►
From YouTube: Jupyter/IPython Dev Meeting June 5, 2017
Description
Meeting of the Jupyter/IPython development team, June 06, 2017
Meeting Notes: https://paper.dropbox.com/doc/June-2017-Jupyter-Weekly-Meetings-hsOrcAavxaIrg1ZiSZq9G
A
B
C
B
D
B
It's
wonderful
to
have
you.
Finally,
working
with
authentica
has
been
insanely
patient.
The
process
of
sending
back
column
vector
optic,
a
lot
of
efforts,
because
it's
kind
of
administrative,
hassles,
Damien
work.
A
lot
of
that
Jessica
comes
from
a
background
and
fitness
to
conceal,
earning
and
actually
have
extensive
experience
using
Jupiter
as
a
research
tool.
B
So
it'll
be
it'll,
be
really
great
to
have
someone
who's
kind
of
deeply
seeped
into
the
tool,
but
will
also
brings
a
lot
of
expertise
in
terms
of
technical
writing
to
the
project,
an
area
where
we
needed
that
so
I
think
the
plan
would
be
per
kind
of
to
focus.
First
on
do
bottom
lines
and
areas
that
right
now
has
very
little
on
the
way
of
more
user.
D
B
Documentation
as
a
starting
point,
something
that
we
want
to
start
presenting
to
the
public
more
quickly,
I
knew
that,
as
the
first
kind
of
concrete
deliverable
for
the
process
of
building
a
set
of
user
facing
dog
that
we
can
also
use
in
the
community
that
can
be
used
this
to
teach
tutorials
and
think
that
there
is
like
two
gray
days
or
some
of
the
LVP
Tercel.
So
we'll
Jessica,
I
think
officially
joined
us
last
week.
Thursday
or
Friday
was
sort
of
first,
the
first
of
all.
B
They
are
happy
having
to
pay
for
a
fine
life,
and
it's
great
to
have
you
on
thirteen
deskah.
What
kana
do
you
want
to
keep
us
in
check
with
the
addendum
I
mean
for
everyone
else
who,
just
last
week
our
All
Hands
meeting,
which
or
many
many
hands
not
all
hands
work
on
that,
but
our
many
hands
meeting
and
it
was
I-
think
it
was
fun
and.
A
B
B
A
E
E
E
A
Thank
him
and
let's
see
a
heads
up,
Ian's,
not
sure
or
Mattias.
If
you
can
keep
an
eye
on
the
chat,
my
interface
has
frozen
and
I
am
going
to
assume
you
guys
can
hear
me
and
see
me:
okay,
but
I'm
not
able
to
interact
with
the
UI
right
now.
A
F
G
A
B
I
A
B
Advanced
for
what
the
permissions
on
these
files
are,
but
we
should
make
we
should
make
these
notes
completely.
Public
I
mean
they're
part
of
the
public
meeting.
These
are
standard
public
notes.
MatÃas
summaries
are
very
nice
posting
to
the
main
list,
but
there's
other
many
links
and
other
things
and
files
that
we
we
should
make
public
at
least
to
space
it
on
the
Gator
channel,
and
hopefully
they.
A
B
So
the
hackpad
migration
I
think
it's
all
done.
We
were
moved
onto
paper
and
Jamie
was
Jamie
had
worked
on
getting
the
download
of
everything.
I
think
that
stuff
is
all
done.
I'll
double-check
the
per
but
I
know
he
was
working
on
on
getting
it
downloaded,
but
non
focus,
but
hi
Donna
talk,
toss
aside
out
of
Seattle,
possibly
on
wheels
and
collaboration.
I
think
Ian,
Chris
and
Brian
have
been
coordinating
on
that.
B
So
it's
taken
care
of
and
Jamie
mentioned
that
folks
have
been
asking
her
about
at
Python
about
the
different
last
position
sort
of
how
fast
we
would
expect
users
without
fit.
Obviously
it's
it's
very
hard
to
know
exactly
how
quickly
a
transition
will
happen
out
in
the
community.
It
seems
like
Jamie
Gates
gave
kind
of
a
ballpark
estimate
that
probably
it
would
take
about
a
year
after
one
point.
Os
release
or
for
sort
of
the
vast
majority
of
community
has
transitioned.
B
That
sounds
to
me,
like
a
reasonable
estimate,
would
be
happy
with
you
in
Conroy's,
when
I,
when
I
think,
for
example,
of
like
Android
is
now
on
whatever
heard
in
their
active,
seven
or
six
or
eight
or
whatever,
but
there
is
I
hope
speed,
see
found
in
the
wild,
using
very
old
versions.
There's
always
a
long
tail
people
who
just
don't
updated
stuff
right
and
so
the
nodes
of
classic
notebook
has
a
large
user
base.
It'll
it'll
be
awhile
before
everyone
transitions,
but
I
think
once
we
hit
1.0
with
good
pip
and
Conda
updates.
B
F
You
say
in
our,
but
what
them
one
thing
we
might
want
to
do
is
if
we
want
to
be
able
to
track
this
since
to
get
user
feedback
and
stuff,
rather
than
just
thinking
about
downloads,
actually
have
a
just
a
flag
in
the
metadata
that
doesn't
at
the
notebook
level.
That
doesn't
mean
much
of
event.
This
was
created
and
edited
in
a
in
lab,
because
I
was
just
thinking.
What
are
the
statistics
that
we
could
get?
F
B
It's
not
a
bad
idea
and
also
I
mean
even
over
time,
seeing
the
presence
of
that
on
github
I
mean
we
could
query,
could
query
ideas
on
github
and
get
us
at
least
a
relative
sensing
and
the
time
a
little
bit
of
sort
of
I
wouldn't
I,
wouldn't
make
it
so
so
specific
to
create
a
lot
of
rather
something
like
that
created
by
or
a
client
playing
creation,
ID
Act,
which
is
optional
back
bad
idea.
I.
E
E
I
have
a
notebook
somewhere
on
one
of
them,
sir.
Maybe
even
a
cron.
That's
been
a
bit
too
much
kitty
and
put
out
some
some
graph,
and
we
can
even
do
that
for
my
use
project.
We've
done
that
at
PyCon
for
iphone
5,
vs,
iPad,
m6
and
I
was
thinking
that
something
for
more
projects
and
how
many
terminals
to
have
her
birthday
could
be,
could
be
useful
to
the
graph
and
we
looked
in.
It
doesn't
accept
partial.
F
F
B
We
have
a
G
I,
don't
know
that
we
have
it
that
we
haven't
set
up
the
porters,
but
we
can
yeah.
We
can
certainly
look
into
that,
but
but
it
can't
prove
it
I'm,
starting
to
do
some.
Some
basic
individually
is
not
a
bad
idea
and
certainly
recording
recording
the
client
creation
ID
so
that
we
can
track
creation
with
with
quantum
notebook
with
Jupiter
lab,
which
will
interact
over
time
is
is
yes
that
sounds
like
a
reasonable.
E
F
F
C
B
B
Id
I
mean
in
order
like
what
file
systems
do
with
this
kind
of
problem
is
basic
or
creation.
Typically,
files
have
two
kinds
creation
time
and
last
last
modified
time,
so
you
could
record
this
creation
and
last
notification
or
you
in
description
it'll,
be
annoying
statistics,
but
a
little
bit,
but
they'll
still
be
useful.
A
B
Absolutely
and
Jessica,
no,
obviously,
no
no
pressure
and
right
now
publicly
expected
anything
majors
that
have
happened
by
today,
but
it's
worth
to
make
sure
that
we
have
a
chance
to
check
on
you
and
that
we
don't
forget
often
unless
it's
written
down
in
the
standing
kind
of
skeleton
of
the
agenda.
It's
easy
to
forget
to
check
in
on
people
Remington
for
you
Jessica.
Sorry,.
D
That
makes
total
sense.
Yeah
no
I
mean
for
me,
I'm,
I'm,
literally
beginning
to
ramp
up
and
wrapping
my
head
around
the
juices
mo
right
now,
so
I
don't
have
Mimi
yeah.
B
A
A
A
I
So
so
forgiver
lab
we
will
be
working
on
bundling
the
mime
renderers
for
things
like
the
json,
renderer
and
potentially
geo
geo,
JSON,
Bega
plotly.
So
in
tandem,
I
would
like
to
implement
the
same
thing
in
the
class
notebook
and
it's
going
to
be
different.
It's
a
different,
API
and
stuff
like
that,
but
that'll
be
one
of
the
things
working
on
on
triaging
issues
as
usual
and
I'd
like
to
start
organizing
and
coordinating
a
5.1
release,
so
I'm
not
sure
the
mime
renderer
stuff
will
be
included
in
that.
I
B
F
B
F
Agreed
a
couple
weeks
ago,
I
released
the
indicator
to
5.2.1,
which
has
a
bunch
of
features
that
I'm
not
remembering
off
the
top.
My
head,
because
there's
a
couple
weeks
ago
additionally
knows,
is
no
longer
in
our
test
base.
We
cut
it
off
to
spite
our
face,
but
it's
we're
now
just
using
PI
test
there.
But
those
are
the
major
things
in
my
mind
or
that
that
happened
and
I'm
want
to
implement
the
tag
based
filtering
operations.
F
G
G
Lots
of
or
the
best
got
to
accept,
icon
and
the
and
the
meeting
in
Berkeley
there's
been
a
lot
of
progress
on
finder
related
things
and
Daryl
and
hoaxes
have
been
working
on
the
documentation
for
deployment
tutor
hub
as
part
of
binder
or
just
on
zone
with
coup
Brandis
using
helm,
and
so
that
that's
going
very
wrong
there.
People
are
writing
that
up
and
I'm
doing
him
some
good
stuff
there.
G
F
G
Should
be
able
to
do
it
automatically,
but
at
least
we
just
going
through
the
steps
of
doing
an
update.
This
is
now
a
free
show
for
it
that's
good
and
then
there's
also
been
some
work
on
docker
stacks
related
to
the
kubernetes
stuff.
Some
work
on
docker
stacks
to
make
it
possible
one
of
the
things
that's
been
an
impediment
to
using
docker
stacks
in
some
environments
is
the
sector
key,
install
stuff
in
the
home
directory
and
quit
users
work
in
a
subdirectory
of
home
called
work.
G
And
since
we
want
to
allow
people
to
note
home
as
a
volume,
we've
been
totally
systematically
getting
everything
out
of
home
and
making
sure
its
installed
in
system-wide
locations
so
that
if,
if
I
use
your
mouse,
there
is
the
employment
amount
home
for
the
users,
then
can
go
work
and
that
the
last
left
bits
of
that
peter
merged
last
week
and
then
just
need
to
deploy
a
new
tag
for
the
docker
section,
and
you
can
start
using
those
in
a
few
more
courses.
This.
B
A
H
Now
so
yeah
so
last
week
we
released
version
0.23
and
that's
got
the
early
support
for
our
RTC
that
is
been
working
on.
We've
been
rethinking
how
the
launcher
works.
Basically,
we
want
it
to
be
a
bit
more.
We
wanted
to
remember
what
things
you
use
and
then
just
want
to
rethink
that
user
experience,
so
that
network
is
ongoing
and
we
setting
system
feature
is
really
close
to
an
initial
release.
H
In
fact,
after
this
call,
if
some
people
have
some
free
time,
I'd
like
to
do
a
just
a
quick
run-through
on
look
at
sort
of
big
evening,
just
on
a
separate
branch
that
people
may
not
have
seen
so
all
right,
so
we
we
decided
that
one
of
the
public
API
is
of
Jupiter
lab
ought
to
be
a
set
of
reusable
components.
H
H
I
I
think
what
you
said
about
the
virtual
Dom
stuff
is
is
as
good
yeah,
Brian
and
I.
We
actually
started
to
take
a
look
at
what
it
would
take
to
like
what
changes
would
have
to
be
made
to
phosphor
in
order
to
enable
this-
and
it
looks
like
it's
literally
like
one-
a
couple
lines
of
code,
but
then
eh
pumpkins
when
you
walk
with
us
I'm.
K
I
Be
changed
yes,
so
as
we
can,
okay
things,
amam,
colors
and
so
yeah.
H
It's
great
yeah,
so
I
guess
that's
all
right.
Do
you
want
to
talk
about
the
mine
renders
I'm
not
exactly
sure
yeah
yeah.
I
So,
like
I
mentioned
guarding
the
notebook,
the
plan
is
to
bundle
the
mind
renderers
in
to
jupiter
lab,
so
that
any
output
that
has
you
know,
application,
json
mime
type,
for
example-
will
render
as
a
tree
and
that
will
not
require
users
to
install
additional
extensions
on
that
will
be
built.
Then
same
for
Vega
and
plotly
and
potentially
geo
JSON.
L
It
actually
also
potentially
widgets
as
well,
but
something
started
to
talk
to
Jason
about
and
it
may
make
sense
an
elite
need.
Some
of
the
pain
in
basic
ship
will
have
the
notebook,
with
a
version
of
which
is
part
of
where
that
one's
less
there's
widgets
is
changing
much
faster
than
all
these
other
libraries.
And
so
it's
not
clear.
What's
version
we
would
ship,
but
it's
something
that
that
we're
starting
to
think
about.
B
Before
we
move
on
before
we
move
off
laughs,
maybe
I
didn't
want
to
give
Ian
a
chance,
maybe
to
say
more
words
about
the
real-time
collaboration,
because
that
I
know
that's
something
that
everyone's
been
waiting
for
a
long
time
and
Darien
mentioned
it
in
passing.
But
it's
something
kind
of
big
enough
and
if
we
all
care
enough
about
that,
it's
probably
worth
saying
a
few
more
words
about
what
what
went
in
and
how
it
works
in
the
context.
B
M
Sort
of
did
a
soft
release
last
week
to
start
playing
with
it.
So
there's
a
0.1
release
of
an
extension
to
do
better
lab
that
uses
Google
Drive
in
the
Macan.
You
can
store
notebooks
and
text
files
and
markdown
files
and
Google
Drive,
and
collaborate
on
the
data
and
edit
them
collaboratively.
So
there's
a
fair
amount,
that's
missing
in
terms
of
UI
elements
and
like
cut
the
facial
smelly
is
there
and
people
want
to
take
it
out.
Person
try
to
see
what
the
install
process
is
like
see.
E
M
B
B
H
H
Just
saying,
thanks
for
calling
it
out,
it's
yeah,
it's
amazing,
to
see
like
I
could
get
blown
had
a
pretty
positive
reaction
last
week,
playing
with
it
together
and
it'll,
be
good
for
lots
of
people
to
use.
It
know.
L
Yeah
we
started
to
discuss
a
couple
topics
related
to
the
notebook
format
at
our
team
meeting
last
week.
The
first
one
is
a
work
to
formalize
the
metadata
for
hiding
input
and
output.
Yeah
and
there's
been
some
discussion
about
whether
or
not
that
would
be
in
the
tags
section
of
the
cell
metadata
or
in
the
metadata
one
level
up
in
the
metadata
section
itself.
Kyle
has
a
PR
open
and
we're
we're
discussing
the
options
there.
L
The
second
is
that
we
would
like
to
formalize
our
understanding
of
what
we're
calling
Jupiter
markdown
and
it's
that
that
makes
it
sound
fancier
than
it
is
at
this
point.
What
we
want
to
do
is
simply
document
the
precise
markdown
syntax
that
we
support
in
the
classic
notebook
and
we
we
may,
as
we
documented
discover
some
weird
edge
cases
that
we
want
to
take
a
step
back
on.
L
I,
don't
think
they're.
Quite
there.
They
were
doing
some
final
package
and
repository
refactoring
last
week,
but
I'm
guessing
that
very
soon
that
will
be
released.
It
may
be
delayed.
I.
Think
Jason
is
actually
a
away
for
two
weeks,
so
it
may
slow
down,
but
but
the
release
is
coming
soon
in
either
case.
Okay,.
B
And
one
I
think
one
one
useful
way
for
for
us
to
communicate.
You
kind
of
set
it
but
I
kind
of
want
to
emphasize
it.
The
the
jupiter
markdown
thing
is
that
in
a
sense
we
realize
that
our
we've
been
very
precise
in
our
documentation
of
the
data
structure,
part
and
affordable
and
Jason's
back
and
adjacent
schema,
and
that
was
very,
very
carefully
and
precisely
documented,
and
we
realized
that
we've
definitely
left
the
loophole
in
that
the
markdown
part
was
just
a
black
box
like
well
it's
markdown
and
who
knows?
B
Who
knows
what
happens,
and
so
this
is
in
a
sense,
basically
just
finishing
that
job,
it's
sort
of
finishing
the
job
of
completing
the
specification
of
what.
What
do
we
mean
by
a
Jupiter
Notebook
document?
That
means
something
at
the
data
structure
level,
and
it
also
means
something
at
the
content.
Okay
and
we're
just
finishing
that
are.
B
N
N
Been
rumbling
for
a
long
time,
I
mean
you
got
people
that
are
doing
like
the
genebank,
notebook
and
stuff,
like
that.
I've
done
a
lot
of
aggressive
things
that
overload
magics
and
stuff
like
that
and
hide
it
at
the
content.
Anyway,
source
doors,
:,
open
quote
or
whatever
you
know,
Albert's
represented
down
there,
that's
I
would.
B
This
is
just
documenting
the
status
quo
with
precision
really
it's
just
clarifying
clarifying
with
some
given
the
fact
that
there's
github
flavored
markdown,
there's
common
mark
and
there's
actually
our
live,
render
there's
a
conversion
that
nd4
that
in
different
words,
it
does
there's
the
fact
that
people
view
notebooks
on
github,
which
has
a
different
render
and
ours
with
different
constraints
on
javascript
and
math.
It's
just
a
matter
of
trying
to
stay
precisely
for
even
for
users
and
ourselves.
What
is
the
lay
of
the
land
today
and
then
from
there?
We
can.
B
We
can
lose
slightly
more
invasive
or
aggressive
changes,
and
at
least
we'll
know
it
nothing
else.
We
will
know
what
they
in
flock
right
now.
Some
of
those
changes
land
in
the
fuzzy
area
that
is
under
specified
so
for
now
it's
a
matter
of
drawing
those
boundaries
more
precisely
and
I
would
run.
I
would
want
to
finish
this
job
first
before
we
whole
wheat.
Ask
all
the
questions
you're
talking
about
me,
yeah.
L
F
It's
also
the
case
that
we
should
really
do
this,
because,
with
lab,
we
actually
are
now
creating
not
just
a
markdown
renderer
in
the
context
of
the
notebook
from
Merck
themselves,
but
because
the
same
renderer
can
be
used
to
render
actual
markdown
documents
like
we're
effectively.
Yet
another
merkon
lender
that
should
be
included
in
like
bable
bable
mark
or
whatever
so
trying
to
be
precise
and
understand
what
our
state
is
is
going
to
be
helpful
and.
B
B
A
Okay,
I
think
we're
doing
we're
doing.
Okay,
we
have
Jupiter
hub
and
ipython
updates,
and
then
we
can
go
quickly
through
the
events
and
then
the
new,
the
releases,
the
past
ones,
and
then
they
expected
ones
coming
up
so
min.
Do
I
want
to
give
an
update
on
Jupiter
hub
Carroll
is
traveling,
I,
think
she's
keynoting
pike
on
Taiwan
I.
G
B
D
G
G
E
Quickly,
so
I
am
released,
ipython
5-4,
which
is
backing
to
and
pythons
we
compatible
last
week
as
well,
and
six
one
I
also
immediately
released
a
time
for
one
because
I
for
was
brought
up
with
ambition.
Please,
real
green
I
will
look
at
the
download
statistical
by
d-I
to
see
how
long
migrations
Python
3
only
is
going.
We
already
had
some
curve
that
Python,
but
because,
when
you
have
kid,
9
say
well,
there
is
no
compatible
version,
I'm
not
going
to
agree
that
we
don't
have
any
participation,
people
downloading
and
we
don't
have
good.
E
Typically,
some
people
is
emitting
from
by
don't
you
and
so
I
hope.
That's
raising
six
one
and
five
four.
At
the
same
time,
animals
to
compare
the
curve
of
the
node
on
both
item
two
and
Python
three
and
give
them
give
us
more
insights
on
transition
and
I
will
probably
work
with
em
to
have
the
density
kicks
for
all
the
project
at
once,
I'm
working
on
a
heading
called
signature
in
in
ipython,
and
so
that's
when
it
completes
the
computer.
E
We
use
a
third
argument
of
the
function,
you're
you're
completing,
and
so
it
can
also
impact
ID
like
ml
and
the
new,
the
new
private
key.
We
added
that
have
lock
up
formation
on
completion
and
in
particular
about
especially
for
really
really
longs
to
measure
how
to
turn
that
into
the
UI,
because
it
can
looks
like
quite
ugly
if
you
have
a
function
with
a
signature
which
is
like
hundreds
of
characters.
E
Long
as
there
are
also
some
question
of
naming
of
these
keys
to
be
more
cross
language
compatible
right
now,
if
we
reality
are
again
I'd
like
caramel,
that
which
is
proposing
to
call
that
signature.
I
was
thinking
of
maybe
renaming
it
system
single
general
like
right
extra,
so
we
can
have
a
less
extra,
which
is
something
which
is
shown
either
on
the
left
or
on
the
right
of
the
completion
in
the
UI.
They
would
like
to
have
feedback
on
tax
I,
put
a
link
in
a
box
of
purple
nodes.
E
Well,
yes,
I
can
display
no
images
and,
more
generally,
you
might
want
to
display
a
given
line
type
in
the
terminal
using
an
external
viewer,
unlike
resin
traditional
micro,
as
you
can
use,
you
can
trigger
for
comically
quick,
look
or
you
can
call
the
if
you
have
keeping
sewing
scape
or
whatever,
and
one
thing
we're
thinking
about
is
because
trippitorial
and
the
end
transports
are
working
on
having
my
mine
renderer.
Can
we
have
a
standalone
electron
axle
to
display
all
the
vine
pipes
at
our
lab
and
and
tracks
can
can
happen?
E
And
so,
if
she
still
has
a
control,
we
can
register
it
in
their
configuration
and
then
ipython
cells
can
display
all
the
collide
plots
and
everything
not
in
line,
but
at
least
open
the
side
window,
which
is
able
to
show
it
is
rich
to
review.
Of
course,
it
would
have
to
be
opt-in,
but
I'm,
just
reading
that
here
for
people
to
consider
tile
and
I've.
Already
played
with
something
algorithm
called
sidecar,
but
now
nice
melodic
architecture
will
sink
and
shared
between
content
and
like
do
something
need
to
do.
E
L
Tackle
it
with
another
thing
that
we
thought
about,
and
that
is
enabling
the
notebook
server
to
connect
to
Colonels.
It
did
not
start,
and
so,
for
example,
if
someone's
running
a
command
line
terminal
that
has
the
actual
kernel
machinery
running,
you
could
connect
the
notebook
server
to
that
and
instantly
use
any
phone
and
then.
E
Yes,
but
I,
but
in
short
terminal
ipython,
is
not
static
and
editing
process
commands.
There
is
no
D
in
communication
because
of
other
signings
like
in
the
to
process
can
manage
accounts,
for
example,
zoobank
being
a
nice
thing
to
a
family,
so
I
think
it
makes
sense
to
also
have
a
standalone
application.
Also
because
most
people-
don't
many
producers,
don't
use
notebook.
We
still
have
wide
majority
of
purely
I,
just
replied
right
now:
I
don't
have
a
20
ml
and
it
just
wants
to
register
external
reviewers
I'm.
E
E
And
if
you
keep
your
might
on
your
code
using
Python
for
inline
and
nobody
else,
we
needs,
but
I
was
thinking
just
like
justice
on
the
loan
application,
and
especially
since
now
there
is
this
new
lightweight
electron,
which
seems
to
start
to
be
a
thing.
We
might
have
something,
not
hundreds
of
megabytes
to
the
mill
I.
B
E
B
Don't
think
one
precludes
the
other
by
the
way,
in
the
case,
in
the
case
where
we
have
proper,
proper
kernels,
the
idea
of
attaching
attaching
actual
notebooks,
which
can
be
very
useful,
but
but
yeah
I
agree
that
something
lightweight
purely
as
a
terminal
level.
When
there's
no
0
mq
in
the
picture,
it's
really
going
to
be
used
for
a
lot
of
people.
B
Yes,
though,
to
be,
though,
to
be
honest,
even
an
electron
app
that
has
to
pop
a
window
is
something
that
that
still
has
some
requirements
and
I
actually
think
that
the
idea
of
displaying
inline
images
and
things
like
your
SSH
into
a
real,
a
remote
server.
But
you
have
a
modern
terminal
emulator
and
having
a
little
bit
richer,
display
right
in
the
terminal
that
still
would
be
great
yeah,
very
nice
purely
in
process
purely
in
terminal
terminal.
Emulators
have
come
a
long
way
and
advantage
of
balance
without
popping
off
separate
windows
would
be
really
lovely.
B
E
Unfortunately,
we
got
auto
detect,
so
it
has
to
be
your
users
that
preference
or
it'll
be
highness
and
terminal
terminal
which
is
able
to
play
mp3s.
Oh
yeah,
you
could
play
mp3,
so
you
could
you
could
listen
to
the
Lego
Lego
netbook
mini
tamiya,
so
yeah
you
can
you
can
push
through
or
something
like
that.
I
was
just
going
to
giving
a
heads
up
at
something.
I
got
is
coming
we're
working
on
that
which
you
have
inputs.
Please
come
on
the
chart
and
let's
try
to
not
keep
the
making
too
long.
E
I'm
going
to
go
fast
and
heads
up
as
a
new
Mac,
OS
and
file
system
might
give
us
some
hard
time,
especially
for
users
are
using
both
the
finder
and
the
command
and
command
line
arguments,
because
the
normalization
of
filenames
we
apparently
not
be
the
same.
She
goes
with
the
UI
and
goes
to
terminal,
so
you
can
have
two
files
that
look
exactly
identical.
That
may
not
be
stealing
beta
to
the
mic
success.
We
should
get
Unicode
reports
on
microwaves
issue
because
it's
bottomless
for
the
next
couple
of
months.
A
Right
anything
else
on
ipython
before
we
move
on
okay,
so
I
think
I'm
going
to
ask
people
to
go
ahead
and
read
through
the
agenda
on
any
other
technical
topics.
I'm,
not
sure
who
put
the
note
in
about
the
trait
let's
5.0
release,
Fernando
or
Brian.
If
you
want
to
add
some
notes
into
the
governance
update
section,
that
would
be
good,
otherwise
related
to
conferences.
A
Please
continue
updating
the
project
calendar
on
github,
there's
a
link
provided
there,
so
you
can
keep
adding
new
events
that
you're
speaking
out
or
that
you
want
to
share
with
the
community
quick,
a
note
or
highlight
for
Jupiter
Khan
the
financial
aid
application
for
students
who
are
interested
in
applying.
If
you
know
any
students
who
need
some
help
getting
to
some
financial
help
getting
to
Jupiter
con,
that
application
is
due
June
16th,
there's
a
Google
form,
that's
been
set
up
and
that's
linked
within
the
agenda.
So
please
push
people
to
to
apply.
G
A
D
B
A
A
F
We
wish
we
had
had
actually
released
ipython,
50.3
or
5.4
I
forget,
which,
one
at
the
time
that
we
released
ipython
6.0,
so
that
we'd
be
able
to
detect
the
changes.
Many
people
updating
to
both
of
those
having
some
handle
on
on
this,
not
just
downloads
though,
but
the
use
in
the
creation
of
notebooks
in
the
different
clients
would
be
a
good
idea.
G
E
Think
there
might
be
two
or
three
day
left
to
make
a
call
to
shoot
a
tutorial
at
5:00,
Bay
and
I.
Don't
think
we
have
anything
to
the
certificates
beginning
of
August
yeah
and
you
probably
will
already
have
it
already
for
end
of
August.
Do
you
want
to
make
a
submission
about
ticket
elaborate
and
collaboration?
That's.
L
A
Is
there
anything
else,
just
in
maybe
a
minute
anybody
you
want
to
talk
about
any
of
the
releases
for
I.
Think
some
of
this
is
already
or
maybe
all
of
it
has
already
been
talked
about.
Doesn't
look
like
there's
any
releases
for
this
week,
anything
else
on
releases
before
we
go
ahead
and
end
the
recording,
no
okay,
well
great,
it
was
good
chatting
with
you
all
and
Ryan
or
Fernando
as
a
possible
for
or
Matias
can
one
of
you
and
the
recording
I'm
still
frozen.
I
got.