►
From YouTube: GitLab 13.1 Kickoff - Manage
Description
The Product team for GitLab's Manage stage (including the Access, Compliance, Analytics and Import groups) assemble to showcase what's ahead for us in 13.1:
- Access: 0:25
- Compliance: 4:26
- Import: 13:45
- Analytics: 21:30
Agenda and issue links are publicly available at https://docs.google.com/document/d/e/2PACX-1vRnN91k9Rx1AyF8cIQga3KGFpjkI-sheTB16Jg-4e8fO3snZas2Th4hJC5C5wKIrNWbzr8aYDjI-T7-/pub.
A
It's
so
great
to
see
you
all
thanks
a
lot
for
helping
on
here
to
talk
a
little
bit
about.
What's
ahead
for
the
manage
stage
in
15.1,
I'm
excited
to
check
with
the
manage
stage
and
all
the
product
managers
here
about.
What's
ahead
and
13.1
is
releasing
to
you.
22Nd
2020
and
I'm
really
excited
to
hear
more
about.
What's
the
store
so
Melissa
over
to
you
to
talk
a
little
bit
about
what
access
is
flames.
B
Sure
so,
for
access,
we
have
some
follow-up
work
for
the
NBC
to
disable
inheritance
by
default,
so
an
initial
version
of
that
I
send
13.0
behind
the
feature
flag,
but
there's
a
couple
of
follow
to
make
sure
that's
a
really
solid
experience,
so
we'll
be
continuing
work
on
that
there's
another.
A
couple.
Other
features
that
were
working
on
that
are
basically
around
enterprise
readiness
for
comm
and
the
first
one
being
preventing
forking
outside
of
a
private
group.
B
So
this
was
functionality
that
was
part
of
managed
accounts,
but
we're
making
it
basically
available
for
anybody
with
a
private
group,
and
the
intent
of
this
is
so
that
enterprises
can
feel
confident
that
users
in
their
group
are
not
inadvertently
leaking
under
IP
and
around
that
same
theme
is
whitelisting
for
multiple
domains.
So
we
have
that
functionality
now
where
you
can
basically
restrict
access
to
a
group
based
on
a
specific
domain,
and
this
is
just
allow
for
multiples.
B
This
came
up
as
a
customer
request
due
to
merger,
but
there's
a
lot
of
different
use
cases
for
this,
like
consultants
or
just
global
companies
that
have
multiple
domains
and
the
next
one
is
one
that
Jeremy
did
the
legwork
on
for
approval.
So
thanks
Jeremy,
but
it's
moving
IP
level
restriction
down
to
premium
for
comm,
and
this
came
up
as
part
of
a
sales
process
of
just
basically
making
comm
competitive
in
allowing
companies
to
use
this
at
a
lower
level.
A
Awesome
thanks
a
lot.
That's
that's
exciting
stuff.
The
preventing
forking
outside
of
a
private
group
I
think
that
so
this
change
really
makes
sense
like
a
lot
of
customers.
So
it's
really
surprised
me
like
how
important
this
is
to
customers
like
as
Melissa
and
I
kind
of
dug
into
the
reason
behind
people
wanting
that
managed
accounts.
It's
actually
been
really
insightful
because,
like
I'd
say
like
60
to
70
percent
of
the
need
is
really
around
people
controlling
Forks,
which
is
like
I'm
nervous
that
someone's
gonna
fork
something
to
their
personal
namespace
and
they're
gonna.
A
Do
something
with
that
fork:
I,
don't
have
insight
or
access
to
so
I
think
that
the
approach
there
is
like,
instead
of
iterating
on
managed
accounts
list.
There
are
some
questions
there.
Let's
start
with
a
simple
first
step
of
just
like
preventing
working
outside
the
private
groups.
I
think
that's
that's
a
great
MVC
and
approach
to
to
just
take
the
most
important
part
of
GMA
and
then
iterate
on
that
separately.
So
I
think
that's
that's
awesome
to
see
it
being
prioritized.
I
have.
A
Because
all
of
these
organizations
that
want
this
they're
all
private
groups,
they're
all
companies
that
are
like
we
just
can't
imagine
having
a
public
or
internal
group.
This
is
like
behind
the
private.
You
know
group
on
get
live.com
and
we
want
to
make
sure
that
that's
why
we
control
that
so
yeah
definitely
cool.
A
C
The
first
one
in
the
first
video
that
we're
excited
about
is
I
think
it
was
in
twelve
nine.
We
released
the
ability
for
self
manage
administrators
to
enable
merge,
request
approval
rules
at
the
instance
level,
and
that
was
part
of
a
longer-term
rollout
to
provide
a
holistic
solution
where,
after
this,
in
thirteen
one
shifts,
we
can
say
that
I'm
gonna
define
those
rules,
I'm
gonna
scope
them
only
two
regulated
projects
leveraging
the
compliance
framework
project
labels
that
we
released
in
12.10,
and
then
only
administrators
will
be
able
to
change
those
settings
once
set.
C
So
this
is
based
on
a
lot
of
feedback
we
received
from
many
enterprise
customers.
That
said,
I
want
to
be
able
to
control
this
at
the
instance
level.
Now
we'll
need
to
explore
how
to
do
this
for
gitlab
comm
in
the
group
level,
but
this
is
where
we're
starting.
So
we
can
figure
out
what
are
the
technical
challenges
and
how
we
can
then
iterate
to
be
more
inclusive
of
the
gitlab
comm
customer
base.
Would.
C
This
one
here
this
is
the
current
design
prototype,
and
this
is
what
we're
moving
forward
with
in
terms
of
how
we're
gonna
implement
it.
So
I
just
need
to
update
the
description
there
yeah.
So
that's
the
premise
is
that
these
three
particular
settings,
and
then
you
can
specify
what
specific
framework
labeled
projects
you'd
like
to
apply
them
to.
C
We
also
captured,
if
authors
or
committers
were
allowed
or
denied
from
approving
the
mr,
so
now
we're
further,
supplementing
and
closing
the
loop
by
adding
the
granular
granularity
of
detail
to
show
that
when
the
rules
are
changed,
here's
the
approval
group
that
was
modified,
and
so
that
should
provide
a
better
data
integrity
for
the
audit
log.
For
this
we
have
an
MVC
that
we're
shipping
in
support
of
a
report
that
will
show
the
list
of
all
group
and
project
memberships
for
users.
C
So
this
started
as
a
larger
issue
that
we've
now
broken
down
into
this
MVC,
but
this
will
manifest
as
a
custom,
API
endpoint,
that's
specifically
tuned
for
this
use
case,
and
this
is
tied
into
the
audit
events,
performance
issues
that
I
mentioned
just
a
minute
ago,
but
this
also
made
sense
as
the
first
step
towards
providing
a
one-click
export
of
this
data
in
a
couple
milestones.
So
this
lays
the
groundwork
for
that,
but
it'll
basically
make
the
audit
events
API
more
useful
for
our
customers.
C
So
this
one
ought
to
export
audit
events
from
the
instance
level.
The
CSV
is
a
stretch
goal
for
13:1,
because
there
are
some
technical
considerations
about
pulling
this
data
because
of
the
performance
issues.
There
is
a
working
proof
of
concept.
It's
still
possible
that
we
can
do
what
we
need
to
do
to
get
this
prepped
and
chip
for
13:1,
but
I
have
some
reservations,
just
knowing
that
this
has
evolved
into
a
more
complex
challenge
than
we
originally
thought.
C
So
we're
gonna
strived
to
bring
this
in
so
that
we
can
get
two
features
in
for
the
audit
reports
category
and
then
last.
We
have
this
epic
for
filtering
audit
events,
there's
a
lot
of
work
that
has
been
done
and
there's
a
lot
of
work
to
be
done
all
in
in
service
and
the
it
looks
like
a
lot,
but
these
are.
C
These
are
broken
out
to
make
this
more
manageable
for
the
engineering
team,
but
the
intent
here
is
that
by
we
estimate,
maybe
13.3
that
will
have
this
drastically
improved
user
experience
for
all
of
the
audit
events,
Tables
instance
group
and
project
level
and
they'll
be
unified,
so
you
can
search
and
filter,
but
then
it'll
be
the
same
experience
across
the
board,
rather
than
the
disparate
experiences
that
we
have
now.
The
intent
here
is
that
it's
too
painful
to
find
and
search
through
audit
events
within
the
application.
C
It
requires
custom,
tooling
and
leveraging
the
API
to
get
what
you
need
quickly,
but
then
there's
an
upfront
investment
there.
So
we
want
to
make
sure
that
non-technical
users
and
just
people
who
prefer
not
to
build
the
custom
scripts
and
tooling,
can
find
the
information
they
need
quickly
within
the
application.
A
Awesome
thanks
a
lot.
Matt
I
have
a
question
on
the
custom
API
endpoint
for
listing
all
group
project
permissions
for
users
well,
I
think
NBC
makes
total
sense.
I
definitely
agree
with
the
need.
I
definitely
hear
about
a
need
for
this.
One
need
that
I
hear
is
like
is
specific
to
get
lab
comm
and
one
question
that,
like
organizations
typically
have
on
get
lab.
Comm
is,
as
a
group
owner
I,
want
to
understand.
A
C
Absolutely
the
second
issue
in
the
epoch
for
this
particular
implementation
is
summarizing.
The
data
within
the
application
I
have
to
double
check,
I
believe
it's
at
the
group
level,
but
I'll
need
to
double
check
that,
but
pretty
much
all
of
this
we
are
bringing
to
get
lab
comm
because
the
needs
don't
change
between
self-manage
or
Gilad
comm
customers.
C
I
know
a
challenge
there
as
far
as
API
would
be
easy
enough,
but
rendering
it
in
the
application
has
two
challenges,
which
is
the
performance
issues
from
audit
events
and
then
how
we
manage
the
actual
UX,
where
you
have
all
of
this
data,
where
one
user
in
larger
groups
might
be
part
of
you
know
hundreds
of
projects
and
that's
just
not
a
very
friendly
UI
experience.
So
that's
something
that
we're
mindful
of
as
well.
C
A
C
A
D
Here
so
let's
look
at
what
import
we'll
be
doing
in
13:1.
The
first
time
I
cannot
want
to
talk
about
is
a
carryover
from
previous
release,
and
it
is
the
UX
improvements
for
project
creation
I'm
showing
the
screen
before
last
month,
and
this
is
sort
of
the
entry
point
for
project
creation
that
we're
going
for.
We
want
to
have
an
explicit
way
for
the
user
to
decide
which
entry
point
for
project
creation.
D
Do
they
want
to
engage
and
kinda
may
can
help
them
make
the
decision
very
early
to
create
a
new
project
or
create
from
template
or
infer
project.
The
plan
here
is
to
increase
the
visibility
and
discoverability
of
templates
and
importers
as
well,
so
that
is
in
progress
and
the
feature
may
actually
be
deployed
to
be
clever
Congre
in
13:1.
D
We
will
be
using
a
be
testing
to
confirm
that
the
new
flow
is
an
improvement
over
the
old
one,
and
that's
that's
the
first
time
we're
doing
that
and
in
court.
So
it's
exciting
to
be
able
to
do
that.
The
next
issue
that
I
want
to
talk
about
is
also
a
carry
over
from
the
previous
release,
and
it
is
the
UI
and
the
user
experience
around
group
export
import.
D
We
were
able
to
deliver
the
export.
You
weigh
in
13
Oh,
however
13
the
the
import
UI
is
not
targeted,
4:13
ones,
so
the
export
is
already
gonna
be
available.
The
import
will
come
as
part
of
13
1
and
again
it
may
be
available
earlier
than
any
to
13.
One
end
date
on
github.com,
which
might
be
helpful
for
for
users
migrating
into
github
comma.
D
There
are
also
a
couple
of
smaller
issues,
but
they're
part
of
a
larger
trend
each
one
of
those
that
I
want
to
discuss.
One
is
this
issue
to
have
an
overview
dashboard
for
the
bitbucket
server
importer.
We
have.
We
have
started
slowly,
adding
dashboards
for
each
one
of
our
importers
so
that
we
can
track
their
usage
and
we
have
an
epoch
with
the
16
quarters
that
we
want
to
track
in
this
way
and
so
far
would
deliver
it.
One
of
those
and
taking
the
next
2013
one,
which
is
gonna,
be
the
bracket
server.
D
This
is
what
that
dashboard
looks
like
for
us
for
the
bitbucket
important.
We
can
see
the
how
many
projects
were
imported
and
kind
of
be
the
makeup
of
those
projects,
as
well
as
the
duration,
so
we'll
be
analyzing
this
for
each
importer,
but
also
across
all
the
importers
to
see
if
we
need
to
make
further
improvements
in
performance,
the
other,
the
other.
A
series
of
issues
I
want
to
discuss
is
merging
translations
from
our
Crowden
crowd-sourced
application.
D
A
Awesome
thanks
a
lot
for
prioritizing
internationalization
improvements.
I
know
that
it's
a
very
painful
process
for
the
current
process
of
merging
in
those
translations.
It's
not
easy!
So
there's
a
lot
of
improvement
there
that
we
can
make
for
engineering
staff.
It's
so
thanks
for
prioritizing
that.
What
is
your
expected
outcome
from
the
improved
project
creation
UX?
So
you
said
that,
like
we
will
use
the
a/b,
maybe
testing,
to
confirm
that
the
new
flow
is
an
improvement
over
the
old
one.
What
do
you
watch?
What
does?
What
do
you
mean
by
improvement
in
your
mind,
okay,.
D
There
are
minor
usability
improvements
as
well,
which
we
have
validated
in
the
solution.
Validation
that
we
know
are
gonna
just
be
a
benefit
once
we
switch
this
view,
but
we
also
want
to
be
able
to
measure
the
discoverability
and
ensure
that
you
know
that
it
doesn't
go
down.
For
example,
like
that
would
be
a
number
to
look
for,
and
then
with
our
a/b
testing
and
with
this
being
behind
a
feature
flag,
we
would
be
able
to
turn
off
the
experiment
if
it
actually
failed.
A
B
D
E
E
We've
got
some
some
big
things
in
mind
for
the
way
that
the
the
flow
is
depicted
for
new
metrics
to
display
for
revisiting
the
way
some
of
our
calculations
are
made,
but
we
want
more
users
trying
it
out
and
in
particular
customers
that
have
a
mono
repo
way
of
working
where
they've
got
multiple
teams
working
and
the
same
repository
have
a
hard
time.
Using
this
feature
in
a
way
that's
appropriate
for
each
particular
team.
E
Every
team
has
their
own
sort
of
workflow
their
own
measures
that
they
want
to
look
at
and
and
right
now,
the
the
most
granularity
you
can
get
is
at
the
project
level.
Without
you
stream
analytics
by
adding
these
filters,
will
then
be
able
to
use
labels
for
individual
teams
to
then
find
out.
What
does
that
flow
of
work?
Look
like
for
a
given
team
that'll
be
really
useful
for
us
at
gate
lab,
as
we
want
to
make
use
of
this
feature
more
broadly
internally
and
it'll
also
allow
you
to
break
down
different
kinds
of
work.
E
For
example,
your
bugs
may
move
through
the
development
pipeline
differently
than
features
and
by
using
those
labels,
you'll
be
able
to
slice
and
dice
that
data
data
differently.
So
you
think
this
is
a
first
move
that
will
unlock
a
lot
of
new
usage.
The
second
we're
not
taking
on
in
this,
but
something
we're
looking
at
to
follow,
is
the
possibility
of
creating
vise
dreams,
multiple
value
streams
inside
a
particular
group
or
project,
so
that
it
now
that
we've
got
customizable
value
streams
where
you
can
define
your
own
stages
and
own
flows.
E
E
Some
an
improvement
to
contribution
and
analytics
contribution
analytics
is
actually
one
of
our
best
loved
analytics
features,
believe
it
or
not.
But
when
I
look
at
this
big
data
table,
I
think
you
know
wow,
that's
a
lot
of
a
lot
of
data.
Our
customers
actually
really
like
being
able
to
see
how
much
activity
and
volume
of
activity
is
happening
in
their
projects
and
groups
and
which
users
are
involved
in
and
which
parts
of
the
workflows
and
contribution
analytics
gives
you
that
all
of
that
data.
E
However,
well
we've
got
a
lot
of
data
about
issues
and
pushes
to
your
repo
and
so
on,
and
even
about
mrs
people,
opening
new
merch
requests,
actually
merging
them,
and
so
on.
One
thing
we
are
not
tracking
today
is
who's
actually
approving,
merge,
requests
and
that's
of
interest
to
teams
that
are
really
trying
to
improve
the
turnaround
time
on
merge,
requests
and
make
sure
that
they've
got
enough
people
helping
to
approve
them.
E
So
they
move
through
the
system
in
a
timely
way,
so
we're
adding
a
new
event
to
our
event
tracking
system
to
be
able
to
track
that
and
then
surfacing
that
new
event.
Here.
In
the
contribution
analytics
page,
so
behind
the
scenes
that
behind-the-scenes
work
will
allow
us
to
do
some
other
interesting
things
with
that
data
and
then
what
will
be
surfacing
it
and
users
will
be
able
to
benefit
from
that
right
away.
E
All
right
and
the
next
item
I
want
to
cover
in
the
last
today
is
a
change,
we're
making
to
do
dues
and-
and
this
may
seem
like
a
very
strange
thing-
for
the
analytics
team
to
do
to
be
messing
with
to
do
this.
Let
me
explain
a
little
bit
what's
going
on
here,
so
the
analytics
team
is
responsible
for
bringing
the
power
of
value
stream
measures
to
across
the
get
lab
product
in
a
way
that
helps
us
to
work
smarter,
faster,
better
when
customers
go
to
buy,
get
lab.
E
One
of
the
main
things
they're
looking
to
do
is
is
to
move
faster
and
to
be
more
effective,
and
that
happens.
That
can
sometimes
happen
by
looking
at
a
dashboard,
reflecting
on
results,
spending
up
improvement
initiatives,
but
that
can
also
you
can
also
help
create
an
environment
of
continuous
improvement
by
putting
power
into
the
hands
of
people
that
are
actually
doing
the
work
right.
E
This
is
the
idea
of
the
and
on
cord,
you
know
from
the
Toyota
way
that
allows
you
to
make
many
small
improvements
throughout
the
day
so
by
by
taking
a
look
at
an
individual
users
inbox
the
work
that
they're
actually
doing,
how
does
that
work
come
to
them?
Are
they
being
overburdened
with
too
much
work?
Are
they
being
given
notifications?
They
can
act
on
in
an
efficient
and
effective
way
without
having
to
hunt
for
information.
We
can
actually
help
our
customers
achieve
that
vision
of
going
faster
in
so
from
the
grassroots.
E
So
this
is
one
place,
we're
looking
to
help
with
that.
There
are
others
in
the
products
we'll
be
looking
at
as
well.
So
our
first
step
in
what
we're
calling
inbox
analytics-
and
we
have
a
whole
epic
here
to
describe
sort
of
our
vision
for
that
we
would
love
your
feedback
on
inbox
analytics
and
some
of
our
ideas
around
that
here.
E
The
very
first
step
is
actually
for
us
to
conduct
some
research
and
we're
doing
two
different
kinds
of
research,
we're
doing
both
quantitative
and
qualitative
research,
so
we'll
be
picking
that
off
in
31.
This
is
part
of
how
we
support
the
quantitative
research.
So
today,
one
of
the
things
that
we
have
observed
by
looking
at
the
data
is
that
many
users
to
do
inboxes
fill
up
and
get
rather
full,
and
then
they
begin
to
have
to
hunt
for
the
things
that
are
important
and
so
what
we're?
What
we
want
to
know
more
about
that?
E
E
So,
in
order
to
help
us
distinguish
that
we're
creating
some
new
States
in
the
data
to
distinguish
between
things
that
you've
actually
acted
on
and
things
that
you've
just
dismissed
because
they
were
in
the
way
and
we're
doing
that
with
a
new
button
on
the
front
end.
So
well
now,
we've
had
at
this
mark
all's
done
button
that
allows
you
to
clear
your
to
do's.
If
you
just
get
to
too
much
you,
you
have
seven
or
eight
of
them.
You've
looked
at
them,
you're
finished
with
them
and
you
want
them
to
go
away.
E
We
can
do
things
like
calculate
the
cycle
time
for
2news
see
to
do
that's
been
waiting
a
certain
amount
of
time.
When
does
it
cease
to
be
relevant
because
you're
not
likely
to
ever
act
on
it
and
so
on,
and
then,
of
course,
we'll
be
doing
qualitative
research
on
that
as
well?
So
that's
what
I
have
to
share
with
you
about
thirteen
one.
A
That's
awesome
thanks
a
lot
Joe
Mason
I'm
interested
in
like
adding
in
the
approved
event
for
Mrs
I'm
curious
to
know
I,
wonder
if,
like
compliance,
could
benefit
from
from
this,
and
if
we're
considering
like
being
able
to
fetch
approval
events
via
API
or
any
kind
of
any
other
changes,
I'm
sure
that
that
might
be
interesting
to
customers
who
are
really
interested
in
approval
events
being
expose
us
in
other
ways.
They
can
fetch
that
through
it
through
an
endpoint,
do
we
like
considered
that
at
all
or
we're
kind
of
introduced
this?
C
Think
there's
also
implication
for
the
compliance
dashboard
in
some
way
where
we
can
maybe
aggregate
that
data
at
a
high
level,
and
just
say
you
know
this
project
has
why
total
number
of
merge
request
approvals
in
the
last
X
timeframe,
or
something
like
that,
not
sure
exactly
what
that
looks
like,
but
I
do
know
that
to
be
one
of
the
you
know,
key
compliance
data
points
that
would
help
at
least
the
compliance
persona
hone
in
on
specific
projects
more
efficiently.
So
definitely
some
value
there.
A
Awesome
and
I
also
had
another
question
around
the
to
do
introduced.
Close
state
I.
Think
that's
everything,
Steen
kind
of
move,
but
I
am
kind
of
curious.
Do
you
think
that
there
may
be
some
confusion
between
the
term
done
and
closed
when
I
was
looking
at
the
the
mock-up?
It
felt
a
little
interesting
to
me
because
we
had
like
I
have
two
buttons
now
that
are
both
equal
visual
weight
and
one
says
done,
and
then
there
says
just
just
as
closed.
A
What
do
you
think
about
that?
Do
you
think
that
there
might
be
an
opportunity
to
either
provide
use
of,
like
maybe
changing
the
visual
style
or
like
having
like
a
tooltip
or
something
that
kind
of
explains
the
difference?
Do
you
think
that
folks
might
kind
of
get
a
little
confused
between
the
difference,
yeah.
E
I,
actually
I
I
think
that's
very
likely.
We
think
that
the
future
state
of
that
capability
is
actually
probably
a
drop-down
button,
so
that
the
only
one
shown
is
closed
and
mark
all
done
is
kind
of
hidden
behind
it.
We're
taking
this
a
sort
of
an
incremental
step,
so
we're
not
too
jarring
to
people
who
are
used
to
that
mark
all
done
button
and
with
the
instrumentation
we're
putting
in
place
we'll
actually
be
able
to
see
whether
people
are
kicking
clicking
on
the
button.
E
We
expect
tooltips
a
great
idea
as
well
we're
actually
in
the
midst
of
discussions
about
whether
close
gets
the
default
color
or
not
to
try
to
nudge
people
in
that
direction,
because
we
think
that's
the
most
common
use
case
for
that
I'm.
Looking
forward
to
being
able
to
look
at
the
data
as
it
rolls
in
and.
A
E
I
could,
since
you
sort
of
opened
a
can
of
worms
there,
Jeremy
I'm
gonna
take
the
opportunity
to
show
one
more
thing:
that
users
who
are
really
interested
in
inbox
and
live
axilla
future
to
do
is
by
the
way,
there's
a
lot
of
research
and
work
going
into
this
area.
Besides
that
from
the
Atlantic
analytics
team,
the
the
planning
team
has
also
done
a
fair
bit
of
research,
project
planning,
team
and.
E
E
One
of
the
things
we'll
be
trying
to
do
to
validate
in
the
research
is
some
of
the
ideas
for
where
to
do.
This
might
go
in
terms
of
having
sections
where
urgent
items
bubble
to
the
top,
where
recent
arrival
activity
shows
up
and
then
the
backlog
of
items
here,
none
not
likely
to
ever
get
to
can
stay
collapsed
at
the
bottom,
so
that
they're
not
littering
the
view,
and
we
can
give
you
concrete
information
about
your
capacity
and
and
specific
goals
to
achieve
based
upon
historical
at
capacity
like
hey,
just
try
to
get.
E
You
know
a
few
items
you
know
in
this
case,
9
to
13
items
checked
off
from
this
section
and
if
you've
got
some
time
left
work
down
here
to
help
the
user
be
be
really
more
more
focused
about
that.
We're
now
taking
this
on
on
13
one,
but
it's
some
of
the
ideas
here
that
were
exploring
with
that
research.
D
D
I
thought
that
the
chrony
tab
management
system
was
going
to
help
me,
but
I
just
help
me
propagate
or
like
enable
my
addiction,
so
I'm
glad
somebody's
looking
into
this
and
saw
the
vision,
I
yeah
I
can
probably
live
with
the
steps
a
little
bit
better.
I
did
I
did
one
they're,
similar
I,
similar
kind
of
questions
as
Jeremy
did,
and
it
was
more
about
the
UX
affordance
like
what
would
you
exa,
Fortius
or
UI
affordance?
E
Well,
they're
close,
so
here's
sort
of
the
assumption
behind
the
research
that
we're
doing
it
and
that's
really
the
primary
reason
we're
doing
this
is
that,
if
you've
hit
the
done
button
on
an
individual
to
do,
you've
had
a
meaningful
interaction
with
it.
You've
decided
to
dismiss
it
because
you
don't
think
there
is
anything
to
do
and
what
we're
truly
trying
to
separate
out
is
sort
of
the
batch
closure
of
I'm,
not
even
looking
at
this
stuff.
D
A
A
C
A
So
John
makes
one
question
that
I
had
so.
The
dichotomy
that
you
just
described
made
me
think
that
maybe
the
right
place
to
put
the
the
separation
between
done
and
dismissal
is
on
the
individual
to
do
level.
So
in
the
in
the
issue,
it
looks
like
you
have
the
existing
approach
for
individual
items,
which
is
there's
a
done
button
next,
every
item,
and
at
the
top
we
have
the
dismiss
all
versus
mark
all
is
done,
while
you're
speaking,
you
said
that
you
wanted
to.
A
A
E
Oh,
when
we
were
doing
sort
of
our
first
pass
analysis
of
the
feature
we
encountered
a
user
that
was
very
passionate
about
marking
things
done,
meaning
that
they
were
really
done
that
and
this
user
I
I
got
the
impression
that
they
were
sort
of
an
inbox.
Zero
type
of
you
know
took
a
type
of
approach
and
they
really
wanted
it
to
mean
done,
and
so
we
were
trying
to
find
out.
E
We
think
you
know
my
inspection
of
the
data
that
we
have
thus
far
suggests
that
that
usage
is
a
small
minority
of
the
way
that
to
dues
are
approached.
But
since
we
encountered
it,
we
thought
well,
let's,
let's
accommodate
that
scenario,
and
then
let's
look
at
the
data,
because
what
we
can
do
is
we
can
compare
what
buttons
were
pressed
with,
how
many
to
use
were
affected
at
once,
right,
so
yeah,
and
that
that's
actually
some
of
the
work
we
did
last
iteration
is
to
be
able
to
make
that
kind
of
distinction.
E
We
used
to
not
be
able
to
distinguish
that
so
we'll
be
able
to
see.
You
know,
I
think
it's
somewhere
between
twenty
and
thirty
two
dues
you
can
safely
assume
the
user
is
not
really
meaning
finished.
You
know
they,
they
mean
getting
it
out
of
my
way,
and
so,
if
we
see
that
you
know
those
buttons
are
used
equally
for
those
different,
you
know
you
know
under
thirty
over
thirty
will
be
able
to
reason
a
little
bit
more
about.
Do
people
understand
those
buttons
do.
E
A
That
makes
that
makes
sense
and
I
have
you
bring
up
data
and
I
have
one
last
question:
I
apologize
for
harping
on
this
so
much,
but
it
such
an
interesting
topic
to
me
so
I
mentioned
earlier
that
you're
shipping
you're
gonna,
look
at
the
data
after
this
MVC,
and
you
know
I
really
appreciate
you
considering
instrumentation
and
measuring
things
like,
as
you
can
have
rolled
this
out.
What
what
do
you,
what
kind
of?
A
E
Well,
we're
collecting
the
data
not
so
much
to
analyze
whether
we
should
have
one
button
or
two
or
the
verses
of
thanks,
but
to
see
with
to
validate
the
problem
that
we've
observed
that
people
find
a
lot
of
noise
in
the
issue
in
the
two
dudes
that
show
up
in
their
inbox
and
that
for
many
people,
that
lists
grows
to
an
unmanageable
you
know
size
and
that
they
could
use
help
managing
that.
So
so
there's
this
kind
of
problem,
validation
and
what
I
would?
E
What
I
expect
to
see
is
that
this
is
a
pretty
widespread
problem
with
active
users
of
to
do
not
everybody
uses
to
Do's
for
sure
a
lot
of
people
don't
for
a
lot
of
different
reasons,
but
we
want
to
make
sure
that
this
is
really
a
problem.
We
should
go
after
that.
There's
going
to
be
a
sufficient
user
community
that
this
is
a
good
place
to
start
so
that
that's
really
what
what
this
is
all
about.
Cool.