►
From YouTube: WG Component Standard Office Hours 20200825
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
Good
morning,
everyone
welcome
to
the
tuesday
august
25th
working
group
component
standard
office
hours.
This
is
our
hour
for
folks
to
drop
in
ask
questions
and
get
help
with
things
that
they're
working
on
I'm
mike
tauffin
and
I'll
be
hanging
out
in
the
channel
to
answer
questions
today.
C
A
B
A
Okay-
let's
see
instance,
configuration
merge
so
just
to
summarize
last
week
I
think
we
talked
about
we're
gonna
try
and
do
it
on
the
yaml
itself
sort
of
like
a
strategic
merge
or
something
like
that
right
right.
All
right,
let's
see.
A
Did
you
run
into
any
problems
on
the
way
there
or
was
it
pretty
straightforward?
No.
A
Straightforward
great:
did
you
have
a
chance
to
confirm
like
whether
we're
able
to
specify
a
strategy
for
each
option.
B
A
A
A
And
then
we
just
strategic,
merge
patch.
Okay,
that's
awesome!
If
it's
that
simple,
that
is
really
great.
I
guess
let's
go
read
about
strategic,
merge
patch
and
like
make
sure
it
meets
all
our
requirements
but
yeah.
If
it's
this
straightforward,
let's
look
at
cubelet
and
see
like
what
happened
in
cuba.
B
A
A
Maybe
let
me
let
me
give
you
co-hosts
and
you
can
share
that'll,
be
easier.
A
If
it's
too,
if
you
can't
switch
over
to
being
able
to
share
right
now,
I
can
do
it,
and
you
can
just
tell
me
where
to
go.
If
that's
easier,
let
me
show
you:
okay,
cool.
A
I
see
scheduler
instance
configuration
merge
and
that's
all
I
see.
Okay,
that's
that's
fine!
I
see
well,
I
see
your
dock
too,
but
that's
just
pictures
all
right.
A
B
B
That's
where
you
define
the
stacks
of
the
strategic,
merge
patch
right,
okay
and
probably
we
need
to
do
in
the
external
instead
of
the
internal
to
because
they're
already
we
already
have
text
there.
So
not
a
problem
too
cool.
B
B
A
Guess
I
guess
the
zero
values,
but
those
should
just
get
overwritten.
B
Oh,
did
this
rule
return
a
byte
array.
B
A
A
I
would
just
make
it
a
new
cap
so
that
you
have
control
of
the
proposal.
Okay,.
B
A
A
B
A
Right,
like
it
just
goes
through
the
same
decoding
path
like
it's
yeah,
because
we're
doing
everything
straight
on
the
ammo
at
the
start,
and
then
we
get.
I
guess
we
get
a
bite
out.
Do
we
does
does
strategic
merge
patch
like
deserialize
it
and
then
serialize
it
again
like
how
does
that
work
on
the
inside.
B
A
Even
if
it
even
if
it
did
go
through
a
round
trip,
I
think
that
would
well
that's
the
thing
we
got
to
watch
out
for
I
guess,
if
it,
if
it,
we
have
to
make
sure
that
the
output
of
strategic,
merge
patch
is
limited
only
to
fields
that
were
set
in
the
inputs,
if
for
some
reason
that
if,
for
example,
if
it
does
deserialize
everything
into
that
object,
you
pass
in
and
then
re-serialize
it
on
the
way
back
out,
then
it's
possible.
A
But
I
just
would
want
to
double
check
that
we
don't
get
any
extra
fields
like
set
to
zero
on
the
way
out,
because
then,
if
we
run
it
through
the
same
path,
that
would
be
treated
as
explicit
user
input,
even
though
it
wasn't
right
like
if
you,
if
you
throw
like
you,
know
something
with
field
a
set
and
a
patch
with
fuel
they
set
into
the
merge
patch
and
your
patch
gets
applied,
but
does
it
come
out
with
field
b
set
to
zero?
A
Just
because
that
object
had
zero
values?
That's
something
to
double
check.
A
I
don't
think
it'll
happen.
I
think
I
think
emit
empties
should
be
set
correctly
and
I
don't
think
it
should
happen,
but
I
would
want
to
definitely
like
test
corner
cases
around
that,
for,
like
just
you,
know,
basic
values,
but
also
like
substructures
or
like
things
that
are
pointers.
You
know,
do
you
because
you
know
some?
A
Sometimes
the
mid-empties
are
set
weird
and
you
get
like
a
yaml
output
where
the
like
the
key
will
be
set,
and
you
know
like
like
it's:
it's
basically
like
an
embedded
struct,
but
it
just
shows
up
as
like
empty
basically
in
the
output.
So
I
I
don't
know
something
to
double
check
when
we
test
it
but
yeah.
This
looks
really
slick.
B
Okay
in
this
scheduler
could
merge
substructures.
You
have
like
deeper
structures.
A
Version
and
it
works
yeah
wow,
but
and
it
should
I
I'm
hoping
we
can
set
for
each
one
of
those
as
like
an
option
of
you
know.
Maybe
we
interleave
it
or
maybe
it's
just
a
false,
like
replacement
of
the
top
level
structure,
depending
on
on
what
the
what
it
is.
B
Okay,
we
can,
we
can
bring
in
the
document
in
doc
of
the
cap
doc
this
corner
case.
Now
we
should
treat
these
these
things
yeah.
So
the
the
deal
with
the
cubelet
one
is
the
dynamic
configuration
right
that
I
need
to
pass
to
the
controller
the
instance
config
file
as
well.
A
Inject
it.
So
do
you
see
if
you
scroll
up
where
the
controller
is
initialized,
that
cubic
config
flag
precedence
function
is
already
attached?
Basically,
so
I
wonder
if
we
can
call
the
merge
there,
oh
here
in.
A
B
B
So,
that's,
that's,
that's
the
only
complicated,
more
complicated
stuff,
but
we
just
put
passes
and
I
think
yeah.
A
So
that's
it
that's
yeah!
This
is
way
simpler,
definitely
way
simpler.
A
Yeah,
I
think,
let's
propose
this
approach
and
we
can
make
sure
that
we
address
like
those
corner
cases,
while
we're
drafting
the
new
proposal.
A
A
Okay,
you
so
let
me
find
the
link
to
the
original
cap.
A
A
I
would
basically
just
copy
this
literally,
I
mean
maybe
change
the
date,
because
your
proposal
will
be
a
new
one,
but
literally
like
put
it
at
the
same
path,
and
you
know
you
can
copy
off
like
the
this.
A
Basically,
the
summary,
the
problem
statement,
why
we
want
to
do
it,
what
the
goals
are
all
that
kind
of
stuff,
but
just
modify
it
so
that
it's
your
approach
with
the
strategic
merge
patch
instead
of
two
kinds,
and
you
can
point
to
the
this
cap,
the
original
one
and
say
like
we
prototyped
this
and
you
can
point
to
your
prototypes
and
and
say,
like
it's
very
complicated
and
look.
How
much
simpler
strategic
merge
patch
is.
A
A
So
what
does
your
time?
Look
like
the
next
couple
weeks.
A
Right
we
have
the
weekends
I'm
gonna
be
on
vacation
on
my
honeymoon
after
september
20th
until
october,
so
I
will,
I
won't
be
around
like
or
working
on
anything
that
time,
so
I
I
don't
know
not
super
urgent
like
it,
but
and
until
then,
I'm
like
swamped
with
work
stuff
to
handle
before
I'm
gone.
A
B
A
Yeah
that'd
be
awesome
if
possible,
but
you
know
don't
work
too
hard.
You
know
yeah.
A
Yeah,
that
sounds
good.
Do
you
have
any
other
questions
before
you
get
started
on
the
proposal?
No,
that's
all
michael
awesome.
Thank
you.
I
mean
take
care
and
good
luck.
Thank.