►
From YouTube: Backdrop Weekly - Jlu 23rd, 2020
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
C
Hello,
I'm
greg
I'm
joining
from
central
greece,
I'm
on
holiday,
but
I'm
looking
to
help
in
any
way
that
I
can.
If
I
have
time.
A
For
lambton
also
oakland
california,
first
on
our
agenda
today
is
that
we
have
new.
We
have
a
new
contributed
theme.
The
para
hallerine
theme
not
entirely
sure
to
pronounce
that
was
released
this
week.
So
anyone
interested
in
that
stuff-
and
that's
all
I
have
for
other
than
core
stuff,
because
our
core
stuff
list
is
quite
long
today.
So
let's
turn
it
over
to
nate
to
talk
about
those
changes.
B
All
right
we'll
see
how
much
I
actually
end
up
talking
about
these
issues,
since
a
lot
of
them
are
ones
that
jen
has
been
active
in
so,
first
of
all,
we'll
go
through
the
issues
that
are
affecting
the
current
release
of
backdrop,
which
is
116..
B
1162
is
our
current
version
with
116.3
being
our
next
bug
fix
release,
we
don't
really
have
any
driving
factors
to
push
out
a
release
in
the
immediate
future,
although
if
some
of
these
accumulate,
then
it
might
be
worthwhile
making
a
release
for
them.
So
first
up
is
update.php.
B
B
Jen,
it
looks
like
you
filed
the
pull
request
and
I
think
that
it
currently
just
needs
review.
C
Yeah
it's
pending
on
me
as
well
to
test
this
I'll,
try
to
reproduce
the
the
actual
use
case
when
this
is
caused.
So
I
need
to
reproduce
that
on
a
vanilla
and
then
try
with
sense
thanks.
C
So
so
code
looks
okay,
just
testing.
B
All
right
next
up
is
site,
gets
stuck
in
maintenance
mode,
and
specifically
this
is
when
you
are
running
a
module
self
update
or
an
in-core
update.
It's
issued
39.56
looks
like
this
is
some
mixing
up
between
using
a
state
system
and
using
the
config
system.
I
looked
over
the
code
and
it
looked
reasonable.
B
The
corrections
that
it
made
there
may
be
some
change
in
behavior
from
how
it
was
originally
intended.
It's
a
little
bit
hard
to
tell
what
it
was
intended
because
we're
making
these
changes-
and
you
know
it
behaves
differently
from
the
way
it
did
before,
but
we're
not
quite
sure
what
it
was
intended
to
do
before
and
what
it's
doing
now.
B
One
takeaway
from
that
kind
of
experience
is
that
you
know
this
is
a
situation
that
can
occur
when
we
don't
have
test
coverage
on
something,
because
we
can
make
a
change
and
nothing
passes
or
fails
because
we
don't
have
it
covered
by
tests.
I'm
guessing
that
you
know
a
module
update
is
not
an
easy
thing
to
test,
but
it
might
be
it
it
not.
B
Might
it
is
worth
investigating
if
we
can
somehow
mock
doing
a
module
update
within
the
simple
test
processes
to
see
if
we
can
get
that
user
interface
under
the
control
of
you
know,
test
coverage.
C
Yeah
then
it
seems
that
there's
conflicts
on
that
pr.
So
if
you
can
just
you
know,
resolve
the
conflicts
I'll
review
again.
B
Okay,
next
up
visual
regression
and
text
format,
order
on
the
overview
form
of
text
formats,
issue
4407:
this
is
the
regression
that
we've
had
since
116
was
released.
There's
still
no
code
behind
this
so
4407,
I
I
don't
think
it's
even
going
to
be
difficult,
but
that
needs
a
pull
request
behind
it
to
fix
the
presentation
of
those
text
formats,
so
they
show
up
in
the
right
order.
B
Let's
see,
there
are
a
number
of
things
that
could
use
some
review
in
addition
to
that,
2766
restore
the
views
pager
by
date.
This
issue
has
been
needs
review
for
a
couple
of
weeks
now
and
it
really
just
well.
It
needs
some
testing
of
the
contrib
calendar
module
to
check
it
out
with
with
the
core
enhancements
and
doc.
Wilmot
already
reviewed
this
and
said
that
the
problem
seems
to
be
resolved.
B
All
right,
let's
see
so,
let's
see
we
also
have
phpcs
that
we
would
like
to
automate
our
code
checks,
issue,
32
13.,
no
updates
on
that
this
week
we
have
tried
a
couple
of
different
approaches
and
I
think
that
we're
rolling
back
to
kind
of
the
let's
just
find
one
that
works
and
go
with
it,
but
I
think
there's
a
lot
of
prerequisites
for
that
that
we
need
to
get
our
our
coding
definitions,
updated
with
encoder
module,
because
I
really
don't
think
that
we
want
to
build
out
some
automated
integration
that
is
dependent
upon
an
old
version
of
software,
so
kind
of
before
I
mean
we
can
work
on
the
automated
portion
of
this.
B
You
know
having
a
task,
runner,
download
and
install
coder
and
then
run
the
code
style
checks.
I
think
that's
worthwhile
work
there,
but
before
we
really
can
deploy
this,
we
also
need
to
update
coder
module
to
make
it
so
that
it's
using
the
latest
version
of
phpcs,
because
I
don't
want
to
build
something
into
our
processes
that
is
already
deprecated.
B
B
A
Sure-
and
I
was
just
trying
to
remember
when
the
I
think
I
added
some
of
these
last
weekend-
some
of
them
this
week,
so
my
mom,
where
we
might
might
not
be
updating
all
of
them,
so
the
first
one
is
to
set
a
reasonable
value
for
a
php
variable
called
max
input
vars,
and
this
is
the
thing
that
allows
you
to
have
a
certain
number
of
form
elements
on
a
page.
A
A
I
would
like
to
prevent
everyone
else
from
running
into
the
same
problem,
and
so
this
is
just
a
quick
fix
that
should
allow
us
to
resolve
that
problem
for
other
people.
C
Yeah,
we
should
specify
it's
been
a
while
now,
but
we
should
specify
what
reasonable
is.
So.
We
have
been
bitten
by
this
in
the
gov
cms
platform
in
the
menus
menu
management
screen,
where
they
have
the
check
boxes
for
enabled
and
disabled
and
then
sas
customers
do
not
have
access
to
the
permission
page,
but
we
do
as
backend
legenders
and
if,
if
the
site
has
like
six
or
seven
user
roles
and
with
all
those
permissions,
it
really
becomes
an
issue.
C
We
have
been
increasing
those
to
like
five
thousand
or
ten
thousand
for
some
sites,
but
I'm
not
sure
if
that's
considered
reasonable.
So
some
people
might
hit
that
limit.
But
we
need
to
specify
what
we
consider
reasonable.
So
I
think
that
we
should
look
at
the
default
roles
that
we
have
or
maybe
a
couple
more.
What
do
you
guys
think.
B
C
Yeah
yeah,
so
so
for
the
menu.
Specifically,
we
looked
into,
including
the
big
menu
which
has
nothing
to
do
with
mega
menus.
The
big
menu
module,
which
sort
of
like
exposes
only
a
portion
of
the
menu
section
each
time
so
that
it
doesn't
have
a
lot
of
checkboxes
in
a
single
page
in
a
single
form
but
yeah.
C
A
It
it
looks
like
that,
pull
request
still
isn't
a
needs
work
state,
because
we
need
to
include
that
number
in
the
messaging
I'll.
Take
a
look
at
it
looks
like
nate.
You
also
had
some
recommendations
about
checking
at
modphp7.c
in
the
htaccess.
B
B
I
think
that's
the
only
comment
I
had
there
and
then,
rather
than
just
saying,
hey,
this
number
is
too
low.
We
should
link
to
documentation,
saying
how
someone
can
fix
it,
because
I
think
what
we're
trying
to
do
is
make
it
so
the
htaccess
sets
that
limit
for
them,
which
is
great
if
htaccess
is
being
respected
and
if
it's
not
being
respected,
it's
overridden
by
their
host
or
they're
using
nginx,
or
something
like
that.
That's
when
the
status
report
message
will
kick
in.
C
Notice
that
now
I've
done
it
qrikets,
so
my
only
comment
with
a
pull
request
is
why
1007
so
the
pull
request
has
the
value
set,
200
1007.
B
Says
it
for
php5
yeah,
I
don't.
I
don't
know
why
that
is
I.
I
would
guess
that
it's
to
make
it
clear
that
that's
where
the
value
is
coming
from,
because
it's
such
an
unusual
number
yeah.
A
A
Moving
on
the
next
issue
here
was
to
by
defaults
on
new
installs
from
backdrop
make
it
so
that
the
search
block
doesn't
say,
search,
search
search.
If
we
just
hide
the
title,
so
it
says
search
search,
then
that
makes
more
sense,
but
if
it
has
the
title
and
the
label
and
the
button
all
saying,
search
in
the
same
box,
it's
a
lot
of
repetition
of
the
word
search,
so
that
was
just
hide
it
by
default.
There's
a
setting!
You
can
turn
it
off
we're
just
going
to
set
the
setting
by
default.
A
If
you
want
it
back
on,
you
can
turn
back
on
that
needs
review.
The
next
one
is
to
fix
a
bug
with
views
ors,
which
I
think
it
bit
a
bunch
of
us.
When
we
first
introduced
the
change
to
views
that
broke,
it
there's
now
a
change
to
views
that
fixes
it
that
we
need
to
introduce
back
again.
A
So
this
issue,
I
think,
needs
another
review.
We
looked
at
it
a
long
time
ago,
but
I
don't
think
it's
gotten
any
attention
since
march.
C
A
Yeah
so
the
problem
was,
we
introduced
some
change
to
abuse.
That
was
like
a
security
fix
that
happened
to
break
the
way
or
swerved.
So
if
you
like
have
one
condition
or
another
condition
that
stopped
working,
and
then
there
were
a
couple
of
fixes
in
the
drupal
queue
that
didn't
actually
fix
the
ore
problem
under
all
of
the
tests.
A
C
Makes
sense
yeah?
I
do
recall
going
back
to
that.
The
patches
that
were
posted
in
the
drupal
queue
and
people
were
going
back
and
forth.
Yeah.
A
Fix
it-
and
this
is
here,
it
does
fix
it,
but
it
reintroduces
whatever
yeah
that
was
complicated
and
then
the
next
issue
we
have.
These
are
three:
the
indigo
xla
posted
the
first
one
is
for
ck
editor,
not
respecting
language,
so
there's
just
a
parameter
that
we're
passing
into
the
sequential
settings
incorrectly
we're
still
testing
language
instead
of
lane
code.
A
Another
one
is
the
text
on
the
user
register
form
wasn't
wrapped
in
t.
So
this
is
a
correction
to
that
and
then
the
last
one
is
really
exciting:
php
syntax
highlighting
so
on
github.
If
you
have
a
file
that
ends
in
php
the
syntax
highlighting
works,
but
we
have
a
bunch
of
files
in
backdrop
that
do
not
end
in
dot
php
that
are
not
properly
highlighted
on
github,
but
there's
not
a
way
to
tell
github
which
of
your
files
are
php
files,
and
so
this
pull
request
includes
that
change.
C
I
reviewed
that
it
only
there
was
a
we
needed
to
add
the
dot
profile
ones
as
well,
and
that's
it
and
that's
how
tbc.
I
have
marked
it
as
an
earthy
position,
that's
on
a
radar,
but
it
needs
this
tiny
fix
and
I
think
nate
you're
able
to
fix
things
before
you
commit,
and
so
it's
just
it's
straightforward.
Maybe
you
can
just
do
that.
C
The
funny
thing,
if
you
look
at
these
last
comments,
is
that
it
wasn't
working
for
lauren
when
he
was
testing
it,
and
then
I
tested
it
as
well
and
I
go,
but
I
saw
it
and
then
I
couldn't
see
it
two
minutes
after
and
then
it
turns
out.
If
you
refresh
the
page
enough
times,
then
it
comes
up.
Maybe
it
has
to
do
with
the
way
that
github
caches.
A
C
I
didn't
know
what
it
was
and
I
go
what
changed
between
when
I
tested
and
when
lauren
reported
the
thing
and
I
go.
Oh
yeah,
I
added
a
comment
to
the
full
request.
Could
that
affect
it?
So
I
removed
the
comment
and
then
I
read
it,
but
it
didn't
affect
it.
It
was
it's
completely
random,
like
you
keep
refreshing
the
page
when
it
comes.
C
C
So
this
this
improves
on
the
feature
that
bw
panda
has
added
recently
in
the
autocompletes.
So
if
you
start
typing
your
see
how
we
have
a
post,
that
is
called
your
first
post.
So
if
you
start
typing
the
word
your
it
doesn't
come
up,
you
have
to
type
sorry
post,
it
doesn't
come
up.
You
have
to
specifically
remember
that
it's
the
thing
or
the
word
first.
B
Oh,
oh
yeah
that
looks
really
simple,
so
the
queries.
I
was
thinking
that
you
couldn't
make
this
better.
Yet
if
it
did
well,
I'm
not
sure
about
better,
but
you
couldn't
make
it
less
strict
by
allowing
any
word
to
match
regardless
of
its
order,
so
you
could
say
post
space
first
and
then
it
would
match
on
that
as
well.
B
B
Allow
a
percent
sign,
allow
it
to
match
the
phrase
anywhere
in
the
middle
of
the
sentence
rather
than
right
at
the
beginning,
and
it
seems
like
an
improvement,
I
would
say
maybe
well
I
feel
like.
Maybe
we
should
get
some
more
feedback,
because
it's
kind
of
a
change
in
behavior
of
like
how
do
you
want
it
to
work?
You
know
it
depends
on
if
you
have
a
whole
crap
ton
of
content,
whether
or
not
this
is
helpful
or
not.
B
Exactly
so,
as
you
get
like
more
loose
with
the
requirements,
you
get
more
content.
That
might
not
be
the
content
that
you're
looking
for
and
so
on,
a
site
that
has
a
modest
amount
of
content.
It's
going
to
be
helpful
to
be
loose
with
things,
but
I'll
say
that's
like.
Would
you
a
lot
of
content?
Strix
might
end
up
being
better.
A
C
B
Well,
I
would,
I
would
be
tempted
gregory
I
mean
we
are
targeting
towards
you
know
smaller
sites,
I
would
be
tempted
to
say,
we've
had
it
this
way
for
several
releases
and
we've
never
tried
it.
The
other
way
like
let's
have
it,
be
wild
card
matches
and
and
then
see
and
then
see.
If
feedback
comes
in
saying
I
can't
find
the
content
that
I'm
looking
for.
C
I
can
I
can
put
a
setting
in
either
the
settings
php
file
or
the
core
config
that
doesn't
have
anything
in
the
interface
and
then
we
decide
later
so
that
we
can
still
allow
people
to
override
it
if
they
want.
C
Where
would
you
have
it
leave,
though?
Settings
php
wow?
We
we
can
overwrite
settings
config
settings
in
settings,
yeah,
yeah
or
yeah.
If
you
can
add
a
comment,
if
you
come
up
with
a
good
place
where
that
should
leave,
I
can
do
it
because
it's
like
the
request
is
pretty
simple
and
then
and
then
the
other.
The
other
issue
is
double
481
and
team
reviewed
that
and
he
asked
the
question
so
I'll
respond
to
that
question.
C
This
is
again
to
improve
on
bw
pandas
feature,
so
the
way
that
the
text
is
presented
in
that
drop
down
menu
is
like
completely
like
everything
has
the
same
text:
size
and
color.
I
guess
so.
It
gets
a
bit
distracting
and
I've
provided
like
a
few
variants
there
and
I
ended
up
doing
a
last
one,
but
if
you
guys
prefer
something
else
like
just
let
me
know
it
should
be
pretty
straight
forward
to
change
the
css.
C
I
think
the
main
reason
for
the
pr
the
main
goal
for
the
pr
was
to
add
the
classes
around
the
the
actual
match
and
the
description.
So,
after
that,
we
can
style
it
and
it
allows
people
to
style
it.
However,
they
want,
with
their
admin,
team
and
team,
to
answer
your
question
I'll
respond
there
as
well.
It
affects
every
path
field
that
is
enabled
to
do
autocomplete.
It's
in
multiple
places.
C
B
C
Yeah
so
initially
I
started
doing
mockups
and
then
I
thought
of
the
places
that
we
have
the
machine
name
as
a
description,
and
then
I
try
to
sort
of
like
mock
the
size
from
that
which
is
secondary
information.
But
anything
you
guys
like
think
it's
better
like
you
can
start
playing
with
in
the
sandbox
and
yeah.
If
you
think
that
the
colors
should
change
or
the
font
size
or
whatever
then
yeah,
we
can
just
provide
a
screenshot
and
we
can
decide.
B
Yeah,
I
yeah,
I
think,
maybe
the
classes
just
need
to
be
tidied
up
a
little
bit,
but
this
screenshot,
here
of
it
being
like
smaller
and
gray,
for
the
small
small
text.
B
So
in
particular,
so
we're
using
classes
like
dot
suggestion
and
dot
description,
and
I
can
just
feel
wes
rivakaba,
like
gazing
upon
me,
frowning
at
that,
because
if
you
we
use
dot
description
in
a
bunch
of
different
places
like
in
form
elements
and
if
you
generically
styled
dot
description
site-wide,
you
would
accidentally
hit
this
dot
description
that
is
used
in
an
autocomplete.
B
C
B
So
what
wes
would
call
selector
wars?
You
know
where
it's
like
you
get
increasingly
specific,
because
you
can't
get
it
targeted
closely
enough,
so
you
want
the
the
stuff
that's
made
to
be
overwritten
to
be
the
stuff.
That's
in
court
to
be
the
least
specific
as
possible,
which
is
usually
just
one
level
of
complexity,
just
a
class
level.
C
A
D
D
Twelve
one
one,
two,
five
nine:
this
is
when
you
make
a
custom
block
and
you
make
a
change
in
source
mode.
It's
a
ck,
editor
issue
and
you
make
a
change
in
source
menu
source
mode.
If
you
change
it
back,
your
change
does
not
get
recognized.
D
I
run
into
this
fairly
often
myself
and
this
issue
bugs
me
so
much
that
I
find
when
I'm
talking
to
people
about
backdrop
or
like
that
that
are
working
with
custom
blocks
and
stuff.
I
will
you
know,
like
pre
pre-emptively
mention
this
to
them,
just
to
save
them
time
and
anyways.
So
I
guess
what
I
would
be
say
would
my
question
would
be
like
some
people
it
looked
like
laralyn
and
and.
D
B
D
D
C
D
Okay,
so
this
doesn't
it
doesn't
it's
only
on
blocks,
it
doesn't
seem
to
be
a
problem
on
nodes,
at
least
like
if
I
tried
a
blog
post
in
a
page
today,
and
that
was
not
a
problem,
but
if
you
create
a
custom
block,
that's
where
the
problem
is,
and
olaf
like
was
suggesting.
D
B
B
The
second
to
last
comment
in
here:
let's
do
a
stack
overflow
post
has
an
interesting
suggestion
that
says
like
on
the
submit
of
the
form
just
manually,
update
the
text,
areas
that
have
ck
editor
on
them,
and
that
would
be
a
good
workaround
for
us
that
would
make
it
so
that
we
don't
need
anything
from
seek
editor's
side.
It
also
wouldn't
hurt
ifc
candidate
or
change
their
behavior
in
the
future
that
if
we
had
something
that
just
synchronized
seek
editor's
value
with
the
underlying
text
area,
doing
it
again
isn't
going
to
hurt
anything.
B
D
B
B
Well,
maybe
if
it's
not
seek
editor,
if
it's
not
a
global
seek
editor
issue,
then
maybe
has
something
to
do
with
the
way
that
we're
submitting
forms
within
dialogues
and
I'll
post
a
comment
there,
because
it
gets
a
little
bit
crazy
because
what
happens
is
the
submit
button
for
a
form
isn't
shown
inside
of
the
dialog,
because
the
jquery
ui
dialogs
have
a
special
location
for
where
buttons
should
go
and
it
moves
it
outside
of
the
normal
dialog
area
into
the
buttons
section
of
the
dialog,
and
so
what
we
do
is
we
make
a
copy
of
all
of
the
buttons
put
them
into
the
special
button
area
and
then
with
javascript,
when
you
click
the
save
button
in
the
dialog,
the
save
button
copy
you
using
javascript
clicks
the
original,
submit
button
and
possibly
what's
happening.
B
There
is
our
javascript,
that's
submitting
the
form
on
behalf
of
the
user,
submitting
the
form
it's
not
triggering,
seek
editor's
behaviors,
and
I
bet
we
if
we
fix
that.
I
bet
that
would
solve
that
that
problem.
If,
if
it
is
true
that
this
problem
only
seems
to
occur
in
dialogues,
I
bet
that's
what
it
is.
B
Okay,
so
tim,
maybe
you
don't
need
to
go
hunting
if
it.
If
it
really
is
just
in
one
specific
place,
it
might
be
good
just
to
clarify
that
the
problem
doesn't
occur
when
saving
a
node
or
anywhere.
That's
not
in
a
dialogue,
and
if
you
can
find
a
second
place
that
occurs
in
a
dialog,
that'll
really
say
just
tested.
D
B
Yeah,
you
could
do
that
because
you
can
create
a
custom
block,
not
within
a
dialog
as
well.
You
can
go
to
you
know
the
blocks
ui
and
then
add
a
custom
block
that
isn't
within
a
layout
and
if
that
saves
as
well,
I
I
think
that
it's
clear
that
it's
not
a
problem
with
that
particular
form.
It's
it's.
The
dialogue,
that's
causing
the
problem.
B
Okay,
yeah:
these
are
all
great
issues.
Let's
see,
what
number
is
that
1259
was
that
particular
issue
I'm
going
to
put
that
into
the
agenda
so
that
we
have
a
record
of
that.
B
Okay,
well,
let's
go
on
to
117,
which
is
the
next
minor
release.
So
that's
where
we're
putting
in
our
new
features
and
functionality.
B
We've
lost
one
on
here.
There's
there
had
been
one
that
doc
bulma
had
been
working
on,
but
it's
not
in
this
list.
I
can't
remember
quite
what
it
was
I'll
pull
that
in
but
the
context
one
yeah
the
context,
one
that's
right:
custom.
B
Yeah,
okay,
so
jen
you've
been
working
on
using
the
administration
form
for
the
login
page,
and
I
know
it's
ran
into
a
number
of
snags.
What's
the
status
on
that
issue,
which
is
4410.
A
I
think
it's
on
hold
until
we
can
resolve
two
blocking
issues,
one
of
which
is
removal
of
the
view.
The
administration
administration
theme
permission
and
the
other
one
is
the
ability
for
a
single
layout
to
affect
multiple
paths
which
I
don't
think
I
have
linked
here,
but
I
think
if
we
can
get
both
of
those
issues
resolved,
then
we'll
have
a
path
forward.
On
the
other
one.
A
B
Oh,
I
see
down
here,
okay,
so
yeah,
let's
go
right
into
removing
the
administration
theme.
What's
what
are
your
thoughts
on
that
issue?.
A
So
I
set
up
a
pull
request.
That
removes
the
permission.
I
thought
I
grabbed
it
everywhere,
but
some
tests
are
failing.
So
I
think
I
might
have
missed
a
couple
of
places.
The
only
places
that
I
found
were
using
it
is.
We
had
defined
that
permission
on
a
bunch
of
tests
and
there
was
a
couple
places
and
layouts
that
checked
it,
and
those
are
the
only
places
that
we're
using
it
currently.
So
the
patch
was
pretty
clean.
A
It
was
like
remove
it
from
three
places
in
layout,
it's
in
18
places
and
tests
and
we're
good
to
go,
but
because
tests
are
now
failing,
I'm
gonna
go
check
other
things
and
make
sure
that
was
all
that
needed
to
be
changed.
I
also
created
a
contrib
module
called
view
administration
theme
that
provides
the
permission,
but
nothing
else
so
that
if
your
site
needed
that
permission
to
do
something
else
or
dependent
on
it,
you
would
have
a
way
to
get
the
permission
back,
but
the
module
doesn't
do
anything
other
than
other
permission.
A
So
I
was
wondering
if
that
module
should
alter
the
behavior
to
check
the
theme
for
the
permission
before
it.
Does
it,
in
addition
to
providing
permission
rather
than
only
providing
information,
so
we
do
the
thing
that
it's
supposed
to
do.
B
C
A
B
I
wonder
how
it
weighs
against
our
policies
on
backwoods
compatibility,
because
it
is
definitely
a
change,
but
it's
like,
but
it's
a
good
change.
B
Yeah,
so
it's
a
good
change
and
I
I
think
that
a
compromise
that
might
make
me
feel
better
about
it,
and
hopefully
you
know
bring
us
in
line
with.
What
we
try
to
achieve
is
that
when
you
update
you're
at
least
notified
that
that
permission
has
been
removed,
if
you
want
it
back,
then
you
can
install
this
contrib
module
so
kind
of
as
a
prerequisite.
I
think
we
need
to
have
the
contrib
module
and
a
release
for
it
in
order
to
incorporate
this
feature
into
core.
A
B
A
B
A
But
then
we
wouldn't
fix
our
problem
that
was
blocking
the
other
module.
The
other
change.
B
C
C
So
there's
a
few
things
where
the
parts
of
the
veil
module
are
being
required
or
shown
in
the
the
user
interface
of
views
like
use
ui
the
settings
where
we
we
ask
which
region
and
it's
called
it
is
a
debug
region
that
was
provided
by
anyways
long-standing.
I
always
see
it.
It
annoys
me,
it
wasn't
a
problem
when
views
was
in
contribute,
because
that
was
a
contrib
module,
but
we
got
it
into
core
and
that
we
inherited
that
yeah
anyway,.
A
We
have
two
minutes
left.
B
Okay,
all
right:
well,
we
can
keep
going
so.
A
So,
let's
not
talk
about
layouts
in
multiple
paths,
because
there
hasn't
been
any
progress
on
that
one:
okay,
leo's
support
for
custom
contacts.
Did
we
talk
about
that
already.
B
D
Well,
my
last
comment
was
I'm
gonna
get
out
of
this
because
I
don't
understand
it
so
every
time
I
think
I
do.
I
do
have
one
or
two.
What
I
think
could
be
help
clarifying
questions
so
jen
you
made
some
comments
about
taking
splitting
the
you,
the
ui
off
and,
and
I'm
wondering
does
that
mean
you're
suggesting
that
we've
not
had
in
this
pr
there'd
be
no
ui
at
all.
A
No,
I'm
suggesting
that
we
create
a
follow-up
issue,
that's
like
fix
the
ui,
but
that
this
change
is
mostly
like
uncommenting
all
of
the
code
that
was
in
there
before
that
had
been
commented
out.
It
should
enable
the
feature,
but
we
know
it's
going
to
be
confusing,
so
we
have
a
second
issue
where
we
can
actually
go
through
what
language
should
we
use
here?
How
should
this
appear?
Should
we
go
first
or
second,
but
sort.
D
D
Okay,
well,
my
only
concern
about
that
is
that
I
think,
if
we
put
it
in
with
the
existing
ui,
it
would
just
be
confusing
right.
So
do
we
want
to
make
it
any
better?
I
understand
that
there
might
be
like
minor
changes
and
major
changes
in
a
separate
ui,
or
is
there
no
changes
right
but
like
putting
it
in
the
way
it
is
to
me
is
just
like
if
I
saw
this.
D
A
So
I'm
just
wondering
which
part
it
is
that
you
think
makes
that
makes
the
interface
so
much
more
confusing.
Isn't
the
fact
that,
before
this
change,
you
didn't
see
context
at
all
until
they
were
added
automatically
and
now
there's
a
button
or
a
link?
That
says
add
context,
and
you
don't
know
what
that
does?
Yes,
that's
the
problem.
Okay,.
D
I
think
there
are
some
minor
changes
in
that
that
could
be.
That
could
be
fixed.
For
example,
doc
wilbon
says
most
wild
cards
in
the
path
will
be
associated
with
the
context
automatically
to
me.
It
wasn't
at
all
obvious
that
he
was
talking
about
the
path
above
right
like
that's,
describing
something
in
a
different
field,
and-
and
I
think
we
could
reword
that
to
be
more
clear.
C
At
some
point,
the
thing
the
doc
will
not
provided
a
an
explanation
of
what
it's
going
to
do
and
it
was
in
zulip.
So
my
first
reaction
was
to
add
it
as
a
comment
in
the
issue.
I
didn't
do
that
I'll
go
back
and
do
it
because
in
the
future,
each
time
that
we
close
the
ticket,
I'm
not
sure
if
anyone
else
does.
But
I
do.
I
think
that
this
should
be
a
follow-up.
A
Okay,
so
tim,
I
think
if
you
want
to
do
recommend
those
changes
in
this
issue.
I
think
that's
fine,
but
I
think
that
we
might
need
another
follow-up
issue.
That's
like
contexts
are
confusing.
A
D
I
mean
I'm
still
in
to
to
suggest
anything,
because
I
just
don't
every
time
I
feel
like
I
understand
this,
then
I
learn.
Oh
I'm
missing
something,
and
so
like
trying
to
to
make
changes,
I
feel
will
be
futile
because
I
don't
really
even
understand.
What's
going
on,
I
mean
the
big
big
misunderstanding:
I'm
not
having
right
now
just
to
throw
this
out
there.
It's
because
I
haven't
elaborated
the
issue
yet
is
jen.
D
D
So
it's
like
you,
don't
understand
what
you
mean
when
you
say
it's
restricting
the
blocks
that
are
available
and
also
you
seem
to
be
saying
you
know
like
it
seems
to
me
that
obviously
doing
is
it's
field
blocks
available,
depending
on
the
context
of
the
situation
right
you,
I
think,
you're
saying
that
it's
doing
more
than
that
right.
A
In
what
you've
noticed
that
it's
doing,
but
you've
only
only
noticed
one
like
subset
of
things.
That's
doing
so.
I
think
that,
like
your
understanding
of
that
part
is
correct,
it's
just
there's
stuff
that
you
can't
see.
That's
also
happening
like
a
context.
Not
only
provides
you
know
the
field
block,
but
it
provides
all
the
other
blocks
that
also
require
that
context
and
what
those
other
blocks
might
be.
A
It
could
be
different
on
any
site
based
on
modules
or
you
know
what
configuration
they
have,
but
it
also
provides
visibility
conditions
so
like
when
you're
on
node
percent.
It
gives
you
options
to
choose
like
what
node
type
that
option
for
choosing
what
node
type
isn't
available
on
user
slash
percent.
Instead,
there's
like
a
user
role
option
so
there's
other
stuff,
that's
kind
of
hidden
until
those
contexts
are
there.
That
also
shows
up
so
there's
a
bunch
of
different
stuff
that
it
does.
A
That
I
think,
is
important
and
I'm
not
sure,
like
context
is
like
everything
and
you're
like
oh,
I
found
one
thing
and
you
you're
recommending
changing
the
label
from
everything
to
one
thing,
but
I'm
like
but
there's
other
things,
and
I
don't
know
this
necessarily
important
that
everybody
understand
in
every
interface,
all
of
the
things
that
are
going
on,
but
finding
some
way
to
make
sure
that
we're
not
limiting
what
people
think
it
can
do
is
good.
A
But
I
do
think
like
contexts
are
confusing
in
general.
So.
D
C
So
what
you
were
saying
now,
then,
about
visibility,
conditions
being
some
of
the
visibility
conditions
being
tied
to
context,
makes
sense
and
I've
never
thought
about
it
before
it
totally
makes
sense,
and
I'm
wondering
how
we
can
capture
all
this
information
into
a
documentation
page
and
then
once
that
would
ever
everyone
feel
comfortable
if
we
already
have
some
help
text
under
the
context
bit
like
the
table,
if
we
added
a
link
to
the
documentation
with
some
examples-
and
we
mentioned
that
a
few
examples
where
this
can
be
used
is
one
the
thing
that
you
just
mentioned
like.
C
If
you
want
to
provide
a
visibility,
condition
that
is
tied
to
the
user,
then
you
need
the
user
context.
If
it's
not
provided
that's,
why
you
would
do
it
and
then
doc
willmott's
example
from
zulip.
So
if
we
make
sort
of
like
a
collection
of
three
or
four
examples
like
that
and
put
it
in
documentation,
then
we
can
link
to
it
and
people
would
be
able
to
understand
it.
B
I
like
that
idea
of,
I
mean
we
do
it
in
a
lot
of
places
where
there's
something
too
complicated
to
explain
within
the
application
and
plus
linking
to
online
documentation,
also
makes
it
easier
to
update
you
know
and
managed
by
the
community
yeah.
C
B
Okay,
well,
let's
yeah
kind
of
wrap
things
up.
We
have
a
wish
list
here
that
we
can
leave
for
next
week
to
discuss
possibly.
A
Yeah,
I
think
all
I
really
wanted
to
highlight
today
is
that
this
is
a
list
of
things
that
I
would
really
like
to
work
on.
A
lot
of
them
are
small
and
we
can
probably
get
them
into
117..
One
of
them,
hashtags
is
probably
not
small,
but
I
would
love
some
prioritization
on
what
or
in
what
order.
I
should
work
on
them
in
terms
of
what
you
guys
think
is
useful
between
now
and
117.,.
C
D
C
Yeah,
especially
now
that
we
don't
have
a
lot
of
features
for
117,
we
could
make
this
a
release
that
is
focused
on
fixing.
You
know
annoying
bugs
that
have
been
long-standing
and
you
know
the
long-hanging
words.
So
I
really
like
that
section.
There,
okay.
B
Okay,
we
don't
have
our
initiative
leads
here
with
us
today,
jeff
for
telemetry
and
look
for
ready
to
wear
and
since
we're
already
overtime.
Let's
just
table
that
for
today
and
say
that's
it
for
the
meeting.
Thank
you
guys
for
coming
and
thanks
everybody
out
there
who
watches
these
videos.
So,
okay,
yep
see
you
guys.
Okay,.