►
From YouTube: Retrospective 12.9 (Public Livestream)
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
With,
like
the
first
note
that
we
have
seven
pages
of
content,
so
a
lot
of
good
feedback,
a
lot
of
good
interaction
and
a
lot
of
good
notes
as
we've
changed
our
process,
I've
put
in
notes
to
where
we
want
to
verbalize
associated
with
this
and
I
want
to
start
off
with
follow-up
items
from
the
previous,
the
previous
retro.
So
John
duplicity
were
available.
Could
you
please
verbalize
the
first
one.
B
So,
as
part
of
the
initiative
to
improve
code
review
efficiency,
which
we've
been
updating
on
the
last
few
brief
perspectives,
I'm
happy
to
announce
that
we
ran
a
domain
experts
pilot
during
12.9
with
two
backing
groups.
The
overall
feedback
was
very
positive.
There's
a
link
in
the
document
where
you
can
see
the
key
take
outs
from
it
and
we've
also
based
on
that
feedback.
As
well
as
various
other
conversations.
We
have
introduced
a
new
'mark
that
proposes
introducing
too
many
experts
into
a
code
review
process
which
is
currently
open.
B
A
John
just
to
know,
we
definitely
are
focusing
on
reviewer
efficiency
and
also
review
efficiency.
It's
something
I'll
cover
tomorrow
in
our
group
conversations.
So
this
is
a
super
important
initiative
and
I
appreciate
everybody's
participation,
including
developing
training,
as
well
as
figure
out
the
right
as
we
go
about
this
so
really
appreciate
everybody's
efforts.
Next
up,
John
Hampton,
you
want
to
talk
about
some
improvement
in
UX
and
fern
and
back-end
coordination.
Yeah.
C
Sure,
thanks
mister
I'm,
sorry
are
12.8
retro
contained
a
recurring.
What
can
we
improve
going
forward
theme
that
was
focused
around
front-end
and
UX?
In
my
reviews
we
identified
some
some
keep
pain,
points
we're
currently
discussing
those
and
then
lake
tissue.
So
please
take
a
look
there
and
feel
free
to
add
any
comments
or
thoughts.
Our
second
point
is
around
creating
a
concept
of
a
UX
definition
of
done.
C
A
John
just
a
note
for
other
engineering
managers.
This
is
something
we'll
be
monitoring
and
taking
track
up,
because
we
may
want
to
do
something
consistently
across
the
board.
Our
definitions
of
done
are
super
important
to
our
effectiveness
and
I
want
to
make
sure
that
that's
included
as
part
of
our
process.
Next,
we'll
move
to
the
section
where
what
went
well
this
month
and
Koli
cow
there's
a
lot
that
went
well
this
month.
In
spite
of
our
situation
globally,
we've
seen
a
lot
of
good
progress
in
a
lot
of
areas.
A
D
Everybody
David
Fernandez
called
out.
There
are
several
oh
wow
moments
when
releasing
NPM
package
forwarding
feature.
The
example
was
pulling
sub
packages
from
a
group
with
NPM.
We
yarn
from
a
project
empty
of
any
NPM
package
will
work
even
better
as
an
example
is
at
Babel,
slash
core,
we'll
pull
other
sub
components.
We
get
loud,
bang,
PM
registry
book
forward,
those
requests
properly
and
NPM
yarn
will
pull
in
without
complaining,
and
so
this
is
versus
a
pretty
surprising
discovery
as
we're
implementing
this.
This
forwarding
feature
that
we've
delivered.
E
A
F
Am
Kyle,
that's
great
news
about
the
review
apps.
Thank
you
for
bringing
that
up,
and
so
John
Hampton
already
talked
about
collaboration
between
engineering
and
us,
but
I
wanted
to
bring
up
also
collaboration
between
UX
and
our
product
managers,
and
lots
of
great
comments
about
how
UX
and
cams
are
working
even
more
closely
together.
F
One
person
said
being
paired
with
their
PM
has
been
helpful
in
focusing
their
work
and
then
the
release
management
team,
in
particular,
so
going
back
to
John's
comment
and
talked
about
how
it's
really
helping
them
to
hit
their
groove
working
well
together.
So
these
are
really
good
things
and
so
I'll
turn
it
over
to
Craig
Gomez.
C
Thanks
Christine,
actually
quick
correction.
Last
name
is
pronounced
gomes,
but
I
think
we're
in
an
hour
comm
and
that
common
mistake
happens
99%
of
the
time
so
yeah.
So
we
recently
have
the
opportunity
to
work
with
Nate
Berkus
peck
on
a
Ruby
on
Rails
performance
workshop
within
gitlab
and
how
this
all
started
was
through
Twitter.
So
nate
is
nate,
run
speech
SEO,
which
is
a
great
site
on
Ruby
on
Rails
performance.
C
He
is
also
a
maintainer
of
Puma
and
saw
in
Twitter
that
you
know
gitlab
was
slowly
rolling
out
to
pooh
min
once
we
got
it
rolled
out.
He
said:
hey
in
a
tweet
me:
I
should
work
with
get
loud
and
work
on
some
of
their
other
performance
bottlenecks
that
they
have.
The
tweet
was
forwarded
to
me
and
I,
followed
up
with
him
said:
hey
we're
hiring.
He
said
no
thanks,
not
looking
for
a
job
right
now,
but
I'm
happy
to
work
with
your
team
and
figure
out
some
kind
of
training.
C
That
would
be
helpful
for
you
and
through
a
series
of
conversations
he
actually
tailored
a
training
session
for
us
it
was
over.
We
are
sorry
for
weeks.
Two
sessions
a
day
depending
on
timezone
and
it
was
custom
tailored
to
get
lab
and
it's
fantastic.
We
had
a
lot
of
really
good
feedback.
We
still
have
to
have
some
there's
a
link
to
the
internal
training
that
was
there
the
folks
that
attended
it
either
synchronously
or
asynchronously,
found
it
really
valuable.
C
We
did
a
survey
afterwards
about
feedback
from
all
the
attendees
and
Nate
even
joined
our
internal
slack
and
was
is
still
doing
office
hours
with
us
and
worked
on
some
em
ours
that
we're
actively
working
on
with
get
lab
comm.
So
it's
been
a
fantastic
interactive
session
and
for
anybody
it's
not
working
at
get
lab
right.
Now
that
I
was
looking
for
any
kind
of
rails.
Training
Nate
was
fantastic
and
he
has
other
offerings
that
are
available
on
his
site.
So
highly
recommend
him.
A
Very
just
want
to
highlight
a
German
culture,
camp,
trainings
and
success,
or
two
key
aspects
that
our
engineers
are
looking
for
help
on.
So
if
we
have
other
ideas
for
additional
trainings
that
we
need
to
do
we're
also
starting
things
like
the
iteration
training,
reviewer
training,
really
looking
for
suggestions
on
how
to
make
sure
that
we
have
a
really
good
developer
experience
here
at
K
lab.
So
I
really
appreciate
you
taking
this
initiative
Craig
and
also
look
forward
to
other
folks
giving
us
options
in
the
future.
G
You
so
in
our
secure
group,
our
12
8
retro
is
a
bit
light
with
only
a
couple
of
topics.
We
attributed
that
to
not
enough
reminders,
not
enough
promotion
and
so
in
in
12:9.
What
we
did
is
just
weekly
as
well
as
within
our
meetings.
We
were
raising
this
topic.
More
often,
please
contribute,
and
so
we
ended
up
getting
15
topics
so
much
to
discuss.
We
actually
had
to
prioritize
several
of
those
topics
and
then
the
group
had
continued
conversation
around.
How
do
we
keep
the
sinc
sessions
which
we
all
being
really
valuable?
G
A
G
E
E
In
here
so
I
encourage
everyone.
Who's.
Looking
at
the
doctor
read
through
it
across
the
quality
department.
The
one
item
I
wanted
to
highlight
is
there
were
two
new
end-to-end
tests
added
as
a
result
of
a
root
cause
analysis
incident
involving
a
customer
install
pipelines.
So
this
will
help
prevent
these
type
of
issues
in
the
future,
from
impacting
our
customers
and
catching
them
earlier
in
the
feedback.
Loop
I'll
hand
it
off
over
to
Dan,
who
I
believe
is
next
under
one
whip
wrong
one
seat
thanks.
D
For
that
guy
yeah
so
was
a
common
theme
coming
up
in
our
weekly
retrospectives,
as
well
as
the
monthly
retrospective,
the
twelve
nine
contribute
being
cancelled,
as
well
as
the
worldwide
situation.
Carbon
19
is
resulted
in
reduced
morale
and
productivity,
Nico
and
there's
a
para
called
this
out
in
our
retrospective
and
there's
a
lot
of
support
for
this
I
think
it's
a
really
important
point
to
recognize
with
our
teams
that,
while
our
day-to-day
may
not
have
changed,
this
is
something
that
will
impact
people's
happiness.
Focus
supporting
their
families
will
take
time.
H
Thanks,
we
had
the
largest
numbers
of
delivered
issues
this
milestone,
but
we
also
had
the
largest
number
of
miss
deliverables.
This
milestone,
some
of
the
things
that
we're
planning
on
doing
is
to
trim
back
the
amount
of
work
that
we're
planning
for
each
milestone,
as
well
as
we're
starting
to
weight
issues
more
deliberately,
to
make
sure
that
before
we
take
on
any
work
that
it's
weighted
and
we're.
H
Also,
thanks
to
the
iteration
planning,
your
iteration
training
issues
that
Christopher
made
iteration
and
making
the
minimum
viable
changes,
become
much
more
to
the
forefront
in
our
grooming
sessions
and
we're
trying
to
make
sure
that
the
issues
that
we're
creating
are
as
small
as
possible
as
well.
We
think
that
will
help
address
the
missed
issues.
Curt
milestone,
thank
you
Chris
Wow,
so
now
I
think
has
to.
I
Their
ears
will
develop
early,
acknowledge
of
it,
but
there's
also
some
some
different
opinions
and
reviewers
during
the
process
that
caused
the
feature
to
slip.
The
milestone.
We
have
some
ideas
in
the
retro:
how
to
fix
this
going
forward,
but
probably
the
root
cause
is
just
too
large
of
an
issue.
I
need
to
break
the
issues
down
further
before
moving
forward
with
large
new
issues.
I
A
Chris,
just
a
note
also
we're
talking
about
if
we
see
a
lot
of
back
and
forth
potentially
pushing
to
synchronous,
so
I
don't
know
if
we
have
an
exact
rule
of
thumb,
but
it's
something
to
also
think
about
I've
seen
a
number
of
groups,
who've
used
that
technique
to
just
say
hey
if
it
gets
above
tens,
back-and-forth
sore,
five
back
and
forth
sore
whatever
the
number
may
be.
Just
think
in
those
terms,
so
additional
thing
to
to
consider
there
Cole
I
think
Lindsay
is
up
under
training.
So.
J
Is
discussed
during
the
defend
retrospective
that
we
had
learned
the
lessons
over
twelve
nine
around
how
feature
flags
work
get
lab.
We've
got
a
very
new
team,
so
unfortunately,
this
was
a
learning
by
doing
but
just
understanding
what
feature
flags
are
there
which
are
powered
by
unleashed
and
which
are
powered
by
flipper
and
just
the
difference
between
those
two.
So
you
know
as
a
team,
we're
taking
time
to
stop
and
ensure
that
we
have
a
thorough
understanding
of
not
just
feature
flags.
J
K
So
we've
been
hearing
from
our
product
counterparts
that
they
don't
have
the
visibility
they
need
on
issue
progress
once
it
goes
in-depth
in
order
to
provide
that
visibility.
We're
trying
to
update
issues
weekly
with
the
progress
that
has
been
made.
We've
also
started
we're
also
going
to
start
using
the
health
status
update
feature
just
just
added
in
twelve
nine
I
linked
the
issue
describing
the
future
in
the
doc,
but
it
essentially
allows
us
to
easily
flag
issues
that
are
at
risk
of
going
off-track,
Christopher.
H
So
one
of
the
interesting
challenges
we
face
and
the
APM
team
is
teams
are
getting
really
empowered
to
break
things
down
into
smaller
iterations,
which
is
great,
but
sometimes
the
PMS
are
not
aware
of
what
scope
is
being
cut
and
decisions
are
being
made
which
creates
an
interesting
situation
where
PMS
feel
like
maybe
we're
we're
dropping
the
bar
in
quality,
but
really
we're
trying
to
iterate
faster.
So
we've
kind
of
create
a
new
process
around
this.
To
ask
engineers
to
make
let
PM's
be
the
DRI
for
the
final
decisions
for
cutting
scope.
H
They
will
probably
be
more
than
willing
to
cut
scope
but
making
sure
that
they're
aware
of
it.
It's
a
big
part
of
figuring
out
how
to
schedule
the
follow-ups
and
we're
also
experimenting
with
a
better
process
for
improving
our
backlog,
refinement
to
make
sure
we're
reducing
the
cutting
of
scope
during
active
development
and
try
to
do
that
before
hand.
A
A
For
holding
stuff
accountability
comment
commence
one
of
our
Champions
of
iterations,
so
we
really
want
to
thank
them
for
all
hard.
The
effort
they're
associate
with
Sarah
really
want
to
make
sure
to
emphasize
here.
Transparency
is
super
key
and
communication
is
super
key.
We
really
want
to
make
sure
that
product
management
is
in
the
loop
on
any
changes
in
content
associated
with
it.
It's
okay
to
collaborate
and
have
a
discussion
around
it.
A
It's
a,
but
we
need
to
make
sure
that
those
are
happening
because
we
don't
want
them
surprised,
particularly
if
we're
thinking
about
scope,
considerations
around
this
and
as
part
of
our
iteration
process
post
process,
always
look
back
and
do
an
iteration
retro
and
how
we
could
have
potentially
caught
this
up
earlier.
So
I
really
appreciate
the
transparency
comment
and
the
further
refinement
in
communication
and
other
aspects:
great
work,
cool,
we'll
move
to
Darva
if
she
is
available
under
item
4
for
collaboration.
Yes,.
L
Thanks
Christopher
I'd
like
to
say
that
we've
seen
some
success,
working
with
QA,
with
quad
planning
and
also
with
security,
and
we're
hoping
to
start
to
measure
that
success
to
see
how
effective
our
collaboration
has
been.
I
myself
have
already
seen
a
few
issues
and
12:9
that
an
12:10
that,
without
their
help,
we
would
have
missed
so
I'm,
really
excited
about
how
well
it's
going.
A
Well
and
a
special
note,
Kyle
in
engineering
productivity,
probably
is
highly
interested
in
how
you
might
measure
this
to
figure
this
out,
because
this
is
something
that
we
may
want
to
consider
across
the
organization.
So
please
share
your
findings
with
Kyle
in
that
process.
Cool
number
five
under
planning,
Nick
yup.
M
Yeah
yeah,
the
G
team,
can
can
better
align
the
work
that
ends
up
getting
done
in
a
milestone
with
the
plan
deliverables.
At
the
outset,
the
team
does
a
weekly
scheduling,
call
where
we
go
through
our
issue
board
of
active
geo
issues
and
there's
concern
that
the
the
team's
effort
is
not
showing
correctly
through
this
board
and
that
having
a
boards
of
mert
bus
excuse
me,
a
board
of
merge
requests
could
help.
With
with
this
visibility,
issue,
I
went
to
an
issue.
M
A
Cool
and
judge
Rome
you're
up
next,
so
during
12.9
there
were
times
where
engineering
was
waiting,
more
blocked
by
design.
One
of
the
things
which
we
want
to
do
is
ensure
that
product
and
design
is
always
working,
one
milestone
ahead
and
that
we've
got
a
backlog
built
which
we
can
simply
pick
off
of
to
fix.
This
Christopher
actually
pointed
me
to
Sam
Goldstein's
team
and
the
planning
process,
which
they
use,
where
they're
pretty
explicit
about
the
timing
and
feedback
process
for
designs.
A
So
we're
gonna
look
at
implementing
something
like
that
in
our
teams,
yeah
and
just
to
highlight
here.
We
really
want
to
try
to
keep
as
close
as
we
can
to
the
official
process,
but
teams
are
flexible
to
consider
options
associated
with
it
and
the
clear
aspect
that
we
want
to
make
sure
is
that
you're
documenting
them
and
you're
treating
them
as
experiments.
If
we
think
overall,
we
need
to
move
to
it
or
a
particular
team
has
a
way.
A
A
So
please
keep
that
in
mind
as
you're.
Considering
this
Sam's
is
a
great
example,
where
he's
asking
for
a
little
bit
more
time,
upfront
for
effectively
iteration
associated
with
the
efforts
and
I
think
this
is
a
good
good
thing
for
us
to
consider,
is
kind
of
we
move
into
this
process
cool
and
then
go
ahead.
I'm.
A
N
Easy
was
before
me,
but
I'll
just
go
really
quick.
There's
been
a
lot
of
comments
about
how
confusing
feature
flags
are,
and
even
in
this
retro
in
the
last
retro
over
and
over
all
sorts
of
different
issues.
So
I've
created
a
training
issue
here.
Please
please
participate
in
this
issue
because
my
team
has
not
experienced
the
same
issues
as
other
teams
and
I
really
need
collaboration
there,
Lindsey
I
think
some
of
your
lessons
learned
would
be
really
valuable
to
to
provide
training
on
these
different
areas.
A
O
A
Easy
just
the
last
section,
the
kind
of
copper
improvements
for
the
next
release,
the
track
I
selected
a
few,
if
folks
have
other
ones
that
they
want
to
track
either
for
themselves
or
for
individuals.
Please
feel
free
to
add
in
the
review
with
the
next.
The
next
retro
ones.
I
highlight
is
the
feature
flag,
training
because
I
think
that's
a
good
one
for
us
to
understand.
Overall,
as
an
organization,
a
couple
items
that
kind
of
made
it
through
the
cracks
since
the
last
retro
one
of
them's
mine,
so
I'll
take
responsibility
for
that.
A
One
Darva
is
on
the
quad
planing,
since
that's
new
to
us,
I'd
love
that
your
feedback
on
how
that's
going
or
not
going
well
and
also
not
going
well
from
the
prospective.
So
we
can
continue
to
iterate
on
there
proving
that
process
and
then
cz
I
think
this
is
an
important
from
an
elastic
search
perspective.
A
Basically,
how
do
we
enable
a
future
across
comm
but
in
a
systematic
and
same
fashion
as
the
way
I'll
kind
of
describe
it,
so
that
you
know
we
minimize
the
impact
and
basically
give
customers
proper
isolations,
so
they
are
impacted
by
problems
or
challenges
that
we
may
have
in
the
process
of
rolling
out
a
new
future,
a
piece
of
functionality,
cool
that
hits
all
the
items
in
the
retrospective.
Are
there
any
items
that
folks,
with
the
like
to
highlight
in
the
one
minute
and
30
seconds
left
before
time.