►
From YouTube: 14.5 Create:Code Review prioritization exercise
A
B
C
A
C
B
B
Back,
I
don't
think
we
had
any
progress
since
we
last
chatted
about
this
five
months
ago.
The
only
difference
like
we
have
some
customer
activity
right,
so
we
took
some
stabs
at
this.
This
was
far
more
complex
than
we
expected.
I
think
right
now.
What
needs
to
be
done
here
is
just
we
go
at
it
again.
We
have
some
merge
requests
already
starting
starting
work,
but
we
didn't
merge
if
I'm
not
mistaken,
we
closed.
C
B
It
was
kind
of
a
dead
end
yeah,
it's
called
it's
not
merged,
and
I
think
that's
where
we
we
should
pick
up
from
is
like
look
at
the
code,
get
ourselves
acquainted
with
the
problem
again
and
I
think
at
the
time
the
problem
was
dealing
with
all
the
versioning
and
making
sure
that
we
had
the
com
the
right
comparison
to
show
off
the
lines,
but
that's
always
computed
on
the
back
end.
So
yes,
this
would
be
requiring
some
back-end
support
so
yeah.
A
Okay
yeah
anyway,
I
think
we're
all
aware
of
the
reach
impact
and
confidence
that
we
have
here.
Although
the
confidence
on
the
solution
technically
the
technical
part
of
the
solution,
I
assume
it's
not
very
high
because
we
encountered
a
lot
of
problems
while
trying
to
fix
this
but
yeah.
If
anyone
has
questions
about
this
issue,
let
me
know-
and
you
can
interrupt
next
issue-
is
use
viewed,
state
from
file
header
in
mrs
file
browser
so
yeah
right
now.
We
bold
all
of
the
file
names
in
the
file
browser
and
merge
requests.
A
So
yeah.
I
think
it's
not
a
lot
of
reach
and
I
don't
think
it's.
It's
a
very
big
impact.
I
think
it's
more
of
a
aesthetic
thing
than
anything.
It
helps
when
viewing
a
large
merge
request.
I
don't
think
I
don't
think
it's
a
huge
impact
so
anyway
yeah
I
wanted
to
know
if
you
have
any
questions
or
thoughts
about
this
one
front
end
only.
C
Had
we
previously
when
we
previously
talked
about
this,
I
think
we
talked
about
blocking
it
on
an
api
right
like
because
the
the
next
step
was
to
add
the
graphql
api
to
make
the
viewed
state
sort
of
persistent
and
give
us
a
way
to
use
it
in
multiple
places.
I
think
my
I
think
this
is
fine.
I
think
the
logical
question
is
going
to
be
like
if
I
click
on
a
file
on
the
file
tree,
is
it
gonna
click?
The
check
box
in
the
ui
like?
B
B
Whenever
we
update
the
code
from
now
relying
on
the
local
client-side
storage
to
rely
on
that,
you
will
be
focused
on
the
checkbox,
so
we
don't
at
that
time.
We
won't
even
have
to
worry
about
this,
because
that
will
be
looking
at
that
checkbox
state
and
as
long
as
we
update
the
code
that
updates
that
to
come
from
the
end
and
not
from
the
client-side
store
and
just
change
the
store.
Basically,
this
will
automatically
work
as
expected.
B
A
A
A
All
right
this
one,
I
think
this
one
we
can
kind
of
skip,
because
it
will
require
back-end
work,
but
anyway,
this
is
I'm
passionate
about
this
issue.
Well,
not
sold
on
the
solution
itself,
but
just
making
some
progress
here
on
having
less
and
less
auto
collapsed
files,
so
yeah,
maybe
in
the
future,
we'll
get
to
it
this
one.
I
think
at
some
point
yeah.
C
A
A
When,
for
example,
you
go
through
the
flow
of
editing
a
file
in
a
merge
request,
then
you
go
back
to
the
merge
request
immediately
after
you
submit
changes
in
the
single
file
editor,
for
example,
and
the
commits
tab
has
the
wrong
counter
number
for
the
number
of
commits.
It
still
says
one
for
example,
in
this
case,
there's
only
one
commit
in
this
image.
B
B
The
solution,
I
think
I
probably
proposed,
is
then
to
when
rendering
the
commits
list
being
given
an
information
about
how
many
commits
are
there
and
then
updating
the
counter
there
to
always
keep
it
in
sync,
I'm
not
sure.
If
we
have
that
information
available,
I
think
we
don't
so
we'd
have
to
add
that,
but
I
feel
like
that's
the
one
of
the
cases
where
we
can
get
by
updating
it
back
in
ourselves
too.
If
we
need
to
yeah
that's
kind
of
it.
A
A
A
So
let
me
remind
you
like,
in
this
case
nick
changed
the
line
of
the
diff
based
on
remy's
feedback,
and
today
you
have
to
click
on
this
link
to
go
to
the
specific
version
where,
where
the
change
was
made
and
this
reloads
the
whole
page
right,
it
re-fetches
everything.
It's
a
refresh,
completely
refreshes
the
page,
and
this
issue
is
to
avoid
doing
that.
A
So
you
would
just
go
to
the
div
tab
and
the
diffs
update
in
place
instead
of
reloading
the
entire
page,
and
this
would
have
the
benefit
of
what
we
recently
did
with
the
tabs.
I
don't
know
if
you
any
of
you
have
noticed
that
we
keep
the
location
of
where
you
were
before.
So
if
you
go
to
the
div
tabs
to
see
sorry
to
the
changes
tab
to
see
hey,
this
is
what
nick
changed.
You
then
go
back
to
the
overview,
tab
and
you're
scroll
down
to
this
thread.
B
B
Whenever
we
want
to
switch,
you
tear
down
the
app
build
again,
the
new
state-
if
you
do
this
a
couple
of
times
with
like
memory
usage,
might
be
creeping
in
some
memory
leaks
in
there,
so
the
performance
will
suffer.
Largemrs
we've
seen
this
in
the
past,
where
tearing
down
a
large,
mr,
a
large
state.
Sorry,
a
large
mr
from
the
state
will
give
browser
a
hell
of
a
lot
of
work
so
and
then
vue
is
not
particularly
performant
at
tearing
down
components.
B
So
that
means
that
you
have
to
wait
for
everything
to
be
turned
down
and
be
updated
with
the
new
state.
So,
in
terms
of
time,
it's
usually
more
effective
to
just
reload
the
page,
because
you
get
a
blank
slate,
you
start
from
scratch
might
be
a
cop-out,
but
we
can
look
into
it
and
write
a
proof
of
concept
to
see
how
how
bad
that
would
look
like.
B
B
A
B
C
Mean
like,
but
we
tried
that
with
like
in
the
context
of
the
settings
drop
down
in
the
top
right.
We
yeah,
like
only
if
you
switch
from
like
side
by
side
to
inline,
like
we
tried
to
which
I
guess
we
don't
sort
of
just
naturally
redo
the
page,
I'm
trying
to
think
like.
I
thought
we
had
the
same
sort
of
thing
where,
like
below
the
tabs.
C
Basically
we
talked
about
reloading
just
that
section
which
is
similar
to
what
this
is
asking
like
below
the
tabs
reload
just
that
section,
as
opposed
to
like,
if
you
click
the
latest,
if
you
click
the
drop
down
thing,
then
we
reload
almost
everything,
but
the
sidebars
are
so
fast.
Like
I
don't
know,
I
wonder
like
what
the
perceived
performance
gained.
This
feels
like
perceived
performance,
game,
great
pedro.
C
A
I
thought
that
for
that
drop
down
on
the
top
right
corner
of
the
changestep,
the
preferences
dropdown
for
all
of
the
modes
except
for
white
space
changes,
we
would
just
re
change
how
yeah
things
were
drawn
in
the
by
the
browser
and
with
white
space
changes.
We
would
have
to
re-fetch
all
of
the
diffs.
A
A
A
C
D
D
A
B
A
A
A
A
B
A
D
It's
yeah.
It's
definitely
weird
timing
right
now,
so
I
think
that's
a
big
part
of
it,
so
I
think
we
should
revisit
it
once
these
once
we
have
a
little
more
once
back
and
at
least
has
a
little
more
time
to
focus
on
it.
Yeah
yeah,
I'm
not
sure,
we'll,
have
to
see
at
that
point.
We
can
decide
whether
or
not
we
want
to
keep
going,
but
I'd
like
to
at
least
keep
trying
it
a
little
bit
after
when
things
theory
are
back
to
normal
a
little
bit.
Hopefully.
C
B
So,
okay,
on
that
note,
we
are
looking
at
the
front-end
level.
We
are
looking
at
those
global.
I
am
having
a
discussion
with
the
team
right
now,
then
we'll
share
it
with
the
other
ems
and
we'll
make
a
decision,
a
joint
decision
so
that
we're
not
like
each
group
picking
a
very
random
topic
and
we're
trying
to
rally
around
the
same
ones.