►
From YouTube: WG LTS biweekly Meeting for 20200414
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).
B
So
too
and
I
ping
Nick
last
night
it
was
still
his
day
time
because
I
thought
he
was
gonna
update
the
calendar
event
for
that
and
he
thought
we
were
going
to
monthly.
But
I
don't
know
if
he
recalled
that
he
was
the
one
who's
gonna
change.
The
invite
and
I
didn't
see
anything
else
after
that,
but
I
partly
suspected
it
was
my
calendar
that
was
screwed
up
because
I
don't
get
calendar
events
correctly
often
so
I
figured
I
would
join
and
see
what
happened.
B
A
B
B
B
B
B
C
C
C
C
B
B
A
They
didn't
really
think
that
was
going
to
be
a
serious
problem.
They
thought
that
just
adding
additional
instances,
you
know
additional
configurations
of
the
existing
test.
Jobs
was
good
enough,
so
there
was
supposed
to
be
a
comment
on
the.
She
was
supposed
to
made
a
comment
on
the
pr2
that
regard.
A
A
A
B
Though
I
wanted,
since
reign
of
Luminara,
wanted
to
chat,
lumineers
had
some
discussions,
we
said
arch
and
things
have
been
back
and
forth
between
cluster
lifecycle
and
release,
also
around
cube
ADM
splitting
out
of
tree
and
that
I
feel
like
that
kind
of
discussion.
The
whole
splitting
the
model
that's
being
able
to
run
at
different
release.
Cadence's
relates
to
this
broader
discussion
of
how
we
support
what
for
how
long
and
I
know
rumors
had
a
lot
of
thoughts
on
this
and
the
discussion
and
sig
architecture.
D
B
Can
throw
some
plots
in
the
discussion
from
a
release
perspective,
I
I
feel
like
right
now
we're
all
sort
of
kind
of
pointing
like
that
way.
Release
can
go
along
with
a
preferred
route.
Like
you
said,
closer
life
cycle
can
go
along
with
a
preferred
route,
but
the
the
global
preference
we
we
don't
have
a
decision
process
for
that.
B
D
Yeah
I
think
we
should
have
a
more
discussion
with
the
other
six
as
well,
but
also
with
seek
release
again.
It
should
be.
My
point
is
that
this
should
not
be
a
some
discussion
because
they
turned
into
bike
shading
and
I
mean
that
the
discussion
of
the
mailing
lists
can
also
to
dance,
but
I
wanted
to
see
concrete
opinions
with
paragraphs
of
text
written
by
people
and
compare
what
we
get
from
those
people
and
thus
far
I
haven't
seen
any
of
that.
I
was
really
expecting
like
people
acting
Hawking
to
jump.
B
D
D
D
And
I
also
wanted
to
help
this
particular
topic.
I
wanted
to
say
what
tip
said:
Claire
and
just
in
Santa
Barbara
think
about
extracting
media,
and
they
said
we
told
one
cube
medium
to
be
the
first
component
to
extract
I'm.
Also
supportive
of
that
they
are
argument
here.
Is
that
when
you
don't
have
a
process
definition,
you
don't
want
to
be
the
guinea
pig
and
if
another
component
is
extracted
with
a
clear,
decisive
process,
we
can
follow
that
process,
but
they're,
basically
saying
that
they
know
what
you
may
be
able
to
be.
B
D
Yeah
they're
timid,
Justin
in
particular,
very
confused
how
cacao
is
going
to
be
decoupled
for
the
both
the
voluntarily
cycle,
because
it's
the
default
client
of
kubernetes.
We
are
supposed
to
test
something
in
our
release.
We
have
to
cover
the
foot
client
so
once
you
decouple
the
default
client,
you
can
end
up
in
a
situation
where
we
release,
but
there
are
breakages
because
of
potential
drift
between
clients
and
servers.
D
This
can
happen
if
the
schedules
are
decoupled
of
the
cooperating
core,
and
this
is
consider
the
Express
that
we
have
been
talking
about
this
for
a
very
long
time
since
Tiger
that
cannot
know
what
to
follow
and
a
problem
I
have
with
Jordan
in
particular.
This
topic
is
that
he
is
very
supportive
of
the
idea
to
not
the
couple
components
with
commodities
release,
because
their
argument,
netic
dibs,
is
also
in
the
same
club.
So
basically
they
want
to
know
companies
to
release
their
own
on
their
own
schedule
in
a
sense
that
they
shouldn't
be.
D
D
Imagine
that
it
has
a
critical
bug.
So
how
do
you
handle
this
situation?
You
have
to
bundle
a
new
version
of
keep
Cairo
in
a
new
kubernetes
release,
so
Kukoc
have
to
release
and
cooperate,
discuss
release
so
the
whole
decoupling
does
not
work
in
practice,
so
I
thought
she'll
really
sure
what
they
had
take
him
like.
That's
that's
one
of
my
reasons.
I
have
this
document.
Is
that
I
want
to
see
this
like
the
tile
description
explaining?
What
is
the
thought
process
about
this?
D
This
recommendation
to
decouple
the
version
schedule
and,
if
cube
Cairo
extracts-
and
we
start
seeing
a
lot
of
these
problems,
you
know
mass
user
confusion.
What
is
my
cube
conversion
supposed
to
be
to
support
kubernetes
119?
If
we
see
one
of
these
flawless
that
the
extraction
and
policy
is
not
successful,.
D
B
Think,
realistically,
any
other
components
have
this
general
issue
of
what
is
kubernetes
as
as
an
entity
and
what
what
does
version
skew
mean
across
decoupled
components?
Is
it?
Is
it
now
based
on
feature
flags?
You
can
use
any
cube
cuddle
with
any
that
is,
do
they
talk
to
each
other,
but
certain
functionality
doesn't
work.
D
Subtitle
spot
here-
maybe
maybe
I'm
overthinking
this,
but
with
the
whole
decoupling
idea.
Maybe
we
are
going
to
a
place
where
there
is
not
going
to
be
a
couple
in
this
release.
There's
not
going
to
be
a
kubernetes
distribution,
although
you
know
users
are
going
to
have
documents
on
how
they
are
supposed
to
create
their
own
distribution.
Now
pull
these
cooperates.
Maybe
we
can
have
tooling
that
checks
for
compatibility
somehow,
but
this
with
this
decoupling
idea,
I
think
we
are
going
into
that
direction
and
I
think
sig.
D
B
I
think
that's
realistically,
one
of
the
outputs
that's
possible
here
is
the
sig
release
just
ceases
to
exist.
There's
there's
an
automatic
release
when
something
is
tagged
on
the
git
repository,
a
tarball
is
generated,
nothing
else.
Yeah.
B
Mean
the
the
whole,
then
it
gets
back
to
the
point.
I
think
Josh
was
one
of
the
early
ones
to
say,
like
all
of
this
stuff
is
typically
done
by
distributors
and
we've
we've
kind
of
we
have
this
implicit
distribution
weekly
of
some
things,
but
it
it's
in
a
fuzzy
in-between
state.
It's
not
a
real
strong
community
distribution.
D
Yeah
I
provided
some
examples
because,
after
all,
kubernetes
is
not
the
center
of
the
universe.
There
are
some
massive
projects
out
there
that
have
been
doing
some
of
this
for
20-plus
years
the
whole.
How
do
we
release
comprehend
southside
of
core
and
I
gave
some
examples
in
the
document,
so
basically,
people
have
developed
mechanics
to
release
all
the
companies
with
the
same
tag
and
all
the
copper
is
follow
the
same
versioning
code.
You
know
component
details
pretty
much
on
point
where
a
new
release
is
happening.
Everybody
is
active.
D
Pushing
the
appropriate
toxic
leads
to
prepare
a
conference
for
the
Ponto
release,
and
people
are
managing
this
pretty
fine,
no
I'm,
not
talking
about
you,
know,
five
cooperates.
Couplet,
keep
karo,
keep
a
game
and
sub
control
and
I'm
talking
about
hundreds
of
components
with
the
same
version
and
people
are
managing
this
in
the
big
communities
driven
by
open
source.
Another
example
looking
at
KD,
for
example,
people
Balaji
this
way.
If
we
go
to
the
proposal
of
Jordan
and
deems
it
used
to
be
like
we
are
getting
into
the
space
where
we
have
to
disbarred.
C
Yeah
I
think
sorry,
yeah,
Jose,
yeah
I
am
I.
Think
that
yeah
it's
fair
to
say
that
when
we
get
to
that
state.
Yes,
that
would
be
the
case,
but
I
do
think
that
you
I
think
that
if
nothing
else,
the
value
that
in
this
working
group
is
out,
it
is
in
a
place
where
we
could
start
asking
these
questions
and
end
up
with
a
you
and
end
up
with
a
place
where
you
can
go
to
say
you
know,
what's
up
with
this
house,
is
that
what's
the
support
thing
on
that?
How
does
that?
C
B
C
A
C
Like
it
does
like
me,
it
feels
like
there's
no
I
agree
that
that
at
that
point
is
worthwhile,
but
like
the
you,
but
the
path
to
get
there
from
here
is
not
clear
and
one
of
it.
This
goes
back
to
why
we
would
have
been
discussing
why
we
have
been
discussing
the
you
know:
one-year
support
thing
as
you.
C
The
meantime,
like
the
9-month
thing,
is
not
working
for
a
lot
of
people
like
you
know
that
you
I,
think
that
that
the
one
thing
that
we
know
from
the
from
the
sofa
that
we
did
do
is
that
yet
the
nonlin
thing
is
not
working
well,
a
lot
of
girls,
don't
like
it,
and
you
talk
to
people
and
you
the
current
really
schedule
doesn't
work
for
anybody
really
and
so
like
yeah.
We
should
try
and
change
something
and
yeah
that
was
that
was
the
whole
point
of
this
was
to
me
like.
C
Well,
let's,
let's
make
it
a
year
so
that
you
get
the
tie
in
with
the
let's
make
the
support
it
in
a
year
which
is
I,
don't
think
anyone
is
really
arguing
is
long-term
support
a
year
is
not
long-term
support.
Right,
like
a
year,
is
not
LTS
as
what
anyone
it
means
this
was
just
a
year
is
something
that
we
that
is
possibly
achievable
in
a
reasonable
timeframe,
and
you
know
possibly
achievable.
We
knew
was
always
a
stretch
and
the
effort
has
borne
that
out
right
like
that
that
even
getting
people
to
that
may
be.
B
C
Yeah
I
mean
I
think
everybody
has
been
every.
I
I
think
ii
think
the
thing
that's
the
fairest
to
say
about
the
feedback
we've
had
is
that
everybody
has
sort
of
said
seems
okay.
Maybe
he
should
talk
to
them
pretty
much
like
it.
Isn't
like
a
big
circle
of
everyone
being
like
yeah,
hey.
You
should
talk
to
those
people
next
door
yeah.
So
you
may.
B
Be
also
with
an
implicit
like
you're
talking,
you
shouldn't
be
talking
to
me
because
I'm
not
the
one
who's
gonna
have
the
problem.
It's
them
there's
no
way
they
can
do
it,
but
the
like
is.
We
talked
to
them
in
like
yeah.
We
could
do
this,
but
there's
no
way
they
can
do
it
and
it's
yeah.
It's
not
quite
one-to-one
circular,
but
there's
sort
of
a
bigger
circle
of
consistent
disbelief
at
the
possibility
of
this.
Yes,.
C
I
think
that's
that
that's
where
he
has
sort
of
felt
to
me
like
one.
If
we
want
the
support
period
proposal
to
go
ahead
like
it's,
gonna
need
I
increasingly
feeling
like
we
will
need.
You
know
some
top-down
push.
If
you
like
we're
gonna
do
is
and
yeah
you
as
everyone
needs
to
get
on
board
nicely
I.
Don't.
A
D
A
A
C
A
A
A
C
C
Yeah
I
mean
I
think
once
we
hit
so
once
so,
my
feeling
is
once
we
hit
the
Oviatt
most
obviously
affected
SIG's,
then
I
do
feel
like
it
would
be
important
to
get.
You
know
some
cigars
people
or
something
like
that,
maybe
to
just
sort
of
be
like
look.
If
no
other
Sikhs
have
a
problem,
then
we're
okay
with
it
or
something
like
that
like
then,
we
can
be
like
well.
You
know
that
we've
got
we've
got
no
objections
left
like
yo.
C
B
And
maybe
maybe,
since
we've
talked
to
folks,
we
could
just
summarize
that
on
the
cap
we
have
no
objections.
Generally,
everybody
that
we
reached
out
to
in
terms
of
Sikh
leaders
chip
was
supportive
of
moving
forward.
This
is
now
implementable,
ready
to
merge.
Can
we
can
we
say
that
today,
at
this
point,
it's
factually
correct.
B
D
C
With
four
months,
I
mean
I
think
that's
an
implementation
detail
to
be
honest,
I
actually
I'm
having
to
scrub
through
the
cap
and
check
that
I
mean
it
says
everywhere.
It
says
we
are
moving
to
a
year
of
support,
not
we're
moving
to
four
releases
of
support.
So
the
decision
about
whether
or
not
we
do
you
know
three
releases
a
year
or
four
releases
a
year
is
actually
orthogonal
to
this
cap,
because
this
cap
is,
let's
do
a
year
of
support.
B
It
had
been
in
discussion
on
and
off
for
a
couple
of
years
like
should
the
cycle
be
shorter
or
longer,
and
in
this
writing
group
we
said
we'd
leave
that
to
another
kept,
there
have
been
voices
from
within
cig
release
saying
well.
Maybe
we
cig
release
should
go
ahead
and
try
that
kept
for
discussion
even
just
a
week
and
a
half
ago.
There
is
a
fair
amount
of
discussion
because
of
right
now
that
the
119
cycle
is
starting
up.
B
Posting
it
drafting
the
cap.
Up
from
my
perspective
personally,
I
would
like
Tim
Sinclair,
also
had
another
proposal
that
we
shortened
the
cycle
he
had
he's
play
devil's
advocate
in
multiple
directions:
I
like
the
idea
of
a
shorter
release.
Cadence.
This
is
something
also
the
Tim
Hawken
had
wanted,
but
somehow
there's
kind
of
a
major
minor
set
of
harmonics,
maybe
within
that
and
I,
would
potentially
be
willing
to
draft
it
kept
along
those
lines.
B
It's
maybe
more
in
the
direction
of
Tim
Sinclair's
a
year
year
and
a
half
ago
do
I
commit
that
he
wrote
up,
but
just
a
simple
shift.
Two
to
four
month
releases,
I
I
feel
like
that's
a
slippery
slope
in
the
wrong
direction.
Personally,
where
it
encourages
because
of
the
elongation
it
encourages
entity,
is
to
try
to
get
things
in
that
aren't
ready,
because
if
you're
gonna
base
something
off
of
this
trimesters
release,
you
you
need
stuff
to
be
in,
but
maybe
it's
really
only
at
an
alpha
or
something
state.
B
But
you
you
want
to
get
it
promoted
sooner,
because
you
want
a
product
or
a
service
to
base
on
it
and
it
snowballs
forward
then,
and
we've
seen
this
on
in
this-
is
the
classic
downside
of
the
long-term
support.
People
know
that
that
longer
support
term
is
is
coming.
The
next
one
is
about
to
start
so
they
they
push
to
get
extra
stuff
in
and
that
invariably
decreases
quality
for
a
while,
and
it
takes
longer
to
stabilize
the
thing
into
a
support
of
bull
state
and
it's
a
it's.
B
C
In
regard
to
the
current
cap,
though,
do
you
like
my
feeling
is
that
the
current
cap
is
about
moving
to
one
year
of
supportive
releases
here
and
then
the
that,
whether
or
not
that
is
done
with
a
three
releases
or
four
releases
per
year
is
orthogonal
to
the
decision
about
whether
or
not
we
want
it
to
be
a
year.
You
know
like
it's.
You
know.
The
important
part
here
is
to
get
everyone
to
agree
in
principle
that,
yes,
we
want
to
support.
A
C
Three
or
four
releases
per
year
is
then
a
separate
discussion
because
he
is
a
separate
discussion
because
it
has
all
of
those
other
flow
on
effects.
But
if
we
can
agree
that,
yes,
we
should
support
for
a
year
because
of
all
the
other
things
that
at
least
means
that
we
can
have
got
a
wide
agreement
across
we've
got
a
template
to
a
Trotter.
To
then
do
that
change
in
who
you
need
to
talk
to
how
you
talk
to
them?
C
How
you,
how
you
actually
push
a
cap
like
this
through
you
know-
and
this
is
again
that's
the
part
of
the
point
of
this
cap-
is
to
push
a
cap
that
is
this
big
and
spans
across
this
many
SIG's
through
the
project
and
have
it
you
and
I.
Have
everyone
be
in
a
yeah
and
and
have
it
roll
out
and
have
it
and
have
everyone
agree
that?
Yes,
that's
the
thing
that
we
did.
B
We
go
ahead
and
just
should
we
just
try
and
move
things
forward.
We've
had
so
much
conversation
back
and
forth
on
the
different
things.
Should
we
just
state
on
the
cap
that
we
believe
it's
ready
based
on
stakeholder
input?
There
were
no
objections
to
the
viability
of
it
that
we
believe
it's
implementable.
At
this
point,
I
mean.
C
I
think
the
strong
caveat
that
we
need
to
do
is
to
make
sure
that
that
the
wording
says
that
this
is
a
one
year
and
that
you
have
something
in
there
saying
yeah.
We
acknowledge
that
today's
discussion
about
whether
or
not
how
many
releases
there
are
per
year
and
how
often
that
releases
are
yeah,
but
I
was
but
we
kill,
but
that
the
minister,
this
kept
feel
that
these
back,
that
discussion
is
orthogonal
to
this
one.
D
D
The
foxes
out
of
their
holes,
yet
so
that's
a
possibility
to
move
forward.
I
mean
having
the
adult
aspect
in
front
of
different
six.
You
know
here
is
the
link
for
the
cap.
Just
have
a
look.
That's
sufficient,
see
Mike
from
anesthetic.
If
people
forget
or
don't
have
the
time
to
review,
we
can
just
continue
with
merging
yeah.
B
So
we're
shifting
to
a
monthly
cadence
would
it
make
sense
to
go
ahead
and
we
could
scrub
that
double-check
that
there's
that
the
document
is
just
about
the
time
and
then
post
cave
and
say
we're
looking
for
lazy
consensus
by
this
deadline
and
state
maybe
a
week
before
the
next
meeting,
so
that,
based
on
whatever
feedback
we
get
in
that
discussion,
we
can
then
at
our
next
meeting,
close
the
loop
and
here's
a
that
would
give
like
if,
if
over
the
next
week,
we've
scrubbed
it
to
believe
that
we're
we're
happy
with
what
we
believe
we're
asserting.
B
C
C
C
C
B
C
Started
before
we
do
go
Tim,
could
you
check
that
I
have
updated
the
calendar
correctly
and
that
the
next
meeting
is
on
my
I
think
I
have
okay.