►
From YouTube: SIG Chair/TL March Meeting Thurs 20200312
Description
SIG Chair/TL March Meeting Thurs 20200312
A
If
I
could
have
everybody
take
notes
and
then
also
put
your
name
down
for
attendees,
let's
try
to
get
better
to
with
like
putting
our
names
down
as
well,
and
then
also
please
interact
with
the
notes.
This
is
a
a
sync
meeting.
You
can
see
the
notes
from
tuesday's
call.
So
please
definitely,
you
know,
put
your
put
any
kind
of
like
either
notes
or
talking
points
or
plus
ones.
Anything
like
that,
because
we
do
have
a
lot
of
things
going
on.
A
A
T
is
for
triage
and
Thursday,
and
we
do
have
a
theme
today
and
that's
the
image
triage
and,
as
I
mentioned
before,
please
put
your
name
on
the
agenda
and,
as
always,
we
do
have
a
code
of
conduct.
Please
be
excellent
to
each
other.
It
looks
like
we
have
about
11
of
us
on
the
line
today,
which
is
a
smaller
group
which
is
great,
so
feel
free
to
respectfully.
You
know
interject
it
a
pause
but
or
also
you
know,
raise
your
hand
and
we'll
try
to
get
to
you
as
well.
A
Alright,
and
since
it
is
a
small
crew,
do
we
have
any
first-timers
with
us
that
have
not
done
an
introduction
to
the
chair
or
technique
crowd
all
right?
Let's
do
a
quick
roll
call
then
just
to
get
us
all
Joan's
stuff
about
this
say
your
name
say
what
say
your
representing
and
that's
it.
Okay,
let's
hear
you
here
record
here
we
hear
roughy.
B
B
E
A
Made
it
through
roll
call,
I
was
like
my
first
time
actually
like
doing
a
roll
call.
I
could
keep
a
reading
like
all
of
these
like
work
from
home
meeting
things.
That
was
one
of
the
things
anyway.
Thank
you
for
a
Pete.
Thank
you
very
pleasing
me,
okay,
anyway,
all
right,
so
announcements
that
I
wanted
to
breeze
through
and
also,
let
me
share
my
screen
in
case
I-
know
folks
are
like
walking
dogs
and
stuff
like
that.
A
So
let
me
share
my
screen
for
you
really
quick
announcements
and
because
I
want
to
breathe
by
this
and
get
to
all
the
good
stuff
right.
These
are
things
that
you
have
to
know
as
a
chair
tech
lead
this
week
this
month,
whenever
the
EU
contributor
summit,
if
you
have
not
heard,
is
obviously
being
postponed,
cube
con
is
being
postponed.
The
contributor
summit
team
is
hard
working,
trying
to
figure
out
what
the
heck
that
means
they
do
want
to
do.
A
Some
virtual
things,
you're
gonna,
have
a
lot
of
communications
shortly
to
all
the
regular
channels,
which
is
most
notably
kubernetes
dev.
So
nothing
necessarily
is
for
you,
except
if
you
were
speaking,
you
should
probably
talk
to
young
and
some
of
the
other
people
that
were
coordinating
in
the
content
and
that's
it
for
that
cool.
Any
questions
about
the
contributor
summit
I
mean
end
or
cube
con.
We
might
know
a
little
bit
about
that,
but,
okay.
A
Again,
it's
a
lot
of.
Obviously
everybody
knows
here
waiting
to
see
what
the
deal
is,
so
lots
of
people
hard
at
work
there,
thanks
to
everybody
on
the
line.
That's
working
on
that
we
have
some
new
tech
leads
and
sake
docs.
We
have
new
chairs
and
are
any
of
those
folks
online
right
now,
I,
don't
think
so.
I
think
Ilana
and
Han
we're
on
Tuesday
is
call
so
anyway,
please
in
the
in
the
sig
chair
slot
channel,
say
hello
to
them.
A
That's
Ilana
and
Han
with
seconds
fermentation
and
Karen
and
Tim
with
sig
Docs
and
the
cool
thing
about
sig
Docs
is
this
is
the
first
time
they've
ever
had
a
tech
lead
role,
so
they
are
getting
the
inspiration
of
splitting
out
those
roles
and
responsibilities
to
have
their
own
boarding
experiences
and
better
succession
planning,
and
things
like
that.
So
also
big.
Congratulations
to
sig
Docs
for
opening
that
window
as
well
and
the
there
is
a
cap.
Is
anybody?
Did
anybody
that
put
this
cap
on
on
the
announcements.
D
That's
the
way
I
mean
I,
don't
know
what
all
you
discussed
on.
Tuesday
was,
but
in
the
other
instance,
but
just
to
sort
of
reiterate,
we
have
been
talking
about
a
change
to
the
support
stance
right
now.
We
do
three
trailing
releases
or
getting
patches,
try
to
pick
back
to
them
and
community
support.
D
So
the
the
key
part
of
this
cap
is
a
proposal
that
we
change.
Our
support
stands
such
that
we
can
encompass
an
annual
support
cycle
for
people.
What
that
practically
looks
like
we've
scoped
this
really
really
minimally
we're
not
trying
to
change
everything
about
how
we
do
release
and
patch
back
support,
so
the
the
minimal
approach
that
would
be
instead
of
doing
three
trailing
support
branches.
We
do
four.
Instead
of
removing
one
from
our
support.
D
We
just
keep
it
for
another
three
months
or
so,
and
then
that
kind
of
gives
that
that
year
span
there's
a
whole
bunch
of
other
proposals
of
how
we
might
accomplish
this
and
a
lot.
The
discussion
of
things
that
have
been
considered
in
the
cap
there
so
I
just
wanted
to
encourage
folks
to
read
that
and
think
about
implications
it
might
have
on
your
stake
and
we'd
love
feedback.
D
You
want
to
you
can.
This
is
a
maybe
a
little
bit
of
an
unusual
catch
in
theory,
every
sig
is
a
stakeholder
on
this,
so
any
sig
could
block
it
from
being
declared
implementable.
It'll
be
interesting
to
see
how
we
decide
to
to
say
that
this
one
has
been.
The
consensus
was
formed
around
declaring
this
is
implementable
or
not
so
if
you've
got
a
few
minutes
to
bring
it
in
your
agenda,
just
get
a
little
more
eyeballs
on
it.
We'd
certainly
appreciate
it
just
to
make
sure
we
have
as
broad
a
possible
exposure.
G
This
one
certainly
makes
sense
because
the
major
cloud
providers
default
to
the
last
version
or
the
the
latest
version
that
is
no
longer
supported,
and
then
the
the
latest
version
most
of
them
have
is
the
oldest
version
that
supported
two
weeks,
yep
and
so
they're,
just
adding
the
next
one
now.
So
that
way,
when
their
version
falls
out
of
support,
they
still
got
the
last
one
that
supported
as
an
option
and
and
I
would
really
love
it.
G
A
I
So
yeah,
so
we've
been
talking
about
some
enhancements,
the
caps
it's
been
happening
on
these
sega
architecture,
mailing
lists,
so
I
think
Tim
I
can
kick
that
off.
He
opened
it
kept
around
enhancing
caps.
So
a
few
things
around
the
ecosystem,
like
moving
cats
to
a
directory
structure,
changing
some
of
the
names
splitting
the
metadata
out
of
the
top-level
out
of
the
front
matter
for
the
for
the
readme
stuff,
like
that,
it's
I
think
it's
pretty
minimal,
but
because
it's
something
that
affects
everyone,
I
do
want.
I
Sig
chairs
technical
leads
to
take
a
look
at
it.
We're
at
the
point
where
I
think
we're
pretty
happy
with
what's
been
laid
down.
At
this
point,
we
wanted
to
try
to
do
that
first,
so
that
there
it
wasn't
gonna
get
to
bite
Chevy
before
opening
the
the
to
the
broader
group,
so
we're
gonna,
I'm
gonna,
send
an
email
out
to
sig,
leads
and
chairs.
It's
likely
it's
and
all
that
good
stuff
and
then
open
a
timeout
period
for
that
cap.
So
questions
comments.
Please
take
a
look
at
the
cap.
Leave
them
there
thanks.
A
All
right
does
anybody
questions
or
comments
like
30
seconds
for
any
of
the
announcements,
so
you've
got
two
things
on
there
that
require
feedback
on
comments,
saying
hello
to
new
sig,
chair
or
sake
chairs
and
tiel's
in
the
slack
channel
and
waiting
for
the
summit
team
to
let
you
know
what
your
actions
are
for
the
contributor
seulement
all
right.
Let
me
check
chat,
really,
quick,
okay
and
alright.
Let's
do
a
last
meeting
last
minute.
Recap
really
quickly:
I,
don't
I'm,
not
gonna.
A
A
This
is
the
sort
of
a
chair
rulebook
handbook
that
I
glued
together
and
I
say
glued
together,
because
I
didn't
make
up
a
lot
of
this
stuff.
This
stuff
comes
from
the
cig,
governance
and
or
governance
MD
or
some
other
governance,
doc
that
we
have
here
and
I,
just
really
put
it
into
bullet
form
and
then
also
Steven
and
Zach,
and
quite
a
few
other
chairs
were
giving
comments
on
the
once
upon
a
time.
Word
not
word
doc,
wow
that
was
ancient.
A
Google,
doc
and
I
put
their
comments
in
here,
and
it
says
things
like
what
I
let
me
get
so
little
pin
suggested
by
so.
If
you
have
comments
or
questions
or
things
that
you
think
would
be
added
here
again
think
about
what
you
do,
and
also
not
only
what
you
do,
but
take
off
any
TL
hats
that
you
also
wear
any
tech
lead
hats
that
you
wear.
This
is
just
for
the
chair
role.
Many
of
you
I
think
it's
the
majority
of
you
actually
carry
a
chair
and
tech
lead
hat.
A
So
it's
important
to
think
about
that
role
separately
for
right
now,
for
this
exercise,
right
and
and
that's
it
they're
questions-
concerns
comments
about
this
and
I
know.
A
lot
of
other
people
did
comments
and
questions
about
this
off
record
as
well.
So
we
can
always
punt
this
to
discussion
too,
but
at
least
wanted
to
bring
this
up
cool
all
right.
Now,
I
gotta
figure
out
how
to
get
out
of
the.
How
do
I
get
out
of
the
annotation
okay?
Here
we
go,
I
got
out
of
it.
Somebody
in
chat
said:
oh
nevermind.
A
A
A
preferred
channels.
Plea
prefer
contact
list
for
chairs
and
tech
leads.
Please
fill
that
in.
It
looks
like
about
half
of
you
right
now
have
so
this
is
just
saying
that
you
would
prefer
to
be
contacted
from
us,
meaning
steering
or
contrived
X
or
somebody
or
another
chair
somebody
working
on
the
project.
You
would
like
to
be
contacted
and
one
of
these
means
anyways
first,
so
some
people
say
hey
get
a
hold
of
me
wherever
I'm
on
everything
all
the
time.
Other
people
say:
oh,
hey,
I'm.
A
A
Right
just
wanted
to
make
sure
it
wasn't
about
us
so
yep
this
is
comms.
Does
anybody
have
any
questions
concerns
comments
about
what
we're
doing
here?
Why
we're
doing
it?
You
hate
it
literally
anything
you
want
to
share.
Let
us
know
you
all
right,
so
that
was
really
the
the
rap
from
some
of
the
conversations
that
we
had
last
time.
A
A
C
A
What
the
current
state
is,
we
all,
we
obviously
have
some
notes
already
in
there
from
Tuesday,
but
Steven
still
gonna
give
you
a
full
rundown.
Also,
you
can
even
see
some
of
the
other
chairs
questions
and
their
if
you
have
questions
go
ahead
and
put
them
on
the
agenda
while
Stevens
talking
and
then
we
can
get
to
the
questions
all
right.
I
A
I
All
right
so
so
this
I
think
we've
been
talking
about
issue
triage
for
a
little
bit
in.
You
know
in
a
variety
of
venues,
and
you
know
it's
definitely
something
that
comes
up
for
the
release
team,
like
triage
teams.
Specifically,
they
work
on
I'm.
Sorry,
there's
a
garbage
truck
in
the
background.
Sorry
about
that,
the
the
bug
triage
team
is
responsible
for
triaging
issues
ends
PRS
that
come
across
in
in
the
zubur
Nettie's
kubernetes
repo,
so
I
think
I
want
to
say
in
1:14,
maybe
earlier.
I
Even
there
was
an
idea
what
if
we
had
a
what
if
we
started
talking
about
what
the
workflow
actually
looked
like
for
people
to
bring
something
from
a
newly-opened
issue
to
a
completed
issue
or
PR
emerged,
PR
right,
so
super
quick
stats.
So
first
thing
to
keep
in
mind
like
this
cap
is
intentionally
light.
I
was
very
sparse
on
details
because,
like
this
is
something
that
we
want
to
try
to
move
through
quickly
and
it's
been
I
guess
in
progress
for
years.
There
have
been
discussions
on
this
for
years.
I
If
you
look
at
the
implementation
history,
there's
a
mailing
list
discussion
going
back
to
2016,
so
this
is
intentionally
life.
But
currently
there
are
about
900
something
poll,
requests
and
2000,
plus
open
issues
and
kubernetes
kubernetes,
and
there
are
a
variety
of
them
in
various
states
that
are
either
stale
rotten
are
frozen,
so
we're
looking
at
about
20%
or
so
of
those.
If
you
combine,
if
you
smush
all
those
statuses
together
that
are
in
some
sort
of
stable
earnest,
so
the
idea
is
like
how
do
we
get
SIG's
to
the
point?
I
I
What
a
working
group
user
group
organizer
sub
project
donor
I
want
to
be
able
to
easily
triage
my
groups,
backlog
of
open
issues,
an
NPR,
just
accelerate,
merge
velocity
any
issue
resolution
right
for
a
reviewer
approver
I
want
a
simple
system
that
allows
me
to
categorize
issues
and
PRS
in
my
purview.
So
I
can
prioritize
what's
a
review
first
right
and
as
a
contributor,
what
the
weather
this
be
a
non
reviewer,
approver
or
tried
by
a
contributor
I,
want
to
understand
the
way
how
the
community
works
to
address
new
submissions.
I
I
want
to
have
some
assurance
that
they'll
be
routed
to
the
correct
group
and
I
want
to
have
them
addressed
in
a
timely
manner.
Right
and
I.
Think
all
of
this
makes
sense.
Garbage
truck
cut
it
out.
So
there
is
so
there's
one
non
goal
for
this,
and
it's
prescribing
an
issue.
Triage
workflow
for
all
of
the
projects
within
kubernetes
works
right
right
now
we're
trying
to
tackle
kubernetes
kubernetes,
because
I
think
that's
the
biggest
main
point.
I
I
One
one
label
would
be
needs
triage
right
and
needs
triage
would
operate
in
the
same
mutually
exclusive
way
that
the
needs
kind
needs
priority
needs
sig
labels
operate
where
by.
If
you
apply
a
kind
label,
a
priority
label,
our
sake
label,
it
would
remove
the
other
one
right.
So
the
second
one
is
lifecycle
ready,
but
I'm
thinking.
It
makes
more
sense
to
call
this
triage,
ready
and
you'll
see
that
in
some
comments,
both
in
the
cap
itself,
as
well
as
the
meeting
notes
that
it
might
makes
more
sense
to
call
it
like
triage.
A
I
And
the
idea
is,
you
would
apply,
you
could
use
a
bot
command
or
if
you
have
write
access
to
you,
the
repo
you
can
manually
apply
the
label
to
remove
to
remove,
needs
triage
right
from
there.
We
don't
want
to
prescribe
workflow
for
everyone
just
yet.
The
idea
would
be
you
at
the
point
at
the
point
which
you
knew.
Your
issues
were
on
triaged
you
could
search
by
the
needs.
I
Triage
label
do
whatever
your
sig
does,
to
move
them
to
a
triage
state
and
then
hit
it
with
the
triage,
ready
bot
command
and
then
from
there.
You
can
search
all
the
issues
or
PRS
that
are
our
issues
specifically
for
triage
that
are
in
the
triage,
ready
state
and
move
on
those
move
on
determining
when
or
how
you
should
do
those.
So
that's
it
I
think
shortened
to
the
point.
Let's
do
let's
see
questions.
H
H
H
You
know
we
need
more
information
and
people
are
unresponsive,
we'll
just
close
it.
So
it's
worked
pretty
well
to
cut
down
our
backlog
substantially
over
the
last
you
know,
six
months
or
so
so
I
think
definitely
some
sort
of
process
and
assistance-
and
you
know
in
proud
would
be-
would
be
great
for
everybody.
H
A
F
A
A
I
That
was
good
news
there,
so
yeah.
So
I
think
this
is
definitely
due
for
some
update,
but
I
had
drafted
kind
of
a
state
machine
of
what
it
looks
like
to
move
through
open
triage
clothes,
complete
backlog
and
so
on
and
so
forth.
I
didn't
want
to
include
this
in
the
cap,
because
it's
pretty
easy
to
get
hung
up
on
these
details.
This
is
due
for
like
a
rework,
but
the
entry
criteria
you
can
see
here
for
like
what
I
consider
to
be
triage
is
like
has
a
cig.
I
It
has
the
needs,
triage
label
or
something
right,
I
think
the
triage
labels
and
then
to
get
out
of
this
you
would
you
would
close
it.
You
would
apply
a
few
things
or
yeah,
so
this
would
be
like
applying
the
triage
ready
label
so
yeah
so
I
agree
there.
I
also
don't
want
to
define
exactly
how
things
get
out
of
the
triage
state
so
that
they
can
feel
free
to
develop
their
own
process
right.
I
If
there's
something
that,
if
that's
there's
something
that
works
starts
to
work
for
multiple
people,
and
we
can
you
can
gain
consensus
on
that.
Then
we
can
consider
applying
it
at
a
larger
levels,
but
I
think
what's
hung
this
up
for
so
long.
It's
the
fact
that
no
one's
agreed
on
just
a
few
things
so
I
want
to
bring
two
things
to
the
table
which
are
which
is
the
the
bottom,
a
shin
for
needs,
triage
and
no
longer
needs
triage
right,
and
this
is
something
that
was
brought
to
contributor
experience,
maybe
eons
ago,
cursed
off.
C
A
D
E
E
E
A
Next
up
on
the
agenda,
we've
got
a
little
bit
of
a
walkthrough.
We
asked
two
SIG's
to
come
to
the
meetings
and
just
give
us
a
walkthrough
of
what
they're
doing
for
triage.
Why
they're
doing
it
different
because
you'll
see
different
approaches?
You
know
different
strokes
for
different
folks.
So
let's
see
what
some
of
the
stuff
that
cluster
lifecycle
is
up
to
today,
Libre
me
or
did
you
need
to
either
share
your
screen
or
anything
like
that?
If
you
don't
that's
cool
I
just
wanted
to
make
sure
yours
that.