►
From YouTube: PWA Community Update May 24, 2019
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).
B
C
D
B
E
F
Sounds
good
alright,
so
who
do
we
have
on
the
call?
Unfortunately,
I
did
not
do
enough
preps
promotion,
but
I
know
we
have
Jordan
and
Todd
and
Artyom.
Thank
you
very
much
for
coming.
I
think
we
may
wait
for
another
couple
of
minutes
to
see
if
we've
got
some
kind
of
a
quorum
when
you
think
of
that
guys.
F
G
C
H
F
I
So
good
morning,
good
afternoon,
good
evening
to
everyone
in
the
various
time
zones
hope
everybody's
had
a
good
week.
This
was
the
first
time
I
attended.
Imagine
it
was
a
really
great
experience
to
be
able
to
meet
so
many
people
in
the
community
and
have
direct
interaction
and
conversations
about
products
that
we
work
on.
We're
relatively
cocooned.
You
know
in
the
Austin
office,
especially
meeting
merchants
is
not
something
that
happens
every
day,
so
it
was
kind
of
nice
to
to
get
some
perspective
on
that.
I
My
main
goal
for
a
using
experience,
standpoint
going
into
imagine-
and
the
team
has
heard
some
of
this
before
so
I
apologize
for
repeating,
especially
since
you're-
also
gonna-
get
a
deep
dive
recap
next
week,
but
the
main
goal
was
to
sort
of
gauge
the
perceptions
of
the
products
out
in
the
wild
like
what
people
think
about
it.
Is
it
being
communicated
the
right
way
and
then
sort
of
the
assumptions
that
we
make
about
agencies
in
the
community
or
March,
and
some
adopting
technology
like
this?
I
Is
that
valid,
and
also
to
get
some
understanding
around
theme?
How
people
think
about
building
a
store
from
today?
They
think
about
it
in
terms
of
components
in
terms
of
page
layout
like
what
is
their
theme
customization
like,
so
that
was
one
part
of
it
and
the
other
part
of
it
was.
We
had
a
lab
where
I
actually
got
to
observe
people
using
the
tooling
and
doing
the
setup
and
going
through
tasks,
which
is
great.
So
we
ran
a
survey.
I
It
was
kind
of
a
me
in
person
with
the
survey
on
a
tablet
taking
it
around
to
folks
and
having
conversations
with
them
before
or
after
they
filled
out.
The
survey
and
I
think
that
it
was
nice
to
to
kind
of
realize
that
there
is
definitely
excitement
and
desire
for
this.
The
type
of
improvements
that
we
would
be
big
in
bringing
to
Magento
is
content
and
the
fact
that
people
will
have
greater
level
of
flexibility
in
customizing
that
it
was
also
nice
to
see
people
being
able
to
go
through
the
lab
tasks
with
relative
ease.
I
I
was
also
part
of
a
site
reviews
activity
where
we
had
pre
sort
of
approved
people
who
are
interested
in
having
their
sites
reviewed
from
a
user
experience
standpoint
to
come
and
meet
with
certain
members
on
the
team,
and
it
was
fantastic
to
realize
that
there's
an
overwhelming
kind
of
need
and
desire
for
people
often
merchants.
They
want
that
kind
of
UX
voice,
because
not
all
of
them
have
somebody
like
that
on
the
team.
I
A
lot
of
them
have
visual
designers
or
UI
designers,
but
they
they
do
look
to
Magento
or
the
Magento
relaxed,
integrated,
put
best
practices
and
recommendations
on
how
their
sites
can
be
improved.
So
what
so?
Those
conversations
were
really
valuable.
We
were
able
to
send
out
kind
of
a
report
to
them,
and
then
we
have
the
opportunity
to
schedule
a
follow-up
conversation
and
between
the
survey
and
the
lab
and
the
site
review
sessions.
D
I
Really
exhaustive
report
by
Nielsen
Norman,
economist,
UX,
best
practices
so
I
think
that's
just
what
came
out
of
imagine
is
that,
yes,
we've
done
a
few
things
and
we've
done
a
lot
in
some
areas,
but
there's
still
lots
more
to
be
done
and
achieved
and
to
know
that
we
have
the
support
of
community
to
do.
That
is
really
fantastic
and
everyone
is
so
open
and
willing
to
provide
feedback.
That
I
was
really
glad.
I
was
able
to
attend.
I
D
F
Thank
you
very
much.
Sam
yeah
I
really
appreciate
the
sort
of
ad
hoc
update
and
IQ.
Was
it
imagined
and
agree
with
everything
you've
said
we
certainly
had
in
general
like
a
lot
of
insights
from
people,
some
of
which
we
expected,
because
we
were
sort
of
adjusting
our
approach
and
talking
about
that
and
some
of
which
were
real
surprises
to
us.
So
you
know
Eric
and
I
and
Jimmy,
who
was
also
present.
We
kind
of
collect
our
personal
experiences,
but
Sami
has
going
a
little
further
and
collecting
real
data
and
really
appreciate
it.
F
And
I'll
just
quickly
review
what
Sam
you
said
later,
but
it
looks
like
what's
up
next.
Is
a
category
page
books
review
with
Tommy
and
for
just
in
context
there
is
a
pull
request
open
that
is
right
on
the
way
into
develop,
where,
as
we
are
going
to
do
with
the
rest
of
Ania,
Tommy
has
refactored
the
category
page
so
that,
instead
of
using
class
components,
it
uses
the
new
functional
components
with
spare
time
that
react
uses
that
gonna
enable
for
all
of
you
with
pyramid
sometime,
you
want
to
take
it.
I'll
take.
F
G
Right,
so
this
is,
as
James
hinted
at
AIDS
I'm,
calling
it
a
simple
refactor
because
as
much
as
possible,
I
try
not
to
introduce
any
new
functionality.
So
this
is
really
just
the
category
page
here.
That
is
just
a
product
to
list,
but
we
call
in
a
category
page
so
you'll
see
here
that
all
of
it
is
still
functional,
pagination
still
exists
and
a
lot
of
the
functionality
that
was
previously.
There
is
still
there.
G
Nothing
has
really
changed
in
the
UI,
except
for
a
little
smoother
scrolling
there
when
you
navigate
between
pages
so
on
to
the
refactor
one
pattern
that
you're
gonna
see
is
when
we
identify
functionality
that
we
predict
that
would
need
to
be
shared
between
components.
We're
gonna,
pull
it
out
into
custom
hooks.
So
pagination
was
one
of
those
hooks
so
you'll
see
here
that
that
simple
pagination
functionality
that
was
previously
in
redux,
we
pulled
out
put
into
this
to
use
pagination
hook.
E
G
Export
our
statement
and
then
you'll
see
that
used
here
in
the
category
component.
Now
so
you'll
see
I,
don't
know
if
a
diff
makes
sense
here,
but
you'll
see
nothing's
gonna
change
like
we
remove
the
class
declaration,
and
now
it's
just
a
functional
component.
We
extract
props
instead
of
all
the
prop
type.
The
default
prop
stuff
will
now
be
at
the
bottom
of
the
declaration,
and
then
all
the
functionality
now
has
just
converted
over
to
using
hooks
so
you'll
see,
as
advised
by
the
react
team.
G
G
This
is
us
using
our
brand-new
hooks
from
Peregrine's
for
using
the
graph
QL
query
API.
So
this
is
kind
of
how
you
gain
access
to
those
pieces
of
functionality
instead
of
wrapping
your
component
in
a
query
component
using
equal
space
baby
guide,
and
then
this
is
kind
of
how
you
vet
data.
Now
this
will
be
the
pattern
for
fetching
data.
You
have
an
effect
that
you
want
to
run
as
soon
as
the
component
renders.
We
also
have
some
dependencies
here
that
will
refire
this
hook,
and
then
we
just
kind
of
set
our
state.
G
Like
anything
else
in
react,
you
just
set
state
and
then
your
component
should
react.
So
that's
what
we're
doing
here
with
loading
data.
This
is
also
where
we
declare
our
scroll
behavior
and
then
our
component
itself.
It's
really
simple,
so
we
run
through
the
different
states
that
we
want
to
handle
and
then
render
out
what
things
sets
there.
So
here
it's
just
either
we
didn't
get
data,
we're
still
loading
or
we
have
our
data.
We
can
go
ahead
and
render
our
category
content
component
here.
G
So
in
this
same
vein,
since
this
was
contained
inside
of
the
category
root
component,
we
also
refactored
this
guy.
This
one
you'll
see
was
much
simpler
because
it
there
it
really
wasn't
doing
anything
besides
taking
props
and
rendering.
So
this
was
just
removing
classified
using
the
classified
or
the
merged
classes
hook
that
we
added
there
and
then
changing
into
a
functional
component
and
then
one
tiny
piece
that
we
were
able
to
do
so.
There's
total
pages
now
is
available
in
the
graph
QL
API.
G
K
K
G
All
right
and
these
Peregrine
hooks
are
actually
coming
up
in
r3
release,
so
that
search
component
that
was
demoed
a
while
ago
by
Jimmy
contained
a
lot
of
this
backing
functionality
for
the
query,
API
and
all
that
so
with
300.
You
can
start
experimenting
with
hooks
and
then
very
shortly
after
that
we
should
have
a
very
exciting
release.
That
starts
getting
a
lot
of
these
reef
actors
in
there
you
can
start
seeing
how
you
can
write
or
refactor
your
own
components
to
start
using
these
hooks.
E
F
F
We
talked
about
the
toasts
last
week
and
they
had
we
the
minutes.
We
would
have
known
that
so
folks,
we're
keeping
these
minutes
now
and
the
github
wiki
we're
trying
to
use
the
github
wiki
a
little
bit
more
because
it's
completely
public
and
because
people
can
propose
edits
and
comment
on
it
in
actually
comment,
but
you
can
propose
edits
on
it.
F
F
So
this
is
some
me
code
which
there
ought
to
be
less
and
less
of
this,
but
this
is
the
stuff
that
I
put
out
as
part
of
the
projects
to
create
the
pipa,
create
ppw
ACL
that
we
had
from
last
two
weeks
ago
that
I
prepped
for
imagine.
So
the
idea
behind
the
unified
system
for
environment
based
configuration.
It
was
based
on
the
fact
that
our
configuration
was
kind
of
all
over
the
place,
though
I
have
a
little
file,
which
is
meant
to
kind
of
review.
This
and.
F
Hey
wait
there.
It
is
I
gotcha
I've
already
received
some
feedback.
I
did
not
review,
but
yeah,
so
configuration
management
that
we're
that
we
were
doing
up
to
this
point
was
kind
of
all
over
the
place,
but
the
plan
from
now
on
is
to
use
environment
variables
as
much
as
possible
and
there's
some
explanation
for
why
we
want
to
do
that
here.
The
main
reason
is
just
that
environment
variables
work
everywhere.
F
F
You
run
a
project
build.
It
should
tell
you
a
few
things
about
your
local
environment
variables,
because
you
probably
already
have
a
dot
env
file
and
that
dot
e
and
the
file
probably
already
has
some
variables
in
its
set
and
you
kind
of
know
what
they
do.
So,
if
we're
going
to
migrate
and
use
different
variables,
it's
important
for
us
to
tell
you
if
we're
centralizing
these
variables,
it's
important
for
us
to
kind
of,
let
you
know.
What's
changing.
C
F
F
Expected
to
be
narrating
the
process
of
a
yarn,
install
and
build
during
that
time,
but
you
might
see
that
in
this
branch
with
configuration
management,
it
checks
the
env
file
and
then
logs.
Some
new
developments
in
our
environment,
variables
and
settings,
and
the
most
pertinent
ones,
are
that
some
of
the
names
have
changed
that
we
haven't
changed
these
names
for
random
reasons.
We
are
trying
to
approach
a
sort
of
a
namespace
type
of
thing.
Some
of
these
environment
variables
are
really
long.
F
F
C
F
It
and
say:
look
in
this
namespace,
it
will
produce
an
object
that
looks
like
this,
and
this
is
something
consistent
now
throughout
the
codebase,
something
that
you
can
use
and
we
haven't
changed
all
of
our
tooling-
to
pass
arguments
to
each
other
based
on
this
new
system.
But
the
basic
idea
behind
it
is
that
you,
you
always
want
to
use
the
outer
function
to
pass
configuration
to
an
inner
dependency,
and
what
that
means
is
the
entry
point
of
a
program
is
where
you
should
do
it.
F
C
F
Configuration
directly
from
a
nicely
formatted
configuration
object,
rather
than
have
everything
throughout
the
entire
project.
Continue
to
parse
environment
variables,
parsing
environment
variables
manually
throughout
the
project
is
nice
because
it
allows
for
broad
and
down
to
specific
configuration
of
individual
functions
from
outside,
but
it
is
very
annoying
to
do,
and
it
requires
that
you
remember
longer
and
longer
name
space
environment
variables.
So
what
we
do
here
is
we
use
the
automatic
name
spacing
feature.
F
Shion's
dev
server
image,
service,
custom,
origin,
it's
pulling
out
just
those
portions
and
then
name
spacing
them
so
that
dev
server
service
worker
enabled
becomes
service
with
camelcase.
So
all
of
these
are
now
specified
in
an
end.
Var
definitions,
JSON
file,
and
we
are
still
kind
of
working
on
where
this
thing
should
be
stored,
but
because
build
pack
and
PWA
studio
are
meant
to
be
building
a
kind
of
standard
environment
between
projects.
F
F
We
specify
the
actual
individual
variables.
We
add
defaults
if
they're
appropriate
and
we
even
add
type
signatures
and
the
configure
environments
function
will
read
from
here
use
defaults
and
then
read
from
your
dot,
env
file
and
then
read
from
the
local
environment.
And
then,
if
there
are
any
problems,
if
you
are
using
any
deprecated
environment
variables,
there's
a
section
down
here
or
specifying.
F
F
F
G
G
F
F
So
when
you
say
portable
like
when
you
generate
bundles,
then
they
are
going
to
use
and
hard-code
environment
variables
from
your
environment.
So
we
try
not
doing
that
very
much,
because
that
each
time
you
do
that
it
makes
the
bundle
like
less
portable.
But
it's
something
that
you
sometimes
want
to
do
so
that
the
open
bug
is
what's
the
bug.
F
Our
expectation
needs
to
be
actually
clarified
on
that
point.
I
think
yeah.
So
inside
the
JavaScript
bundle
environment
variables
are
all
available
at
Build
time.
So
if
you
site
an
environment
variable
inside
a
build,
the
way
that
we
do
in
braintree
and
I
guess,
I
can
show
that
off
here
hold
on
minutes.
F
So
you
can
see
that
in
here
we
actually
refer
to
process
and
check-out
braintree
token.
So
we've
changed
this
variable
name.
Can
you
guys
hear
me
all
right
or
if
I
started
to
cut
out
because
I'm
screen
sharing
it
so
far?
Okay,
good
so
process
and
check
out
Braintree
token
will
actually
get
turned
into
the
Braintree
token.
So
it
defines
this
thing
as
literal
on
this
line
then
passes
it
on
later.
C
F
Build
time
which
means
that
the
bundle
that
is
built
should
be
only
usable
for
a
specific
deployment
of
magenta,
which
is
what
we
expect
if
we
want
to
reuse
bundles,
and
we
do
eventually
want
to
reuse
bundles
between
deployments
and
systems.
So
we
may
end
up
taking
things
like
this
and
embedding
them
in
some
other
way,
maybe
spitting
them
out
of
some
dynamic
system
or
embedding
them
as
JSON.
F
F
F
F
H
J
E
G
F
I'd
love
I'd
love
to,
but
since
I'm
calling
in
I
I'm
not
going
to
be
able
to
share
so
if
somebody
wants
to
share
screen
and
just
go
to
the
layered
navigation
PR,
all
that
I
wanted
to
do
was
say:
hey,
there's
a
layered
navigation,
PR
open.
We
had
one
of
our
most
beloved
community
members
working
on
it,
but
he
is
he's
gone
to
pursue
another
opportunity
and
we
don't
imagine
that
he's
gonna
have
a
lot
of
free
time
to
work
on
this.
F
So
I
wanted
to
put
it
to
the
assembled
that
we
would
love
to
have
some
help,
and
if
anyone
has
some
interest
in
or
already
some
familiarity
with
this,
this
work
to
add
filtering
faceting
to
category
displays
and
search
result
displays
lists
of
products.
Then
we
would
be
really
appreciative
and
you
get
a
lots
of
points
if
you
took
it
on
and
if
you're
not
familiar
with
it,
then
you
can
talk
to
me
in
slack
or
otherwise,
and
we
can
bring
you
up
to
speed.
F
Was
all
I
just
I
didn't
want
to
talk
about
it
in
great
detail?
I
want
you
to
know
that
the
grass
to
L
team
approached
us
yesterday
to
ask
if
what
the
graph
QL
API
currently
provides
is
enough
to
do
a
rich
interface,
and
it's
not.
Can
we
make
suggestions
to
them?
So
we
can
do
that.
We
can
open
issues
both
from
our
community
and
from
our
core
team
against
the
graph
QL
CD
repository,
and
so
we
can
work
with
the
graph
QL
team
to
get
to
an
amazing
faceting
experience.
F
F
C
D
D
C
L
L
B
L
So
you've,
probably
all
seen
our
custom
project
before
and
we've
been
building
a
better
mobile,
the
new.
Then
we
had
before
so
or
a
desktop.
It's
just
really
simple:
we
call
it
the
mega
menu,
but
for
mobile,
it's
pretty
to
probably
start
the
I've
still
loaded.
So
it's
basically
the
same
trawler,
but
we
did
some
meaningful
transitions
in
it.
L
C
C
L
F
Okay,
thanks
again
for
your
attendance
in
for
your
patience
with
my
connectivity
problems,
and
this
will
be
the
last
time
I
try
to
run
one
of
these
remotely
I'll
just
come
to
the
office
bit
earlier
and
change
the
plans
around
well.
Thank
you
very
much
and
I'll
see
all
you
guys
next
week
and
some
of
you
guys
in
the
next
hour,
yeah.