►
From YouTube: Plan | Weekly Sync 2022-08-03 (Part 1)
Description
B
Sure
so
I'll
share
my
screen
and
demonstrate
a
recent
change
that
I've
made.
So
let
me
zoom
in
so
we
are
basically
adding
the
ability
to
mark
a
task
list
item
as
confidential.
B
B
I
I
do
have
a
couple
of
questions
around
it
which
I'll
ask
later
so
once
you
turn
on
the
confidentiality
and
close
the
model,
we
should
also
be
updating
this
list
to
reflect
it,
which
would
possibly
involve
updating
our
banzai
filters
so
that
we
can
add
a
relevant
markup
to
show
the
icon
within
this
list.
That
a
taskless
item
is
confidential.
B
B
For
now,
I've
added
the
loading
animation
and
I've
placed
it
right
next
to
this
drop
down,
but
I'm
not
sure
whether
that's
the
correct
way
to
go
or
do
we
need
to
show
the
loading
animation
or
some
progress
that
the
confidentiality
toggle
action
is
in
progress
anywhere
in
ui.
So
right
now
it
shows
right.
Next
to
this
drop
down,
I
feel
it
works,
but
obviously
you
will
need
confirmation
from
ux
so
yeah
that
is
it
for
the
demo.
Any
questions.
C
B
B
C
B
Yeah,
so,
even
if
we,
even
if
we
call
the
mutation
from
the
work
item,
detail
component,
which
is
the
root
component
here-
that
should
still
allow
us
to
identify
that
something
was
changed
with
regards
to
work
item
and
we
should
be
able
to
update
the
description,
yep
and
yeah
I'll
stop
the
screenshare
and
next
is
natalya.
Sign
is
widget.
A
B
Element
for
gi
yeah,
so
there's
so
there
are
two
issues
regarding
the
confidentiality
one
issue
is
showing
the
confidential
status
on
ui
and
the
other
one
is
obviously
adding
support
for
changing
the
confidentiality
status.
Now,
as
far
as
showing
the
status
is
concerned,
it
currently
just
covers
model
dialogue,
but
there's
an
ongoing
conversation
like
how
we
are
going
to
show
the
same
for
the
listing
part,
and
if
we
were
to
do
it
within
the
same
milestone,
then
we
would
need
to
update
the
issue
description
to
reflect
the
confidentiality
of
the
item.
D
We
don't
need
to
do
it
during
the
same
milestone
because
we're
removing
the
listing
of
tasks
from
the
issue
description
anyway,
so
yeah,
so
there's
not
gonna,
be
a
need
to
update
the
description,
at
least
for
fifteen
three.
Okay,
then.
B
E
I
think
we
do
need
to
update
the
child
items
section,
though,
ideally
in
real
time
after
you
switch
it.
So
it's
not
confusing.
E
D
A
E
B
Yeah,
so
the
thing
is
that
I
I
don't
know
whether
this
site
these
items,
I
think
yeah.
These
items
are
tied
to
this
one,
but
there's
no
way
to
update
the
confidentiality
from
this
list.
It
only
happens
from
this
model
so
yeah.
We.
We
can
in
fact
update
this
list
to
show
the
icon
reflecting
confidentiality,
and
that
should
be
possible
because
this
list
is
I'm
not
sure
whether
it
is
using
subscription
or
not
but
yeah
when
the
model
is
dismissed
after
confidentiality
is
updated.
We
can
certainly
refresh
the
system
and
do
it.
B
Yeah-
and
so
I
I
haven't-
had
much
chance
working
on
the
child
items,
but
when
I
looked
at
all
the
feature
flags,
I
got
confused
in
getting
this
model
dialogue
to
show
up,
because
I
had
to
first
enable
work
items.
Then
I
had
to
enable
work
items
mvc2.
B
Yeah-
and
I
was
I
was
looking
at
our
documentation
around
the
task
list,
which
did
reflect
like
what
to
expect
when
you
hover
this
and
turning
a
task
list
item
into
a
work
item.
But
then
the
feature
flag
mentioned
in
the
documentation
says
just
enable
work
items
and
it
should
work.
But
that
wasn't
the
case
so
within
this,
mr
I'll,
be
sure
to
update
the
documentation
so
that
I
basically
in
in
future,
I
don't
forget
like
which,
which
which
flags
are
responsible
for
what.
A
Yeah
I
copied
in
the
issue
there
which
floor
is
working
on
to
open
the
model
when
you
click
on
one
of
the
items
in
the
widget
she
mentioned.
So
it's
related
to
my
next
point
is
that
she
mentioned
this
morning
that
she
has
to
re-implement
the
modal
for
this,
I'm
not
sure
like
of
the
ins
and
outs
of
it,
but
it
may
be
neither
like,
since
you've
merged
this
there's.
She
also
has
to
include
your
newest
work.
B
A
That's
great
thanks,
yeah,
like
it's
just
that,
if
I
understand
it
correctly,
she
has
to
implement
re-implement
the
modal
entirely
for
the
widget.
Is
that
does
that
sound
right,
or
does
that
sound
like
I've
got
this
horribly
wrong.
D
I
I
think
it's
about
when
the
modal
is
or
which
modal
we're
using,
because
I
think
we
attached
the
like
modal
container,
essentially
within
the
issue
description
originally,
but
when
we
move
that
behind
the
feature
flag,
we
no
longer
have
access
to
that
modal
from
outside
of
the
issue.
Description-
maybe
way
off
here.
So
I
I
I.
B
Think
this
may
be
around
what
heinrich
mentioned
earlier,
like
in
the
in
the
child
items
task
list
clicking
on
individual
items
doesn't
open
the
model,
so
I
think
maybe
flori
was
referring
to
that
behavior,
where
clicking
the
child
items
from
the
task
list
should
be
able
to
launch
the
model
and
show
the
relevant
information,
because
right
now.
D
E
Yeah,
so
I
looked
at
her
mr
this
morning
and
it
looks
like
she's
be
using
the
model,
that's
that
we
used
for
when
you
click
on
the
task
item
in
the
description,
and
so
I'm
just
thinking
of
yeah.
What
john
might
have
been
by
harvey
doing
the
model.
It
doesn't
seem
to
be
the
case
so
far,
but
yeah,
I'm
not
sure.
I
don't
have
other
context
around
in
her
struggles
because
she
mentioned
that
she
had
to
spend
time
redoing
some
stuff
and
but
yeah.
I
guess
we
could
get
clarification
from
flori.
B
Yes,
I'll
get
in
touch
with
her
tomorrow
and
basically
sync
on
how
we
should
move
forward
with
regards
to,
because
my
changes
are
also
touching
model
and
I
may
end
up
causing
more
conflicts
for
her
when
she
ends
up
refactoring.
The
models.
D
One
last
question
on
that:
so
for
linked
items
or
linked
issues.
What
what
model
were
we
using
there?
Were
we
creating
a
new
modal
from
within
that
widget
or
where
we
were
using
the
description.
B
So
we
we,
we
have
a
common
gl
model
component,
which
we
just
show
depending
on
when
user
clicks.
So,
given
that
it's
a
description
body
which
is
not
a
view
application,
then
we
are
basically
binding
event
listeners
to
those
items
from
the
description
and
that
basically
propagates
into
the
work
items
have
that
we
have,
and
that
ends
up
launching
the
model
and
shows
the
information
by
fetching
graphql
first
so
right
now,
obviously
we
are
having
a
communication
between
a
non-view
part
of
the
page
and
the
view
part
of
the
page.
E
D
So
john,
we
probably
need
to
and
cushion
sink
back
with,
flory
to
get
a
bit
more
details
on.
A
C
Okay,
chinese,
so
this
branch
actually
uses
the
latest
changes
on
both
backhand
and
front
end.
It's
based
on
heinrich's
branch,
where
he
adds
the
input
permutation
and
I've
added
front-end
changes
here
as
well.
There
are
a
few
things
that
probably
are
not
perfect,
but
let's
look
at
the
ui
right
now,
so
assignments
have
no
login
state.
Whenever
you
just
change
the
assignments
list,
mutation
is
happening
on
the
background
right
now
and
we
don't
show
any
kind
of
disabling
the
input
or
anything.
C
You
may
ask
like
what
happens
if
I
do
a
few
mutations
very
fast.
So,
for
example,
if
I
remove
this
one
and
then
remove
this
one,
there
will
be
probably
okay.
My
mutations
are
super
fast,
but
normally
you
would
see
the
old
result
for
a
second
and
then
a
new
result,
but
there
is
no
colliding
in
terms
of
problems.
Issues
with
back
end
anything.
You
just
see
the
result
of
the
first
mutation
and
then
a
result
of
the
second
mutation.
C
It
has
a
quite
nice,
I
would
say
interface
when
you
work
with
keyboard,
so
you
just
type
anything
and
then
you
can
remove
anything
from
the
keyboard
too.
That's
a
good
difference
with
our
current
drop
down
and
the
only
problem
that
I
see
from
the
ui
point
of
view
is
that
we
have
this
clear,
all
button
and
it's
visible
on
all
the
states,
because
this
is
gl
token
selector.
C
So
maybe
I
will
just
remove
it
and
show
it
only
when
we
are
editing
so
in
this
state
whenever
the
input
is
focused
but
rest
should
be
pretty
much
the
same.
Yeah
sign
myself
works
too
just
works,
and
I
really
like
their
ux
here.
We've
been
working
with
all
assignees
a
lot,
and
this
one
is
feels
much
better
faster
with
a
few
visual
glitches.
Maybe
but
we're
working
on
that.
E
Now
I
have
a
question
on
the
okay,
so
you're
saying
the
front
end
like
you're
doing
the
mutation
on
the
back
and
right
so
question
is:
do
you
try
to
reconcile
the
result
of
the
mutation
to
the
ones
that
are
currently
shown,
or
do
you
just
ignore
the
result
of
the
mutation.
C
The
the
result
is
not
ignored
whenever
we
have
their
response.
Actually,
we
update
the
assignees
again.
So
first
we
assign
they
have
the
specials.
C
It
is
a
copy
called
local
assignees,
but
whenever
we
have
the
response,
the
result
is
updated
all
the
time
and
when,
if
mutation
errors,
for
example,
we
just
remove
all
the
changes
we
made
locally
and
we
roll
back
to
the
previous
assignments
that
we
have
so,
for
example,
I
have
non-assignees
and
I
assign
myself,
we
see
ourselves
in
the
sinus
build,
but
if
mutation
fails
it's
removed
and
we
have
no
assignments
again,
so
we
synchronize
the
state
with
a
back
end.
All
the
time.
E
Yeah,
I
ask
because
sometimes
when
you,
for
example,
send
a
mutation
to
assign
user
one
and
user
two,
the
back
end
could
reply
that
only
user
one
was
assigned
because
we
checked
for
permissions
and
other
stuff,
like
only
users
that
can
read
the
issue,
can
read,
can
be
assigned
or
something
so
it
might
like
partially
fail
and
only
some
users
are
assigned
or
something
but
yeah.
I
think
the
other
thing
with
this
reconciling,
I
think,
is
like
do
you?
E
C
C
C
D
Yeah,
so
my
question
was
was
exactly
what
heinrich
was
wondering
around
the
order
like
what
happens
if
we
send
four
delete
mutations
quickly
and
the
first
one
errors
out,
do
we
like?
How
do
we
yeah?
How
do
we
handle
that
first,
one
failing
and
then
the
following
three
succeeding?
C
We
send
them
to
science
but
assigning
ideas.
First
of
all,
if
mutation
fails
that
there
will
be
a
roll
back
to
the
previous
state
like
before
the
mutation,
but
the
following
mutations
will
be
resolved
and
we
will
update
the
sinus
list
according
to
the
mutation
response.
F
I'll
write
this
in
the
doc
in
a
second
one
of
the
things
I've
I've
been
thinking
about.
As
I
was
looking
at
the
new
widgets
is
we
probably
need
to
define
like
a
standard
sort
of
pre-interactive
state
that
for
for
any
widget
right
and
then,
which
could
be
basically
that
everything
starts
as
like
a
like
a
button
or
something
that,
like
immediately
on
interaction
like
switches
to
that
active
state
like?
Would
that
would
that
help
resolve?
Some
of
the
like,
you
know
like.
F
I
know
right
now,
you're
hiding
a
lot
of
stuff
and
then
showing
a
lot
of
stuff.
Would
it
be?
Does
that
seem
like
a
reasonable
approach
to
instead
of
hiding
and
showing
we
have
like
a
default
state?
That's
maybe
more
of
like
a
structured
component
and
then
on,
like
focus
or
on
enter
like
switch
to
the
to
the
input.
C
It
will
bring
a
trade-offs.
Usually
when
you
replace
components
you
will
need
to
synchronize
the
state
between
the
component
and
the
input
both
and
focusing
the
state
and
especially
keyboard
interaction,
can
be
a
pain
if
we
have
different
components.
So,
for
example,
like
blur
and
focus
handling
this
with
replace
an
input
with
some
default
component
is
complicated.
So
for
now
I
would
probably
stick
with
just
hiding
the
clear
role
on
the
input,
but
we
can
revisit
it
later
if
it
becomes
too
complicated.
F
Okay,
in
the
other
consideration,
there's
going
to
be
consistency
with,
like
I
think
I
I
initially
noticed
this
with
like
due
dates,
because
due
dates
are
not
because
it
has
to
have
like
this
other
panel.
It's
not
strictly
like
an
input
that
we
can
utilize,
so
we
might
want
to
look
at
like
how
can
we
make
everything
sort
of
feel
the
same,
but
we
can
probably
take
different
approaches
to
getting
to
that
feeling
depending
on
the
type
of
widget,
if
needed.
C
Okay,
no
more
questions
on
the
assignments.
C
D
I'll
I'll
demo,
because
I
didn't
I
didn't
warn
or
anyone
or
asked
to
demo
so
I'll-
walk
through
kind
of
the
rest
of
the
recent
work
items
or
test
progress.
D
Okay
cool,
so
I
am
going
to
demo
from
production
within
our
plan
stage
group
on
the
test
project.
Within
that
plan
stage
group.
The
first
thing
to
show
is
the
listview
now
filters
by
tests
or
has
now
shows
tests
and
has
the
ability
to
filter
either
by
tests
or
by
not
tests,
so
it
works
shows
all
of
our
tasks.
There.
D
I'm
often
trying
to
find
out
just
issues,
because
I've
created
a
lot
of
tests
within
that
project
and
you
can
filter,
filter
out
tap
them.
You
can
still
do
the
oops.
Oh
no
hotkeys.
D
Should
show
everything
that
we
showed
before
this
change
all
right,
so
not
a
demo,
an
actual
work
item
I'll
create
another.
E
D
I'm
going
to
add
a
test
list
here,
just
because
of
where
we're
kind
of
at
with
feature
flags.
So
we
have
the
the
test
issue
here.
D
We
currently
still
have
the
ability,
within
gitlab,
org
and
gitlab.com,
to
create
a
task
from
the
issue,
description
that
is
behind
a
separate
feature
flag
now,
which
we're
going
to
turn
off
once
the
once
the
ability
to
create
a
child
from
within
the
hierarchy.
Widget
is
done
and
tested,
and
we
add
one
more
feature
to
that.
So
within
the
next
couple
days,
I'm
going
to
turn
off
this
feature
flag,
so
we
can't
create
it
from
from
the
issue.
Description
that
feature
flag
is
now
called
work.
D
Items
create
from
markdown
just
to
kind
of
level
set
where
all
of
our
are
relying
on.
Where
all
of
our
feature
flags
are
currently
have
the
work
items
feature
flag,
which
is
primarily
the
core
of
tas.
D
We
have
the
work
items
hierarchy
feature
flag,
which
is
child
items.
Here
we
now
have
the
work
items
create
from
markdown
future
flag,
which
is
the
issue
crate
from
markdown
that
I
was
talking
about,
and
then
we
have
the
work
items
nbc2
feature
flag
which
includes
all
of
the
widgets
outside
of
the
core
task
experience.
D
D
D
Create
it
and
it
creates
it
here-
the
feature
that
I
was
talking
about
that
we
wanna
make
sure
is
in
before
we
switch
to
feature
flags.
Is
the
ability
to
to
click
into
this
task,
which
we
currently
can't
do,
but
the
mr,
I
think,
is
in
review
so
that
should
be
in
the
next
couple
days.
D
D
Okay,
wait:
widget
is
in
it's
integrated,
the
front-end
back-end
is
integrated,
so
this
works
and
changes
saves
it
there
close
it,
reopen
it
still
there,
if
I
refresh
still
there,
so
the
weight
persists
there.
D
Labels
is.
I
don't
think
this
is
fully
integrated
with
the
with
the
back
end
yet
so
this
will
not
persist,
but
this
reuses.
D
It
looks
like
it
is,
but
this
reuses
a
lot
of
the
at
least
token
display
logic
that
that
natalia
demoed
on
assignees,
so
it
should
be
pretty
similar
to
what
we
have
there
and
just
search
doesn't
look
like
it's
working
yet,
but
again
that
does
not
persist
yet
and
I
think
tapping
out
of
it.
I
think
unfocusing
saves
it.
D
That's
nice
or
it
may
just
save
once
we
once
we
click
or
auto
complete.
D
Let
me
refresh
just
to
confirm
that
it
doesn't
purchase
yeah,
it
doesn't
persist
yet
all
right
labels
and
then
the
last
one
is
this.
This
call
out
here
with
the
information
about
tasks
that
the
pika
created.
D
D
A
All
right
can
I
do
them
out
of
order,
since
we're
already
on
this
view,
like
what
happens,
if
you
try
to
make
the
issue
confidential
in
life
that
you've
created
some
public
tasks.
D
So
if
we,
if
we
turn
this
to
confidential
and
the
parent
issue,
is
public,
then
we
will
get
in.
Oh
that's
a
lot.
That'll
work
once
right.
D
D
A
So
that's
what
I'm
wondering
like.
What's
the
experience
like
now,
I
know
we
have
the
back
end
validation,
but
I
don't
think
we
have
the
front
end
like
the
front
end
needs
to
send
automatically
the
confidential
state
when
it
creates
the
task,
and
I
don't
think
that's
right,
so
I'm
kind
of
wondering
like
what's
the
user
experience
like
at
the
minute
like.
Does
it
just
fail
with
no
indication
of
why
or
is
there
like
a
flash
message
with,
like
you
know
what
I
mean.
D
Already
got
the
error
there
yeah.
So
let
me
let
me.
F
D
A
D
That's
not
right,
265
is
the
id.
I
guess.
D
E
A
D
E
There
was
a
discussion
around
that
I
mean
and
we
actually
changed
that
for
making
creating
a
task
from
the
description
on
a
task
list.
That
mutation
actually
uses
the
parent's
confidentiality
to
to
create
the
task.
So
the
discussion
was
around
if
we
should
do
the
same
for
the
child
items,
mutation
is
still
missing.