►
From YouTube: Magento PWA Backlog Grooming, 21 September 2018
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).
A
We'll
give
everybody
just
a
minute
before
I
get
started,
so
I
think
for
those
who
saw
the
saw
the
updates,
and
so
we're
gonna
go
through
two
items
here
for
some
Romanians,
oh
really
want
to
get
the
basics
of
search
coverage
at
a
bit
even
have
saw
me
go
through
some
of
the
latest
work
that
she's
doing
in
that
area.
But
the
idea
there
is
to
create
a
very,
very
basic
search,
nothing
fancy
without
a
complete
nothing.
A
Fancy
was
even
pagination
yet,
but
the
idea
is
that
be
able
to
test
the
current
API
and
show
and
display
something
that
would
pull
some
full
catalogue,
data
that
you
would
expect
and
so
the
same
spirit.
They
were
talking
about
and
had
done
with
checkout,
but
apply
that
to
search,
and
that
makes
great
fodder
or
for
someone
to
pick
up
in
the
and
the
new.
The
Moodle
folks
were
taking
in
terms
of
self
self-sufficient,
but
also
business
oriented
stories,
and
so
it.
A
Be
so
we're
gonna
cover
cover
that
a
bit
and
so
again
same
philosophy
there.
You
know,
obviously,
now
that
we've
got
authentication
that
opens
the
doors
to
new
experiences
purchased
otherwise
known
as
the
order.
History
is
a
great
one,
and
so
we'll
go
through
that
that
you
know
gold
they're,
not
necessarily
again,
to
also
go.
You
know
right
into
and
then
even
pulling
up
an
order,
details
screen.
Yet
that
would
come
back
to
be
able
to
retrieve
that
information
in
some
meaningful
way.
That
would
be
the
beginnings
of
what
Zambia
has
created
today.
A
C
Actually
yeah
I
did
have
a
quick
question
yeah,
so
I
was
just
having
some
issues
with
the
circle
see
I
integration
on
my
latest
poll
requests
the
account
creation.
One
specifically
I
thought
something
in
a
chat
about
master
being
broken.
Somehow
as
wondering.
If
there
is
something
going
on
there,
I
didn't
know.
If
is
related.
A
A
D
D
So
I
looked
at
it
and
it
looks
like
you
need
to
update
your
package.json,
which
is
a
new
requirements.
I
recognize
I'm.
Sorry,
your
package
lock
JSON,
which
is
a
new
requirement,
I
recognized
what
I
would
say
is
the
easiest
thing
for
you
to
do
is
to
pull
in
latest
master
and
then
do
the
whole
like
clean
and
rebuild
the
environments.
And
then,
if
there's
changes
that
you
see
to
package.json
commit
those
changes,
and
that's
one
thing
that
I'm
actually
going
to
talk
about
today.
Briefly,
is
those
changes.
D
C
What
was
going
on
in
circle,
CI
yeah,
it
did
say
the
package
lock
in
the
package.json
weren't
matching
up
and
when
I
was
trying
to
run
npm
install
just
on
the
roots.
I
was
getting
some
recursive
like
infinite
error,
kept
trying
to
install
the
same
package
over
again
I.
Think
I,
probably
just
need
to
do
like
a
complete,
clean,
install,
like
you
said
so,.
D
Do
a
clean
is:
do
a
clean,
install
and
pull
or
be
based
completely
on
master.
What
what
that'll
do
is
get
you
to
the
changes
that
we've
made.
One
change
we've
made
is
that
the
learner
repo
is
no
longer
bootstrapping.
Every
time
you
reinstall,
we
I
switched
from
a
learner
workflow.
That
requires
that,
where
you
declare
these
dependencies
and
then
it
sort
of
cross
links
them
and
detects
them,
I
change
it
from
that
to
used
learner
converts
to
kind
of
hard
code
file
based
dependencies
into
the
package.json
sense.
D
So
that
means
that
now
npm
can
cross-linked
by
itself
and
learn
as
responsible
just
for
topo
sorting
of
the
build
so
long
story
short
when
you
pull
master
you're
gonna,
see
all
of
the
package
locks
in
the
package.
Directories
themselves
are
gone
package,
locking
is
now
the
responsibility
of
the
root
and
the
dependencies
that
are
common
have
all
been
pulled
down
to
the
root.
D
The
only
dependencies
you'll
see
in
the
individual
packages
are
runtime
depths
or
dev
depths
that
need
a
command
that
have
a
command
line
command
that
they
expose
that's
a
little
bit
of
a
change
so
make
sure
that
when
you
pull
that
in
that,
you
do
a
fully
clean,
install,
which
is
just
as
simple
as
running
NPM
or
NPM,
run
clean
and
then
NPM
install
and
then
just
make
sure
that
you
run
through
the
whole
test
suite
before
before
pushing
that's
it.
Okay,.
D
A
A
Right,
let's
go
again
started
again,
as
promised.
We're
gonna
be
covering
the
basics
of
search
and
the
basics
of
a
purchase
history
here,
and
so
so
for
those
of
you
take
a
look
at
the
in
advance,
so
the
idea
here
is
to
cover
again
some
extension
around
the
the
venya
theme
right
and
so,
as
you'd
expect.
You
know.
We've
made
some
good
progress
on
Browns,
but
we're
gonna
need
a
search.
A
So
the
story
here
is
as
follows:
the
idea
again,
as
a
shopper,
you
wanna
be
able
to
search
for
products
on
this
menu
storefront
again
using
live
data
by
using
basic
keywords
like
hoodie
that
I
could
use
to
make
an
informed
purchase.
So
what?
Of
course,
it
would
imply
updates
to
to
the
venue
header-
and
we
got
some
ideas
around.
That
also
implies
that
this
could
be
and
would
be
initiated
by
a
keyboard
again.
A
The
idea
is
using
got
a
keyboard
entry
to
commit
again
the
details
themselves,
of
course,
coming
through
by
way
of
UX,
but
one
of
the
assumptions.
The
assumptions
here
are
things
like
filtering
sorting,
not
part
of
this
first
effort.
Of
course,
every
great
search
has
something
like
that,
but
we
want
to
iterate
on
that
one.
Those
would
be
forthcoming
stories,
but
the
top
I'd
you
so
the
development
and
build
will
be
done
with
that.
A
Coming
down
the
road
pagination,
which
we've
covered
a
little
bit
last
week,
we'll
continue
to
refine
that,
but
not
necessarily
part
of
this
as
well
as
autocomplete.
Again,
all
the
complete
scan
is
shown
right
is
going
to
be
a
pretty
pretty
sizeable
story,
so
we'll
do
that
one
in
the
future,
but
but
essentially
the
basics
of
what
will
create
here,
which
is
really
just
a
search
results,
page
and,
of
course,
we're.
You
know
where
the
API
is
covered.
A
We've
use
as
much
as
we
can
around
graph
to
oh
and
maybe
lucky
on
this
one,
but
all
the
rest
where
were
appropriate.
So
what
are
some
of
the
tasks?
These
are
ones
I've
identified,
but
definitely
going
to
need
your
help
on
this
one.
The
idea
here
again
is
that
we
want
to
retrieve
products
from
product
catalog
data
bed
by
keyword
via
those
api's.
We
want
to
identify
again
some
of
those
future
considerations
for
for
the
faceted
navigation,
faceted,
inner
filtering
that
would
occur
and
and
then
just
the
implementation
itself,
and
of
course
you
know.
A
E
E
E
I
do
have
gratings
and
products
thumbnails,
and
things
on
here
need
to
know
how
much
of
that
is
which
is
going
to
start
with
the
text
and
then
I
guess
I
did
have
a
pagination
search
results,
as
opposed
to
just
cover
me
on
a
categories
page,
but
for
just
a
quick.
We
can
just
do
the
show
and
I
either
corrected.
A
So
yeah
I
think
so
in
the
spirit
I.
Think
of
really
showing
what
good
and
great
is
on
this
and
I'd
love
to
keep
this
in
mind
at
the
epic
level.
So
you'll
see
the
story
we
have
here.
Creatives
come
to
master
for
what
this
will
be,
but
Tommy
I
are
going
to
talk
about
this
a
little
bit
this
afternoon,
though,
so
that
we
don't
need
full
versions
of
the
same
reality.
A
A
That
autocomplete,
that
you
would
have
there
sure
it'll
be
great,
but
that
becomes
a
little
bit
more
work
from
that
it
would
be
initiated
and
then
again
to
a
very
similar
experience
that
you've
got
here.
You
know
the
the
ratings
which,
which
believe
that
doesn't
even
exist
yet
in
terms
of
API
coverage
and
I,
like
the
idea
of
show
more,
at
least
for
the
time
being
until
we
until
we
button
up
pagination.
A
Even
then,
even
if
that
becomes
a
an
issue,
we
can
even
push
that
off
to
a
to
a
to
a
future
sprint,
although
that
would
create
some
interesting
cases
for
if
you
added
the.
If
you
could
say
you
use
the
letter
A
in
a
search
in
the
entire
catalog,
it
comes
out
and
that's
that
down.
That's
that's
not
awesome
either
for
that.
So
that's
that's
kind
of
what
we're
thinking
super
straightforward.
I
do
like
the
idea
of
kind
of
pulling
down
from
the
header
opening
this.
A
E
A
Like
that
personally,
I'm
certainly
open
to
feedback
I
like
the
idea
of
being
able
to
kind
of
go
quickly
through
a
couple
permutations.
Let's
say
you
missed
all
buddies
with
that
with
the
Y
versus
an
IE
and
and
then
you
want
to
kind
of
quickly
go
through
some.
Some
of
those
experiences
I
think
to
me
I
think
that
makes
the
most
sense
for
now.
You
know
in
part
of
what
we're
doing
here
also
is
doing
and
conducting
user
research
as
part
of
it.
A
C
E
Definitely
certainly
could
be,
I
mean
the
more
we
can
reduce.
Effort
on
the
part
of
the
shop
are
the
better
that
they've
already
indicated
that
they're
within
a
certain
category
and
they're.
Looking
so
maybe,
let's
say
they're
in
pants
and
they
type
it
would
only
show
them
pants
in
blue
or
shades
of
blue
certainly
be
nice.
However,
I
don't
know
that
I'm
specific.
E
A
Search
is
an
easy
thing
to
get
over.
Complicate
that
my
hopes
for
this
one
is.
It
could
be
very
based,
we've,
actually,
some
interesting,
fully
active
base,
working
them
internally.
That
pulled
this
off
fairly.
You
know
great
for
and
I
think
the
work
is
going
to
be
as
we
get
into
scenarios
or
just
discussed
I
think
that's
five.
We
want
to
find
those
we
get
before
we
get
too
far,
but
we
want
to
fold
some
of
these
scenarios
with
live
day.
A
A
Well,
look
we
if
we
are,
if
we're
clear
on
what
needs
to
be
done
for
this
one
again,
I'll
work
with
down
Mia
on
the
right
way
and
again
you
get
D
here
to
us
the
link
to
the
right
concept.
Here
we
want
to
be
dynamic
with
with
what
we've
got
and
what
we're
showing
here,
assuming
everything
we
just
saw.
Do
we
feel
like
we're
in
a
good
place
to
make
some
estimates
I
have.
F
F
D
The
heterodox
like
in
results,
I'm,
sorry,
that's!
That's
a
twenty
dollar
word
when
I'm
looking
for
a
ten
cent
one
you
wanted
to
just
have
multiple
types
in
line
and
the
results
that
you
want
to
be
a
blowed
categories
as
well
as
products,
and
so
that's
what
graticule
doesn't
yet
support
category
tree,
for
instance,
does
not
let
you
just
search
through
the
tree
for
categories
right
now.
The
query
only
lets
you
specify
a
category
and
then
walk
the
tree
by
running
recursive
queries,
which
is
just
untenable.
D
So
you
know
there
are
a
couple
of
things
you
could
do,
but
I
would
say:
what's:
simplest
is
to
use
the
graph
QL,
API
and
just
load
products
and
get
that
working
and
then
because
that
immediately
confers
some
some
value.
Then,
perhaps
during
that
you
can
think
about
ways
to
make
in
a
more
generic
search
so
that
it's
not
entirely
bound
to
products
and
their
particular
data
shape.
D
So
you
know
nodes
that
you'd
have
a
select
so
that
any
given
component
has
a
method
that
translates
it
into
a
search
result,
shape
or
Maps
it
into
a
search
result
shape.
And
then
you
know
you
can
just
do
products
and
and
we'll
accept
that
story.
I
was
saying
it's,
it
would
just
get.
You
know
iterated
later
to
include
other
stuff.
Okay,.
D
We
we
I
would
leave
it,
but
because
it's
not
actually
an
acceptance
criterion.
What
it
is
is
a
guideline
for,
if
there's
any
data,
that
you
absolutely
need
to
complete
the
story
that
you
can't
get
but
you're
perfectly
free
to
take
a
path
through
the
story
that
satisfies
its
basic
user
story:
premise
that
doesn't
use
the
REST
API,
that
is,
that
is
okay
to
do.
Okay,
that's.
A
Good
dealer's
choice,
as
they
would
say:
no,
that's
a
good
one
and
again
remember
if
it's
kind
of
stepping
back
to
what
our
hopeful
and
in
expected
outcomes
of
this
as
we
find
cases
where
graph
fuel
isn't
covered
or
necessarily
working
the
way.
We
need
it's
the
perfect
opportunity
for
us
to
with
them
and
work
with
them.
I
know
the
comrade.
A
Folks
are
going
back
and
forth
kind
of
between
the
two
different
teams
here
and
they
can
bring
that
fairly
quickly
and
get
that
on
there
they're
their
board
as
well,
which
is
that
which
is
great
so
that
sort
of
part
of
what
we're
doing
here
and
part
of
why
we
want
to
make
really
progress,
even
if
it's,
even
if
just
maybe
he's
twenty
percent
of
the
the
intended
vision
of
some
of
the
UX
work
that
Sally
and
those
guys
have
done.
So.
Yes,
that's
up
how
we
feel
about
estimate.
A
D
Again,
bearing
in
mind
that
you
want
to
give
a
generous
estimate
for
the
minimum
viable
thing,
I
would
guide
you
guys
again
to
think
about
just
doing
products
for
now
and
that
reduces
the
plumbing
obligation.
Now
you
just
got
to
think
about
the
UI,
and
you
know
you
want
to
just
go
use:
Kent,
C,
Dodds
component
for
rendering
autocompletes
plug
that
in,
if
you've
not
seen
it
before
it's
pretty
nice.
D
A
That
would
say,
at
least
from
a
product
standpoint.
This
feels
relatively
size
to
others
that
we've
declared
as
5
in
this
category
and
so
that,
in
terms
of
scope
and
size,
get
nothing
the
person
who's
going
to
be
building
those
for
that
steel.
It
feels
right
for
what
that's
worth
great,
alright
notion
notion
to
make
this
a
five
then
so
any
objections.
A
Alright,
very
good,
we'll
make
it
so
excellent.
So
so,
what's
going
to
update
this
one
again,
not
a
lot,
not
a
lot
of
changes
and
I
think
we
had
a
really
good
discussion
of
what
this
is.
What
this
isn't
again,
we'll
add
the
the
link
to
the
prototype
here
shortly
afterwards,
which
will
match
exactly.
We
talked
about
quick
to
an
end
process.
A
You
may
notice
a
couple
changes
here
in
the
pipelines,
and
so,
if
you
were
familiar
with
a
graph
QL
project,
this
will
look
extremely
familiar
for
this,
and
so
one
of
our
hopes
and
James
will
cover
a
little
bit
on
some
of
the
work,
that's
being
done
of
the
releases
themselves.
But
we
want
to
make
our
work
easier
to
pick
up
and
you
know
declaring
things
that
are
indeed
ready
to
start.
So
that
could
mean
anybody
here.
A
Anybody
elsewhere
in
the
community
or
even
internally,
can
pick
this
up
and
start
making
progress
on
that,
so
we're
still
being
bound
ourselves
by
the
by
the
sprints.
So
how
big
the
the
the
cadence
and
the
determined
is
that
we
do,
but
we
want
to
make
these
a
bit
more
autonomous
than
where
we
are.
It
works
quite
well,
but
but
with
the
scale
that
we're
continuing
to
get
many
thanks
to
the
folks
here,
but
also
with
a
with
the
rise
and
the
inclusion
of
a
feature
team.
A
That's
will
be
joining
us
here
internally
in
a
week
we,
you
know
this'll
help
us
quite
a
bit
and
so
in
this
example,
we've
groomed
it.
It
was
ready
to
groom
it's
ready
to
start
now,
but
did
you
know
we
might
run
into
a
situation
again?
You'll
see
a
few
of
these
stories
here
for
Jimmy,
where
we're
hitting
some.
You
know
hitting
some
road
bumps
some
roadblocks
in
the
coverage
right,
and
so
that
puts
it
in
an
investigation.
But
ultimately,
once
that's
picked
up,
you
know
we'll
making
good
progress
to
them.
He's
in
progress.
A
We've
consolidated
we'll
be
in
QA,
which
there
was
there
were
some
questions
on
the
differences
to
do
there
and
then
we'll
continue
to
build.
A
queue
of
this
does
need
to
be
updated
with
your
the
New
Orleans
there,
but
the
one
thing,
islands
that
need
to
be
merged
right.
We've
got
a
lot
of
great
force.
That's
out
there.
We
are
addressing
things
like
master
and
again.
A
James
are
looking
at
a
release
strategy
here,
but
will
clearly
see
the
items
that
have
been
completed
but
just
need
to
be
merged
after
they've
gone
through
review
and
then
once
once
the
merge
they're.
Not
so.
Hopefully
that's
is
pretty
straightforward,
but
just
a
quick
note
on
process,
as
we
scan
things
out
a
little
bit
both
or
a
relief
standpoint,
but
also
a
also
a
a
backlog
standpoint
as
well.
Any
questions
on
that
before
before
we
switch
gears
that
into
purchase
histories.
A
All
right,
very
good
and
again
also
part
of
the
a
part
of
the
reason
here
too,
is
to
be
consistent
amongst
the
other
active
higher-profile
projects
in
the
community.
So
the
idea
that
you're
contributing
the
one
you
might
be
contributing
to
others
and
those.
So
this
may
make
your
your
lives
a
bit
easier,
as
you
understand
how
things
work.
So
that's
one
part.
So
let's
talk
about
purchase
history,
otherwise
known
as
order
history
again.
A
You
too
much
again
we're
gonna
go
with
a
very
minimal
approach,
whether
the
shopper
I
want
to
find
previous
purchases
on
venue
by
order
number
so
that
I
can
make
a
repeat
purchase
which
is
an
m2
they
also,
but
also
request,
help
from
customer
service.
You
see
an
order.
You
have
a
question
about
if
you
should
be
able
to
pick
up
the
phone
and
talk
about
that
order,
and
so
that's
some
really
interesting
things
here
happening
around
authentication.
We
get
be
authenticated
to
pick
this
up.
A
You've
got
to
be
able
to
pull
and
understand
real
order,
numbers
which
is
great
and,
of
course,
layout
and
the
data
beyond
that,
as
grouped
by
some
of
the
things
you'll
see
in
the
prototype.
So
what
are
the
assumptions
again?
It's
gonna
require
a
level
of
authentication
if
we're
gonna
use
Veronica
on
this
one
as
well
to
access
data.
We
like
to
try
to
try
to
take
a
stab
at
grouping.
You'll
see
this
in
the
US.
The
idea
of
the
last
30
days
and
kind
of
earlier
are
desirable,
but
noted
stretch
goals.
A
We
can't
get
there.
That
adds
a
little
bit
too
much
complexity,
that's
okay!
If
we
really
want
to
make
this
by
a
more
minimal
week-
and
we
can
just
have
a
straight
list
of
the
recent
orders
placed
that's
okay
order,
details
will
hit
be
handled
separately,
and
so
we'd
have
to
worry
about
linking
on
this
one
simply
a
hash
tag.
A
We
do
in
your
in
your
link
with
us
we'll
cover
that
as
a
as
a
separate
story
as
that
gets
refined,
and
things
like
filtering
much
like
what
we
just
covered
in
search
is
going
to
be
consider
in
the
future.
Of
course,
there
should
be
build
with
that,
knowing
that
is
going
to
be
in
mind
that
that
will
be
more
of
a
time-based
search
ads,
especially
in
cases
they're,
lots
and
lots
of
lawyers
and
and
they're
a
bit
of
our
boilerplate
on
the
Left
versus
grass.
You
all
will
use
graft
you
all.
A
Note
that
this
might
have
been
copied
and
need
to
make
a
change
to
the
basic
search
piece
to
this,
and
then
also
and
on
the
testing
and
documentation
for
this
as
well
and
so
want
to
make
those
updates.
That's
what
we
got
here.
So
those
are
yeah
I,
think
those
are
the
basics
again,
a
really
just
creating
a
list
of
the
orders
that
were
placed
by
the
account
on
this
one.
That
I
think
they're
all
very
nicely.
A
B
A
Mean
the
product
person's
here
I
would
love
those
of
course
that
take
it
eat
it
too,
but
but
we
would
separate
that
as
a
separate
story,
because
theoretically
you
could
create
this,
be
informed,
handle
things
offline,
kind
of
the
spirit
of
what
we're
talking
about
and
some
of
the
tasks
or
the
reset
scenarios
per
se.
But
then
this
is
a
follow-up
story
likely
not
get
to
it
until
probably
next
week,
but
then
handle
that
once
we
feel
good
about
the
design
there.
That
would
give
you
a
link
to
the
details
of
the
order
itself.
A
A
E
E
The
first
way
is
based
on
this
is
actually
the
second
iteration
of
it
and
is
based
on
internal
feedback,
and
the
approach
it
takes
is
to
display
orders
in
groups,
items
that
were
ordered
together
as
the
unit,
and
then
you
dig
in
to
the
order
to
see
individual
items
with
the
options
to
purchase
again
and
other
things
like
other
details
and
so
on.
So
this
is
the
findings
or
authenticated
version
of
my
account.
You
would
go
through
purchase
history
to
a
list
now
one
of
the
problems
there's
a
glitch
on
the
screen,
never
mind.
E
A
E
If
there,
if
there
are
in
excess
of
four
items,
my
order
now,
if
you
did
click
on
the
the
dot
similar
to
what
we
have
in
mini
cars,
that
shows
you
options
you
would
have.
If
it
is
an
active
order
in
shipment,
you
could
track
the
order
or
you
could
view
order
details
and
then,
when
you
did
choose
to
view
order
details
you
would
have
you
know.
E
You
would
have
the
individual
items
with
the
opportunity
to
buy
again
share
or
review
you
could
track.
It
will
see
the
payment
details
and
on
to
something
that's
approach.
One
approach
to
is
the
way
Amazon
currently
does
it
and
it's
not
on
in
the
prototype
yet
I
removed
it
because
it
was
getting
confusing
for
me
to
manage,
but
essentially
what
they
do
is
they
have
a
list
of
individual
products
when
you
dig
into
a
product
purchase,
it
shows
you
what
else
would
purchased
with
that
product.
E
Hang
on
I
guess:
I
got
rid
of
that
screen,
but
essentially
it
would
be
the
same
screen
right
with
a
little
bit
of
modification,
I
kind
of
see.
Why
Amazon
does
it
this
way?
I
think
people
have
a
far
greater
likelihood
of
remembering
an
item
that
they
purchased
as
opposed
to
maybe
an
order,
but
I
think
that's
where
the
thumbnails
may
help.
E
It
certainly
keeps
the
the
parent-child
structure
a
little
bit
more
logical
when
you
go
through
the
order
to
an
individual
item
is
supposed
to
the
other
way
around
I'm,
not
necessarily
tied
to
one
or
the
other
I
need
some
guidance
on
which
way
which
side
of
the
wall
we
want
to
the
land
on
to
go
and
I
can
I
will
definitely
be
testing.
This
I
just
haven't
had
a
chance
to
do
it
yet.
E
So
it
just
depends
on
which
might
be
easier
to
change.
I
guess
I,
don't
know
like
if
they
stopped
started
with
a
grouped
approach
and
went
individuals,
and
then
it
turned
out
that
the
individual
order
approach
was
in
fact
the
more
easily
understood
or
e
or
the
better
experience
when
tested.
Would
it
be
relatively
easy
to
flip
that,
or
would
it
be
a
huge
amount
of
effort?
Is
the
question.
D
C
If
I'm
looking
at
my
purchase,
history,
I'm
just
thinking
I
want
to
make
sure
I'm
seeing
all
those
items.
The
thing
that
concerns
me
is
the
low
like
+4
button
or
whatever.
If
I'm
not
seeing
my
item
exactly
I'm,
not
gonna,
remember
which
card
it's
in
me
personally,
I
don't
make
organized
cards
or
anything
so.
C
E
What
else
you
purchase?
Because
if
you
are,
you
know
if
you
are
going
to
return
something
I
mean
I,
don't
know
that
there's
you
know
whatever
you
return,
so
I
think
that
I'm
in
kind
of
inclined
to
go
that
way.
I
just
wanted
to
make
sure
that
I
address
the
internal
feedback
I
had
received
and
make
sure
that
we
kind
of
but
I'm
proceeding
with
the
Amazon
sort
of
model
of
going
with
individual
products.
A
Yeah
I
would
I
would
second
that
and
that's
coming
from
someone
who
challenged,
probably
the
initial
thinking,
knowing
that
amazon
has
a
different
style
of
purchase
than
probably
what
we
would
want
in
a
reference.
Storefront
I,
do
like
the
simplicity
of
the
item,
focused
approach.
It
doesn't
mean
we
could
come
back
and
maybe
even
suggest
a
filter.
I'll
go
to
go
back
to
an
order
of
put
me
on
Facebook
pervert
well
well
into
the
future,
but
I
do
like
that
and
again
some
of
it.
A
A
We
we
always
reserve
the
right
to
pick
this
back
up
at
a
later
day,
but
I'm
inclined
to
keep
this
one
simple
for
now,
which
is
kind
of
where
we
started
and
then
put
on
hold
with
what
we've
got
the
order
based
approach,
and
so
you
can
think
of
a
puzzle
and
or
the
research
itself
may
may
show
it
differently.
What
we
could
do
is.
E
A
E
B
This
is
more
of
a
future.
I
said
this
is
more
of
a
future
consideration
one,
but
presumably
the
benefit
that
you
have
from
the
view.
My
order
is:
is
that
related
to
kicking
off
things
like
returns
or
seeing?
I
think
this
is
more
common
in
the
EU,
so
I'd
like
seeing
the
invoice
that
purchase
as
well.
A
Yeah,
that
was
that
was
the
initial
thinking
and
my
feedback.
I
I
gave
family
about
a
week
or
so
ago,
which
was
it
represents
a
moment
in
time
and
so
again
we're
not
talking
about
at
least
again
with
a
reference
door
front
type
of
throughput
than
the
type
of
you
know,
purchased
frequency
that
you
would
see
with
some
of
that,
and
so
there
there
may
be
merit
to
that
again
also
merit
to
at
the
moment
in
time
around
customer
service
in
returns.
E
Okay,
thank
you
indeed
and
then
just
to
clarify
the
workflow
for
this
for
the
screen
system,
some
questions
about
it.
What
you
guys
saw
an
envision
here
is
our
intent
of
working
copy.
So
what
happens
is
once
I
discuss
changes
with
the
team
here
and
we're
in
a
comfortable
place
to
sort
of
shared
it
out?
We
put
it
in
the
shed
for
the
diet
that
you
all
have
access
to
the
link
to
which
I
had
pasted.
Maybe.
B
E
A
E
Rolling
just
to
close
the
loop
with
you,
you've
provided
some
feedback
on
the
quantity
selector.
So
we
edit
the
cart
item.
I
did
implement
that
I'm
happy
to
send
that
to
you.
Offline
I
went
through
the
defaults
for
Android
and
iOS
and
with
the
Android
one
I
checked
about
three
or
four
people's
devices,
and
it
all
seemed
to
be
the
same,
but
I
did
have
questions
if
it
was
done
in
any
other
way.
So
hopefully,
if
you
have
time
you
could
provide
some
feedback
on
that.
A
A
C
G
D
D
D
C
A
Yeah
I'd
rather
be
more
conservative
on
this,
when
the
orders
we
haven't
played
as
much
in
orders
as
we
have
with
products
and
so
yeah
if
we
come
in,
come
in
as
a
file
fantastic,
but
this
would
be
a
good
I
feel
better
about
a
draftsman
here
for
what
it's
worth
as
well.
A
A
A
Get
two
of
these
done
in
time
and
well
well
well,
faster
than
I
had
expected
today,
and
so
that's
really
all
we
have
our
weeks
early
can
go
into
the
depths
of
order
details.
You
guys
can
start
to
see
the
beginnings
of
that,
but
I'd
rather
become
into
the
Dreaming
a
little
bit
more
for
parents.
Oh
I
think
we
can
cue
that
up
for
the
next
time
around
getting
family
a
little
bit
of
time
to
us
that
kind
of
didn't
finish
them
the
things
we
just
talked
about
as
well.
A
D
It
would
be
valuable,
and
not
everyone
who
is
maybe
relevant
to
this
discussion
is
on
the
call
but
I'm
going
to
follow
up
with
them
separately.
So
I'll
speak
briefly,
because
I
really
want
to
make
sure
this
isn't
just
me
talking,
but
that
this
makes
it
into
documentation
in
the
readme
on
pwace
to
do
the
repository
in
the
wiki
and
enforced
or
at
least
encouraged
by
danger
and
RCI
process,
and
it's
this
we're
gonna
switch
to
a
model
called
git
ship.
It's
a
variant
of
git
flow
that
is
simplified.
D
That
does
not
have
developed
branch,
even
though
there
is
currently
a
develop
branch.
Those
experiments
will
be
about
merging
everything
from
there
and
part
of
the
reason
we
selected
it
is
that
the
originator
of
the
idea
of
get
ship
is
part
of
the
Magento
community
and
is
willing
to
volunteer
his
time
to
help
us
get
it
set
up
the
best
way.
D
D
D
Lastly,
the
way
that
we're
gonna
manage
releases
first
of
all
is
we're
going
to
kind
of
retro
actively
take
the
existing
PWA
studio,
codebase
and
call
it
zero
point
X
and
we're
going
to
issue
a
point
release
kind
of
at
the
tip
of
Master.
Once
we've
got
some
basic
stuff
integrated
and
working,
some
PRS
from
bar
greens,
some
from
others,
are
going
to
make
it
in
because
they
pertain
to
the
existing
set
up
and
then
the
sea
change
that
we're
gonna
make
is
that
it's
no
longer
a
theme.
D
It
relies
on
an
upward
server
and
that's
going
to
be
kind
of
the
centerpiece
of
1.0.
So
you'll
see
a
release,
1.0
branch
that
large
changes
are
going
to
start
getting
merged
into
so
that
we
can
accelerate.
All
this
stuff
will
be
documented
and
that's
it
questions
I'm,
not
hiding
from
your
questions.
I
dropped.
My
pen.
A
Around
there
others
involved
in
others
who
would
be
interested
in
hearing
about
this,
as
well
as
the
good
stuff.
This
is
another
same
theme
as
the
backlog
and
we're
continue
to
get
a
lot
more
contribution,
which
is
great,
we're
continuing
to
get
help
internally
as
well,
which
is
great
and
so
a
little
bit
part
of
that.
The
release
strategy,
as
well
as
what
we're
doing
on
the
story,
is
a
big
part
of
that.
So
thank
you,
of
course,
all
right.
Well,
any
other
questions
or
anything
else
that
we
wanted
to
cover
before
we
break.
B
D
We
are
confident
that
we'll
be
able
to
integrate
those
into
workbox
by
using
its
middleware
structure,
but
presently
we're
using
a
serviceworker
wrapper,
plugin
that
simply
delegates
down
to
work
box
with
a
couple
of
bits
of
auto-generated
config
like
where
the
manifests
should
go
and
how
to
prefix
a
domain.
We
have
that
abstraction
there
so
that
we
can
Rev
work
box
under
it
and
so
that
we
can
add
things
without
being
obtrusive
to
people's
existing
webpack
configs.
So
that's
the
place
for
it
to
go.
D
Naturally,
it's
not
there
yet,
and
it's
not
backlogged
and
I'm
excited
to
do
it,
but
I
think
it's
it's
it's
fairly
large
candidly.
One
of
the
reasons
that
we're
not
pursuing
it
is
that
there's
a
lot
of
vania
to
get
done
and
the
people
who
would
perhaps
have
the
strongest
jeaious
skills
and
then
take
the
most
interest
in
doing
something
cool
like
that,
our
other-directed.
So
there
you
have
it.
However,
you
know
with
caching.
In
the
you
know,
the
stuff
you
get
from
work
box
for
free
you
get
catalog
offline,
close
to
free.
G
B
Yes,
they're
very
much
kind
of
you
know,
one-off
right,
yeah,
I
I
do
think
that
I
offline
catalog
I
think
is
tempting,
but
I
do
worry
that
there's
hidden
complex
the
around
stock
availability,
pricing
and
so
on,
where
you
want
to
be
very
sensitive
about
things
like
invalid
items
being
in
the
cache
the
yeah
order.
History
would
just
be
a
great
one
for
kind
of
going
yep.
G
D
Right
and
you
know
that
brings
to
mind
that
there
ought
to
be
a
way
to
positively
identify
ahead
of
time
or
while
you're
working
in
a
way
that
is
machine,
readable
views
that
pertain
to
essentially
write
only
data.
If
there's
a
way
for
us
to
just
quickly
tag
that
like,
oh,
then,
the
cache
can
find
it
separately.
D
The
way
to
sort
of
address
volatility
in
the
midst
of
catalog
information
like
stock
availability,
is
something
we've
been
talking
to
thinking
about
a
while,
Sammy
and
I
spoke
about
this
like
to
Magento
offices
ago.
It
feels
like
did
you
remember
that
conversation
song
mmm-hmm,
yes,
I
mean
I?
If
you
have
arrived
at
some
more
definition
on
it,
please
feel
free
to
talk
about.
I
have
not
but
I.
Just
remember.
D
We
discussed,
but
definitely
gonna
need
to
think
about
what
that
UX
is
beyond
just
having
an
indicator
of
offline
mode
to
have
some
way
of
showing
that
data
is
volatile
in
in
an
intuitive
way,
see
it
seems
like
new
ground.
So
you
know,
if
you
felt
more
about
it,
be
my
guest
to
talk
about
it
now.
But
you
know
point
is
it's
something
that's
in
our
in
our
minds,
circulating.
E
And
also,
what's
really
valuable,
to
have
in
an
offline
mode.
What
I
think
people
most
likely
do
want
to
do
the
thing
about
the
fashion
retail
vertical
is
that
they're
not
critical
purchase
in
intrinsically
speaking?
It's
not
like
medication
or
something
else
that
it
has
a
sense
of
urgency,
or
maybe
even
importance
in
most
contexts.
D
A
And
I
think
there's
a
lot
to
be
said
around
the
checkout
experience,
as
we
refine
that
even
more
to
make
sure
something
is
completed
as
intended,
of
course,
making
sure
the
things
that
are
not
intended,
maybe
maybe
have
fires
or
more
so
that
time
you
come
back
online.
Do
you
think
about
that
right?
Those.
E
B
A
All
right,
well,
very
good.
Well,
hey!
Listen!
Thank
you
again,
as
always
for
your
time,
your
help,
your
feedback
and
participation.
It's
always
inspiring
every
every
week
with
some
of
these
to
get
70
questions
with
whoever
you
know
where
to
find
it,
but
thank
you
again
and
then
have
a
great
weekend.
Okay!
Thank
you
thanks.
Everyone
thank.