
they
have
missing
packages
or
they
need
to
figure
out
what
labor,
what
version
of
a
particular
library
was
installed
to
reproduce
the
results,
and
in
this
case
they
can
just
load
up
the
project
and
run
so
maybe
I
went
through
that
kind
of
quick,
but
that
about
sums
up
the
demo.
I
have
for
today,
if
you
want
to
look
at
any
of
the
code.
It's
all
open
source
on
github.
D
A
F
D
Con
we've
considered
binder,
but
at
least
for
our
use
case
with
our
users
that
binder
essentially
means
launching
from
a
gear
repository
and
a
lot
of
our
users
try
to
do
things
without
a
lot
of
like
direct
coding.
So
this
allows
more
persistent
workspace
without
the
need
to
rely
on
git
commits
and
stuff.
At
the
back
end.
D
I
can
I
can
talk
a
bit
about
that,
so
we're
working
in
a
domain
really
where
we
have
a
mix
of
people
who
are
say
bioinformaticians,
who
are
very
good
with
coding
and
the
command
line
stuff,
and
we
also
have
people
who
are
bench
biologists,
who
are
very
good
at
the
science
of
what
they're
doing
on
working
on,
but
they're
they're
they're,
not
like
primarily
programmers.
You
know
like.
Maybe
they
can
hack
out
a
line
of
r
or
tube.
That's
that's
about
the
extent
of
their
the
programming
experience.
D
D
We
do
a
lot
of
widgets
so
that
they
can
just
point
and
click
and
arrange
the
workflows
without
having
to
directly
write
the
code
themselves,
and
this
is
a
way
to
ease
the
management
of
environments
and
dependencies
for
users
in
that
full
spectrum
without
having
to
do
a
bunch
of
creating
content
environments
or
something
like
that
from
the
command
line.
A
D
Yeah
yeah
we've
found
it
at
least
in
a
lot
of
a
lot
of
these
labs.
We
work
with
the
bioinformaticians
end
up
becoming
a
bottleneck
because
they
have
the
coding
experience
necessary
to
run
these
things,
and
it
helps
basically
remove
that
bottleneck
so
that,
hopefully
you
know
everyone
can
run.
D
So
we
do
have
a
jupiter
lab
version.
That's
currently
in
beta
it
hasn't,
it
does
doesn't
have
a
full,
stable
release
yet,
but
that
should
be
hopefully
be
coming
in
the
next
month
or
two.
If
you
pull
up
our
gene
pattern,
notebook
repository,
you
can
find
it
there's
a
branch
called
lab,
which
is
the
jupiter
lab
support.
A
D
Yes,
I've
I've
actually
been
working
on
on
that
very
recently,
we
do
a
number
of
widgets
that
we've
made
work
with
vlad
very
recently
and
we're
trying
to
figure
out
things
that
we're
doing
or
currently
experimenting
with
how
best
to
integrate
that
into
you
know
our
setup
and
just
different
use
cases
for
that.
But
yes,
we
that
is
actively
underway.
Full
integration.
C
D
Yeah,
that's
something
that's
actively
under
investigation
and
we're
we
we
have
this
jupiter
lab
version.
That's
currently
in
beta
I'm
up
this
week,
updating
it
trying
to
update
it
for
jupiter
lab
3
compatibility
since
jupiter
lab
3
just
came
out
yay
and
I'm
also
just
I'm
actually
really
happy
with
some
of
the
the
different
ways
that
extensions
can
be
packaged
in
three.
D
So
I'm
looking
forward
to
that
and
fix
some
of
the
some
of
the
some
of
the
the
the
tricky
parts
that
I
ran
into
with
earlier
lab.
Two
support.
D
And
I
see
one
comment
here
about
a
good
in
notebook
genome
browser
with
track
support.
Actually,
there's
a
there's
a
project
out
there
called
igv
that
has
a
jupiter
integration.
It's
pretty
good
for
a
genome
browser
with
track
support.
Just
it's
not
my
project.
I
just
wanted
to
recommend
it
since
it's
something
I've
worked
with
in
the
past
and
it
does
a
pretty
good.
A
A
Okay,
well,
I
am
reviewing
the
agenda
and
I
think
somehow,
we've
actually
made
it
through
everyone
on
here
who
wants
to
present
this
month.
So
if
anyone
has
any
more
discussions,
they
want
to
bring
up
or
a
very
impromptu
discussion
or
demo.
That's
welcome.
Now,
if
not,
I
can
wrap
up,
but.
J
I
just
want
to
say
a
quick
thank
you
to
isabella
for
resurrecting
these
meetings
and
having
just
a
great
turnout
and
great
demos
today
from
everybody
who
presented
so
I'm
really
excited
and
regenerate
energized
by
the
meeting.
So
thank
you.
A
Thank
you,
that's
very
kind.
I
really
appreciate
it
these
meetings.
I
know
I
think
they're
really
important,
so
I'm
really
glad
to
see
other
people
thinking
they're
important
too
and
taking
the
time
to
show
up
and
share,
because
I
know
I'm
asking
a
lot
of
people
also.
So
thank
you
all
for
being
here.
Jason
mentioned
q.
A
with
everyone.
Do
you
mean
just
in
general,
any
questions
that
people
have
I'm
happy
to
do
that?
I
don't
know
what
questions.