►
From YouTube: JupyterLab Dev Meeting, September 16, 2016
Description
Meeting of the JupyterLab development team for September 16, 2016
Meeting notes: https://jupyter.hackpad.com/JupyterLab-Weekly-Meetings-UUJ3gIQ3iBS
A
A
B
Yeah
working
on
this
morning,
actually
just
put
in
a
PR
to
great
upgrade
prospered
to
type
to
get
to
know
which
is
nice
is
a
pretty
straightforward
change.
It
wasn't
entirely,
they
could
dig
files
straight
out
and
that's
because
the
compiler
caught
one
thing
where
we
were
faking
creating
keyboard
events
and
the
compiler,
the
updated
compiler
caucus
baking
that
so
I
just
have
to
use
a
bigger
hammer
to
get
to
work.
B
You
know,
option
bags
and
that
kind
of
stuff
that
should
all
be
read-only.
We
can
now
go
ahead
and
use
that
keyword
so
I
like
those
updates
and
then
the
other
big
feature
that
I
think
we
can
make
you
so
is
the
strict
Mel
checking
which
is
going
to
be
a
bit
harder
to
roll
into
the
code
base
or
just
a
bit
more
effort.
It's
not
going
to
be
as
straightforward,
but
that's
really
nice,
because
that
will
free
us
from
having
to
do
a
lot
of
checks
to
see.
B
A
Quick
question,
so
we
ran
into
the
same
problems:
updating
typescript
for
the
Jupiter
lab
code
base
where
it
caught
some
actually
caught
a
place
or
two
that
we
were
doing
things
incorrectly
and
then
other
places
would
need
to
be
cameras.
But
the
read
only
issue
was
a
big
thing
as
well.
I
left
the
read-only
tests
in
because
I
didn't
want
somebody
to
later
on
put
a
set
handler
asset,
pant,
learn
in
the
for
an
attribute,
and
so
I
I
just
coerced
links
to
any
before
trying
to
set
it
and
then
checking
a
javascript
error.
A
C
D
There's
an
outstanding
PR
that
I
think
to,
and
then
I
tried
to
see
if
there
any
outstanding
PRS
that
had
just
been
hanging
there
for
a
while
that
could
could
be
reviewed
and
a
couple
women,
and
also
today,
oh
good
dance
on
one
dan
and
I,
had
a
conversation
about
the
barn
councillor
stuff
that
he
can.
He
can
talk
about.
That's
pretty
much
it.
D
I
hope
to
finish
the
completer
stuff
today,
I'm
just
stuck
in
the
middle
of
a
bug,
that's
a
little
tricky,
but
but
after
this
one
last
bug
fix
I
think
we
should
probably
merge
and
iterate
from
there.
With
the
there
are
some
finer
behaviors.
We
might
still
want
to
nail
down.
Matthias
and
I
have
been
talking
on
the
on
the
actual
thread,
but
I
think
we
aren't
one
hundred
percent
clear
on
some
of
those
issues
yet,
but
I
think
we're
close
so
yeah,
that's
it
for
me.
I.
E
F
C
D
A
F
A
E
E
A
F
A
G
Don't
have
too
much
I,
just
you
know,
we'll
be
sending
some
materials
out
today
about
schedule
for
next
week
again
I'm
really
looking
forward
to
seeing
you
guys
work
on
Jupiter
logs
out
there,
I
did
I
surveyed
the
team,
like
the
non
Jupiter
lab
groups
on
the
team
to
see.
If
there's
anything
they
wanted
me
to
listen
for
specifically,
so
I
could
give
you
know
just
kind
of
keep
people
updated
on
the
goings-on
next
week.
So
I
just
have
a
few
things
for
men
and
hadn't
heard
back
from
anyone
else.
A
And
I
have
two
things:
I've
been
working
a
lot
on
the
third-party
extensions,
so
I,
padua
chitz
is
working
fine
at
least
minimally,
but
what
I
really
know
is
that
you
can
get
a
slider,
that
you
can
move
back
and
forth.
I
need
to
test
more,
but
it
seems
like
it's
working,
fine,
floating
CSS
and
everything
I'm
currently
working
on
BQ
plot
and
it's
bringing
up
an
issue
Steve.
Maybe
you
can
comment
on
this.
How
are
we
treating
custom
loaders
right
now?
I'm
I'm
using
the
plugin
infrastructure
or
the
the
third
party
extension
bundling.
A
If
structure
and
I
have
a
less
loader
4bq
plot,
it's
coming
out
with
a
big
huge
string,
that's
could
being
required.
That
has
a
bunch
of
exclamation
points
and
you
know
it
loads.
The
less
letter
than
a
CSS
loader
did
we
I
know.
We
at
one
point
did
something
with
the
custom,
loaders
and
exclamation
points
in
the
syntax.
C
C
The
the
default
settings
of
the
Builder
will
use
the
existing
default
loaders,
none
of
which
are
less
loader.
So
you
would
provide
that
as
external
web
pack
config
and
that
that
would
result
in
a
module
that
looks
like
something
exclamations
and
that
that
would
get
parsed
in
each
part
would
get
a
sin.
C
Ver
version
angled
version
a
version
of
that
import,
that's
that
it
should
ok
and
then
that
should
just
transparently
work
as
long
as
there's
only
one
of
those
arm
matching
full
strings
with
the
the
bash
separated
pieces,
then
it
won't
be
d,
duped
and
then
that's
what
you
get.
But
I
would
expect
that
to
just
work
as
I'm,
saying:
ok,
ok,
so
I
think
that'll
go
Ralph,
the
loader
itself
will
be
a
module
and
then
the
thing
that
uses
the
loader
they
require
statement
will
have
the
loader
bash
and
then
something
else.
Ok,.
A
And
I
think
that's
what's
happening.
Ok,
I'll
dig
deeper.
Ok,
your
input
about
what
you
think
should
be
happening,
and
second
thing
so
Paul
even
off
has
been
working
on
a
Jupiter
lab
sitting
beside
me.
This
last
week
he's
working
on
a
launcher
as
a
sort
of
ramp
up
up,
get
up
to
speed
project
and
so
he's
taking
the
current
page
that
has
the
new
notebook
console,
file,
etc
and
turning
it
into
a
service
that
has
a
registry.
A
B
Covered
it
yep,
I
would
say
Paul
if
you
could
keep
notes
about
like
you're
learning
process
and
like
as
you
has
your
since
you've
got
fresh
eyes
on
this
and
you're
like
a
new
user.
If
I
keep
notes
about
what
you
experience,
what
you
like,
what
you
don't
like
you
know,
because
a
lot
of
us
might
have
blinders
on
for,
like
the
view
user
experience,
since
we've
been
in
it
for
so
long,
so
that
would,
I
would
definitely
be
good
feedback
to
have
yep
totally
totally
will
do
that.
E
F
A
That
was
enough:
yeah
watch
your
email
for
Jamie's
email,
sharing,
the
google
doc
that
has
a
bunch
of
planning
information
dan.
If
you
could
email
us
about
what
pays
you'd
like
to
join
us
that'd,
be
great
and
we'll
see
a
lot
of
you
in
New,
York
and
then
Chris
I
guess
some.
Some
of
you,
like
Chris,
will
catch
up
on
video
conference.