►
From YouTube: WG Component Standard 20190723
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
Okay,
good
morning,
everyone
and
welcome
to
the
tuesday
july
23
2019
working
group
component
standard
meeting.
We've
got
kind
of
a
small
crowd
and
a
short
agenda
today,
so
it
doesn't
look
like
lumiere
is
here
yet
so
I'll
leave
that
first
one
just
you
know
in
case
he
shows
up
and
then
jorge.
B
So
one
day,
one
of
the
one
of
the
issues
is
that
I
wasn't
particularly
sure
about
all
the
things
legacy
flags
where
they
were
supposed
to
be
used
for
and
based
on,
the
comments
and
the
mention
of
the
other
pr.
That's
our
work
in
progress
for
the
x
with
a
k1,
hey.
A
Hey
I
already
your
mic
or
maybe
your
connection
might
be
dropping
out.
Do
you
want
to
just
double
check.
B
B
It's
not
I
checked,
and
there
is
no
way
of
actually
configuring
the
there's
a
there's,
no
way
of
telling
a
copper
command
to
use
this
particular
type
of
facet.
B
So
the
only
way
of
modifying
the
only
way
of
working
with
the
integrating
the
legacy
flagship
with
the
ap
flagship
is
to
call
the
corporate
command
a
flags
function
which
is
going
to
return
a
reference
to
the
to
the
flag
set,
and
then
from
that
we
can
have
a
I.e.
I
essentially
just
follow
the
the
k
log
x,
a
the
k
log
example
to
write
a
to:
u
to
use
it
in
cube
proxy
to
register
to
register
all
the
flags.
So
that's
a
huge
ad
flex
function
that
I'm
referring
to
that.
B
Yeah
so
yeah
sorry
so
wait.
One
way.
One
of
my
big
blockers
was
the
fact
that
I
never
actually
looked
at
p
flags
under
the
hood.
I
just
been
using
it.
You
know
I'm
a
normal
user
until
until
last
week,
so
so,
for
example,
for
a
lot
of
cube
proxy,
he
has
some
a
funny
way
of
setting
flags,
so
I
had
to
go
back
into
the
actual
code,
see
how
it
does
it,
and
that
and.
A
B
See
how
to
how
I
could
best
use
legacy
flags
to
actually
create
a
bunch
of
a
bunch
of
apply
functions,
so
I'm
working
on
that
actually.
Actually
this
morning
I
got.
I
got
to
a
point
where
I,
where
I
could
actually
compile
your
proxy
and
the
flags
were
still
registered
and
I
was-
and
I
was
creating
and
now
in
and
I
replicated
for
the
most
part,
the
structure
of
the
k.
Log
example,
where
you
register
a
flag
from
a
using
legacy
flagship.
A
Do
you
want
to
send
me
a
link
to
like
just
like
a
github
commit
hash
of
kind
of
what
you're
working
on?
I
just
want
to
like
make
sure
we're
not
doing
it
in
a
more
complicated
way
than
we
have
to.
B
B
So
so
the
k
log
example
it
creates.
It
creates
a
simple
it's
in
one
of
the
comments
in
my
pr
yeah.
It
creates
a
small
configuration
it
currently,
it
creates
a
small
configuration
and
then
it
registers
the
flags
using
legacy
it
and
it
it
iterates
through
all
the
return
values
to
yeah.
So
if
you
go
to
the
comments
sure.
A
B
Yeah
and
for
example,
if
you
go
to
files
change,
I
think
there's
a
file.
That's
called
example.
A
B
In
that
one,
you
call
the
so,
for
example,
looking
at
the
main
function
you
create
the
flag
set
with
k,
flags
new
flagship,
then
apply
flags
calls
options,
add
flags,
so
they
add
flex
function.
I've
been
trying
to
mirror
it
as
much
as
possible
in
my
in
my
and
I
think
the
code
for
that
should
be
right,
either
right
below
or
right
about,
the
file.
A
A
A
Yeah,
most
of
the
core
components
follow
a
pretty
similar
structure
like
that
and
then
you'll
see
the
new
cubic
command
here
has
cobra
in
it,
but
we
actually
parse
the
flags
ourselves
and
run
so
there's
actually
a
flag
set
that
we
maintain
outside
of
the
command
that
don't
attach
to
cobra,
because
cobra
does
some
things
where,
like
there
are
a
number
of
sort
of
not
really
edge
cases
but
cases
in
cobra
where
it
will
automatically
merge
in
the
global
flag
set
with
whatever
you're
parsing,
we
were
trying
to
maintain
control
over
that
flag
set,
because
there
are
some
libraries
that
kubernetes
depends
on
that
register
flag.
A
Is
that
we
don't
want
exposed.
So
in
order
to
avoid
that
we
kind
of
have
to
do
it
ourselves.
B
A
A
Some
things
in
here,
like
you,
had
to
take
shortcuts,
necessarily
have
a
good
way
to
do
it
and
like
legacy
flag,
is
one
of
the
things
we're
hoping
we'll
fix
some
of
those
shortcuts
now
but
yeah
in
general.
This
is
a
good
thing
to
read
if
you're
interested
in
how
things
work
or,
if
you're,
just
trying
to
figure
out
solutions
to
some
of
those
problems.
B
B
Okay,
awesome,
yeah
I'll
start
pushing
I'll
start
pushing
it
out.
Today.
I
really
spent
here
my
world
and
there
we
are
three
or
four
times,
but
I
think
I
think
I
think
I'm
getting
the
gist
of
it.
A
Cool
all
right
so
ross,
you
have
an
item
on
here,
organized
component,
config
design
meeting.
Do
you
want
to
talk
about
that.
C
This
is
basically
an
idea
that
I
had
today,
so
we
probably
need
to
like
do
a
like
a
broader
sync
up
between
c
cluster
lifecycle
and
c
api
machinery
folks,
so
we
can
actually
try
and
sort
of
figure
out
where
the
conversion
violation
boundaries
like
and
which
we
should
actually
take.
C
So
basically,
qb
is
hitting
this
problem
right
now
and
we
we
don't
know
what
direction
to
take
exactly,
and
I
figured
out
that
this
is
probably
going
to
be
like
a
question
that
other
cluster
life
cycle
tools
are
going
to
like
face,
for
example,
cops
or
even
the
cluster
add-ons,
and
we
probably
need
some
broader
discussion
and
more
like
focus
discussion
on
what
what
steps
to
take
in
order
to
like
convert
and
validate
component
conflicts
from
the
cluster
lifecycle
perspective.
C
And
probably
this
meeting
is
going
to
produce
some
sort
of
consensus
in
order
to
like
get
the
gap
in
place,
and
I
start
implementing
these
features
around
the
cluster
lifecycle
tools.
A
Okay,
I
think
that's
a
good
idea.
It's
definitely
worth
talking
to
this
folks.
Do
you
want
to
take
point
on
setting
that
up.
C
Yeah
I'll
email,
the
sex
and
we'll
group
components
alert
and
probably
try
to
set
up
some
doodle
to
figure
out
the
the
best
possible
time
slot.
A
All
right:
well,
I'm
gonna
move
lubemir's
item
to
next
week,
so
he
can
actually
be
here
for
that
review
and
then
thanks
so
much
everyone.
You
have
a
nice
day.