►
Description
B
A
A
This
starts
to
get
really
important
as
we
get
closer
to
burn
down
and
burn
down
when
we
start
like
making
decisions
because
of
the
amount
of
time
it
takes
me
to
process
in
the
blood
recordings.
It's
so
much
better
if
we
have
the
decisions
written
down
someplace
first,
but
we
can
look
back
on
okay,
I
think
we
have
everybody
here
and
nobody's
told
me
in
my
I'm
on
mute.
Yet
so
I
think
you
can
hear
me
so
I'll
do
the
doodly-doo
where
I
say
hey!
A
Everybody
today
is
Monday
February,
11th,
2019,
I'm
Ehren
of
sig
beard,
and
this
is
the
kubernetes
114
release
team
weekly
meeting.
This
is
week,
six
we're
participating
in
a
publicly
recorded
discussion
about
the
release,
so
we're
going
to
be
adhering
to
kubernetes
code
of
conduct,
which
boils
down
to
don't
be
a
jerk,
and
we
can
all
watch
ourselves
not
being
jerks
on
YouTube
but
later
so
we
are
at
the
halfway
mark
of
the
release
yay.
How
does
everybody
feel
about
that?
I
see
quite
a
golfclap
from
death?
A
A
There
are
probably
a
few
more
of
those
to
come
when
I
set
up
calendar
and
writes
for
the
burned
down
meetings.
I
took
a
best
guess
at
the
appropriate
level
of
Ackles
for
the
sundry
mailing
lists
that
we
have
so
historically,
we
used
a
mailing
list
called
kubernetes
milestone,
burned
down,
which
included
people
release
team
people,
not
in
the
release
team
people
who
seek
release
people
bought
into
place,
which
beautiful,
trying
to
change
it
around
his
time.
Thomas
that.
A
Okay,
thank
you
thank
you
for
that.
So,
basically,
let
me
find
the
issue.
I
tried
to
sort
of
write
down
what
I
think
all
the
ackles
are
supposed
to
be
for
all
of
the
various
mailing
lists,
because
every
time
I
kept
trying
to
change
things
around
I
had
heard
that
people
ask
me
like
why
they
didn't
have
access
so
I'm,
trying
to
document
it
and
then
follow.
It
can
insistently.
A
Please
hold
Internet
so
pasting
a
link
in
chat,
so
I
think
that's
the
way.
It's
supposed
to
be
the
the
real
tricky
thing
here
is
I'm
trying
to
remove
people
who
have
direct
access
to
Doc's
as
part
of
this.
So
if
you
were
listed
as
having
direct
write,
access
to
the
enhancement
spreadsheet
or
the
CI
signal
report,
I
removed
you
because
you
should
be
in
the
kubernetes
release
team
group,
which
should
give
you
direct
access
to
Doc's
and
if
that's
not
working,
let
me
know
and
we'll
revisit
that.
A
But
the
goal
is
everybody
on
the
release.
Team
leads
and
shadows
should
be
in
the
release
team,
Google
Group,
which
should
give
them
access
to
all
the
things
the
shared
calendar
is
editable
by
the
cig
release
chairs,
because
the
release
team
is
a
sig
release,
project
and
myself
and
the
release
lead
shadows
and
then
the
major
milestones
like
code
freeze,
code,
thaw
and
all
that
good
stuff
are
shared
with
the
community
calendar.
So
it
shows
up
with
all
the
other
community
meetings.
A
Mike,
our
paya
is
not
here,
but
he's
going
to
be
leading
an
EU,
friendly
time
zone
version
of
this
meeting,
which
will
happen
Tuesdays
at
4
p.m.
GMT
8
a.m.
Pacific
on
that
seems
to
be
the
option
that
everybody
voted
for
the
most
in
the
doodle,
so
I
created
a
recurrence.
Ooh
meeting
did
all
the
needful.
A
Does
everybody
have
the
invite
for
that?
Does
that
all
look
good
I'm
seeing
general
head
nods
and
then,
as
far
as
how
to
run
the
meeting
my
suggestion,
based
on
what
I've
heard
from
other
release
leads
in
the
past?
Is
you
don't
necessarily
need
to
copy
paste
the
full
agenda
that
has
call-outs
from
all
of
the
different
roles?
A
The
final
large
thing
for
me-
and
this
is
kind
of
where
I
need
everybody's
help-
is
now
that
we
are
in
week
six.
We
start
to
focus
on
what
is
chief
of
this
release.
What
does
it
actually
look
like
and
is
it
on
target
so
we're
gonna
start
by
looking
at
all
of
the
caps
and
I
asked
everybody
to
put
in
a
test
plan
in
the
cap
and
I
also
asked
their
debility
to
put
in
some
considerations
for
a
big
downgrade
and,
first
importantly,
for
us
a
checklist
that
said.
A
Yes,
whether
or
not
this
feature
is
good
to
go
and
what
I
need
is
everybody
to
actually
read
that
checklist
and
make
sure
it's
workable
for
you
so
enhancements
those
the
cap
makes
sense.
Do
you
understand
what
the
feature
is?
Will
you
be
able
to
talk
competently
about
this
feature?
It
comes
as
we
decide
what
the
top
five
enhancements
are
to
talk
about
for
this
release.
Ci
signal.
Does
the
test
plan
in
the
cap
actually
look
good?
A
Could
you
look
at
that
test
plan
and
say
yes,
I
know
whether
or
not
this
features
or
this
enhancements
tests
have
been
written
and
I
know
whether
or
not
this
enhancement
is
passing
its
tests
issue
triage.
Are
you
going
to
be
able
to
identify
all
of
the
issues
that
are
related
to
this
enhancement
and
then
myself
and
pretty
much
everybody?
Does
the
checklist
actually
make
sense?
Oh
wait.
Let
me
call
out
another
thing:
Docs
and
release
notes
like
our
Docs
and
release
notes,
part
of
the
graduation
criteria.
A
A
D
Hey
everyone,
so
we
have
four
enhancements.
We
are
of
40
currently
tracked
and
of
those
40
38
have
an
a
a
kept
linked
that
the
PR
has
been
merged
for
the
two
that
don't
submitted
exceptions
and
those
exceptions
were
approved
and,
in
addition
to
those
two
there
were
seven
additional
or
addition
to
those
two.
There
were
five
additional
exceptions
approved
for
a
total
of
seven.
D
We
have
12
alphas,
14,
betas,
10,
stables,
and
this
week
would
I'll
be
focusing
on
will
be,
as
Aaron
just
mentioned,
reviewing
caps
the
status
and
content,
and
we
also
want
to
start
following
up
on
open,
PR,
Xin
KK,
for
each
enhancement
to
see
as
we
get
closer
towards
burndown.
How
are
these
open,
pr's
tracking?
Are
they
looking
like
they'll,
be
merged,
or
is
anything
starting
to
look
more
at
risk?
Any
questions
from
anyone
on
the
call.
A
D
So
we're
gonna
follow
up
with
all
12
of
those.
So
far,
I've
hit
a
couple
this
morning
and
the
responses
I've
been
getting
have
been.
Oh,
we
haven't
updated
the
PR.
We
need
to
go.
Do
that
so
continuing
to
follow
up
on
those
just
to
check
if
it
is
all
missing,
PRS
or
if
there
are
some
that
still
need
work
to
become
implementable.
A
E
A
So
I
would
like,
for
example,
I
suggested
it.
Thank
you
for
the
lazy
link
like
I,
suggested.
We
change
the
caps
column
so,
instead
of
linking
to
PR
for
the
caps,
it
now
calls
the
status
out
for
the
caps.
That's
how
I
noticed,
which
was
implementable
and
which
was
something
else
and
I
think,
because
we
have
caps
for
everything.
We
probably
don't
need
proposals
for
everything
and
then
like
from
an
issue
triage
perspective
where
I
had
asked
like.
F
A
I
also
don't
want
to
mess
with
people's
world.
If
it's
like
it's
in
a
place
where
they
like
it,
they
use
it.
They
need
it.
But
it's
confusing
to
me
so
I'm
trying
to
make
it
more
useful
to
me
personally.
I
just
wanna
make
sure
I'm
like
not
stepping
on
people's
totes
and
people
feel
little
as
though
they
can
volunteer
changes
to
make
their
lives
more
useful
or
make
their
lives
better
or
make
these
to
make
the
tools
more
useful
for
them.
A
G
G
A
user
of
the
spreadsheet
this
is
lucky
I,
think
the
changes
you've
made
make
the
user
experience
a
little
bit
easier.
So
thank
you
for
making
them
I.
Think
just
having
the
state
going
to
cap
has
made
it
easier.
We
used
to
track
obviously
community
proposals
before
that
which
added
a
lot
of
confusion.
H
H
H
The
master
upgrade
is
looking
a
little
more
concerning
in
that
there
are
five
jobs
failing
at
the
moment,
they
do
boil
down
to
the
same
test
failure,
so
we
need
to
follow
up
with
sigil,
sig,
Network
and
sequester
life
cycle.
The
reason
that
I'm
a
little
bit
more
worried
for
these
is
that
especially
the
the
two
first
ones
have
been
failing
for
quite
a
while,
so
we'll
work
together
with
the
SIG's
this
week
to
see
if
we
can
unblock
them.
H
I
think
this
is
mostly
it
in
terms
of
say
signal
at
the
moment.
I
did
have
a
question,
because
I
realize
that
I
haven't
seen
this
until
now
in
terms
of
getting
input
for
like
CI
signal
related
input
for
new
enhancement.
Is
that
something
that
happens
so?
Is
that
a
conversation
that
happens
with
a
release
team
or
how
do
we
find
out
the
test
once.
A
All
of
the
test
players
should
be
called
out
in
each
of
the
cups,
so
you
should
go
through
the
enhancements
spreadsheet.
The
caps
are
linked
in
the
cap
state
column,
so
the
word
is
the
kept
statics.
So
if
implementable
great,
you
should
definitely
look
that
and
then
something
that
cap
should
be
a
heading
called
test
plan.
Okay,.
H
H
I
mean
for
these
specific
ones,
after
being
the
the
sakes
and
slug,
but
we
both
sort
of
shared
how
this
happened
like
it
was
I
think
them.
Sorry
I've
talked
to
my
starters
about
the
conversation
they've
seen
it
happen
in
real
time
and
then
my
shutters
have
followed
up
another
issue
so
in
terms
of
following
the
process,
they've
been
very
involved
for
these
two
specific
issues.
I
happened
to
do
it
was
that
the
question
sorry
yeah.
H
So
they've
been
they've
been
sort
of
like
in
public
channels.
At
the
moment
we
haven't
spoken
in
private
shots
or
anything
like
that
until
now,
at
least
I'm
not
sure.
If
there's
anything
that,
if
there's
anything
else,
we
can
do
to
unblock
the
SIG's,
because
I
assume
I
don't
know
they're
quite
busy.
A
A
All
right,
let's,
let's
try
keeping
an
eye
on
that
and
see
if
we
can
get
a
little
bit
more
green
by
the
end
of
this
week.
Up
you
know,
first
half
of
the
release,
I
think
generally
folks
have
been
a
little
more
relaxed
about
this.
It
has
always
driven
me,
nuts
that
we
cut
release
builds
even
though
the
tests
are
failing.
A
I
focused
my
energy
on
making
sure
everybody
had
a
kept,
rather
than
focused
on
everybody,
had
passing
tests,
but
now
we're
going
to
sort
of
start
to
flip
over
to
like
we
really
really
really
need
passing
tests,
especially
as
we
get
closer
to
burn
down
and
code
freeze.
It's
going
to
be
really
helpful
to
know
what
is
actually
failing
for
real,
because
somebody
introduced
a
change
that
broke
things
versus
what
our
failures
that
are
meaningless.
J
K
J
K
The
idea
was
to
define
the
use
case
of
those
dashboards,
so
we
were
thinking
of
just
to
copy
github
tokens
issues,
and
just
noting
over
over
the,
since
they
are
used
in
different
projects,
is
just
taking
a
snapshot
on
a
daily
basis
or
an
even
smaller
interval
like
6
hours,
maybe,
and
instead
of
keeping
those
because
they
are
query
results.
So
we
just
take
them
every
couple
of
hours
and
then
we
can
do
all
of
the
processing
that
we
need
to
do
on
them.
K
Mostly,
we
need
to
correlate
all
of
the
open
issues
in
the
various
repositories,
just
to
make
sure
that
they
are
assigned
to
the
right,
jex
and
right
kind,
and
there
that
we're
following
the
milestones,
keeping
snapshots
of
the
data
is
also
very
beneficial
later
on.
So
we
can
actually
track.
How
well
did
we
do
in
different
intervals
so
based
on
the
release
cycle
that
we
have?
K
A
A
J
B
L
Sure
it's
been
an
exciting
week
for
testin
FRA,
studying
with
the
yaw
streamers.
We
emerged
about
113
PRS,
which
is
a
bit
lower
than
last
week,
but
is
significantly
higher
than
our
initial
weeks.
Then
we
created
the
PR
for
setting
up
the
CI
jobs
for
this
release.
I
have
a
question
regarding
that,
but
I
will
circle
back
to
it
later
other
than
that
the
wall
also
created
a
PR
for
automating.
Most
of
this
work,
so
future
releases
will
have
a
much
easier
time
setting
up
the
CI
jobs
regarding
prowl
status.
A
I'm
sure
I
don't
know
if
anybody
else
he
was
previously
testing
for
us
on
here.
I
have
seen
it
done
that
way
to
be
conservative
about
the
amount
that
we
change
all
at
once,
so
removing
the
old
jobs
and
making
sure
those
are
all
turned
down
is
one
set
of
things
that
you
can
watch
and
then
also
verifying
that
the
new
jobs
are
set
up
correctly
and
work
is
another
set
of
things
and
they're
kind
of
unrelated
to
each
other,
but
so
I've
seen
it
just
for
ease
yeah.
A
It's
like
it's
just
a
smaller
change
and
I'm
a
huge
fan
in
general
of
smaller,
more
frequent
changes.
But
it's
your
test
infra,
you
roll!
How
you
want
if
I
start
seeing
a
lot
of
bumps,
maybe
I'll
insert
some
more
opinions
but
otherwise,
like
whatever
seems
to
make
it
easiest.
You
operate
things
and
manage
change.
L
F
Hey
everybody,
so
it
should
be
pretty
quick.
Nothing
to
a
major
to
update
everyone
on
two
weeks
from
now
is
where
we
say
that
we
need
a
PR
opened.
It
doesn't
need
to
be
the
full
docs
just
needs
to
be
a
placeholder
PR,
so
over
the
next
two
weeks
here,
I'll
be
dividing
up
the
40
or
so
caps
amongst
the
shadows
and
myself
ends
up
being
roughly
you
know
around
8
or
so
caps
per
person.
F
Some
boom
obviously
don't
require
any
any
changes,
but
some
of
them
will
require
a
little
bit
of
prodding
for
docs
and
then
obviously
technical
reviews
and
all
that
fun
stuff,
I,
also
pigs.
The
release
notes
team
today
trying
to
sync
up
with
them,
and
that's
mainly
due
to
advice,
I
got
from
previous
docs,
leads
where
they
said
that
they
saw
the
release,
notes
once
that
went
out
and
saw
some
features
that
maybe
they
didn't
see
before
maybe
kept
solves
this
problem.
F
But
ultimately
the
goal
is
to
sync
up
to
make
sure
there's
nothing
in
there
release
notes
or
from
what
we're
seeing
that
I
might
need
Doc's
at
the
end
of
the
day.
So
we'll
be
looking
to
work,
hand
in
hand
with
them,
making
sure
that
everything
you
know
mens
were
there
I'm
sure
we
pretty
pretty
good
and
then
also
I
was
working
with
Stephan
Augustus
on
creating
a
PR
for
the
docs
handbook.
So
the
release
team
and
myself
we've
been
working
off
a
hack,
MD,
IO
dock
and
I
snuck
into
Steven.
F
F
So
that's
going
to
be
the
long-term
goal
and
finding
the
best
way
to
do
that
kind
of
with
the
community
consensus.
There
will
be
what
that
survey
comes
out
for
so
be
on
the
lookout
for
that
and
I
wanted
to
bring
it
up
here
in
case.
There's
anything
that
somebody
specifically
wanted
me
to
include
in
that
survey
for
feedback
I
generally
plan
on
having
to
be.
You
know
less
than
five
questions
pretty
open-ended
about
how
to
handle
the
shadows
that
weren't
selected
any
questions.
A
D
A
C
I
M
Oh
yeah
I'm
here
just
just
unmuting
so
today,
in
about
an
hour
and
a
half,
we're
going
to
have
a
little
meeting
about
the
release
blog
and
the
Keene
media
messages.
So,
as
you
know,
I've
invited
Enhancement
release,
notes
airing
you
as
a
lead
and
then
the
shadows
for
communications.
Of
course,
anybody
on
the
release
team
can
join
this,
but
those
are
the
key
ones
that
keep
people
that
I
need
to
talk
to.
M
So
we'll
go
out
over
the
1.13
blog
that
we
wrote
as
well
as
a
media,
key
messages
and
then
we're
going
to
go
over
at
the
enhancements
list
and
discuss
what
the
114
blog
should
look
like,
and
what
questions
we
need
to
answer
for
any
media
stories
that
we're
doing
as
well
as
the
five
day
blog
series
and
which
features
that
should
include
again.
The
reason
I
call
this
a
brainstorm
is
a
lot
of
this
will
not
be
really
solidified
until
code
freeze,
but
we
want
to
get
kind
of
a
jump
start.
M
A
M
No
definitely
so
at
the
very
top
of
the
release
blog.
We
have
people
that
need
to
approve
this,
so
obviously
Aaron
you're
on
their
option,
enhancements
leads
is
on
there,
as
well
as
the
different
things,
so
we're
really
calling
out
enhancement
features
under
a
certain
fit
things.
So,
for
example,
Windows
are
storage
that
cig
needs
to
sign
off
on
this
before
we
publish
so
they
can
go
in
and
say.
Actually
this
is
not
accurate.
This
is
not
accurate
and
then
the
enhancements
lead
goes
in
here
and
says
ooh.
M
M
Yes,
yes,
that's
part
of
it.
That's
part
of
the
discussion
and
we
actually
don't
have
to
do
a
5
as
well,
so
for
1.13
one
feature
dropped
out,
I
think
that
was
Windows.
So
we
did
it
for
blogs
and
then
I
want
to
say
around
110,
since
we've
had
a
number
of
features
drop
out
and
so
I
think
he
only
has
three
one.
So
it
doesn't,
we
shoot
for
five,
but
it
doesn't
necessarily
have
to
be
5.
M
A
N
Good
evening,
everyone
so
tomorrow
is
the
cut
for
alpha
3,
nothing,
nothing
special
there
I'm
giving
my
shadows
right
off.
This
call
a
run-through.
What
what
that
is
and
how
you
cut
the
release,
then
the
other
thing
is
the
packaging
story.
I
talked
to
various
people
on
C
cluster
lifecycle
and
they
started
doing
stuff,
which
currently
seems
to
be
like
checking
what
what's
the
way
forward
now
like?
How
do
we
build
the
pack?
How
do
we
want
to
build
the
packages
and
so
on?
That's
mainly
in
the
linked
issue.
N
There
is
a
lot
of,
or
there
was
a
lot
of
discussion
going
on
and
they
said
they
will
track
their
progress
there.
So
I
will
stay
engaged
with
them.
On
this
issue
and
yeah,
we
will
see
how
it
goes.
I
told
Tim
Sinclair
that
I
don't
see
anything
coming
in
for
this
release,
but
my
plan
is
to
set
ourselves
up
so
that
we
can
bring
in
new
release
stuff
in
115,
and
then
there
is
the
related
question
here.
N
So
I
started
based
on
actual
the
issue
where
the
discussion
is
now
going
on,
mostly
based
on
that
I
started
to
writing
the
cap
and
try
to
understand
all
the
bits
and
pieces
and
to
the
best
of
my
knowledge
how
we
could
move
forward
now
see
cluster
life
cycle
does
not
necessarily
want
to
like
work
on
this
cap.
The
cap
is
not
merge.
The
cap
is
not
in
implementable
state,
so
my
question
is:
how
do
we
want
to
go
forward
there?
N
A
N
A
N
It's
the
thing
that
I
wrote
in
the
cap.
What
is
still,
and
the
discussion
is
like
details-
do
they
use
to
lay
or
do
they
use
to
be
and
like
a
lot
of
details
in
the
cap
are
just
things
I
thought
thought
about,
but
not
necessarily
how
it's
really
implemented
in
the
end,
they
are
currently
hashing
that
out
what?
How
does
it
really
look
by
like
experiments
and
and
hashing
on
stuff
on
on
the
issuing,
but
and
so
again,
I
open
the
cap.
Sorry
go.
N
The
thing
is,
I
opened
the
cap,
they
never
thought
they'd
wanna
or
they
never
thought.
They'd
need
a
cap,
so
they
started
out
on
this
issue
and
now
it
would
either
be.
We
tell
them.
No,
you
need
a
cap
or
you
need
to
update
my
cap
or
we
say:
ok,
we
will
close
the
cap,
we
don't
force
you
to
use
a
cap
and
we
just
kind
of
check.
What's
in
the
issue,
if
you
see
what
I'm,
okay.
A
That's
I
hear
that
statement.
I
have
seen
Timothy
Sinclair
get
a
little
pushy
about
this
stuff,
but
again
the
forcing
function
here
is.
If
this
is
a
change
they
want
to
land
in
this
release,
they
really
ought
to
have
it
documented
in
a
cap.
So
my
personal
difficulty
is
I'm
supposed
to
be
on
vacation
this
week.
So
this
isn't
something
I'm
really
supposed
to
allocate
a
ton
of
bandwidth
to
I
can
try
and
have
a
one-on-one
chat,
but
you
have
my
full
authority
to
push
back.
Everybody
here.
Does.
N
A
Don't
it's
not
like
I
want
to
stop
prototyping
and
experimenting
and
stuff
it
purely
comes
down
to.
Is
this
a
change
that's
going
to
land
as
part
of
this
release?
If
so,
I
need
to
see
test
plan,
graduation
checklist,
all
that
stuff,
so
that
I
can
understand
if
it's
actually
on
target
as
we
approach
code
freeze
but.
K
A
I
personally
haven't
had
time
to
look
at
the
scope
of
of
all
this.
Like
I,
don't
know.
My
hope,
though,
is
we
could
it
seems
reasonable
to
ask
for
an
exception.
Certainly
this
long
before
code
freeze
and
talk
about
how
what
they're
doing
ties
in
to
your
cap-
and
we
could
say
that
that
is
the
work
that
is
intended
to
Luria
and
in
114.
A
A
Like
it,
for
example,
in
terms
of
prior
art,
one
of
the
other
caps
that
we
have
discusses
how
we're
going
to
break
up
the
test
star
ball
from
one
mando
tar
ball,
that
has
the
tests
for
every
single
platform
into
the
tar
ball
per
platform.
This
matches
what
we
did
for
the
uber
Nettie's
tar
ball,
but
there
were
still
enough
finicky
details
in
it
that
it
was
really
worthwhile
to
write
out
a
cap
to
detail
what
was
going
to
land
in
this
release
know.
A
A
So
generally
I've
given
y'all
the
remainder
of
these
minutes
to
do
with
whatever
you
want,
and
we
can
do
that
again,
but
one
thought
I
had
was
since
we
are
at
the
halfway
mark
and
we
have
38
people
on
this
call
I'm
kind
of
curious
how
you
think
it's
going,
and
so
we
have
a
retro
doc
that
Jeff,
if
you
could
be
my
Savior
again
and
maybe
help
type
things
out
in
the
retro
doc.
If
we
mentioned
some
things
which
I'm
going
to
link
in
chat.
A
F
In
bug
triage,
we
haven't
really
begun
in
earnest,
so
it
feels
a
little
bit
hard
to
be
able
to
judge
whether
or
not
we're
being
useful
or
not.
I
will
say
that
you
know.
Amico
has
been
good
at
communicating
with
us
and
using
the
time
that
we
have
before
our
work
really
begins
to
try
and
help
out
the
project
as
much
as
possible.
So,
while
I
haven't
been
able
to
contribute
to
the
automation
work
as
much
as
I'd
like
I,
think
that
we've
been
effectively
used
pretty
well.
G
Would
you
know
Claire's
been
great
with
communicating
keeping
us
up
to
date
with
all
the
process
and
then
delegating
out
work
items
to
there's
three
shadows,
so
I
would
say:
Claire
has
done
a
fantastic
job
with
giving
us
all
bits
of
work
and
control
and
then
communicating
the
bits
throughout
the
process,
as
we
all
come
to
learn
it
so
kudos
to
Claire,
but
I
think
I'd
certainly
feel
empowered
as
a
shadow
with
pieces
of
the
process.
So
Claire
has
done
a
great
job
with
that.
O
All
right,
so
this
is
a
Sunday
for
sick
dog.
So
far,
this
is
my
first
introduction
to
cognitive
inside
of
the
cabinet
itself,
but
Jim
are
doing
very,
very
well
in
terms
of
communication,
but
I
see
it
notification
like
1:00
a.m.
in
the
morning
pipe
whoo,
so
it's
been
very,
very
liking
in
terms
of
communication
and
guidance.
That's
Molly.
P
Oh
Milan
with
the
CIA
signal
theme,
a
Maria
has
been
really
amazing
and
funny.
If
anything.
Another
comment
that
I
would
like
to
add
is,
for
example,
when
Jove
s,
when
the
previous
team
was
looking
for
shadows
and
and
everything
for
114.
What
was
it
like?
200
people
showed
up
and
volunteer
for
something,
at
least
in
CI
signal.
It
will
be
I
think
a
little
bit
useful
to
expand
the
team.
P
A
I,
don't
know
if
chalet
I
don't
know
in
terms
of
shadows,
but
if
you
know
actually
a
get
a
lot
more
people
so
and
distribute
the
work.
So
you
know
if
somebody
wants
to
become
a
new
contributor
and
you
can
tell
him
hey.
We
just
open
this
new
issue,
because
something
is
failing.
A
take
ownership
on
that
a
work
with
the
sig
like
directly
work
with
this
again.
You
know
you
get
something
out
of
the
process
and.
A
A
Except
now
we
have
a
CI
signal
role.
That's
now
spread
out
amongst
the
team
of
people
where
you
are
responsible
for
figuring
out
what
tests
are
failing
and
then
who
those
tests
belong
to
you
and
then
asking
them
kindly
to
please
fix
their
tests,
and
part
of
me
wonders
if
this
isn't
a
responsibility
that
should
actually
live
with
the
sig
itself,
where
some
SIG's
have
I
know.
A
Sig
docks,
for
example,
has
a
PR
Wrangler
role,
where
it's
their
responsibility
to
make
sure
that
all
of
the
PRS
are
generally
headed
in
the
right
direction,
but
just
for
their
sig
it
would
be
useful
or
educational
to
understand
how
many
sakes
actually
are
aware
of
what
tests
they're
responsible
for
and
if
they
have
people
dedicated
to
keeping
those
tests
passing
or
not,
because
I
feel
like
George.
What
you're
asking
for
for
a
larger
team
of
CI
signal.
A
People
could
also
be
satisfied
by
having
all
the
SIG's
responsible
for
keeping
their
tests
passing,
but
it's
it's.
You
know
now
push
versus
pull
mechanism,
I
guess,
and
it's
definitely
something
that
would
be
super
helpful
because
again,
this
is
something
we
have
tried
to
automate
a
way
in
the
past.
So
we
have
tons
of
data
about
what
tests
are
passing
and
failing,
and
we
can
have
BOTS
Auto
file
issues
based
on
all
this.
But
people
don't
seem
to
pay
attention
to
that.
For
some
reason
you.
C
F
F
I
think
there
was
probably
like
between
I,
say
thirty
to
fifty
in
Docs,
so
you
know
we
selected
the
five
shadows
and
then
the
remaining
it
took
ons
suggestions
from
the
cig
release,
lack
where
he
was
sent
all
the
people
that
weren't
there
selected,
even
if
they
chose
they,
wanted
to
be
involved
in
multiple
roles
and
said:
hey
just
because
I
don't
have
time
or
capacity
to
take
you
under
as
a
shed
or
a
mentor.
It
doesn't
mean
you
can't
come
to
the
sig
release.
F
Meetings
doesn't
mean
you
can't
hang
out
in
sigdoc,
so
you
can't
pick
up
issues
be
part
of
the
community
and
thinking
that
Hans
is
doing
his
way.
I'm
doing
it
my
way,
I'm
sure
some
of
the
other
leads
are
doing
it
your
own
way
or
you
know,
however,
you're
doing
this
I
think
that
that
process
is
more
of
like
a
sig
release.
F
It
contributes,
you
know,
I,
guess,
question
or
problem
to
solve
where
I
think
the
response
to
those
who
aren't
selected
is
is
a
big
opportunity
to
get
more
people
in
the
community
and
I
wanted
to
see
what
people's
thoughts
were
on,
how
we
move
them
from
generally
interested
into
being
actively
involved,
even
if
they're
not
part
of
the
release
process.
So
that's
that's
where
my
mindset
is
but
I
don't
have
any
questions
to
speak
of
yet,
but
it
won't
be
more
than
five
I.
Don't
think.
A
I
completely
agree
with
that,
like,
like
part
of
me,
wants
to
say
that
it's
almost
a
kinder
gentler
ride,
if
you're
not
on
the
release
team,
because
things
will
get
pretty
intense
as
you
get
to
burn
down
to
actually
keep
up
with
everything.
So
if
you
just
want
to
help
out,
there's
so
many
places
on
this
project
that
people
could
help
out,
I'm,
not
sure
that
there's
anything
super
special
about
the
release
team.
Although
that
said,
we
do
seem
to
have
a
bit
of
a
leads
and
shadows
thing
going
on.
A
But
it's
unclear
to
me
how
much
other
SIG's
do
that
sort
of
the
explicit
mentorship
thing,
but,
as
you
said,
how
how
that
relationship
works
between
lead
and
shadow
is
very
dependent
upon
the
person
who's
the
lead,
maybe
maybe
some
shadows
feel
like
they're,
not
getting
any
mentorship
at
all
from
their
leads
or
some
shadows
feel
like
yeah.
They
have
a
great
you
know,
weekly
standing
meeting
and
in
a
wonderful
relationship
and
they're
being
taught
all
sorts
of
things
kind
of
unclear
me
what
level
of
consistency
we
have
there
and
what
the
right
balances.
A
F
Yeah
and
one
last
thing,
I'll
add:
if
there's
anyone
out
there
as
a
leader
or
shadow
who
has
any
questions
we'd
like
to
garner
some
feedback
on,
feel
free
to
leave
a
comment
on
that
issue
that
I
linked
in
the
meeting
today,
so
there's
anything
that
you're
interested
in
getting
feedback
on.
Like
I,
said
I'm
gonna
send
us
out
to
the
cig
release
team,
or
this
a
release
channel
on
slack,
probably
contributes
and
then
I'm
gonna.