►
From YouTube: Kubernetes Meet Our Contributors! 20190807
Description
When Slack seems like it’s going too fast, and you just need a quick answer from a human...
Meet Our Contributors gives you a monthly one-hour opportunity to ask questions about our upstream community, watch interviews with our contributors, and participate in peer code reviews.
See https://github.com/kubernetes/community/blob/master/mentoring/meet-our-contributors.md for more information
A
A
So
people
always
asked
us:
can
I
have
a
mentor
upstream
and
mentors
unfortunately,
are
few
and
far
between
based
on
time
and
all
kinds
of
other
open
source
time
related
activities,
so
we're
here
once
a
month
giving
anyone
advice
who
said
hey
I
would
I
would
love
a
mentor
whether
you
are
new,
whether
you've
been
on
the
project
for
four
years.
Whatever
it
is,
will
answer
your
questions,
whether
it's?
How
do
you
manage
your
time
with
the
source
projects
all
the
way
to?
Why
is
my
test?
Flaking
we'll
take
it
all.
A
We
might
not
know
it
all,
but
we
will
definitely
get
you
the
resources
to
do
it.
So
how
do
you
ask
of
questions?
We
hang
out
in
a
flak
channel
only
submit
any
flak
if
you
don't
have
an
invite,
that's
flack,
k8s,
dot,
IO
for
that
and
grab
that
invite
there.
The
channel
is
meet
our
contributors
and
if
you
would
like
to
remain
anonymous,
you
can
feel
free
to
direct
message
me
on
black,
which
is
more
than
appropriate.
We
get
about
60%
of
questions
that
way
and
feel
free
to
DM
me.
My
name
is
Paris.
A
You
see
that
on
your
screen,
right
now
same
display
name
in
black.
We
already
have
a
couple
questions
queued
up
for
today,
but
let's
talk
about
some
rules.
First
things.
First,
we
have
a
code
of
conduct
that
we
abide
by.
That
is
through
the
cloud
made
of
seafood
foundation.
This
is
TLDR,
please
be
excellent
to
each
other,
so
whether
you
are
in
a
question
on
slack
or
you
are
having
fun
with
us
on
Twitter
today.
A
A
The
kubernetes
contributor
summit
is
actually
going
to
be
taking
place
on
November,
17th
and
18th.
This
is
a
save
the
date.
So
if
you
do
upstream
contributing
for
the
project
and
interested
and
hanging
out
with
us
doing,
someone
conferences
and
other
really
awesome
things
check
the
kubernetes
dev
mailing
list
for
saved
dates
and
updates.
As
we
progressed,
registration
is
going
to
be
open
in
September.
A
B
So
I
am
heeta
I
work
as
a
software
engineer
at
a
German
startup
called
Itza,
but
I
work
remotely
from
India
in
communities
upstream
land
I
am
the
technical
lead
one
of
the
technically.
It's
pretty
contributed
experience
when
they
github
admin
team.
So
that's
the
team
which
has
owner
privileges
across
all
kubernetes
oryx
and.
B
Well,
I'm
also
involved
in,
like
some
other
states
like
SiC
testing,
Sigma
Lee's
capi
machinery,
all
the
things
with
at
all
and
then
maintain
some
projects
like
publishing
BOTS,
which
is
like
the
bots
responsible
for
sinking
code
between
the
main
kubernetes,
kubernetes,
repo
and
other
repos
like
api
api
machinery,
8.0
and
I
also
do
client
go
Ruiz's
and
yeah.
That's
pretty
much
it
and.
A
C
Well,
I
I
work
at
Red
Hat
as
a
principal
site.
Reliability
engineer
have
currently
a
team
lead
on
the
open
shift
on
Azure
or
the
ad.
Your
Red
Hat
open
shift
team,
so
we've
run
open
shift,
which
is
red
hats,
downstream
kubernetes
distribution
as
a
managed
service
on
Azure,
which
is
kind
of
fun.
I'm
in
terms
of
kubernetes
upstream
I've,
been
involved
in
sig
instrumentation
for
a
little
over
a
year,
working
on
metrics
usability
and
performance
of
various
metrics
collection
things
that
kind
of
thing
and
I
only
recently
became
a
project
member,
not
really
realizing.
C
A
Really
wanted
to
read
your
Twitter
bio
today,
actually
to
be
honest,
like
it's
one
of
the
best
Twitter
BIOS
that
I've
seen
what
Anna
I'm
gonna
I'm
gonna
get
it
out.
If
that's
okay
with
you,
because
I
think
it's
freaking,
awesome
and
I
wanted
to
even
say
it
in
like
a
serious
voice.
It's
like
queen
of
Debian
closure,
Empress
of
symbol,
versioning
Congress
of
Av
eyes,
director,
open
source
or
tree
Red,
Hat,
petition,
tolerant
available
and
not
consistent.
A
D
Jeff
I
work
at
the
University
of
Michigan
I
am
a
research
database
administrator
which
is
not
really
descriptive
of
what
my
day
job
actually
is,
but
in
terms
of
kubernetes
I
am
a
chair'
on
sig
UI,
so
the
kubernetes
bored
help
out
with
contributes
doing
things
like
streaming
and
I
am
a
slack
admin
and
a
moderator
on
YouTube,
thus
the
streaming
and
I'm
on
the
release
team
for
1:16.
Yes,.
A
Jeff
is
making
this
work
today.
So
if
our
stream
sounds
weird
lame,
Jeff,
don't
blame
anyone
blame
less
retros
anyway,
and
last
I
in
Paris
I'm
your
host
today,
I
work
at
Google
and
I
do
fun
stuff.
Like
this
all
the
time.
It's
wonderful
I
try
to
make
sure
that
our
lovely
kubernetes
project
here
has
contributors
who
are
happy
and
wonderful
and
doing
cartwheels.
So
that's
it.
A
C
C
C
It's
a
it's
a
very
friendly
environment,
I
think
to
like
try
to
learn
a
new
language
and
because
you
get
to
go
look
through
this
massive
big
new
codebase.
It
kind
of
gives
you
an
idea
of
like
what
does
idiomatic
go,
looks
like
so
I.
Don't
think
that
you
have
to
have
it
all
figured
out
before
you
jump
in
in
terms
of
finding
a
project.
I
would
recommend
like
get
involved
in
this
stuff
that
you
care
about,
don't
just
be
like.
C
I
have
decided
that
big
API
machinery
sounds
cool
and
I
want
to
get
involved
in
that,
because
it
may
turn
out
that
I
mean
I'm
sure
that
they
are
wonderful.
People
and
you'll
have
a
great
time
there,
but
if
you're
like
I,
don't
even
know
what
that
is,
but
I'm
gonna
like
show
up
and
be
like
mentor
me
tell
me
how
to
get
involved,
but
I,
don't
know
anything
about
you
and
I.
Don't
really
care
I
just
want
to
get
involved,
though
they
might
be
like
well,
we
don't
really
know
how
to
help
you
like.
C
What
are
you
interested
in
and
so
specifically
when
I
got
involved
in
the
project?
I
specifically
got
involved
with
sting
instrumentation,
because
I
was
working
at
a
company
with
these
really
large
on-premise
clusters.
We
didn't
have
any
of
this
cloud
stuff.
We
were
all
running
upstream
code,
I
needed
to
instrument
things
and
I'm,
just
kind
of
like
this
seems
like
the
Wild
West
out
here
like
I
guess.
Maybe
vendors
have
solved
this
problem
for
everyone
else,
because
the
open
source
ecosystem
isn't
looking
really
great
and
so
I
thought.
Ok,
how
do
I?
C
How
do
I
figure
that
out
so
I
went
and
looked
through
the
list
of
things
and
I'm
like
okay,
there's
a
sake
instrumentation,
maybe
I'll
like
try
to
show
up
to
their
meetings
and
ask
them
some
questions
like.
Why
is
this
thing
this
way?
This
doesn't
make
any
sense
to
me.
So
I
did
and
then
I
filed
some
bugs,
and
then
some
people
fixed
some
of
the
bugs
that
I
filed,
which
was
exciting
and
then
I
fixed,
some
of
the
bugs
that
I
filed,
which
was
also
exciting
and
kind
of
went
from.
A
Is
a
really
great
answer
for
those
that
are
just
joining
us?
We
have
a
couple.
We
have
quite
a
few
people
that
are
just
joining
us
live
right
now.
Thank
you
for
joining
me
all
gets
addition
to
meet
our
contributors.
We
aren't
in
the
middle
of
answering
a
question
from
someone
who
is
a
newbie
to
golang
once
the
start,
contributing
to
the
project.
I
wanted
to
get
some
on-ramp
advice,
Nikita.
What
about
you
anything
that
was
missing
there?
It
sounds
pretty
good.
A
A
B
Would
add
like
a
couple
of
more
points,
so
we
have
like
lots
of
issues
in
the
kubernetes
kubernetes
repo,
which
are
like
goal
in
Texas.
A
goal
in
is
a
tool
which
helps
in
like
linting
and
checking
if
the
co
files
are
following
some
format,
so
that
can
help
in
like
fixing
like
C.
You
just
basically
need
to
write
a
script,
and
it
will
tell
you
what
things
to
change.
B
B
There
were
like
lots
of
contributors
who
reach
out
to
me,
and
they
were
like
I,
don't
know,
go
what
do
I
do.
I,
usually
I
mean
most
of
them.
I,
don't
know
if
it's
like
always
this
way,
but
most
of
them
usually
are
into
Python.
So
I
usually
point
them
out
to
the
kubernetes
client.
/
people
write
about
the
repo
name,
so
there.
B
Oh
yeah,
so
there
are
lots
of
repos
in
the
kubernetes
/
client
org.
So
despite
them
is
Java,
Script,
typescript
and
there's
even
poor
I,
don't
know
how
a
poor
client
looks
like,
but
so
all
of
these
are
kubernetes
clients
and
the
Python
client
is
as
far
as
I
know,
the
most
after
one.
So
if
you're
interested
to
reach
out
to
that
they
hand
out
in
kubernetes,
client,
Channel,
I
guess,
and
they
also
have
like
a
public
triage
call
where
they
triage
issues.
I
think
that's
a
really
good
place
to
understand.
A
Shoutouts
to
our
SIG's,
who
do
public
live
triage,
yay,
API
machinery,
that's
a
great
way
to
get
involved
if
you're
interested
in
a
cig
which
you're
like
what's
sig
Paris
sig
is
a
special
interest
group.
That's
how
we
divide
and
conquer
this
project
is
how
people
get
work
done
they
own
individual,
debating
on
individual,
they
own
different
parts
of
our
code
base
and
that's
how
you
can
kind
of
get
the.
Where
am
I
interested
in,
and
what
am
I
interested
in
from
because
each
one
has
a
certain
interest.
A
We
did
have
a
question
in
the
slack
channel
about
the
announcement
and
for
those
that
are
just
joining.
The
announcement
was
that
we
are
doing
a
contributor
summit
in
San
Diego
right
before
Steve
Conn,
November,
17th
and
18th
folks
wanted
to
know
a
little
bit
more
about
that.
What
does
that
mean?
That
means
that
we
are
doing
two
different
kinds
of
shows.
One
show
is
for
new
feature
readers
who
are
interested
just
like
a
lot
of
you
right
now
that
are
asking
questions
in
doing
a
hands-on
intensive
workshop
in
landing
your
first
PR.
A
So
that
is
an
awesome
thing
that
we've
been
doing
now
for
four
or
five
q
cons
and
we've
been
onboarding
several
hundred
contributors.
That
way,
which
is
amazing,
and
then
we
will
have
an
opportunity
for
over
300
current
active
contributors
to
do
and
conferences
other
kinds
of
workshops
like
writing,
caps,
which
is
a
kubernetes
and
Hansen's
proposals.
Many
features
and
it's
gonna
be
a
really
good
time.
We
also
have
a
celebration
of
the
night
before
where
we
all
get
together
and
have
a
social
and
Nikita.
It's
already
excited
about
that.
A
That
is
one
of
my
favorite
parts
of
the
week
where
we
all
just
get
to
hang
out.
This
is
a
humongous
just
humongous
to
share
via
team.
If
you
want
to
put
it
from
a
numbers
perspective,
3,000,
committers,
that's
committers
and
then
31,000
contributors.
These
are
people
that
are
helping
us
triage
on
github,
help
doing
doing,
commits
doing
pr's,
etc.
So
we
have
a
humongous
contributor
base
and
all
of
us
would
hang
out
with
each
other,
which
is
awesome
all
right.
A
Up
yeah,
so
volunteers,
we
have
a
team
of
25
volunteers
y'all.
It's
amazing
amazing
team
and
we've
actually
modeled
it
in
a
very
similar
fashion,
as
we
did
with
the
release
team,
or
they
have
shadows
so
that
it
can
be
a
sustainable
event
for
us
go
forward
and
we
are
not
open
for
volunteer
opportunities
right
now.
However,
we
will
most
likely
near
closer
to
the
event.
A
So
if
you're,
not
on
that,
that's
covered
Eddie's
dev
at
Google,
Groups
comm
get
on
that
for
all
things
upstream
news,
and
then,
as
far
as
registration
is
concerned,
the
reg
will
also
go
out
there
and
that
will
be
in
the
beginning
of
September
and
registration
is
free
all
right
and
then
it
looks
like
we
have
in
slack
PI
AG,
and
then
we
also
have
a
+1
on
that
question
as
well.
What's
the
advice
for
someone
starting
out
when
good
new
issues
or
good
first
issues
are
not
available
or
already
picked
by
others?
A
B
Okay,
I
can
start
with
that.
So
usually
that's
that
shouldn't
be
the
case
on
repos,
because
we
want,
like
you
know.
Three
posts
should
have
good
first
issues,
but
I
know
that
it's
still
the
case
that
good
first
issue
is
a
sort
of
prayer
and
as
soon
as
an
issue
that
as
a
good
first
issue,
they're
they're
taken
up
by
someone
else.
We
also
have
another
label
called
as
how
it
went.
It's
kind
of
synonymous
with
this,
but
it's
kind
of
like
step
2
after
good
first
issue.
B
So
if
that's
taken
up
the
next
step
would
be
to
look
into
the
Help
Wanted
label.
What,
and
if
you
also
like,
don't
find
a
Help
Wanted
label.
What
I
would
usually
do
is
go
to
the
slack
channel,
for
that
say
for
the
project
that
you're
working
on
and
like
think
some
maintainer
czar
paying
some
approvers
and
tell
them
that
you're
interested
in
contributing
one
thing.
That
I
usually
find
hurtful
when
someone
asks
me
that
question
is
talk
a
little
bit
about
yourself.
B
Talk
a
little
bit
about
what
you
already
know
so
that
we
come
to
know
like
where
you're
starting
out
from
so
it's
not
like
completely
random
and
I,
can
sort
of
tell
you
the
next
steps.
Also
one
thing
that
helped
that's
helpful
is:
if
you've
already
done
some
research
that
you
just
like
gone
to
the
documentation
or
just
gone
to
the
existing
issues,
and
you
can
probably
like
I,
mean
just
give
us
an
introduction
about
what
you
want
to
read,
read
or
what
you
are
ready
to
answer
that
I
think
it's
easier!
B
First
away
tell
you
also
attend
meetings.
If
you
can
I
know
they're
at
like
part
time,
sometimes
like
I'm
from
India,
and
sometimes
it's
like
a
2:00
a.m.
3:00
a.m.
so
I,
don't
attend
them.
So
if
you
can't
attend
meetings,
it's
a
good
place
to
understand
what
are
the
current
topics
that
are
going
on
in
the
sake
and
the
current
concerns
so
that
you
can
jump
up
and
those
opportunities
sometimes
SIG's
also
like
put
out
the
call
for
volunteer.
Their
meeting
notes
are
good
place
to
figure
out
if
you
are
able
to
attend
meetings.
A
B
A
C
Think
there's
like
a
key
thing
to
note,
which
is
that
it's
not
evenly
distributed
in
that
sense
so
like
there
are
a
lot
of
different
SIG's
that
govern
the
project
and
they
are
all
of
different
sizes
and
different
levels
of
activity,
and
some
of
them
are
going
to
be
easier
to
get
involved
in
others.
So,
for
example,
I
got
involved
in
sig
instrumentation
and
it's
a
pretty
small
sig.
It
has
a
few
active
members.
It
meets
bi-weekly,
there's
not
like
a
huge
number
of
PRS
that
we
get
requested
to
review
that
kind
of
thing.
C
So
my
vice
there
would
be
try
to
find
the
pockets
of
the
project
that
you're
interested
in
and
that
maybe
aren't
necessarily
like
the
path
that
everybody
else
is
going
down.
If
you
want
to
be
able
to
make
an
effective
contribution,
because
those
smaller
SIG's
that
don't
get
as
much
interest
are
going
to
be
better
able
to
help,
you
contribute.
A
It
was
a
great
question
for
pretty
answers.
I'm
sorry,
yeah
I
feel
like
there's
just
tons
of
ways
to
get
involved
in
this
project,
and
even
if
you
like,
even
to
the
cut
the
goaline
question,
I
feel
like
it's
much
easier
when
you
get
in
and
see
how
the
how
the
cig
special
interest
group
is
run
and
like
kind
of
what
their
norms
are
and
then
it's
gonna
be
so
easy
for
someone
to
say:
oh
they
don't
have
this
or
something
along
those
lines.
A
I
feel
like
I
was
sort
of
a
fly
on
the
wall
to
SIG's
for
a
minute.
It
was
kind
of
this
like
low
stress
barrier
of
entry,
where
I
wasn't
necessarily
saying
I
was
going
to
do
anything
or
making
any
commitments.
It
was
just.
Let
me
see
how
this
thing
works
and
then,
as
after
that,
it
kind
of
you
kind
of
go.
Okay,
I
do
this
and
then
all
of
a
sudden,
you're
doing
more
and
more
I
think
that's
I!
Think
that's
a
good
approach.
A
So,
in
the
in
the
slack
channel
we
had
someone
asked
about
a
kubernetes
aid
project
cube
fed,
specifically
they
had
their.
They
had
someone
answer
their
question,
but
I
actually
want
to
take
this
question
for,
for
others
who
are
listening.
That
might
also
benefit
from
it,
which
is
hey,
I,
found
something
that
I
want
to
contribute
to.
So
in
this
case,
this
person,
saying,
hey,
cube,
fed,
looks
really
fun.
It's
really
cool.
How
do
I
find
out
who
owns
it?
How
do
I
find
out
what
sig
it
belongs
to?
A
B
So
all
repos
are
mentioned,
are
owned
by
a
particular
sexily.
Every
repo
must
be
owned
by
a
particular
sake,
so
and
all
SIG's
have
a
readme
file,
which
mention
which
lists
out
the
sub
projects
that
the
sick
owns
and
all
the
Reapers
that
it
owns.
So
you
will
find
in
one
of
those
agreed
means,
like
the
Reaper
name
would
be
mentioned
there.
So
that's
also
kind
of
helpful
to
understand
which
they
go
and
said.
B
The
other
thing
is
to
like
find
out.
If
who
are
the
particular
individuals
who
own
repo
is
the
owners
files
at
the
root
of
the
repos
owners
files
are
those
files
which
mentioned
if
like
who
are
the
maintainer,
is
for
that
and
who
can
approve
or
will
new
changes
to
that
codebase?
And
so
you
can
look
at
the
usernames
there's
our
github
user
names
mentioning
the
owners
files
and
you
can
sort
of
ping
them
on
slack.
B
The
user
names
usually
correlate,
but
sometimes
you
might
have
trouble
finding
them
and
the
other
thing
that
I
usually
do
to
figure
out
who
actually
owns
that
piece
of
code
as
look
at
the
history
of
that
file.
So
whenever
you
open
a
file
on
github
you'll
find
a
history
button
on
the
top
right
and
you
can
actually
see
who
has
made
the
most
number
of
commits
to
that
file.
B
Who
has
recently
changed
that
file,
so
you
can
probably
reach
out
to
them
or
just
ask
them
for
review,
or
things
like
that,
so
I
find
that
you
more
useful,
because
sometimes
the
people
listed
in
the
owners
files
might
have
touched
that
or
made
big
changes
to
that
codebase,
maybe
two
years
ago,
but
recently
someone
else
has
changed
it
and
they
probably
have
much
more
idea
about
that.
Codebase
today,.
C
So
yeah
yeah
I
also
want
to
mention
Paris.
Could
you
go
and
click
on
a
git
blame
in
a
file?
I
use
get
blame
a
lot
because
you
know
frequently
you'll
have
like
this
two
hundred
line
like
file
of
code
or
something
like
that
and
lots
of
different
people
have
changed
it.
This
one's
quite
old.
It
looks
like
so
those
folks
aren't
even
on
github,
but
so
one
of
the
like
handy
things
about
that.
This
might
not
be
the
best
example.
But
what
are
the
handy
things
about
this?
Is
that
you
know
when
you're?
C
In
this
view,
you
can
see
who
changed
an
exact
line
of
code.
You
can
click
on
that
commit
for
that
particular
line
of
code.
You
can
see
all
of
the
context
of
what
they
were
changing
when
they
made
that
change
and
github
will
even
like
link
you
back
to
the
original
pull
request
that
like
generated
that
change,
so
I,
don't
know
parents
you
want
to
just
like
pick.
C
C
Yes,
yeah,
so
that
that
it,
this
particular
change,
hit
the
master
branch
from
that
particular
pull
request
and
that
that
that
managed
to
end
up
in
whichever
releases
this
is
really
handy.
If
you're
like
when
was
this
feature
introduced,
how
many
releases
is
the
spin
in
like
oh
I,
guess
that
was
introduced
in
114
or
115
or
whatever
I.
B
Also,
like
use
the
flame
feature
on
github
for
understanding
a
particular
piece
of
the
code
so
like,
if
I'm
reading,
a
file
trying
to
understand
it
better
and
there's
something
I
come
across,
which
doesn't
make
sense
to
me.
So
I
usually
like
blame
it
on
github
and
then
check
out
the
product
quest
and
the
full
request
usually
has
way
more
context.
C
Unfortunately,
not
everybody
writes
the
most
descriptive
commit
messages,
so
the
pull
requests
often
will,
but
they
won't
just
give
you
the
context
around
that,
because
there's
like
a
required
PR
template
like
you'll,
see
and
like
the
first
message
of
the
PR
like
this
is
the
issue
that
this
is
fixing.
These
are
any
caps
that
are
related
to
this
and
so
on
and
so
forth.
What
we
did
briefly
say
what
a
cap
was
before:
what
the
caps
for
folks
I,
don't
know
what
it
kept.
C
Is
it's
a
kubernetes
enhancement
proposal
and
it's
basically
where
folks
announce
saying
I
want
to
make
this
big
change.
That's
gonna
affect
a
lot
of
people.
We
should
talk
about
that
and
get
some
community
buy-in
before
we
go
and
break
everyone,
so
I
would
co-author
of
a
cat
wherein
I
was
like
let's
break
everybody's
metrics,
because
they
don't
make
a
lot
of
sense
right
now.
That's.
A
C
Finishing
up
like,
for
example,
that
one
some
of
these
caps
take
a
long
time
to
implement.
So,
for
example,
one
of
the
things
that
I,
like
my
sort
of
mission
in
this,
was
to
fix
the
fact
that
we
had
all
these
Prometheus
labels
on
time
series
that
you
couldn't
do
because
they
were
different.
So
there
was
like
the
standard,
instrumentation
guidelines
that
specified
you
know
you
should
call
this
label
pod
and
this
label
container
and
so
on
and
so
forth.
C
But
for
whatever
reason
see
advisor
was
using
pod,
underscore
name
and
container
underscore
name,
and
so,
if
you
wanted
to
do
a
join
on
those
metrics,
you
couldn't
do
it.
You'd
have
to
like
real
able
them
first,
which
would
result
in
really
ugly
messy
queries
or
a
bunch
of
relabeling
logic.
You'd
have
to
do
in
your
Prometheus
or
like
it.
It
was
kind
of
a
mess.
So
I
said:
okay,
let's
fix
the
labels,
but
you
can't
just
like
go
and
be
like
hey
everybody.
C
You
know
those
like
nice
container,
metrics
labels
that
you're
relying
on
for
all
of
your
queries
and
all
your
dashboards
I'm,
just
getting
rid
of
those.
So,
instead,
what
you
do
in
like
a
proposal
like
this,
is
you
say?
Okay
here
is
like
what
I
want
to
change
these
two
so
suggested
as
part
of
this
larger
cap,
which
involves
a
bunch
of
breaking
changes.
C
We
want
to
rename
these
to
match
the
instrumentation
guidelines
and
in
order
to
do
that,
we're
gonna
deprecated
the
old
labels
and
we're
going
to
have
both
sets
for
a
couple
of
releases.
So
you
have
time
to
switch
over
that
your
queries
don't
immediately
break,
but
then
at
some
point
we're
gonna
get
rid
of
them
and
then
that
will
be
a
breaking
change.
C
You
have
to
switch
by
that
point
when
you
upgrade,
or
else
it's
not
gonna
work
so
previously,
I
think
in
the
1:14
release
we
got
the
new
labels
in
and
now
currently
for
the
1/16
release,
I'm
working
on
removing
the
old
labels.
So
that
way
people
had
114
and
115
to
switch
over
and
then
in
116
they've
got
you
the
new
thing.
What's.
C
It
was
pretty
straightforward,
I
guess
our
the
cig
sort
of
collaborated
in
a
Google
Doc
sort
of
all
together,
because
there
were
a
lot
of
different
breaking
changes
that
went
in
and
so
we
met
and,
like
you
know,
discussed
it
in
a
couple
of
meetings
and
then
the
cig
lead
sort
of
proposed
it.
So
it
was,
it
was
pretty
straightforward
and
then
I
guess
at
cube
con
North
America
last
year.
That
was
when
we
went
and
like
actually
went
and
put
it
in
the
implementable
state.
D
Not
so
much
a
question
more
of
a
comment
along
with
the
caps,
so
for
kubernetes
we
tend
to
release
every
quarter
or
we
strive
to
and
every
release
team
actually
tracks
the
enhancements
that
are
going
into
that
release
of
aiya.
The
caps
and
the
fact
that
your
instrumentation
metrics
thing
is
in
the
116
release.
Just
made
me
want
to
bring
that
up
as
well.
Caps
help
us
track
that
as
well.
A
A
So
that
means
you're
on
the
hook
next,
next
time,
I
don't
118.
Maybe.
A
A
B
A
follow-up
question
that
better
go
shoot
so
I
had
written
a
design
proposal.
It
was
like
the
term
that
we
used
before
caps,
where
think,
and
that
would
quite
a
lot
of
times,
I'm
curious
what
you
think,
Alana
and
Jeff
with
you.
What
do
you
think
are
they
required
or
the
average
time
commitment
for
someone
who
wants
to
write
a
cap.
D
So
I
wrote
a
cap
further
release,
notes
website
which
I
don't
know
if
anyone
has
seen
if
someone
can
drop
a
link
in
the
chat
for
an
example
but
removing
the
all
the
change
log
entries
from
the
change
log
MD
per
release
and
put
it
in
a
website,
so
we
can
search
and
filter.
It
seems
like
better
usability.
I
wrote
a
cup
for
that
because
it
had
to
be.
It
was
a
user
facing
change.
D
So
you
should
write
a
cup
that
took
me
two
hours
and
that
was
like
me
nitpicking
my
own
stuff
and
then
get
over
to
a
friend.
Hey.
Does
this
look
good,
but
that
was
also
to
me
a
fairly
straightforward.
We
are
doing
X
and
then
moving
it
to
Y,
so
I
don't
know
for
a
more
complicated
kept
like
the
metrics
one
or
pull
some
other
complicated
one
I'd.
C
Say
that
it's
the
metrics
overhauled
that
first
one
is
the
one
that
I
was
a
co-author
on.
So
this
is
like
a
pretty
long
cap
and
it
took
like
I
would
say,
like
I,
don't
know,
maybe
a
month
or
two
iterating
on
in
order
to
like
have
this
fully
written,
but
it
really
depends
like
right,
like
as
Jeff
was
saying
you
know
you
could
have
like.
C
This
is
my
very
small
thing
and
not
like
with
this,
which
is
like
literally
like
here's,
a
bucket
list
of
like
all
of
the
things
that
we
want
to
fix
and
break
at
the
same
time
kind
of
thing.
So
there's
a
template
that
you
can
fill
out
and
like
if
it's
a
small
non-controversial
thing
I
imagined
it.
You
know
the
couple
of
hours
seems
like
totally
reasonable
to
me.
C
D
A
And
that's
kind
of
what
we
were
talking
about
here,
where
it's
it's
either
like,
because
for
the
cats
like,
sometimes
you
have
to
stick
around
for
a
while,
but
then
other
things.
It
seems
like
it's
going
so
fast
talk
about
a
little
bit
about
the
about
your
time
management.
That's
what
I'm
also
curious
about
how
you
all
manage
your
time
in
this
kind
of
an
environment
of
fast
slow
and
what
goes
what
thing,
what
kind
of
things
go
slower
and
what
kind
of
what
other
things
do
you
feel
like?
C
Don't
know
I
don't
feel
like
anything
in
kubernetes.
Go
that
fast,
that
I'm
working
it's
only
like
six
months
or
something
like
that.
It's
from
sort
of
like
I,
am
interested
in
contributing
to
getting
like
a
three
line.
Pr
merged
and
I.
Think
it
took
like
four
months
of
that
PR
being
up,
although
granted
the
cap
I
think
had
not
actually
been
out
there
and
approved
in
order
for
it
to
get
merged.
So
that
was
a
little
bit
part
of
it
and
I'm
fighting
with
flaky
tests.
C
And
then,
like
you
know,
the
implementation
phase
again
would
take
a
long
time
and
that
was
very
volunteer.
Limited
kubernetes
at
least
tends
to
have
a
lot
more
like
full-time
paid
contributors.
So
stuff
gets
done.
People's
jobs
are
working
on
kubernetes
full-time,
so
they
tend
to
be
pretty
responsive.
I'm,
currently
like
in
the
situation
where
I've
got
like
the
small
PR.
C
That's
like
the
follow-up
to
my
other
small
PR,
and
it's
been
up
for
like
a
few
weeks
and
it's
like
I
have
to
track
down
three
different
approvers,
because
I
have
to
update
the
label
on
a
test
which
got
missed
previously
or
got
added
in
between
the
releases
or
something
like
that,
and
nobody
from
that.
Sig
is
like
around,
and
so
maybe
you
have
to
go
to
their
sig
meaning
and
be
like
hey.
A
If
sometimes
I
wonder,
if
you,
if
we
get
this
like
fast
slow
because
I
mean
our
Venn
diagram
of
contributors,
is
mostly
all
users,
I
mean
it's
hard
to
I
think
contribute
to
kubernetes
without
having
touched
it,
but
I'm
wondering
if
it's
because,
like
they
see
on
like
the
user
and
it's
just
so
fast,
there's
just
so
many
new
features
coming
out.
So
it's
like
hard
to
I
guess
stay
abreast
that
way,
but
then
you're
also
like
slow
trying
to
make
change
on
the
contributor
side.
C
A
How
do
you
manage
your
time
right,
like
so
both
of
you,
so
you
have
full
time
jobs,
so
all
of
us
have
full
time
jobs.
How
are
you
managing
throughout
this?
Like
so
you've
got
cats
going
on
which
are
gonna
be
slow,
and
then
you
know
that
you're
gonna
need
to
stick
around
for
six
months
to
a
year,
and
then
you
probably
have
some
other
smaller
things
going
on
another
in
other
areas
of
your
life
like.
How
are
you
managing
yeah.
C
Obviously
kubernetes
stuff
is
a
little
more
directly
relevant
to
my
day
job.
So
that's
stuff!
You
know
if
I
need
like
a
little
bit
more
time,
I
can
say:
hey
I
need
to
do
this
thing.
It
affects
these
other
teams
at
work
and
then
it's
you
know,
usually
not
a
problem.
To
get
that
time
allocated
my
last
job.
It
was
much
more
channel
because
you
know
I
was
working
at
like
a
financial
company,
you
know
basically
a
kubernetes
end-user,
and
so
it
was
less
easy
to
justify
to
management
like
this.
C
C
A
C
Taking
more
than
40
hours
a
week,
and
it
has
like
when
at
my
last
job
you
know,
I
had
too
many
job
responsibilities
and
not
enough
time
to
spend
on
upstream
stuff,
because
I
couldn't
get
the
time
allocated
at
work.
I
said:
okay,
yes,
I'm
not
doing
upstream
stuff
right
now,
that
was
fine
and
I
was
able
to
jump
like
totally
back
into
it.
Once
I
had
come
up
to
speed
at
my
new
job
beauty.
A
B
I
mean
I'm,
definitely
ask
me
about
it:
I
remember
that,
so
there
are
few
things
so
I
don't
have
a
set
routine
around
open
source
work,
mostly
because
sometimes
I
will
so
open
source
is
not
technically
as
part
of
my
day.
Job
I
usually
do
it
after
or
the
forward,
and
so
what
I
do
usually
is
in
the
mornings.
B
What's
going
on,
or
just
like,
like
figure
out
if
there
any
new
changes
happening
in
this
area,
but
also
like
that's
also
like
one
aspect
which
helps
me
in
like
keeping
in
touch
with
the
fast
part,
because
things
do
change
fast,
sometimes
and
the
other
thing
that
I
do
sometimes
is
I
check.
The
latest
comments
that
I've
been
added
so
since
I'm
in
India,
most
of
like
whenever
I
go
to
sleep,
things
start
happening
and
whenever
I
wake
up,
things
are
already
happening,
so
I
usually
wake
up
to
see
you.
B
It
just
check
the
recent
comments
side
of
merge
to
understand.
If
there's
something
interesting
or
it's
something
that
I
want
to
go
and
more
about
I'll,
go
check
out
that
or
I'll
just
like
go
and
bookmark
it
and
check
it
out
later
over
the
weekend
or
something
one
more
thing
that
I've
done
is
in
the
past,
and
sometimes
even
now
is
I,
usually
add
words,
as
slack
keyword
highlights.
B
I,
don't
know,
what's
right,
word
for
it,
like
you
get
a
notification
if
someone
mentions
that
word
on
slack
for
things
that
I'm
interested
in
so
I
did
that
for
CR.
These
when
I
was
like
changing
the
API
and
it
was
like
when
series
was
really
really
new
and
then
suddenly,
everyone
started
using
crts,
and
my
notifications
just
blew
up
completely
so
do
that
at
your
own
risk.
B
Make
sure
that
you
don't
had
to
comment
like
keywords
are
too
common,
but
that
helps
me
like
understand,
of
someone's
facing
problems
using
that
particular
component
or
something
and
I
can
answer
those
answer.
Their
questions
quickly
and
also
like
helps
me
find
issues,
and
it
has
helped
me
find
issue
is
pretty
quick
before
so.
That's
that's
really
helpful
for
me
honestly,
and
the
other
thing
that
I
do
is
I
also
watch
some
repos
on
github
so
like,
for
example,
I
watched,
the
community
repo
the
sake
release,
repo
I
might
not
read
all
like.
B
B
A
little
bit
more
pretty
much
yes,
so
there's
also
this.
If
you,
if
someone
can
like
share
the
link
once
like
or
I,
can
do
that
after
this.
But
there's
we
have
recommended
best
practices
for
email
filters
document,
so
I
pretty
much
follow
that
so
it
kind
of
hurts
and
like
if
you're
in,
if
you're
interested
in
some
people,
don't
watch
the
kubernetes
kubernetes
repo,
because
you
are
like
your
email
quota.
B
I
will
like
blow
up
in
a
day,
but
if,
like
community
or
signal
use
are
even
enhancement
is
pretty
pretty
current
these
days,
but
community
or
cig
releases
are
good
repos
to
like
watch
maybe
and
that
kind
of
plate
keeps
me
in
touch
with.
What's
going
on,
I
again,
like
I,
might
not
read
everything,
but
I
will
just
check
the
subject
and
do
that,
and
so
these
are
Sobeys
email.
B
Filters
will
also
tell
you
like,
if
someone's
requested
review
on
a
PR,
or
someone's
mentioned
you
or
they've
assigned
to
you
and
I,
usually
just
like
go
through
the
assign
assignee
label
first
and
then
their
review
comments
and
such
so.
If
you
assign
me
you'll,
probably
get
a
reply
faster
and
that's
I
guess
pretty
much.
It
I
mean
I,
probably
write
code
or
write
code
for
a
feature,
maybe
one
hour
two
hours
a
day
depending
on.
If
I'm
have
the
energy
to
do
that.
B
C
C
C
So
you
will
get
automatically
notified
if
you
need
to
go
and
look
at
a
particular
PR,
you
don't
have
to
like
actively
necessarily
set
up
some
like
monitoring
or
watch
all
of
the
things.
You
can
also
do
some
fancy
filtering
with
labels.
If
you
just
want
to
like
get
notifications
for
your
particular
cig,
you
can
say,
like
yeah
I,
only
want
to
see
like
these
notifications
in
my
inbox
for
like
seeing
instrumentation
or
something
like
that.
Other
things
need
not
apply.
A
So
there
is
a
no-win
situation
now,
keep
going
on
tracking
the
win
is
pick
one
that
works
for
you,
because
we
will
definitely
try
our
hardest
to
over
communicate
in
this
project.
We
try,
we
might
not
always
win
with
the
over
communications,
but
we
definitely
try
all
right.
So
our
last
question
in
in
chat
right
now
the
individual
wanted
to
know.
Is
there
a
diagram
of
all
the
SIG's
and/or
governance
structure?
How
do
the
SIG's
interact
in
the
decision-making
processes?
A
So
we
talked
a
little
bit
about
caps
right.
So
that's
like
a
decision-making
process.
How
I'm,
gonna
and
I'm
gonna
show
right
now
on
my
screen
and
I'm
gonna
share
some
diagrams
that
we
do
have,
but
from
y'all's
perspectives,
how
did
the
SIG's
interact
with
other
things,
for
instance,
especially,
like
you
know,
we
mentioned
testing
earlier
talk
to
you,
talk
to
us
a
little
bit
about
your
experiences
there
and
then
I'm
gonna
share
my
screen.
Yeah
I.
C
Guess
mostly
I
will
like
join
their
slack
channels,
poke
people
but
frequently
I,
find
that,
like
people
aren't
necessarily
the
most
responsive
on
slack,
especially
because
a
lot
of
these
sake
channels,
they
just
get
a
lot
of
like
sort
of
drive-by
newbie
questions
and
people
aren't
necessarily
paying
attention
to
those
or
you
know
people
don't
happen
to
be
around.
When
you
ask
a
question,
then
a
bunch
of
other
people
talk
and
then
there's
like
some
scroll
back
and
nobody
actually
ends
up
addressing
your
thing.
C
The
best
way
that
I've
found
to
interact
with
other
SIG's
has
been
to
show
up
to
their
cig
meetings,
because
that
way,
you
know
you
get
like
an
immediate
response
and
also
they
like
see
your
face.
They
know
that
you
exist,
and
maybe
you
even
get
something
in
their
meeting
notes,
so
that,
like
everybody
else
in
the
community,
can
be
aware
of
it.
I
guess.
If
you
wanted
something
a
little
bit
more
formal
and
like
a
record
of
a
note,
you
could
go
and
email
the
Google
Groups
specifically
as
well.
C
C
A
Sort
of
thing
that's
awesome,
Nikita,
how
do
you
interact
with
other
SIG's
and
how
does
that?
How
does
it
play
in
and
how
do
you
find
out?
Actually,
this
is
like
a
loaded
question.
How
do
you
find
out
what
their
decision-making
process
is?
Sister,
a
document
perhaps
like
a
starts
or
the
C
word
charter.
Oh
yeah,.
B
So
all
those
things
also
have
this
thing
called
a
charter
that
you
can
find
in
the
community
repo.
So
there
are
lots
of
directories
and
each
sig
directory
will
have
something
called
as
a
charter.
So
that
also
mentions
how
the
sake
makes
decisions
usually
are
also
like
who
can
approve
certain
changes.
It's
usually
the
chairs
or
the
technical
needs,
but
sometimes
they
can
like
define
certain
individuals
as
well
as
far
as
interaction
between
SIG's
go
like
Alana,
said.
I.
B
C
People
to
show
up
to
meetings
for
you
to
that's
something
that,
like
sometimes
meetings,
are
an
inconvenient
time
and
so
I'll
be
like
hey.
You
know
other
person
and
say
instrumentation.
Can
you
go
present
the
case
on
this
to
like
cloud
provider
because
I'm
busy
or
it's
not
at
a
convenient
time
or
something
like
that
and
that's
often
pretty
effective.
A
Hey
go
advocate
my
thing
right
now,
all
right,
so
that
really
wraps
us
up
right
now,
but
I
always
like
I,
try
I'm
trying
to
end
with
the
same
type
of
question,
which
is,
if
you
knew
one
about
upstream
kubernetes
when
you
started
it
could
be
like
a
little
bit
about
your
dev
environment.
It
could
be
something
about
love
like
a
cig
meeting
like
what's
one
thing
you
wish,
you
knew
when
you
started
that
you
didn't
know,
and
she
like
later
on
down
the
road
question.
A
C
Was
really
weird
like
I'm?
This
is
definitely
not
my
first
open-source
rodeo,
so
I
like
it
didn't
even
occur
to
me
like
that
I
should
apply
to
become
a
project
member
or
that
the
barrier
to
entry
there
was
really
low,
like
you
only
need
to
have
a
couple
of
PRS
merged
or
be
a
little
bit
active
in
the
project
to
become
a
member.
C
A
C
The
biggest
benefit
is
I
can
tell
the
bot
to
stop
marking
my
issues
as
stale.
That
was
really
annoying,
because
I
have
to
go
ping.
Someone
and
be
like.
Can
you
like
so
I
need
to
like
now?
I
can
be
like
lifecycle
frozen,
stop
killing
my
issue,
I
can
lgt
MPR's
I.
Don't
need
an
okay
to
test
on
my
PRS.
There
are
all
sorts
of
like
small
benefits
and
I.
Have
the
fancy
kubernetes
badge
on
my
github
profile.
B
I'd
actually
+1
to
that.
So
when
I
applied
for
membership,
I
had
like
two
design
proposals
merged
to
API
changes
and
everything,
but
I
was
really
not
sure
if
I
was
worthy
of
becoming
a
member
and
when
I
was
being
added
to
an
owner's
file.
At
that
time,
I
was
like
hey,
wait
a
sec
I'm,
not
a
member.
Yet
he
found
out
me
and
they're
like.
Are
you
crazy?
B
So
yeah,
that's
one
thing
and
also
like
if
you
are
reviewing
code
in
some
component
and
if
you
think
you
should
be
a
reviewer
reach
out
to
the
approvers
and
let
them
know
that
you're
interested
to
be
one
and
chances.
Are
there
be
pretty
okay
with
you
being
added
in
there,
which
I
was
also
pretty
shy,
but
I
was
still
pretty
shy
about
and
I'm
still
learning
how
to
go
about
that.
A
A
Yes,
and
if
you
would
like
a
better
experience
with
the
bots
s,
this
is
for
you,
okay.
Alright,
let's
wrap
things
up.
You
all
have
been
awesome
with
the
information
that
you
have
provided
here
is
so
valuable
for
so
many
people.
We
really
appreciate
it.
We
are
here
same
time
next
month
with
different
panelists,
potentially
different,
YouTube
streamer,
so
everybody
wave
and
just
take
us
out
of
here
thanks
so
much
everybody
see
you
soon.