►
Description
0:00:00 - Discuss CRM features
0:30:45 - Working on the confirm_modal refactoring
A
A
I
guess
ours
read
right,
so
we
can
now
sort
of
see
the
the
contacts
and
the
organizations
that
are
attached
to
a
group,
but
we
can't
create
them
or
update
them.
So
I'm
looking
for
some
direction.
I'm
not
sure
that
I've
got
time
or
I'm
that
fussed
about
us
starting
to
hack
code.
We
can
maybe,
but
more
so
of
that
bouncing
of
ideas
like
okay,
where
have
you
guys
been
recently
in
the
products
where
you've
either
created
a
new
one
or
worked
with
a
recent
cruddy
type
operation
for
creating
records?
A
A
Started
talking
to
kristen
yesterday
and
she
was
telling
me
some
really
cool
stuff.
I
don't
know
if
any
of
it's
top
secrets,
I
I
don't
think
it
is
nothing
at
get
lab
is
but
yeah
I
mean
realistically.
I
think
we
should
assume
that
it's
going
to
be
one
at
a
time.
Obviously,
you've
got
your
data
import,
your
data
migrate,
whatever
you
want
to
call
it
if
you're
moving
from
another
platform
or
setting
up
for
first
time,
we
can
think
about
something
cool
for
that
at
some
point.
A
B
D
B
A
So
what
does
this
look
like?
So
we've
got
a
nice
table,
we've
got
a
new
user
button
top
right.
That
seems
like
a
simple,
simple
way
forward
and
then
we've
got
edit
against
each
one.
So
that
sounds
sounds
like
a
good
approach.
If
you
hit
new
user,
what
is
it
takes?
You
do
you
guys
know
off
the
top
of
your
head.
If
this
is
in
view-
or
I
doubt
it's.
B
A
B
A
So,
okay,
let's
just
rewind
a
bit,
so
I
guess
in
my
head-
I
I
didn't
think
of
it
being
a
separate
page
yeah.
So
so
forgive
me
because
I
always
get
a
bit
confused
with
the
whole
like
single
page
application.
You
know,
I
know
we're
not
there
we're
not
even
really
heading
for
it,
but
I
I'm
intrigued
to
kind
of
know
like
in
my
eyes.
They
don't
really
see
the
necessity
to
kind
of
navigate
away
from
here.
Yeah,
you
know
ad
could
be
more
of
a
modal
type
thing.
A
B
I
want
to
build
the
field
for
this
method
on
the
model
and
if
you've
you're,
basically
guaranteed
that
your
form
name,
all
the
names
of
the
fields
are
going
to
match
with
the
names
on
the
the
rails
model
that
that
you're
looking
at
and
if
you
aren't
building
the
form
in
hamel,
then
building
the
form.
Just
in
view
you
can.
We
kind
of
have
to
you
know,
hope
that
the
names
of
our
fields
are
going
to
match.
The
api
requests.
We're
going
to
send.
B
A
Think,
with
the
sort
of
simplicity
of
the
record
at
the
moment
that
it
seems
like
a
a
good
good
option,
my
only
thinking
that
that
allowed
some
tiny,
tiny,
tiny
complexities
I
went
after
we
add
the
record.
We
should
probably
reload
the
page
or
re-run
the
apollo
query,
which
I
think
could
have
some
sort
of
like
negative
side
effects.
A
I
let's
say
I've
changed
the
sorting
order
or
I've
done
something
I
don't
know
how
well
like
you
know,
there's
still
a
lot
to
do
on
this,
like
pagination
and
other
bits
and
pieces,
so
it
might
be
simpler
to
simply
build
it
as
another
page
and
still
do
it
in
view.
I
guess,
then
I've
got
to
like
figure
out
the
back
end
stuff,
because
I
have
to
have
like
a
new
controller
action
and
a
new
route
and
all
that
kind
of
stuff.
So
now
I'm
I'm
playing
devil's
advocate
here.
D
B
A
Yeah,
so
I
think
that,
like
initially,
I
was
thinking
about
like
refresh
and
worrying
about
stuff
like
that,
but
now
I'm
thinking
about
like
the
the
rails
and
the
controller
and
the
routes
and
that's
complexity
that
I
could
kind
of
do
without
yeah.
I
wonder
yeah,
I
wonder
if
anyone
would
be
upset
if
we,
if
we
stuck,
you
know
if
it
had
to
be
if
it
was
in
this
page
at
the
moment,
and
you
couldn't
navigate
to
something
like
new
or
ad
or.
B
I
think
I
think
there's
when
to
keeping
it
on
one
page,
like
I
think,
that's
thinking
of
the
user,
the
user
may
be
sorting
through
things
or
edited
some
stuff,
and
I
realized.
Oh,
I
missed
this
person.
Let
me
add
them
real,
quick,
like
I
think
that
there
is
when
to
keeping
them
on
this
page.
So
let's
look
awesome.
Let's
wait!
Let's
we
can
just
keep
it
simple
and
just
have
a
nice
little
modal
that
pops
up
when
we
want
to
create
a
new
one
of
these.
B
A
B
A
A
A
B
A
B
A
Mean
my
immediate:
maybe
we
can
go
a
few
other
places
to
see.
So
you
know,
we've
we've
got
users
and
we
know
that
it
was
in
the
top
right,
but
you
know
so:
okay
select
project
yeah,
so
it
looks
like
because
I'm
in
a
group
not
a
project,
but
if
I
go
into
a
project
into
the
issues
list,
I
think
there's
going
to
be
an
ad
button
in
the
top
right.
So
I
think
that
we've
got
a
pattern
there
that
we
can
observe
that
the
the.
E
B
B
A
A
B
A
B
C
C
B
A
A
B
Well,
I
think
you
have
it
where
you,
you
have
live,
reload
enabled
so
it
actually,
I
don't
know
uh-huh
fresh
is
for
you,
I
I
don't
so
I
probably
that's
probably.
Should
I
create
one
so
that
we
can
see
what
it
looks
like?
That
must
be
what
we're
missing
we're
missing
some
sort
of
like
title
thing:
yeah,
let's
create
one.
So
we
can
see
what
it
looks
like.
B
B
A
B
E
B
A
B
A
I
don't
know
all
right.
That's
that's.
Awesome
I'll,
probably
have
a
look,
how
they've,
how
they've
done
this
as
well,
because
I
do
I
do
like
the
idea
of
having
it
in
mode
or
I'm
guessing.
We
can
still
kind
of
keep
the
same
like
the
the
view
component
can
still
look
identical,
but
instead
of
having
to
have
a
because
this
will
be
a
hammer
page
right.
That
mounts
the
view
out.
A
I
just
throw
a
third
option
out
there
before
we
get
get
the
view.
Is
there
a
nice
way
that
we
can
have
the
new
button
not
take
you
to
a
new
like
url,
but
just
kind
of
render
the
component,
so
we
don't
have
to
so
I
don't
have
to
create
another
route
and
another
controller
method
and
another
hammer
page
and
and
and
how
was
I,
how
is
that
different
than
the
model?
A
A
A
D
I'm
I'm
not
quite
quite
sure
about
this
option
because,
like
single
page
applications
are
great
and
we
don't
have
enough
room
there
for
them
at
gitlab.
But
for
me
a
good
single
page
application
would
also
update
the
url
and
still
be
linkable
and,
if
like,
if
we
could
do
that,
but
I
don't
know
like,
then
we
would
have
to
update
the
url
via
view
or
maybe
have
to
deal
with
routes
anyhow.
So
if
we're
able
to
update
the
ul,
then
I'm
totally
hooked.
But
if
not,
then
I
have
some
concerns.
I.
A
B
B
B
Yeah,
the
the
issue
design
the
designs
for
the
if
you've
ever
used,
that
that
also
uses
view
router,
where
it'll
update
the
routes
based
on
whether
you're
viewing
a
design
or
creating
one
or
whatever,
and
you
can
revisit
that
route
like
and
it's
it's
pretty
good.
I
have.
I
have
another
option
for
you:
oh
yeah,
so
go.
Let's
go
back
to
feature
flag
table
and
I'd
love
to
hear
what
yanik
thinks
of
this,
and
I
now
start
to
think
that
all
of
my
ideas
are
you
know.
B
So
what
if
clicking
new
feature
flag
did
not
open?
A
modal
did
not
take
you
to
a
new
page
but
revealed
this
form
in
like
an
accordion
sort
of
fashion
right
below
the
button.
So
you
click
new
feature
flag
and
then
it's
like
boom.
Here's
the
id
status,
whatever
you
need
to
do
to
fill
out
a
new
feature,
flag
right
right
next
to
the
kind
of
right
next
to
the
button,
and
we
do
this
kind
of
in
we
do
it.
B
B
B
A
A
Yeah
I
mean
to
be
honest,
you
even
gave
me
a
another
sort
of
idea
of
close
to
the
feature
effects
and
now,
let's
open
that
back
up
as
well,
even
give
me
another
like
yeah.
We've
now
got
about
six
options
that
the
the
other
thing
it
could
do.
I
think
I
I
like
yours,
it
just
pushes
the
table
down,
but
we
could
even
just
flip.
A
We
could
just
sort
of
hide
it
and
there
could
be
like
one
gif,
that's
hidden
and
one
that's
shown
and
just
flipped
the
visibility
almost
to
kind
of,
because
you
know
the
the
extra
scrolling
on
a
mobile
et
cetera
it
might
just
yeah.
It
might
be
easier
if,
when
you
scroll
to
the
all
you've
got
on
the
page
is
the
form.
B
Yep,
yep
and-
and
that's
definitely,
I
think
that
would
be
a
good
question
for
you
x.
If,
if
we
wanted
to
go
down
this
path
of
in,
I
would
call
this
in
context,
creation
or
in
context
editing
like
we're,
still
keeping
our
contacts,
and
we
want
to
do
it.
We
don't
want
to
leave
the
context.
If
that's
the
path
you
want
to
go
down.
B
B
A
Think
I
think
it
maybe
does
maybe
doesn't
fit
in
with
some
interesting
research
you.
I
guess
it's
kind
of
like
uxc
research
that
pedro's
been
doing
with
regards
to
merge,
requests
and
perceived
performance
versus
actual
performance,
and
you
know
because
you're
not
like
navigating
away
from
the
page.
Probably
in
this
instance,
it
will
actually
improve
actual
performance,
because
you
won't
have
to
wait
for
a
new
route
for
the
controller
for
all
the
the
kind
of
client
server
stuff.
D
D
A
B
A
Yeah,
I
I
think
it
will
be
that
there
will
be
something
there.
That's
I'm
going
to
say
usable,
but
sorry
it's
a
funny
thing.
Well,
I
like
forget:
I'm
like
okay
right.
We
can
now
see
the
contacts
we
can
create
them
and
update
them
all
through
the
graphql.
We
can
attach
them
to
issues.
We've
got
the
quick
actions
I
think
just
being
merged
in
so
you
can
now.
A
B
B
Bye,
oh
man,
yannick,
that's
quite
something
yeah
yeah!
It's
you
know
it's.
I
know
lee
is
the
last
handful
of
times
we've
met
we've.
I
haven't
really
paired
as
much
as
done
this
kind
of
ux
discussions
and
okay
it'd
be
nice
to
have
ux
and
ball
like
involved
in
the
discussion,
and
I
think
they
get
involved.
I
think
we
just
bounce
ideas
of
the
ideas
that
we
would
like
to
present
to
ux,
but
maybe
it
would
be
cool
if
we
actually
had
like
some
ux
pairing
like
that
would
be
an
interesting.
D
B
Yeah,
I
think,
and
even
then
I
don't
know
if
it's
this-
I
don't
know
if
he's
working
with
just
one
designer
or
because
he's
kind
of
a
community
contributor
working
on
this
one
you're
just
kind
of
pinged
whatever
is,
I
don't
know,
and
I
think
definitely
some
of
the
ways
that
we
are
organized
contribute
to,
and
we
were
just
talking.
We
identified
six
different
ways
that
we
create
a
new
thing
in
gitlab.
B
There's
ux
debt
there,
but
there's
you
know,
there's
also,
I
don't
know
I'm
the
kind
of
person
I
like
that.
Inline
like
I
like
I.
I
feel
it
feels
really
nice
when
I
click
a
button
and
then
I'm
not
taking
anywhere
and
right
where
my
mouse
is
is
like.
Oh,
I
could
just.
I
don't
have
to
travel
too
far
to
do
the
thing
I
wanted
to
do.
That's
always
feels.
D
Yeah
yeah
do
agree
and
the
more
the
more
I
use
them,
the
more
I'm,
not
a
fan
of
modal
slicing
place
because
they're,
just
that
is
hard
to
reproduce
in
in
some
cases,
and
you
kind
of
you
only
can
have
one
of
them
and
if
things
get
complex,
what
do
you
want
to
do
if
there's
an
issue
in
your
model
open
up
another
one?
You
rather
don't.
B
D
B
That's
funny
I
remember
like
there
was
always
it
felt
like
at
one
point:
websites
tried
to
model
their
ux
after
the
operating
system
like
we're
going
to
try
to
do
what
the
what
windows
is
doing,
and
so
at
one
point
it
was
we're
going
to
open
a
modal
and
you
can
move
it
around
and
you
can
minimize
the
modal.
Do
you
know
what
I'm
talking
about.
D
B
See
where
we're
going-
and
I
don't
know-
I
don't
I'm
glad
we're
at
this
stage
of
our
we're-
glad
we're
at
this
stage
of
maturity,
of
how
we
build
software
in
the
web.
Where
we're
not
just
copying,
you
know
things
we
see
out
in
the
wild,
but
we're
actually
trying
to
figure
out
and
solve
what
helps
the
user.
What
are
you
know
psychologically
helpful
for
users
absolutely.
D
D
D
D
D
A
D
All
of
these
had
to
be
adjusted
as
well,
so
that
was
going
out
nicely
and
there
was
one
action,
exactly
delete
user.
That
was
a
little
more
tricky.
We
did
that
in
the
last
section
session,
which
basically
was
the
same
thing,
but
it
was
a
whole
another
model
to
work
on,
so
that
was
it
and
what
yeah
it's?
Actually
it's.
It's
it's
a
frankenstein
thing
bigger
than
expected.
D
So
here
is
the
thing
there
is
a
component
called
a
user
model
manager
which
I'm
definitely
not
familiar
anymore
with
what
it
does,
but
the
the
failing
spec
and
this
kind
of
makes
sense.
What
is
currently
failing
is
this
test,
so
it
doesn't
doesn't
render
the
model
anymore
and
not
100
sure
what
is
going
on.
But
if
I
look
at
this
setup
and
knowing
that
we
now
have
the
event
hub,
that
does
not
surprise
me.
So
that
was
definitely
something
we
could
have
have
a
look
into.
Yes,
so
I
think.
B
B
A
D
B
D
So
I'm
kind
of
afraid
that
this
might
be
a
pain
if
you
just
want
to
have
a
couple
of
those,
but
I
don't
really
definitely
but
other
than
that
we
definitely
shouldn't.
So
that's.
Basically,
all
of
this,
like
all
of
these
actions,
are
used
within
the
users
table
and
maybe
on
more
spots
which
I'm
not
aware
of,
but
I
know
about
the
user's
table,
but
that
is
about
that
is
about
it
currently
and
that's
absolutely
cool.
B
B
B
D
B
B
D
D
D
You
know
what
what
I
put
my
idea
on
what
what
might
be
going
on
here,
because
we
are
accessing
the
dom
and
we're
accessing
the
dome
from
the?
U
and
we
are
looking
for
different
elements
and
what
we're
seeing
here
is
like
loading
behavior,
maybe
some
async
stuff,
but
it
could
be
that
in
however,
kind
of
way
beside
the
site
is
loading.
The
element
might
be
already
there
or
it
might
not
be.
B
B
D
B
D
D
Before
I
dive
into
this,
I
want
to
see
one
more
thing,
because
what
I'm
wondering
is
so
obviously
it's
sometimes
there
and
it's
not
I'm
just
yeah.
Now
it's
not
obviously
not
there
well
at
least
not
what
we
wanted,
but
now,
okay,
so
it
is
not
there
and
it
doesn't
get
added
and
even
interesting.
So
it's
not
even
updating
back
to
at
a
later
point
or
something.
D
D
E
B
A
B
Object,
the
event
was
originally
dispatched.
Current
target
is
a
reference
to
the
currently
registered.
This
is
the
object
to
which
the
event
is
currently
slated
to
be
sent.
Let
us
try,
I
would
be
up
for
trying
current
target
and
if
that
still
doesn't
work,
I
have
another
idea,
but
maybe
we
should
try
current
target.
B
D
D
B
D
D
B
B
B
B
D
D
B
B
D
B
Yeah
and
that's
kind
of
the
thing,
so
let's
what
was
causing
that
and
and
here's
a
big
deal
why
I
don't
want
to
just
because
if
this
was
caused
by
our
confirm
modal
stuff,
then
our
confirmed
modal
stuff
could
be
breaking
other
things
like
it's
a
really
expect
confirmed
model
to
be
a
no-op
yep
so
are.
Are
we
not
getting
our
click
listener.
D
B
B
D
D
B
D
B
And
I
don't
know:
what's
I
don't
know
exactly
what's
going
on
and
what
it
might
be
is
because
we
saw
this
was
titled,
something
shared
like
we
may
have
abstracted
something
that
really
can't
be
abstracted.
We
may
have
abstracted
something
that
isn't
worth
it
and
we
need
to
just
have
it
be.
You
know
maybe
like
a
specific
thing:
yeah
yeah
yeah.
So
hopefully
these
these
feature
specs
are
fixed
now
because
we
fixed
the
bug
with
the
icon
clicking.
I'm
sorry.
So
I
was
just.
D
B
D
B
D
B
B
D
E
D
B
B
B
B
B
B
D
B
Got
the
console
error
and
so
the
model
didn't
show
because
we
blew
up,
but
that's
not
great,
so
two
things
I'd
like
I
think
we
should
do
the
one
is
yeah.
We
can't
do
the
event
hub
thing
and
have
that
class
at
the
same
time,
and
the
long
term
goal
is
don't
reference
that
class
anywhere
in
view
just
use
the
event
hub.
But
the
second
thing
is:
let's
have
our
class-based
approach.
Let's
add
a
little
bit
of
defensiveness.
B
B
B
B
D
B
Let's,
let's
go
ahead
and
just
do
path,
maybe
right
now-
and
I
don't
know
what
what
your
preference
are.
I
like
actually
doing
this
feel
free
to
do
whatever
you
prefer.
I
like
to
do
I
like
to
return
early,
so
I
would
consider
this
like
a
guard
clause,
and
so
I
would
like
to
do
like
if
not
path,
return
and
then.
B
Do
the
rest
of
the
behavior
below
the
f?
That's
that's!
This
is
a
hundred
percent,
just
a
preference,
and
if
you
don't
like
it
then
feel
free
to
be
like
hey.
I
don't
know.
Actually
I
actually
do
like
it.
So
just
like
that.
Basically
isn't
it
yep
yep
yep
and
I
think
that's
going
to
be
a
helpful
level
of
defensiveness.
B
A
D
A
D
D
D
A
B
D
B
B
B
Oh,
that
doesn't
make
sense
the,
but
it
uses
the
spring
loading
server,
which
basically
optimizes
how
it
reads
and
caches
the
ruby
files
and
just
makes
tests,
makes
running
tests
a
lot
faster
wow.
You
see
it
especially
with
unit
tests.
The
feature
specs
can
still
be
slow
because
it
requires
the
yeah
yeah.