►
From YouTube: YUI Open Roundtable 01/03/2013
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
E
E
E
Then
we're
going
over
in
clearance
is
here
to
discuss
the
specific
board
quest
and
we
have
also
some
other
staff
for
request
as
well,
including
a
time
we
have
someone
assigned
box
that
we
can
talk
as
well,
but
I
think
the
first
off.
Let
me
get
this
link
out
and
then,
if
we
kick
off
with
a
tivo
doing
this
yeah.
G
Okay,
uh-uh,
so
I
just
wanted
to
so
one
of
the
things
I've
been
working
on
for
the
last
week
and
a
little
bit
is
thinking
about
of
this
concept
of
some
prepackaged
default.css
elements
a
lot
with
Jeff.
So
too,
we've
kind
of
worked
on
two
of
these.
What
is
images
0
us
see
this?
Yet,
let's
not
sharing
your
desktop
and
got.
G
Yeah,
you
can
see
that
there
you
go
yeah,
so
we
were
thinking
about
creating
some
gallery
modules
similar
to
what
exists
now
for
CSS
button
utterly
that's
so
behind.
G
Understand
everything
so
yeah
we've
been
thing:
McGrady's
and
gallery
module
tumor
to
CSS
button,
which
pulls
down
some
CSS
that
you
can
use
with
standard
HTML
elements,
so
I've
made
one
for
tables
and
I've
made
one
for
forms,
so
I
just
wanted
to
go
through
so
we'll
start
with
tables.
I
guess
so.
This
makes
HTML
tables
look
a
lot
like
data
tables,
and
there
are
some
different
customization
options
available,
and
this
stuff
is
right
now
is
on
my
get
up.
G
H
G
To
get
that
up
by
the
end
of
the
day,
I'm
also
and
I,
also
for
these
as
I
go
through
they'd
really
nicely
for
all
older
browsers,
I
haven't
noticed
in
any
issues
or
older
browsers.
There's
some
minor
issues
on
like
IE,
6
and
7
with
some
position,
some
positioning
off
x,
1,
pixel,
sort
of
issues,
but
nothing
that
makes
them
unusable
sense.
So
anyways
I
started
going
to
just
show
you
guys
what
I've
worked
in
what
I've
been
working
on
and
get
some
feedback
possible
so
for
its
CSS
tables.
G
If
you
add
a
Yui
3
table
class
to
a
standard
HTML
table,
you
get
this
this
table.
That
looks
like
this
with
just
the
vertical
lines
on
each
column.
So
it's
pretty
standard
with
like
the
styled
competitor.
If
you
add
Yui
3
table
border,
you
get
a
order
table
with
horizontal
and
vertical
lines.
A
G
G
B
G
So
I
was
actually
talking
to
maggie
s
realized
in
the
afternoon,
so
for
tables
and
floors
of
them.
I
am
using
normalize,
but
if
we
don't
decide
me
to
decide
not
to
pull
it
in,
what
I
was
thinking
is
that
if
normalizes
a
blocker
right
now,
we
can
just
take
the
Norfolk,
especially
for
forms
like
CSS
forms,
the
one.
Why
stop
helps
a
lot
with
vertical
alignment,
and
things
like
that,
so
I
think.
G
Exactly
so
yeah,
so
if
I
didn't
want
like
this
is
going
to
be
the
gallery
for
now,
anyways
so
I
think
it's
okay,
if
it's
delaying
immortalized
it,
but
if
later
on,
if
you
decide
to
package
this
all
a
bunch
of
the
stuff
up,
along
with
the
Skinner,
then
worse,
like
force
case
of
normalizes
a
blocker.
At
that
stage,
we
can
just
take
the
normalized
from
stuff
performs
and
prick
inside
perform
CSS
and
Mason
to
deal
with
it.
That.
G
G
D
G
D
And
then
Anthony
what
I
would
suggest
for
data
table
once
that
that
happens,
if
you
are
providing
a
variety
of
different
options
within
a
theme,
so
we
have
the
night
theme.
For
example,
you
have
a
light
on
dark,
vs.
Sam
skin
is
dark
and
light.
So
if
you
have,
if
you
want
to
provide
options
to
have
striped
table
or
not
striped
table,
then
have
that
be
a
configuration
option.
Tournament
during
the
instantiation
of
the
table.
D
G
H
G
K
Totally
so
if
you
had
a
page
that
was
dark
green,
it
still
get
a
readable
table
without
any
extra,
more
okay.
G
I
think
I
job
from
my
competition
Matt
briefly
yesterday,
one
of
the
things
we
decided
on
we
kind
of
did
this
with
the
great
build
or
two
is
that
to
have
same
defaults
like
people
can
use.
But
if
you
want
to
override
those
defaults
and
customized
stuff,
then
they
should
be
able
to
do
that
and
I
think
that's
where
the
scanner
itself,
like
a
good
burger,
comes
again
so
so
yeah,
that's
an
overview
of
the
of
the
CSS
table.
Style
I
also
go
to
the
phone
stuff
just
a
little
bit.
G
B
B
E
G
G
I
really
call
it
say
in
line
help
box
by
specifying
Yui
3
help
in
line
I
should
actually
think
update
gist,
but
if
you
have
a
Yui
help
in
line
it
just
in
line
wats
it
along
with
the
phone
control,
so
I
just
basically
just
got
some.
This
is
just
very
simple,
like
adding
margins.
Yeah
I'll
get
into
require
a
tree
and
stuff
and
I'm
using
Yui
3
point
in
CSS
to
style
the
buttons
here
so
they'll.
Actually,
so
that's
that
you
can
make
multicolored
forms
using
Yui
periods
if
you
want
so
forms
this.
G
The
way
the
forms
CSS
is
built
is
there's
form
CSS
inner
swarm,
responsive
CSS,
in
this
case
we're
just
using
great.
So
there's
there's
nothing
in
here.
That's
really
from
own
responsibility
for
some.
So
for
some,
some
of
the
forms
that
you
have
an
inline
form
on
a
really
small
screen,
the
submit
button
will
like
go
to
a
new
line
just
because
it
was
my
circus
and
just
hang
on
the
edge.
So
there's
some
responsiveness
that
I've
built
into
this
form
CSS
but
yeah.
G
L
G
So
we
have
in
line
for
me,
I
think
it's
just
going
into
the
way
to
the
media
Greek
here,
because
this
should
actually
just
be
there's
mr.
actually
being
11
straight
line.
So
I
look
into
this,
but
then
my
form,
essentially
just
if
you
add
a
white
with
you
form
in
line
it'll
just
in
light
block
everything,
that's
inside
it
and
then
in
smaller
screens
it
will
unenlightened
block
a
few
things
so
that
it
doesn't
look
really
weird
on
consoles.
So.
B
For
for
the
placeholder,
or
were
you
did
you?
What
were
you
thinking
for
like
internet
explorer?
Yes
and.
G
Now
the
solar
stuff
is
just
for
special
here,
like
what
I
expect
the
most
common
form
to
be
really.
Is
this
next
one
on
horizontal
for
China,
you
can
see
a
thing:
it's
not
even
pull
down
yeah.
G
He
adds
a
width
to
the
label
here
and
then
in
line
box,
the
form
control.
If
you
have
a
really
long
label-
oh
I'm,
noticing
an
issue
do
didn't
post
before
just
the
vertical
line
in
here
is
not
exactly
as
I
would
want
it
to
be.
So
this
is
these
are
some
small
issues
that
I'm
trying
to
sort
through,
but
I
feel
this.
G
This
is
probably
the
style
using
used
all
over
the
place
now
with
the
right
line
label
with
the
with
the
phone
control
and
then,
if
you
put
in
Yui
3
could
I'm
sorry,
but
why
three
controls
of
div
or
quite
a
few
controls
fascinating
one
on
some
some
element,
the
child
elements
get
a
margin
left
of
the
same
amount
as
the
as
the
labels
here.
So
you
get
this
right,
align
behavior,
which
is
useful
for
partisan
to
chat
boxes
and
stuff.
Well,
also,
note
you
increase
your
font
to
that
made
yeah
no.
G
G
Good
tax
infidelities,
like
text
text
password
email-
oh
it
just
reduces
the
margins
of
all
of
them.
Well,
it
actually
positions
one
of
the
relative,
so
that
the
end
has
some
zed-index
fix
in
there,
so
that,
if
you
make
you,
if
you're
active
on
it,
you
still
get
the
upper
the
cursor
Alberto
sort
of
effect
on
all
four
corners,
but
essentially
it
takes
the
first
child
ren
border
radius,
border
radiuses.
G
The
top
takes
the
mother-child
border,
radiuses,
the
bottom
and
it
so
it
works
really
well
for,
like
username
password
sign
in
to
everything
and
I
use,
the
actually
looks
fine
on
older
browsers
to
you,
don't
get
the
border
radius,
but
it
looks
fine
and
I
used
box
sizing
quarter
box
of
these
input
elements
so
that
the
button
is
the
same
width
as
the
elements
are
here.
This
nice.
G
Speaking
of
wigs
and
stuff,
so
you
can,
you
can
style,
you
can
set
these
class
names
to
style.
The
different
widths
of
any
equipment
just
sets
the
width,
so
you
can
actually
zoom
out
buggies,
as
well
as
one
controls,
so
they're,
pretty
valid
inputs
I'm
using
html5
required
attributes.
If
you
have
that
in
you
get
this
invalidate
for
control,
which
is
like
red.
H
G
G
Mandatory
disabled
inputs
to
the
disabled
attribute
rounded
inputs
by
adding
the
lighter
three
but
round
you
class
on
it,
and
it's
there's
some
stuff
cross-browser
stuff
to
do
with
my
celexa
melting
selects
and
just
check
boxes,
radio,
button
alignment
with
their
language.
G
Like
lay
out
the
forms,
ok,
so
like
hollow
lives,
we've
seen
people
use
it.
It
just
makes
it
easier
for
your
lifestyle
them
not
so
much
to
do
with
like
color
eight
now,
but
you
can
change
that
lets
me
put
it
into
the
scanner.
So
if
you
don't
wake
a
dark
background,
you
want
the
forms
backgrounds
to
not
be
white,
so
it's
like
shows
up
change
that
it's
really
one
to
give
you
feedback
on
this.
What's
the
best
way
to
do
that,
so
it'll
be
only
get
so
it'll,
probably
on
the
gallery.
G
N
G
Was
using
less
but
then
I
decided,
I
kind
of
did
it
go
through
with
it
so
right
now
it's
just
it's
not
really
that
much
CSS,
so
I'm,
just
using
straight-up
CSS
right
now,
yeah
for
both
of
them.
G
So
yeah,
that's
this
way.
I
had
is.
B
That
also
that,
once
someone's
using
normalized
as
well
yeah.
G
B
I
mean
one
thing:
you
could
create
a
a
gallery
module
for
normalize
if
there
isn't
already
won
and
just
as
like
a
oh,
you
wouldn't
even
really
have
to
wrap
it
I
guess
it
yeah
just
modulus
and
then
well.
B
G
Someone
just
has
a
normal
sses
s
pulled
from
like
github
directly,
and
it
used
yourself
very
well
right
now,
just
a
manual
test
like
just
really
just
looking
to
make
sure
everything's
good.
The
offender
has
any
talks
about
how
I
can
like
do
any
sort
of
automated
test
it
for
the
stuff.
That'd
be
that'd,
be
good.
O
B
C
C
F
E
Okay,
so
for
week,
two
folks,
I'm
going
to
talk
about
whether
other
demo
we
gained
to
address
for
them
I'm,
not.
P
M
Q
O
L
L
E
E
E
S
To
facilitate
progress
by
hand,
so
let's
walk
through
this
bug
and
see
if
we
feeling
it's
worthy
and
then
see
if
I
did
horner,
the
bag
is
add,
option
to
Jason
P
for
a
sink
when
making
multiple
calls
too
much
weld
or
cluster
kuda.
Please
provide
a
way
to
make
multiple
requests
in
parallel.
So
then
this
is
going
down
to
get
script
which
lost
by
default
would
be
happy
to
have
an
option
and
recipe
not
to
default,
and
so.
S
Request
code
change
adds
up
of
Johnny,
but
tests
are
proving
problematic
and
then
so
is
Luke
here.
Yes,
if
you
say
I'm
quoting
you
from
the
bug,
but
there
are
problems
with
the
test
and
then
you
see
I'm
not
terrible,
invested
in
putting
much
effort
to
Jason
heat
in
it.
I'd
rather
work
on
an
I
or
ii
factory
that
support
of
Jason
P
as
yeah.
H
Adding
a
Singh
to
Jason
peas,
not
going
to
really
help
us
a
whole
lot
unless
you
su
a
batch
of
transactions
because
get
is
blocking
anyway,
get
doesn't
get
get
his
transaction
based.
So
if
you
fire
off
a
request
and
get
gets,
not
gonna
do
a
damn
thing
until
that
one
returns
so
making
it
a
sinks
not
going
to
matter.
Okay.
S
H
B
H
D
H
H
B
I
think
I
mean
so
the
reason
is
probably
because,
when
using
it
for
module
loading,
you
probably
want
this
like.
We
want
a
specific
transaction
q,
four
modules
being
loaded
or
but
if
we're
doing
JSONP
request
to
gather
data,
we
should
be
able
to
do
those
whenever
we
want
right,
like
that,
there's
no
dependency
restriction
there
right.
D
D
D
H
H
Loader
also,
doesn't
fire
off
another
module
transaction
while
one
of
them's
running,
so
it
doesn't
really
matter
whether
gets
doing
anything
else
or
not.
I
think
it's
limited
to
the
logic.
That's
instead
of
get
currently
that
it
needs
to
know
what's
going
on
at
any
given
time,
it
can't
do
it
more
than
once
so.
D
If
you
pass
in
configuration
dacing
true,
it
just
respects
that
configuration
as
it
passes
it
down
to
what
I
get
that
script.
That's
all
I
mean
it's
like.
Oh
I,
think
it's
like
a
one-line
change,
so
it's
very
simple
to
do,
but
that
as
far
as
certainty,
when
making
any
modifications
to
the
core,
you
want
to
make
sure
that
the
test
coverage
is
there
and
the
tests,
because
Jason
P
is
one
level
divorce
from
get
script
and
get
script
is
using
script
tags.
D
S
D
D
S
H
B
The
the
end
goal
would
be
able
to
support
fetching
data
asynchronously
and
like
multiple
things
at
a
time
like
right
now
through.
T
B
The
serial
request,
problem
and
III
think
yeah,
I'm,
sure,
Dave
and
I
were
talking
about
it,
and
that
was
a
thing
we
trace
it
back
and
no,
it's
not
Jason
P.
It's
actually
all
the
way
down
and
get
doing
it's
one
transaction.
At
a
time
thing
Ryan
in
IRC
says
he's
looking
up
looking
at
the
code
now
to
see
if
it
was
intentional
or
not.
L
A
P
P
P
J
A
V
J
J
J
D
J
Now
these
terms
are
the
way
the
modules
are
broken
up.
Probably
better
shown
here.
I've
got
my
own
personal
api
docs
sitting
here,
there's
a
notion
of
axes
base,
which
is
essentially
the
basically
the
data
logic
of
the
axis.
You
know
whether
it's
a
numeric
category
time
and
it
handles
you,
know
the
Parsons,
four
min
and
Max,
and
you
know
different
rounding
values.
J
If
you
go
to
the
effectivity
of
pull
request,
really
the
only
actual
code
change
I've
made
at
this
stage
in
the
refactoring
is
with
the
axis
layer
and
there's
an
image
that
says
current
axis
and
that's
the
current
you
kind
of
a
UML
diagram
of
sorts
of
the
current
of
axis
the
way
and
set
up,
and
basically
you
have
an
access
class
that
extends
with
it
and
it
uses
a
other
top
axis
left
axis
right,
axis
bottom
axis
or
no
access
lay
depending
on
the
direction
of
the
axis
tune.
J
It's
extended
by
a
class
access
type,
which
adds
some
base
data
functionality,
which
is
then
in
turn
an
extended
by
category
axis
time.
Axis
numeric
axis
and
stacked
access
below
is
is
the
newly
the
new
architecture
for
the
access
layer.
There's
no
new
code
per
se.
It's
just
been
broken
out
quite
differently
and
the
first
set
you
have
them
in.
J
There
go
okay,
you've
got
an
axis
base
class,
which
is
extended
by
the
different
types
time.
Category
numeric
and
stacked
extends
in
America
base
of
these
classes
use
either
a
time
category
numeric
or
stacked
implementation
class,
which
contains
actually
the
methods
and
direct
repeats
that
are
germane
to
the
type
of
axis.
On
the
other
side,
we
have
an
access
class
which
extends
widget
again.
It
still
uses
the
top
left
right
and
bottom
axis
strategies,
and
this
class
is
extended
by
time.
J
Category
American
stack
and
they
also
would
use
the
via
base
new
utility
classes
that
contain
two
methods
and
with
this
change
none
of
the
none
of
the
API
for
the
final
charge.
Application
changes
at
all-
it's
just
restructured
so
that
the
classes
can
then
be
broken
broken
or
the
models
can
be
broken
up
into
smaller
footprints.
For
just
your
specific
needs.
You
know
like
the
spark
lines,
example
or
currently
for
a
pie
chart
which
is
very
different
from
everything
else.
You
need
to
pull
everything
in,
but
now
you
can
just
pull
in.
J
B
So
is
the
is
the
base
class
like
usable
on
its
own
where's,
our
use
cases
where
you
would
use
that
instead
of
one
of
the
time.
E
I
J
J
V
J
G
J
Someone
else
can
sugar,
quite
honestly,
I
more
focus
on
the
lower
level
of
oxygen
charge
to
make
it
more
usable.
For
example,
if
you
want
to
build
your
own
charts
application,
there's
there's
really
no
sense
to
go
through
the
top
layer
of
charts.
If
you
only
need
certain
things
or
you
have
certain
business
needs
that
are
various
cific,
the
current
chart
class
is
one-size-fits-all.
J
I
mean
it's
very
customizable,
but
you
know
you
can
probably
get
closer
to
the
metal,
so
I'm
trying
to
get
pieces
of
charts
that
can
be
used
missing
blood
you're
like
a
lot
of
people
when
they
build
a
chart,
they're
very
much
static.
You
know
and
there's
a
lot
of
layout
logic
that
they
don't
need.
You
know,
so
it's
things
like
that.
If
anyone
so
I'm
just
trying
to
make
it
where
you
can,
it
can
choose
what
you
need.
J
J
Oh
I,
don't
fert
I,
don't
see
any
I,
don't
see
it
impacting
anyone
from
an
art
beat
if,
even
if
someone
was
like
extending
a
lot
of
classes
inviting
their
own
because
they
have
to
be
doing
some
really
weird
extending
to
use
24,
10
pack
them
I
personally,
don't
think
I
need
to
defecate
they
old
all
this
one
I'm
invade
into
its
there's.
J
No
api
change
and
architectural
changes
shouldn't
impact
anyone,
but
if
some
disagrees,
I
can
imagine,
okay
and
part
of
the
reason
I
don't
want
to
deprecate
is
because
there's
a
lot
of
things
that
I
want
to
refactor
with
charge
tonight.
I
don't
want
to
just
have
a
deprecation
every
two
or
three
at
least
race.
J
J
S
T
J
J
S
S
J
I
haven't
looked
at
that
yet
I
mean
that
it
probably
does
need
to
be
more
right
now
on
it,
I
cleaned
up.
They,
you
got
quite
a
bit
so
that
it
at
least
makes
sense
if
you're
looking
at
it
but
I
there
should
be
more
information
in
there.
I
hope
you
get
that
in
and
of
course,
there
needs
to
be
updates
to
the
user
guide
as
well,
because.
S
J
V
J
Very
good
copy
that
would
be
very
useful
if
anyone
wants
to
take
a
gander
at
my
bill,
JSON
your
pedophiles
to
see
how
they're
rolled
up.
If
there's,
if
there's
my
guess,
right
now,
obviously
the
first
step,
that's
getting
everything
granular
and
everything,
two
separate
pieces
that
can
be
as
simple
as
you
want
at
that
point.
What
stuff
should
be
rolled
up
together
or
what
rick
roll
up.
S
J
I
E
I
J
I
Yeah
I'd
rather
get
this
solid.
The
pieces
rethinking
that
buck
up
here
too,
or
is
it
more
like
next
version?
Not
sure
yet
I
haven't
done
depends.
S
R
I
T
E
E
D
All
right,
so
the
just
a
quick
update
from
Ryan
and
IRC
is
that
the
way
I
get
jas
and
CSS
returned
a
transaction
object
which
has
an
execute
method
and
that
execute
method
then
jumps
the
queue
and
it
happens
immediately.
So
it
doesn't
cause
the
the
synchronous
processing
requirement,
the
serial
processing,
that
is
the
default
behavior.
So
the
current
bug,
that's
are
the
current
pull
request.
D
That
Jenny
closed
is
now,
if
I
add
in
a
call
to
execute
on
top
of
the
call
to
white
I,
get
that
script,
and
it
should
execute
immediately
and
not
be
not
be
postponed
until
the
serial
response
from
previous
calls.
So
that
means
that
it
is
possible
to
implement
this.
D
D
B
We
have
them
for
testing
stuff.
We
do
have
the
the
new
echoecho
project,
which
helps
with
doing
this
type
of
testing
for
actually
making
requests
to
a
server
that
will
respond.
So
if
you
want
I
can
look
at
that,
so
we
don't
have
to
be
faking
out
these
tests
so
much
I'm
familiar
working
with
it
now
with
IO
and
even
contributed
some
to
it.
One
person
that
I
stuff
so
that.
D
H
D
D
D
B
Well,
yeah
I
can
help
her
and
that
okay.
S
Again,
I
am
not
an
expert
on
this
topic,
but
I
will
facilitate
the
discussion
so
to
shoot
at
recursive
equals
true,
doesn't
work
for
long
templates.
The
following
recursive
substitution
call
will
fail
to
substitute
the
token
be.
The
problem
is
that
why
not
substitute
save
template
link
on
enter
and
does
not
anticipate
the
fact
that
it
may
become
longer
after
previous
position,
renowned
the
solution
would
be
to
recompute.
S
Conversation
here,
20
hours
ago,
Ryan
asks
didn't.
We
agree
that
whiteout
substitute
should
be
deprecated
back
when
we
discuss
the
plan
for
why
that
template
might
as
well
fix
this
bug.
If
the
solution
is
as
simple
as
tomash
says,
but
if
we
are
really
going
to
deprecate
substitute,
we
should
get
on
that
approach.
People
to
stop
using
it
forget
what
we
said
about.
Why
out
substitute
I
feel
like
that
should
be
documented
somewhere
did
we
have.
S
S
Okay,
does
anyone
feel
passionately
that,
yes,
we
should
go
ahead
and
formally
deprecated
it
okay.
So
I
think
what
we
should
do
is
put
this
to
the
mailing
list
for
a
vote
to
deprecate
and
we
can
handle
objections
through
that
venue
and,
if
all
goes
well,
we'll
take
steps
to
formally
announce
the
deprecation
in
the
next
release.
Get
it
in
the
documentation,
get
our
API
Doc's,
documented,
correctly
scrub
our
library
code
of
whiteout
substitute
calls,
which
you
know
well
I'm
talking
I'll,
do
a
quick.
D
S
E
D
Wait,
did
you
say,
set
TN
or
satyam,
I
was.
K
F
B
D
H
E
N
It's
not
that
much
different.
The
only
main
changes
were
I
had
some
people
file
bugs
against
the
gallery,
implementation
of
it,
and
it's
only
like
some
minor
updates
after
that,
so
other
than
that,
all
the
tests,
all
the
documentation
and
everything
is
there.
So
just
let
me
know
there's
like
any
problems
with
it.
E
Look
at
the
code
dirt
is
it
something
that
you
feel
like
is
just
an
update
if
something
it
sort
of
been
discussed,
or
what
would
you
like
this
to
the
dress?
I
mean.
N
B
B
N
R
S
N
Okay,
so
yeah
all
this
is
basically
it
adds
a
synchronization
layer
using
local
storage
instead
of
anything
else,
instead
of
like
rest
or
example,
which
eric
already
has
in
core
and
all
does
it
just.
It
uses
like
an
in-memory
cache
if
local
storage
isn't
available,
such
as
for
like
the
older
Internet,
Explorer,
browsers
and
other
than
that,
it's
pretty
straightforward.
It's
very
similar
to
the
background
version
of
it
and
the
only
main
customization
that
people
can
use
whip-it
is
changing
how
the
IES
are
generated
for
each
different
model.
N
B
So
the
the
one
there's
one
there's
one
potential
issue:
around
an
exception,
throwing
throwing
when
local
storage
is
full
and
you
try
to
add
something
to
it.
That.
N
I,
haven't
really
tried
yeah
I,
but
if
I
don't
know,
I
haven't
really
tried
a
when
local
storage
is
full.
So
that
might
be
something
I
want
to
take
a
look
at.
B
Yeah
I
think
that
that
was
one
thing.
I
remember
other
possibility
that
I
could
come
up
if
somebody's.
Actually,
you
know
pumping
a
lot
of
data
into
it
or
the
local
storage
for
that
domain,
as
already
has
other
stuff
in
it.
When.
S
M
C
N
I'm
not
sure
how
I
would
do
that,
except
for
like
Bill,
maybe
like
simulating
a
full
local
storage
item.
But
is
there
an
easy
way
to
stimulate
that
I.
B
B
N
S
B
W
N
B
B
S
N
N
N
B
So
I've
got
to
forget
if
John
Lendl
might
have
opened
a
poll
request
today
for
this
as
well.
The.
E
B
Yeah
I
yeah
he's
he's
assigned
that
over
to
me.
I'm
not
sure
why
he
assigned
it
back,
but
I
mean
that
this
pull
request.
The
383
has
has
the
same
thing.
F
S
B
Yeah,
I
remember
a
Tony
and
John
I
believe
how
to
talk
about
this,
because
I
basically
have
no
idea
what
the
context
is
around
this.
I
just
don't
really
get
look.
The
problem
is
so
yeah
I
can
make
sure
that
the
test
pass
and
the
code
path
is
going
through,
but
I
don't
really
know.
If
that's,
if
it's
doing
something,
we
want
to
do
or
don't.
S
Well,
if
Tony's
got
more
context
I,
I
would
appreciate
him
I'm
walking
us
through
and
if
he's
handed
off
and
had
a
discussion
with
John
to
just
fill
us
in
and
if
this
is
an
evolution
of
Tony's
original,
then
we
can
move
to
johns
and
close
out
Tony's.
So,
let's
sync
up
with
Johnny
and
have
them
in
on
the
next
discussion
to
walk
us
through
their
change.
S
B
Think,
actually,
if,
if
John
defines
some
tests,
that
would
be
breaking
normally
and
then
show
them
now
fixed,
it
will
help
create
that
context
as
well
for
like
what
the
actual
problem
was
that
you
right
into
and
how
this
fix
is
it
by
just
having
a
test
to
show
it
right.
B
S
E
B
B
Fit
yeah
one
of
the
things
that
came
up
when
implementing
it.
If
we
trying
to
find
it
here,
but
there
to
implement
it,
I
had
to
do
some
gross
stuff
with
aliasing
over
attribute
methods
so
that
I
didn't
have
to
override
them
and
that
smelled
really
bad
and
I
talked
with
sathyan
about
it.
And
we
wanted
to
figure
out
a
better
way
to
not
have
to
do
this.
B
And
so
that's
why
this
thing
ended
up
getting
stale
and
sitting
there,
because
before
doing
this,
it
would
be
nice
to
make
some
changes
in
to
attribute
that
we've
been
talking
about
for
a
long
time
when
model
was
originally
developed.
So
that
model
doesn't
have
to
duck
punch
attribute
where
it
does
and
instead
attributes
flexible
enough
to
handle
what
it
needs
to
do.
But.
B
B
I
kind
of
I
mean
merging
this
in
it
has
like
a
big
section
in
one
of
the
the
files
that
just
has
like
attribute
hacks,
and
it's
just
in
model
observable
I
believe
it
is
yeah.
There's
there's
a
section
that
says
hacks
in
it
says
these
reference
wha
dat,
model-based
prototype.
B
These
references
are
to
preserve
the
attribute
method.
Overrides
model
observable
uses
attribute
observable.
So
when,
when
used
as
a
class
extension
attribute,
observables
implementation
of
these
methods
will
override
basis.
So
the
it's
just
some
ordering
problem
here
and
the
the
root
issue
is
that
model
should
not
be
overriding.
The
public
methods,
a
dadar
set
and
set
adders
of
attribute
and
instead
model
and
attribute,
should
change
such
that
the
features
that
we
want
can
be
implemented
without
having
to
do
these
types
of
overrides,
and
so
that
was
one
of
the
hang-ups
that
I
ran
into.
B
S
B
E
Okay,
if
you
want
to
keep
going,
you
can
go
to
or
a
quest
to
81
this
one
almost
last
updated
two
months
ago,
and
this
is
Luke's
comments
and
ship.
It
so
I
think
this
is
that
question
is
taking
a
look
at
the
code
again
I,
guess
or
finding
out
what
Whitney's
then
move
from
here.
It's
assignment
Tony
and
it
was
a
fix
for
25
3,
2,
1,
3,
4
and
in
table
sort
yeah.
W
S
D
It's
not
ready
to
pull
in,
because
I
need
to
make
some
modifications
to
the
code,
so
it
can
stay
in
pending.
Oh
I
can
either
reissue
a
new
poll
request
with
the
source
code
updated
if
it's,
if
it's
two
out
of
date,
that
cause
trouble
or
I
can
just
continue
on
that
same
flow
request.
It
still
has
a
green
bar
at
the
bottom
I.
Don't
think
you
should
create.
E
E
B
Thought
there
was
some
discussion
yesterday
or
the
day
before,
between
Matt
and
Ryan,
about
this
somewhere
else
and
forget
where,
but
earlier
than
a
month
ago,
which
is
the
last
comment
on
here,
right
I
think
Matt
Parker's
was
showing
maybe
was
an
IRC.
He
showed
ryan
his
branch
or
a
spark
or
something
that.
D
S
E
B
Yeah
this
is
this
is
yet
another
one
for
me,
I
think
we're
talking
to
you
about
this
Andrew
that
there
was
I
wanted
to
get
right,
how
the
interaction
should
be,
and
it's
kind
of
funny
that
we're
talking
about
this,
because
we
were
talking
about
this
stuff
earlier
in
IRC
today
about
data
binding
to
forms
and
being
able
to
handle
to
eight
inputs,
and
essentially
this
ran
into
a
problem
of
like
bi-directional
inputs
in
in
with
a
model
and
view,
and
the
DOM
and
I
didn't
like
the
way
that
that
we
had
to
go
about
it
in
this
pull
request.
B
N
B
B
B
But
there
are
some
parts
about
it
that
I
didn't
didn't
and
it's
not
necessarily
Alberto's
fault,
but
it
I
I
didn't
feel
happy
with,
though
we'd
be
showing
people
how
they
should
know
this
problem,
because
I
think
it's
one
that
actually
come
up
a
lot
and
that's
where
I
wanted
to
try
to
noodle
on
some
better
ways
to
do
it
before
we
go
and
document.
This
is
the
way
that
you
know
people
should
be
handling
this
okay.
Well,.
K
B
E
Q
E
E
S
Yeah
yeah,
okay,
so
362
replaces
34
th
the
westin.
The
closest
one
yeah
just
make
sure
that
there's
nothing
in
the
first
one
that
needs
to
be
over.
S
S
E
E
E
D
B
B
Or
do
we
care
that's
right,
but
yeah?
So
one
pointed
us
to
a
or
somewhere
how.
B
So,
on
iOS
tabs
that
are
hidden
do
not
run
timers
and,
in
other
browsers,
I
believe
the
tabs,
which
are
not
in
focus,
have
a
timer
resolution
of
something
less
than
the
ones
that
are
so
there's
that
similar
behavior
but
I
I
don't
know
if
we
should
try
to
break
that
feature
of
the
device,
because
it
essentially
means
that
a
running
a
long-running
thing
cannot
be
long
running
in
mobile
safari.
When
it's
not
the
active
tab
and
I.
Don't
know
me
as
a
user
on
my
phone.
B
S
For
the
record
Eric,
your
last
comment
here
indicates
that
you
feel
ready.
I'm
her
assistant
to
the
ranch
based
on
the
mobile
is
a
hard
thing.
Does
that
change
your
mind?
Yeah.
D
B
So
the
one
issue,
then,
is
that
I
still
see
remaining
is
what
we
do
about
organization,
and
does
this
need
to
be
its
own
module?
Is
there
something
with
wide
out
later
that
we
can
change
to
rename
to
Yui
dash
timer
or
something
to
that
effect
and
then
include
both
wide
out
later
and
wide
out
soon
in
that
module
in
that
one
module?
B
B
H
B
D
E
H
B
B
D
B
Have
the
one
the
one
exercise
I
never
ended
up
doing
was
to
look
at
core
and
see
where
we
were
doing
set
timeout
0
and
look
at
each
of
those
instances
and
determine
if
it
would
benefit
from
using
said
immediate
or
process
X
tick
or
essentially
having
this
abstraction.
Over
top
of
that,
that
wideouts
you
an
abstraction
but
can.
S
Is
it
too
much
to
ask
who's
this
Steven
Jo.
F
B
But
we
I
think
we
need
to
have
a
plan
of
what
that
is
right.
So,
if
we're
deciding
on
creating
this
timers
module,
deprecating
it
as
being
part
of
core
and
having
it
exposed
wide
out
later
and
why
out
soon,
yes,
but
I
think
that
we
have
to
agree
on
that
before
we
just
say
reorganize.
This,
like
I,
want
to
have
some
plan,
so
you
sent
trying
several
different
ways
of
reorganizing
it.
No
right.
R
H
We
have
to
figure
out
one
or
two
things.
First,
we
have
to
figure
out
whether
we
want
it
in
the
core
or
not,
and
if
we
want
it
in
the
core,
we
have
to
figure
out.
If
anything
can
use
it
inside
of
the
court,
which
would
be
get
and
those
two
set
timeout
0
calls
it
makes.
If
it's
not
going
to
be
any
benefit
there,
then
we
figure
out,
if
we're
going
to
put
it
as
a
module
and
then
we're
going
to
put
it
as
a
module.
H
C
B
I
think
that
the
timeout
yeah
I
mean
the
the
soon
is
implying,
I
think
something
conceptually
different
like
it's
a
hack
when
you
have
set
timeout
0
right.
Like
imagine
the
first
time
that
you
you
run
into
that
when
programming,
it's
you
and
javascript
you're,
like
what
the
hell
is,
this
thing
and
so
I
feel
that
it's
probably
a
good
thing
to
not.
B
To
encourage
people
to
start
using
this
and
that's
why
the
whole
set
immediate
API
is
being
provided
in
the
browsers
now
and
and
and
oj
s
so
delicious.
But
yeah
I
see
your
point
about
having
two
ways
of
doing
it,
but
I
think
we're
kind
of
helping
them
when
they
use
the
old
way
through
wideout,
later
zero
right.
E
S
Mean
the
process
indicates
that
we
have
discussion
captured
in
this
thread.
I
think
that
I've
heard
a
proposal
to
pull
out
wide
out
later
and
different
a
stand-alone
module.
That
includes
why
I
later
went
on
soon.
D
D
If
it
actually
proves
to
be
necessary
in
the
future,
then
it's
then
it
should
be
relatively
simple
to
include
that
going
forward,
but
for
now
separated
out
as
a
module
leave
it
out
of
core
and
then
everything
should
be
as
it
is
now
or
a
little
bit
better.
It's
not
going
to
get
worse
for
not
including
it
in
court.
B
So
we'll
we'll
have
to-
and
that's
probably
that's
probably
going
to
impact
like
a
large
percentage
of
programs
by
pulling
that
out
of
core.
So
we
should
definitely
do
like
a
the
deprecation
policy
we
have
for
real.
You
know
on
that
one
and
alias
the
module
and
actually
include
it
until
like
a
few
months
from
now
when
we
can
remove
it.
S
It
does
steer
Steven
towards
having
a
serological
provider
soon
well,.
H
H
D
S
H
B
B
D
H
T
S
R
E
E
B
W
H
S
Are
about
the
program's,
these
particular
ones,
land
it
on
someone's
plate
recently,
and
they
said
no
I'm,
not
the
right
person
for
this.
That's
why
they
have
to
be
on
this
list,
but
you
can
kind
of
get
a
sense
of
which
modules
are
kind
of
out
in
orbit
right
now,
slider!
It's
now
sheep
are
the
two
that
are
represented
here,
but
there
are
other
ones
that
you'll
start
to
see
crop
up
here.
So
you
know
my
hunch
is
someone's
going
to
care
about
cider
in
the
community
one
day
and
they
they're.
N
C
Yeah
most
of
the
signal
light
stuff
is
there
off
the
top
of
my
head.
I,
don't
know
specifically
what
we're
missing
then
I.
Don't
now
squeals
one
tap
I,
don't
have
a
bit
simulating
despite
to
a
bed
single
okay.
G
S
G
G
I,
usually
listen,
read
and
go
I
didn't
use
Jefferson
and
the
way
the
way
that
app
works
is
that
it
does
different
things
based
on
pretty
nice
in
here.
So
it
does
not
some
house
down
locks
on
desktop
and
so
I
just
I
just
want
I
just
do
assimilate
of
those
of
a
bad
event.
It's
just
not
rolled
in
a
dress
reasonable.
You
just
let
me
send
song
it
I
so
like
why
I
event
does
simulate
like
for
each
separate
emotions,
I.