►
From YouTube: 2023-02-06 Plan and Manage design review
Description
We discuss hidden layers causing poor performance in Figma, and OKR feature in GitLab
A
A
So
I've
been
like
having
really
bad
like
lag
and
delay
with
interactions
and
like
I'll
click
on
something,
and
it
doesn't
do
anything
for
like
10
or
20
seconds,
and
then
it
loads
and
the
whole
content
works.
Fine
and
I
think
it's
something
to
do
with
a
bunch
of
probably
like
hidden
layers
is
what
I
was
reading
and
one
of
the
docs
that
somebody
had
shared
trying
to
find
an
example,
but
like
I,
have
tons
of
like
nested
stuff
and
like
I'm.
A
Basically,
the
question
is:
is
there
a
better
or
smarter
way
to
optimize
this
sort
of
stuff,
considering
the
complexities
I
have,
or
this
just
the
limitation
of
this
type
of
process
or
tool
that,
if
I'm
going
to
make
prototype,
that's
super
interactive?
It's
going
to
be
kind
of
laggy
or
buggy.
I
thought,
like
maybe
I,
could
try
and
improve
this
thing.
That
Dan
had
helped
me
with
early
on
in
the
process.
There's
like
five
versions
of
this
component,
or
this
widget
or
interface,
but
they're,
all
iterations
of
like
one
like
this.
B
A
B
I
overdid,
it
I
think
yeah.
It's
like
the
nest,
the
nested
stuff
that
that
tends
to
slow
it
down
right.
You
have,
like
you
know,
like
a
checkbox,
that's
interactive
in
a
row
that
that's
interactive
in
a
table.
That's
you
know,
that's
where
I've
I
found
at
least
it
gets
really
slow,
and
so
sometimes,
if
you
can
like
strip
some
of
those
down
to
like
a
single
thing
that
changes,
but
it
doesn't
necessarily
sound
like
that,
might
be
enough
to
get
you
what
you
want
well.
A
I
also
thought
maybe
I
could
just
reduce
the
interactivity
in
the
Prototype
and
just
use
static
screens
instead
of
having
to
load
a
component
but
like
that
doesn't
seem
logical.
But
to
me
that
seems
making
it
worse.
That's
where
I
started
and
I
had
the
same
problem
and
then
I
worked
with
Dan
to
try
and
componentize
a
lot
of
these
pieces
of
the
the
Prototype,
and
it
works
great.
It's
really
interactive
and
it
does
exactly
what
I
want,
but
I'm
still
having
problems
with
delay
and
lag.
A
In
the
Prototype
or
the
users,
so
that's
the
thing
is
like
I'm,
going
to
test
with
users
and
I've:
have
them
open
up
a
prototype
when
they
click
on
something
and
there's
like
nothing
happens
for
10
20
seconds
they're
like
that,
do
something
you
know.
It
basically
presents
a
problem
with
the
testing
experience
and.
A
B
D
A
B
As
I
said,
when
I've
run
into
this
before,
like
it's
I
would
say
it's
a
less
optimal
solution,
but
what
I've
done
is
to
sort
of
to
make
it
more
linear
where,
like
your
test
becomes
like
you
know,
set
the
namespace
to
this
option.
So
then
you
can
like
literally
just
make
one
page.
That's
like
the
namespace
has
changed
and
you
don't
have
any
of
the
other
things
interactive,
and
then
you
move
to
the
next
step,
which
is
like
a
new
you
know.
A
A
A
D
C
So
there's
it
was,
it
was
when
I
switched
from
visible,
like
show
me
show
me:
what's
there
to
show
me
wireframes
and
a
bunch
of
hidden
stuff
that
isn't
actually
there
showed
up
and
I've
I've
got
suspicion
that
it's
that
hidden
stuff,
interesting,
okay,
we'll
catch
him
now
before
it
before
it
changes.
But
these
columns
should
just
be
one
two.
Three
and.
B
B
A
A
C
B
C
Create
something
that
only
lives
in
this
similar
to
your
my
table
sale.
Maybe
you
make
a
new
component,
that's
just
three
things!
Next
to
each
other
or
whatever,
because
there's
a
lot
of
hidden
stuff
will
it?
Let
me
do
it
again
because
they're
just
going
to
go
straight
to
crash
now
straight
to
crash,
but.
D
B
Is
there
a
way
to
like
audit?
What's
using
memory
like
I've
hit
that,
like
once
and
I
I
know,
I
had
a
I
had
a
client
that
had
a
design
system
where
they
built
all
the
interactions
into
the
design
system
and
they
were
like
there
was
this
warnings
everywhere
that
was
like
detach
and
delete
stuff
like
do
not
hide,
it
will
crash.
So
this
is
amazing,
I
guess.
E
E
E
Let
me
open
it
first,
all
right
so
and
Nick
you
probably
have
better
knowledge
of
this
than
I
do,
but
basically,
right
now,
Leanne
is
working
to
open
child
objects
in
modals,
instead
of
so,
for
example,
right
now,
I'm
within
an
objective,
if
I
open
a
KR
within
this,
it
opens
in
a
full
page.
Basically
the
URL
changes.
E
B
I
mean
my
understanding
of
like
the
history
of
this
goes
back
to
epics
and
like
a
desire
to
basically
be
able
to
Traverse
an
epic
without
leaving
the
Epic
right,
because
today
what
you
would
do
is
you'd
open
the
Epic.
Let's
say
you
have
20
issues
on
the
Epic.
If
you
wanted
to
like
do
stuff
with
those
you're
gonna
sit
and
open
20
tabs,
and
then
people
are
like
this
sucks.
I
have
like
my
browser's
crashing,
because
I
have
all
my
git
lab
tabs
open.
E
Cool
so
yeah
we're
this
is
behind
the
future
flag
right
now,
but
we've
gotten
some
feedback
from
internal
users
that
are
like.
Why
is
this?
On
a
you
know,
an
immortal
they
want
the
full
page
and
but
the
experience
now,
like
Nick,
said
you
can
kind
of
navigate
through
the
model.
Nick
you've
created
a
lot
of
issues
here.
Thank
you
for
that
like
to
follow
up
on
some
of
the
experience
and
there's
some
kind
of
maybe
weird
things
where,
like
an
objective,
you
know
the
parent
is
open
on
top
of
the
parent.
E
There's
things
we
can
play
with
I
think,
but
the
media
asked
is
that
internal
users
are
like.
Oh,
how
do
I
open
this
in
you
know
in
a
larger
screen.
How
do
I
open
the
full
page?
You
can
still
just
you
know,
open
a
new
tab.
You
know
just
that
or
use
the
keyboard
and
that's
easy
enough,
but
they
are
wanting
to
do
it
from
the
middle
itself.
E
The
proposal
was
just
something
like
you
know:
have
the
have
kind
of
that
expand,
but
to
me,
and
maybe
I'm
just
overthinking,
but
to
me
this
doesn't
really
make
sense,
because
we
are
not
expanding
and
collapsing.
This
view.
It's
like
going
to
another
URL,
it's
another.
Like
the
full
page
view,
it
seems
a
little
different
to
me.
E
A
E
E
I
think
it's
similar,
but
the
comment
feel
of
your
expanding
and
collapsing
something,
whereas
this
and
again
maybe
I'm
just
being
too
into
this
semantics
like
this-
is
not
expanding
it
would.
It
would
basically
be
like
instead
of
the
let's
see,
I'll
show
you
'll
be
opening
up.
Oh
I
can't
do
it
from
there
so
navigate
the
user
to
this
full
view
of
the
child
care
instead
of
the
modal
view,
so
expand
I,
guess
I.
E
A
E
So
that's
what
we
do
now.
So,
if
I
click
on
the
child
object
now
it
just
navigates
to
that
tab,
object
and
that's
kind
of
what
I'm
thinking
here.
Maybe
we
could
still
allow
users
to
do
that,
but
kind
of
like
Nick
was
saying
the
feedback
was
that
the
context,
switching
writtenic,
that
allowing
a
user
to
just
kind
of
quickly
open
up
these
objects
and
navigate
them
from
the
parent
without
navigating
away
yeah
I,
know
maybe,
and
that's
kind
of
where
I'm
getting
Nick
I'm
like
do
I
need
to
kind
of
like
research.
A
I
do
know
that
there's
two
example
or
an
example
of
this.
From
my
experience
where,
if
you
click
on
the
white
space
in
the
container,
for
example,
it
says
child
KR
and
you
were
to
click
on
that
White
space.
It
would
pop
up
the
modal,
but
if
you
were
to
click
on
the
link
directly,
it
would
go
to
the
object
directly.
E
A
E
A
I
just
know
that,
as
like
as
a
user
I,
if
I
know,
the
interaction
exists
and
I
want
to
go
power
mode
and
I
was
like
hold
command
on
my
keyboard
and
then
just
click
on
the
title
of
every
object.
It
goes
straight
to
the
Thing
versus
if
I
just
want
to
browse
real
quick
I
can
click
on
the
white
space
and
then
the
model
pops
up.
Okay,
move
on
to
the
next
task,
whatever.
E
E
B
No,
like
normal
stuff,
you
might
expect
that
could
be
a
way
to
test
some
of
that,
but
yeah
I
mean
I.
Don't
I
think
what
what
we're
going
to
end
up
with
is
like
the
main
goal,
which
we're
not
there
yet
is
to
be
able
to
open
these
things
from,
let's
say
a
board
or
a
list
in
some
sort
of
modular
view
like
a
drawer.
We
just
got
like
customer
feedback,
I.
Think
on,
like
Friday
of
somebody
that
was
really
like.
E
That's
what
I
I
was
thinking
now,
like
maybe
the
rule
is
that
there's
certain
views.
I
know
Daniel
been
looking
at
the
different
views
too,
but
maybe
in
some
views
it
opens
in
a
modal
or
in
a
dialogue
of
some
type,
but
maybe
things
like
the
child
elements
it
it
could
default
to
just
open
in
full
page,
like
maybe
there's
different
needs,
different
jobs.
We've
done
across
these
different
views,
I'm
not
sure
that
this
would
always
be
the
case
that
they'd
want
to
model
here.
C
I
was
just
half
remembering
using
jira
and
what
the
experience
was
and
I'm
pretty
sure.
I've
got
vague
memories
of
just
copying,
the
URL
that
I'm
on
and
then
pasting
it
to
force
myself
into
a
page
view,
and
then,
when
I
saw
this
screenshot
thing
I
was
like.
Oh
maybe
there
was
a
click
on
the
breadcrumb
thing
as
well,
but
I
don't
really
remember.
So
it's
just
a
waste
of
comment
really
thanks
for
coming
yeah.
E
No,
no
you're
gonna
totally
right
like
it
has
like
a
link.
You
could
either
copy
or
if
you
click
on
the
object,
it's
a
little
confusing
because,
like
if
you
click
on
the
current
object,
it
would
navigate
to
the
full
View
and
jira
or
if
you
click
on
one
of
the
ancestors,
though
it
would
navigate
to
that
ancestor
within
the
modal
itself.
So
it's
a
little
mix
and
match,
but
it
makes
some
sense.
C
E
E
D
E
I
think
it's
a
yeah,
it's
a
good
thing
to
investigate
and
I
think
my
PM
is
pushing
to
investigate
this
and
then
some
are
pushing
for
just.
It
makes
sense
that
model,
but
I
think
it's
just
what
they
want
to
care
about.
That's
what
they
care
about.
They
want
to
edit
it
in
kind
of
a
focus
view
which
is,
or
maybe
it's
just
what
they're
used
to
it's.
Something
we'll
we'll
have
to
ask
people
but
I
think
that
and
there's
some
kind
of
interactions.
E
Maybe
what
we
need
to
kind
of
iron
out
within
these
models
that
maybe
are
kind
of
throwing
people
off
and
Nick
you've
already
created
a
lot
of
issues
around
these,
where
it's
just
not
quite
seamless
yet,
and
maybe
that's
also
frustrating
and
they
just
kind
of
want
to
get
back
to
their
their
prior
view.
But.
D
B
For
now,
yeah
because
I
mean
there's
no
way
to
create
like
children
on
tasks,
so
you'd
never
have
a
comparable
sort
of
experience.
There
I
don't
know
Dan
on
I
mean
requirements
can
be
nested,
so
that
would
be
probably
the
next
place
we
will
run
into
this.
I
would
guess
I
think
the
the
use
case
for
the
full
page
is
probably
like
the
one
I
have
heard
is
just
because
people
want
to
have
multiple
tabs
open
and
be
able
to,
like.
You
know,
pull
up,
maybe
like
this
KR.
B
E
E
E
So
let's
say
we're
on
a
key
result:
I
want
to
change
this
to
a
objective
which
is
larger
basically,
so
it's
kind
of
that
promotion
action,
but
we
are
starting
with
from
scratch
with
the
work
items,
so
we
can
improve
that
promotion,
experience
and
I
think,
like
the
general
idea,
is
like
instead
of
closing
the
object.
So,
for
example,
now,
if
you
have
an
issue,
you
promote
it
to
an
epic,
it
just
closes
the
issue
and
still
kind
of
keeps
it
attached
to
the
Epic.
E
E
But
the
idea
of
picking
and
kind
of
like
changing
this
to
this
key
result
to
a
objective,
is
kind
of
the
small
thing
we're
trying
to
solve
now
and
I.
Think,
like
the
most
MVP
kind
of
like
what
we
have
now
would
be
adding
another
field,
so
you
say
type
objective,
key
result
which
I
what
I
don't
like
about
this
is
like
we're
taking
up
more
space
within
these
fields
that
we're
already
getting
feedback
about,
and
there's
only
two
types
to
choose
from
here,
so
really
you're.
E
Seem
like
a
drop
down
would
be
necessary.
I
played
around
with
kind
of
putting
this
within
the
bread
cones
as
well
so
you're
picking
here
what
the
type
is
I
think
what
would
be
the
smallest
change
here
would
be
either
promoting
to
objective
or
to
say
like
change
to
objectives,
maybe
within
the
more
the
overflow
menu
and
change
there,
I
don't
know
what
do
you
all
think
about
that
if
you
wanted
to
just
promote
or
change
this
item
to
its
parent
item.
C
C
E
Yeah
yeah
I,
don't
know
exactly
where
we
are
calling
out
just
okrs.
That's
why
I
want
to
keep
it
as
small
as
possible.
I
think
it
is.
We
need
to
think
to
the
larger
experience
of
like
okay,
that's.
This
is
why
I
tried
to
scope
it
down
just
to
promotion,
because
I
think
like
that
demotion,
has
a
lot
of
larger
questions,
especially
like
what?
If
what
if
this
thing
has
children,
what
happens
to
them?
E
How
do
you
manage
children
when
you're
moving
you're
changing
to
an
object
that
you
can't
have
children
or
the
attributes
are
different?
We
are
preserving
all
of
the
fields
attributes
like
any
data
entered
so
like
when
the
user
transforms
from
one
object
to
the
other.
It
preserves
it
so,
like
you
can
kind
of
roll
back
that
change
more
easily,
but
yeah
I
do
agree.
E
It's
a
larger
discussion,
I
think
there's
an
issue
around
it
for
for
an
epic
around
it
and
project
management,
Nick
or
I'm
changing
the
type
of
work
items.
So
for
this
one,
because
it's
an
ask
for
this
okr
the
seg
project
for
internal
users,
it's
more
around
like
oops
I
created
a
KR
I
meant
it
to
be
an
objective
I
want
it
to
be
a
parent.
E
So
that's
why
I've
been
working
to
scope
it
down
really
small
and
I
was
thinking
that
like
something
like
this
themes,
a
decent
NBC,
while
we
think,
through
the
larger
experience
of
like
just
changing
it,
to
any
type,
both
up
or
down
the
hierarchy.
This
would
be
just
one
level
up
hierarchy.
The
kras
can't
have
children.
If
you
have
anything
and
then
we
need
to
definitely
spend
more
time.
Thinking
through
that
larger
experience
of
tasks
to
Epic-
or
you
know,
epic,
to
task
which
would
be
more
difficult.
C
B
E
E
B
But
well
sometimes
you'll
have
to
delete
the
data
because
it'll
be
like
in
Conflict,
but
if
we
don't
have
to
be
great,
if
we
didn't
yeah,
that's
only
like
consideration
like
that's
that's
my
concern
with
like
the
field
or
one
of
my
concerns,
other
being
the
space
that
takes
up.
Is
that
like,
if
we
like
in
every
other
case
other
than
this,
we're
probably
gonna,
have
to
prompt
the
user
to
be
like
hey,
like
this
isn't
going
to
migrate
or
like
you
might
lose
something
like
confirm,
you're
gonna,
you
wanna!
E
B
Run
into
this,
occasionally
where
it's
like
this
isn't
available
here,
like
what
do
you
want
to
do
with
it
anytime?
You
might
be
destroying
something,
and
it
doesn't
really
feel
like
obvious
that,
like
how
you
would
handle
that
there,
whereas,
like
the
more
like
top
level
actions,
it
seems
a
bit
more
clear
that
you
could
be
like.
Oh
there's
a
prompt
and
then
things
below
are
gonna
change
and
like
it's,
it's
sort
of
restructuring
the
the
thing
you're
looking
at.
E
Mm-Hmm
yeah,
it's
I
always
have
more
weight
to
it
and
then
yeah
you're
gonna
be
like
reconfiguring
all
the
children,
or
maybe
things
that
are
disabled
for
MVC,
but
yeah
there's
a
lot
more
thinking
there.
So
what
I'm
hearing
is
to
keep
this
very
small,
which
I
think
maybe
something
like
this.
This
is
almost
copy
paste
right,
just
like
just
make
it
bigger
for
now,
and
we
have
this
pattern
with
promotion
or
Dan
I
can
get
more
snarky.
Perhaps.
E
C
C
I
think
I
think
it
was
all
tied
up
tied
up
in
that
my
my
two
pence
on
it
all
is
I
think
we
could
probably
have
some
good
opinions
for
what
the
default
behavior
is
and
we
shouldn't
let
that
stop
us
from
doing
doing
the
stuff.
Like
I'm
sure
we
could
say
for
certain
types:
hey
just
make
the
children
a
child
of
whatever
this
thing's
parent
was,
if
possible,
and
opinions
like
that,
I
think
would
be
a
would
be
okay
to
have
yeah
something,
like
I'm
sure
was
in
this
year.
E
B
You
I
do
think.
Alexa
is
the
menu
item
that
I
think
you're
kind
of
like.
Maybe
this
is
the
easiest.
One
is
probably
the
most
extensible
which
I
kind
of
like
like
because
right
now
like
you
could
basically
make
that
change
to
objective
right
now
and
then
it
would
be
like
a
really
nice
transition
to
make
that
change
type.
B
And
then
you
just
select,
you
know
a
type
or
something
like
that,
and
then
we
can
prompt
like
we
can
do
any
of
the
messaging
or
prompting
and,
like
whatever
follows
from
there
I'd
map,
like
the
jira
style
one's
fine
too.
Although
I
find
that,
like
a
fairly
hard
to
discover
pattern,
because
you
noted.
E
E
B
Yeah
change
to
objective
and
then
like
later
we
could
make
that
change
type
and
then
you
know
you
could
pick
whatever
types
are
available.
We
can
explain
why
certain
types
are
not
available.
If
we
need
to
like
you
know
we
can,
we
can
do
that
kind
of
in
like
a
modal
or
something
where
it's
a
bit
more
focused.
E
B
Because
we're
like
one
of
the
scenarios
that
we're
gonna
have
for
that
is,
like,
let's
say,
you're
changing
like
let's
say
we
allow
you
to
change
like
an
objective
to
an
epic
or
something
so
like,
then,
what
do
you
do
with
all
the
KRS
that
hang
off
of
it
like?
We
might
want
to
prompt
them
to
like
keep
it
as
KRS
or
like
convert
them
to
issues
like
there's
a
whole
workflow.
That's
going
to
follow
for
some
of
that.
That
I
think
is
is
going
to
require
a
bit
more
stuff
to
manage.
A
D
Yeah
as
I
was
thinking
about
it
as
well
too
I
feel
like
I,
like
Nick's,
approached
me
to
think
about
thinking
about
it
more
as
like
the
type
we
changed,
the
type
versus
you
know
the
way
it
is
now
but
yeah,
it
just
seems
like.
Sometimes
when
you
know
you're
trying
to
make
a
product,
you
know
more
generic
and
more
flexible,
it
just
seems
confusing,
like
okay.
D
Why
can
I
do
this
in
in
a
task
versus
like
relationship
versus
like
a
KR
versus
okr
versus
like
task
subtask
or,
like
you
know,
stuff
like
that,
so
I
think
maybe
doing
it.
You
know
calling
it
a
type.
That's
changing
is
making
me
makes
more
sense,
but
yeah.
It
could
get
confusing.
Yeah
like
you
when
you
try
to
build
like
a
really
complicated
Swiss
army
knife
you're
like
okay,
which
which
one
should
I
use
for
what
now
you
know
which
knife
so.
E
A
Got
my
feedback.
One
thing
that
I
was
going
to
point
out
is
that
I
think
there
needs
to
be
a
strong
definition
of
what
the
parent
and
child
relationship
is
and
the
idea
of
going
from
okr
to
KR
vice
versa.
That
needs
to
be
strongly
defined
and
I.
Think
once
you
have
that
a
lot
of
this
kind
of
is
easily
it's
more
easily
solved
or
less
less.
Second
guessing
so
to
speak
right,
but
that's
up
for
I
guess.
Perhaps
a
study
or
usability.