►
From YouTube: CentOS Automotive SIG meeting - 2021-12-15 US am
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
Well,
hello,
recording
I'm
here
just
in
case
anyone
else
shows
up.
I
was
going
to
briefly
go
through
pierre's
update
and
invite
an
open
floor,
but
there's
no
one
else
here
right
now:
jeff
rowe,
hope
you're
doing
well.
If
you're
the
one
first
person
to
see
this
we're
thinking
of
you.
A
B
A
Nice
that
same
one,
I
have
the
the
logitech
right,
yeah.
B
Yeah,
the
1080
pro
67
box
of
walmart
yeah,
I
wasn't
gonna,
take
the
whole
sony,
aps-c
mf
capture
or
the
g9,
or
one
of
these
from
my
stupid
boom
micro.
B
A
B
Upshot
is
someone
very
dear
to
me,
went
out
and
had
a
visit
and
they
have
dogs
a
la
black
lab
and
a
shepherd
kind
of
still
a
puppy
and
went
to
the
dog
park.
So
I
walked
along
the
river
with
the
dogs
and
I
have
a
new
appreciation
for
for
dogs.
I
hadn't
had
a
dog
in
the
family
since
I
was
a
kid
and
someone
with
a
well-trained
couple
of
dogs
and
they
get
goofy
with
each
other,
but
not
in
a
painful
way.
B
A
A
I'm
a
big
fan
of
mark
maron
and
he's
famous
for
having
cats.
He
always
ends
his
podcast
with
boomer
lives.
This
is
one
of
his
feral
cats
that
just
disappeared
one
night
and
he's
in
his
in
his
mythology.
Boomer
is
still
alive
somewhere,
but
they're
still
not
just
not
where
he
is
boomer
lives,
but
yeah.
When
he's
asked
like
I
just
stuck
in
my
mind
when
he
you
know
he
was
talking
to
one
of
his
guests,
this,
oh
yeah,
I
don't
have
dogs.
I
don't
want
anything
to
love
me
that
much.
A
D
A
A
B
A
Hey
everybody,
hello,
hello.
I
have.
I
had
not
appreciated
that
jeffree
was
out
so
and
I
don't
have
a
firm
agenda
here.
I
think
we've
got
everyone
we're
gonna
get.
So
I
guess
I
will
just
this
is
like
when
they
when,
when
the
when
congress
is
mostly
vacant,
but
somebody's
still
there
like
reading
into
the
record,
so
let
the
record.
B
B
A
So
pierre
wanted
to
highlight,
if
the
for
those
who
are
interested
in
it,
I
always
feel
guilty
that
I'm
not
as
up
to
date
on
this
as
I
should
be,
but
they
have
done
a
fair
amount
of
work
to
streamline
both
the
the
sample
manifest
that
we
have
in
the
sig
for
these
sample
images,
which
reflect
our
sort
of
early
examples
of
what
customer
use
cases
are
they've,
also
enhanced
the
instructions
for
building
them
so
we'd
set
generally
like
it
would
be
great
if
you,
if
you're
coming
to
the
sig
and
we're
advocating
that
these
are
the
kind
of
things
we
want
to
do,
and
these
are
the
tools
we
want
to
do
it
with,
and
these
are
the
packages
and
whatnot
easier.
A
It
is
to
engage
the
better
and
they've
taken
that
to
heart,
and
he
wanted
to
highlight
that
he's
also
been
very
good
about
giving
regular
updates
on
the
mailing
list.
We
we
do
have
some
progress
on
the
auto,
colonel
and
that's
being
done
in
public.
If
clark
were
here,
I
would
ask
him
to
speak
to
it,
but
he's
not.
A
But
that's,
okay,
hello,
whitney
welcome
we're
just
reading
into
the
congressional
record
here
and
oh
and
now
she's
gone
well,
no
important
representatives
here
so,
but
that
that
is
on
the
cusp
of
being
included
into
our
initial
builds,
but
I
don't
think
we're
quite
there
yet,
but
it
does
exist
and
the
the
the
what
we're
hoping
to
firm
up
in
the
next
few
weeks
once
well.
A
I
guess
in
the
new
year
this
will
be
now,
is
let's
go
from
building
these
repositories
and
showing
them
making
sure
we
have
the
kernel
in
there
and
then
eventually
start
publishing
some
images
and
those
are
a
real
good,
quick
start.
So
images
for
or
vagrant
of
the
raspberry
pi
4
we've
shown
these
repeatedly
and
we'd
like
those
to
be
front
and
center
for
people
who
are
looking
into
the
sig.
And
hopefully
you
know
in
my
dreams
we
would
you
know,
maybe
you
know,
maybe
a
hardware
vendor
would
like
to
add
theirs
to
the
mix.
C
B
B
I
don't
imagine
if
someone
ran
centos
stream
9
as
their
host
instead
of
what
it
says
here.
Cs8
is
that
that
wouldn't
make
much
of
a
difference
right.
A
I
don't
think
I
don't
anticipate
it
would
at
the
time
this
was
initially
written.
I
think
centos
stream
9
was
in
a
little
bit
rougher
shape
than
it
is
now,
like
apple
wasn't
available
for
centos
stream
9,
for
example.
I
think
it
is
now,
but
I
I
would
bet
that
it
would
work
and
yeah,
then
the
manifest
themselves
and
the
images
are
being
built
off
of
content.
That's
mostly
from
centos.
B
I
like
how
it
touches
on
the
things
it
does.
My
I
kind
of
have
a
classical
embedded
in
linux
in
in
the
sense
of
I'm
into
using
a
distro
that
has
you
know,
rpms
or
devs.
I
want
to
get
into
it
like
be
inside
it
install
things
like
do
lots
of
things
as
root,
and
I
didn't
do
a
lot
of
copper.
Although
I've
done
other
some
of
these
alternative,
you
know
fedora
build
kind
of
repos
where
they're
individual
people
things.
A
Yeah
just
to
get
a
sense
of
what
what
that
really
means
and
we'll
say
like
we
simultaneously
with
some
of
our
earlier
engagements
just
speaking
in
the
abstract
here
that
if
someone
is
starting
trying
to
understand
what
development
looks
like
on
an
rpm
based
system,
if
they're
perhaps
not
familiar
with
that
already,
we
want
to
have
a
model
where
people
can
get
a
more.
I
mean
in
the
sort
of
silver
blue
fedora
core
os
model.
A
We
would
encourage
them
to
use
a
toolbox
container
log
in
and
start
doing
so,
they'd
create
a
container
image
and
that's
where
they
could
do
their
rights.
They
could
add
to
their
build
root
and
that
kind
of
thing,
but
that's
a
I
view
that,
as
as
a
bit
of
a
barrier
to
entry
for
someone
who's
looking
at
this
for
the
very
first
time,
so
we
do
also
have
this
isn't
highlighted
in
the
sig
documentation.
Right
now,
but
centus
nine
stream
itself
publishes
a
vagrant
image,
which
is
a
pretty
popular
way.
A
It's
a
bit
like
stephen
in
fact
so
welcome.
I
was
you
you're
the
you're,
the
first
person
who
I
recognize
as
a
as
a
genuine
external
participant.
So
I
will
just
repeat
what
I
said
earlier:
is
that
we're
sort
of
in
the
early
stages
of
the
red
hat
holiday
shutdown,
so
jeff
rowe,
is
off.
A
I
did
not
have
a
specific
agenda
other
than
to
highlight
some
of
the
work
that
had
been
done
in
the
sig
repo
to
improve
documentation.
I
was
going
to
open
the
floor
to
any
other
discussion.
People
wanted
to
have,
but
also
make
it
clear
that
we
could
we
could
end
early.
I
didn't
want
to
leave
anyone
hanging
might
have
joined.
E
Did
you
have
have
we
have
we
discussed
on
during
these
meetings,
the
the
release
cadence
and
doesn't
need
to
be
brought
up
during
this
meeting?
Sorry,
it's
new
to
me
this
this
concept
so.
A
I
I
don't
know,
as
we
have,
I
think,
if
jeff
rowe
were
here,
he
would
highlight
that
this
would
be
the
place
that
we
could
do
it
and
we
have.
I
have
personally
shied
away
for
I'll,
admit
chad
away
from
talking
about
it,
but
now
maybe
the
time
may
have
come
to
start
being
a
little
more
formal
about
that.
So
I
wouldn't
want
to
go
too
deep
into
it.
This
particular
time
gotti,
because
we
don't
have
good
representation
from
our
stakeholders
in
the
city,
but
seems.
A
D
Stephan
here
just
a
question
following
the
oh:
we
want
to
align
on
center
stream.
D
Will
it
be
done
on
a
let's
say,
a
regular
basis,
meaning
for
example,
you
want
to
do
it
once
per
month.
I
don't
know
how
you
would
reconcile
things,
but
or
is
it?
Is
it
done
very
regularly,
so
so
that
centuries,
automotive
would
be
a
kind
of
rolling
release
as
center
stream
is
rolling
that
what.
A
The
the
latter,
what
you
just
described,
is
what,
to
the
extent
that
we're
firming
that
up
that
that's
what
I've
been
guiding
people
towards
that.
I'm
really.
I
really
happy
with
the
centos
stream
model.
I
think
it
is
a
huge
improvement
for
people
who
want
to
be
engaged
with
with
both
rel
and
centos,
and
I
was
hoping
to
mirror
that
so
yes,
we've
talked
about
formalizing.
A
The
fact
that
at
least
from
the
one
of
the
red
hat
contributions
in
the
sig
is
going
to
be
produced
to
produce
a
sort
of
rolling
release
that
mirrors
what
we
anticipate
delivering
in
the
vehicle
product
in
a
way,
that's
as
as
similar
as
possible
to
the
way
that
stream
functions
in
that
way
for
what's
going
to
land
in
the
rel
product.
So
whether
we
you
know,
regardless
of
what
we
end
up
doing
in
terms
of
a
more
formal
release,
cadence
there'll,
be
an
element
of
what
we
surface
in
the
sig.
A
That
will
be
more
of
a
rolling
release.
That
is
the
sort
of
streamlike
artifact
for
the
in
vehicle
os.
Now
I
think
to
be
to
be,
and
by
and
stream
like,
I
mean
in
particular
centos
9
stream,
which
is
the
first
pentos
9
and
rel.
Nine
are
the
first
release
where
we
have
this
model
in
place,
essentially
from
birth,
from
the
ga
or
launch
of
the
rel
release.
A
So
there
are
some
things
that
are
a
little
different
about
the
way
that
centos
9
is
being
managed,
since
it's
nine
stream
versus
rel8,
and
we're
definitely
going
to
take
our
inspiration
from
centos
9
stream.
So
stefan,
if
I
say
I
think
we
would
like
one
of
the
key
things
to
be
a
sort
of
rolling
release
artifact
similar
to
what
stream
is
that
sounds
sensible.
Appealing,
desirable.
D
Yeah,
I
think
so,
at
least
in
the
early
phase
of
the
project,
because
my
my
question
was
a
bit
like
what's
going
on.
If
I
take
the
the
chair
of
a
customer
what's
going
on
for
him,
I
mean
at
some
point
you
will
want
to
stabilize
your
product
right,
and
so
I
would
expect
that
this
customer
will
wait
for
a
kind
of
enterprise
release
line
in
three
years
from
now
to
get
a
stable
product.
D
So
what's
going
on
in
between,
maybe
those
those
people
would
like
to
have,
let's
say,
punch
release
or
something
I
mean,
even
if,
if
if
the
the
distribution
is
rolling,
which
is
a
good
a
good
thing,
in
my
opinion,
at
some
point,
people
want
to
have
something
more
stable
to
develop
and
don't
care
about
the
rolling
stuff
you
see
they
want
just
to
to
build
something
on
top
of
it,
with
the
base
being
stable.
D
So
I
just
I
was
just
wondering
how
you
can
reconcile
that
with
a
central
stream
and
before
that
enterprise,
pops
up.
A
A
That
you
would
and
say,
like
I
said,
what
would
the
suse
equivalent,
for
example,
is
tumbleweed
right
yeah?
Is
it
still
called
that
yeah?
No,
then,
that
that's
rolling
in
the
sense
that
it
rolls
between
the
it
rolls
through
the
changes
that
they
would
make
in
between
major
releases
right,
and
that
sense
is
not
a
good.
That
means
that
you,
you
can't
ever
really
stabilize
on
it,
because
it
doesn't
promising
to
be
stable.
D
Yeah,
so
this
is
a
bit
different
because
it's
a
somehow
they
do
what
what
you,
what
red
hat
used
to
do
with
centuries
I
mean
ascend
to
us,
is
discontinued
next
year.
D
D
I
don't
know
one
in
six
months
from
now,
maybe-
and
so
that's
the
same
model
for
me
as
the
one
that
that
red
hat
used
to
have
with
the
centers,
but
now
as
centers
disappeared,
and
we
don't
have
yet
the
replacement
for
this
kind
of,
let's
say
stable
release
based
on
the
rolling
one.
D
D
When
we
talk
about
about
that
with
customers,
we
always
say
we
try
to
line
on
the
red
hat
and
they
know
that
center
has
stopped
sent
to
a
stream,
is
okay
but
sent
to
us
stopped,
so
they
always
talk
about,
but
yeah,
okay,
I
understand
your
your
your
road
map,
but
what's
going
on
in
the
three
three
years,
the
three
up
upcoming
years,
if
I
want
to
to
start
a
product
earlier
and
with
the
answer
is
your
project
is
probably
longer
than
three
years?
So
it's
not
a
problem.
A
I
just
highlight
again
that
I,
I
would
feel
perfectly
comfortable
saying
to
a
rel,
nine
isv,
for
example,
or
customer
that
absent
the
sort
of
support,
promises
and
delivery
and
whatnot
that
there's
not
they're
not
going
to
be
ill
served
by
looking
at
streamstream
is
not
going
to
make
jump
changes
equivalent
to
what
fedora
does
it's
really
just
an
early
preview
of
what's
going
to
land
in
the
product?
And
that's
when
I
say
I'd
like
to
see
something
similar
for
the
in-vehicle
os
and
we've
yet
to
come
up
with
a
name
for
this.
A
We
by
the
way,
that's
what
I'm
thinking.
I
don't
think
that
a
customer
would
the
notion
that
you
would
do
some
work
on
the
stream,
the
vehicle
equivalent
of
stream
and
then
decide
that
you
really
want
to
formalize
it
for
deployment
on
the
in
vehicle
os.
That's
not
it's
not
going
to
be
a
huge
diff
between
the
two,
much
like
working
with
centos.
Nine
stream
gives
you
a
very
clear
view
into
what
real
nine
is
going
to
be.
D
We
had
some
questions
for
motorcycles
for,
for
example,
small
motorcycles,
low
end
motorcycles
in
india,
okay
on
very
low
end
boards,
okay,
and
so
you
understand
that
those
kind
of
products
can
be
developed
in,
I
would
say,
maybe
not
months,
but
I
would
say
one
year
or
two
something
like
that:
yeah
yeah,
so
it's
cheap
and
they
don't
want
to
invest
more
on
development,
so
it
has
to
be
fast.
D
A
I
mean,
I
think,
if
you
thought
you
were
gonna,
have
an
ecu
and
a
motorcycle
in
two
years.
I
would
look
at.
I
would
look
at
what's
coming
up
here.
I
think
that
would
be
the
candidate,
but
I,
if
you
are,
if
you're
interested
in
that
and
again
this
isn't
so
automotive
specific,
but
I
know
that
there
are
some
changes
and
some
efforts
to
provide
something
that
looks
more
like
well.
A
Let
me
put
it
in
the
past
the
answer
for
what
should
I
do
to
be
aligned
with
what
rel
10
might
be
in
three
years?
The
answer
would
be
follow.
What's
going
on
in
fedora
for
a
year
and
a
half
to
two
years.
This
is
just
my
impression.
I
don't
you
know,
don't
don't
take
this
as
gospel
and
then
watch
for
early
early
availability
or
beta
availability
of
rel
10
about
a
year
or
year
and
a
half
before
it
comes
out.
A
I
know
there's
some
effort
afoot
to
have
there
be
something
that
looks
a
little
bit
more
like
a
rolling,
a
true
rolling
rel
like
release,
but
I'm
not
I'm
not
really
well
up
to
date.
On
that,
I
don't.
I
wouldn't
feel
comfortable
representing,
but
their
efforts
to
smooth
out
the
discontinuity
that
used
to
happen
sometimes
where
fedora
would
progress
for
a
while
and
then
we
would
snapshot
it
and
stabilize
it
in
the
run
up
to
the
next
rel.
That's
evolving
somewhat,
but
I
don't.
I
can't
speak
to
it
in
detail.
B
That
was
an
interesting
market
point.
The
idea
of
a
low-cost
motorcycle
that
that
manufacturer
may
not
care
about
ota,
for
example.
They
just
want
to
get
it
in.
It
works
good,
that's
it,
but
maybe
they
do
a
wi-fi
connected
motorcycle.
Maybe
that's
an
exciting
thing.
F
They're,
wi-fi
connected
scooters,
the
whole
bird
infrastructure
and
the
whole
limewire
infrastructure
are
all
wi-fi
enabled
because
you
have
to
unlock
them
with
your
phone.
A
B
A
B
A
Yes,
well
that
I
was
happy
to
see
what
I
was
less
happy
to
see
is
we
started
watching
a
1970s,
sitcom
or
procedural
on
amazon,
prime,
and
it
started
me
showing
showing
me
ads
for
dietary
supplements
to
prevent
constipation
and
once
I
realized
that
these
aren't
like
these
aren't
the
ads
that
everyone
sees
for
this
show
they're.
These
are
being
these
are
being
catered
towards
me.
I
said
why
are
they
doing
that?
But
no,
it
was
an
iot
sim
card.
A
A
B
Was
a
yeah
yeah
it
there
and
I
forget
what
I've
got
a
link
somewhere
and
basically
the
5g
networks
have
plenty
of
bandwidth
and
some
of
these
low
bands
reach
out
far
and
having
the
power
meters
of
the
world
chirping
over
5g
is
not
really
going
to
do
much
to
the
network,
and
so
they
make
it
accessible
and
they
kind
of
bake
the
you
know
it's
like
it's
like
the
card
is
probably
from
an
mvno
that
uses
you
know
made
a
deal
with
a
carrier
for
just
those
low
beat
protocols.
Perhaps.
B
A
So
it's
50
bucks
for
a
year,
but
when
I
looked
it
up,
the
coverage
map
was
it
was.
It
was
a
different
type
of
they
had
a
name.
There
was
a
term
of
art
which
I
think
is
what
you're
referring
to
bruce,
where
the
rf
is
actually
different
and
it
doesn't.
The
coverage
for
their
lte
network
is
not
the
same
as
where
they
support
this
modulation.
Actually,
but
anyway,
this
is
getting
far
afield
of
but
yeah.
I
guess
the
thought
was.
A
F
We've
got
about
two
minutes
left,
so
I
don't
know
if
I
want
to
surface
this,
but
I
think
on
our
last
meeting
ian
we
had
talked
about
how
do
some,
how
do
one
of
the
questions
that
we
heard
was
as
we're
gathering
requirements
downstream?
How
do
those
requirements
find
their
way
to
this
group
and
I'm
curious
from
your
experience
just
with
open
source
projects
in
general?
You
know
when
requirements
are
coming
in
downstream.
F
Are
there
other
mechanisms
for
capturing
requirements
upstream?
Where
either
you
know
yeah
I
mean
like
it's
right.
It
seems
like
counterintuitive
right
everything's.
All
the
work
is
flowing
upstream
to
downstream,
but
all
the
requirements
are
coming
downstream
and
trying
to
find
their
way
back
upstream.
Just
I
don't
know,
maybe
I'm
over
over
into.
A
No,
I
think
it's
no
it
as
always
it's.
You
know.
You
very
frequently
frame
these
like
thoughtful
higher
level
questions
in
a
way
that
I
you
know,
I
I
I
think
I
know
what
you
mean
and
I
can
immediately
see
where
how
I
think
this
is
happening.
I
would
never
have
thought
initially
to
describe
that
as
downstream
requirements
finding
their
way
upstream.
Like
that's
a
good,
I
think
that's
an
insight.
A
A
Another
thing
that
you
know,
depending
on
how
advanced
the
customer
is,
they
tend
to
move
move
closer
to
doing
that
by
themselves
over
time
like
more
advanced
customers.
Might
we
had
a
rel
customer
that
made
extensive
use
of
the
unix
auto
mounter?
This
is
a
very
old
technology.
This
is
an
old
example,
but
they
got
to
the
point
where
they
knew
they
needed
it,
so
they
knew
their
own
requirements
so
well
that
both
red
hat
and
the
customer
were
active
in
the
upstream
directly
to
transmit
these
to
that
upstream
project.
A
A
I
don't
know
that.
There's
one
answer.
I'd
suspect
they'd
give
at
least
the
two
broad
examples
I
did
and
then
some
other
things
I
don't
know
how
helpful
that
is,
but
that's
my
off
the
cuff
I
mean
you
know,
we've
got
other.
At
least
bruce
has
been
in
the
open
source
red
hat
game
for
a
while.
He
might
have
some.
B
Yeah,
that
was
a
good
little
story
there
that
you
put
out,
because
I've
been
for
where
I
was
I
was
in
consulting,
and
so
I
would
be
out
there
actually
embedded
with
the
client
become
part
of
their
team
for
some
period
of.
C
B
Help
them
create
a
technology
and
for
them,
what
it
looked
like
is
the
ability
to
open
a
case
that
turns
into
an
rfe
and
then
get
exposed
to
a
little
bit
of
the
internal
process
involved
around
building
the
business
case
and
how
and
when
red
hat
might
take
that
thing
up
as
the
creation
of
a
new
feature
and
then
that
allows
the
client
to
not
have
to
get
pulled
into
the
upstream
unless
they're
interested
a
lot
of
clients
are
interested
in
the
oh
yeah
where's
that
where's
that
dialogue
happening
and
others
are
like
hey.
B
Let
me
know
when
the
feature's
ready
thanks
right,
yeah
and
it's
cool,
that
it
can
be
done
at
those
various
kind
of
engagement
levels
through
the
same
same
little
interface,
because
bz
is
open.
The
upstream
projects
pretty
much
have
open
mailing
lists
or
and
or
repose.
A
You've
heard
that
before
buxilla
is
a
tracker
that
we
provide
for
both
rel
bugs
and
fedora
bugs,
and
I
think
it's
used
for
some
layered
products.
Ours
is
an
instance
of
that
by
the
way
the
the
bugzilla
tool
is
used
elsewhere,
but
we
have
an
instance
of
it,
but
yeah
so
bruce.
Usually
your
point
is
that
one
way
they
get
upstream
is
by
them
being
product
sort
of
standard
software,
company
interactions
with
red
hat
and
customers,
and
then
they
trans
red
hat,
provides
the
translation
layer
to
the
upstream.
A
A
I
guess
that's
the
third
example.
Another
way
that
they
get
is
that
people
people
participate
in
the
upstream,
but
not
by
contributing
code,
but
by
participating
in
sort
of
umbrella
groups
like
this
sig
or
other
things,
desperately
trying
to
get
a
read
after
whether
you're
finding
your
your
writing.
You
know
I'm
always
trying.
F
No,
this
is
helpful.
You
know
because,
like
I
said,
I'm
super
new
to
the
to
the
open
source
open
source
way,
and
you
know
as
a
as
a
ux
or
an
ux
researcher
and
a
designer
you
know,
I'm
saying:
okay
are
there
requirements
that
I
can
go
out
and
research
for
the
upstream
community
is
like
my
form
of
contribution
to
this
community.
F
Is
it
you
know?
I
I
see
mike
on
this
call.
Is
it
mike
and
I
going
out
and
like
dreaming
up
a
use
case
and
saying
hey?
This
should
be
really
interesting
in
the
automotive
space.
Here's
our
research
as
to
why
we
think
it
would
be
interesting.
How
would
we
enable
this?
On
top
of
you
know
what
we're,
what
we're,
what
we're
putting
together,
you
know
like
where
do
ideas
come
from?
F
Because
we
have
the
technology
right
like
so
we
have
a
dos
right,
so
I'm
never
going
to
swerve
again.
Potentially,
if
I
have
sensors
on
my
brake
lights,
you
know,
can
I
get
a
maintenance
request
sent
in
advance,
so
they
just
send
me
the
replacement
part
or
my
automotive.
You
know
service
service
team
is
like
we
know.
Your
brake
lights
are
going
to
go
out
three
weeks
in
advance.
You
know
so
you're
solving
like
a
social
problem,
but
with
this
technology.
A
First
of
all,
I
could
be
wrong.
It
could
just
be
that
I've
always
been
at
a
lower
level
and
also
doesn't
mean
that
there
isn't
like
the
opportunity
to
bring
some
of
this
higher
level
design.
Thinking
upstream
conversation
like.
B
B
What
they'll
come
to
the
you
know
the
project
in
this
case
it
would
be
the
centos
automotive
sig,
they
would
say,
hey,
there's,
there's
a
dependency
for
this
really
cool
app
and
we
need
it
to
be
in
the
product.
So
hey
centos,
sig
people
put
that
put
that
package
in
it
instead
of
an
apple,
and
then
there
are
talks
about
hey.
Is
that?
Does
that
create
a
lot
of
work?
B
Is
it
widely
applicable
and
so
that
that
kind
of
discussion
represents
a
little
bit
of
the
of
the
mapping
between
what
nick
talks
about
with
a
high
level
idea
and
a
running
app
is
somewhere
they've
got
some
machine
somewhere
they've,
they
got
this
app
cooking
and
they
want
to
get
it
hey.
Will
this
run
in
the
centos
automotive
sig,
because
that
is
unknown?
B
It's
it's
got
enough
stability
and
enough
movement,
and
we
want
to
take
it
for
a
spin
so
that
it
sort
of
sketches
out
how
wha
what
that
would
mean
to
the
centas
group.
I
suspect,
and
in
a
best
case
scenario
the
person
actually
engages
they'll
they'll
come
into
this
group
and
and
say
especially
notional
stuff.
This
is
this:
is
the
the
data
that's
the
hardest
to
get
it's
it's
one
thing.
F
F
So
how
does
that
come
bubble
up
to
this
group
right
where
you
know
it's
coming
from
a
potential
partner,
but
it's
probably
relevant
to
a
lot
of
partners,
so
just
exploring
like
how
ideas
and
requirements
could
come
from
the
upstream
channel
and
then
how
those
requirements
find
their
way
from
the
downstream
requirements
that
are
being
captured
to
the
upstream
communities.
So
this
is
really
helpful.
Thank
you.
I
know
we're
a
few
minutes
over.
So
thanks
for
taking
10
minutes.
D
Just
just
to
complete,
I
would,
I
would
add
that
you
have
to
think
about
that
with
the
long
run,
maybe
because
what
you,
when
you
design
the
system
for
for
a
car.
B
D
Has
to
to
last
for
maybe
10
years
and
from
it
changes
the
perspective
from
your
upstream
or
downstream
project,
because
it's
a
bit
different
for
even
small
companies
or
even
bigger
companies
to
support
something
for
10
years,
and
in
that
case
you
want
to
push
it
to
upstream
projects
so
that
the
the
effort
to
to
support
your
product
or
your
components
on
the
long
run
would
be
split.
D
Probably
so,
there
is
a
mitigation
to
compromise
to
find
between
what
you
want
to
keep
for
you
and
what
you
want
to
upstream
and
also
how
what
you
have
stream
can
be
used
by
by
other
people.
That's
that's
the
point.
There's
three
different
factors
for
me
that
can
help
to
take
the
decision
and
obviously
the
time
is
one
another
another
factor
also
at
the
end,
four
factors
at
least.
B
Interesting
yeah
that
that
longevity
thing
invokes
all
kinds
of
we've:
we've
had
discussions
about.
You
know
how
long
of
an
expected
lifespan
some
of
you,
because
some
people
drive
their
vehicles
30
years,
okay
and
then
now
we're
dealing
with
flash
wear
out.
Do
we
say
the
car
doesn't
work
anymore
because
we've
got
too
many
errors
on
the
flash,
or
do
we
say:
okay,
there's
some
features
that
don't
work
and
some
of
that's
going
to
be
decided
by
the
the
end.
Application
developer,
probably
some
guidance
in
the
project.
B
The
pro
you
know
there
might
be
a
flashware
monitor
that
says:
hey
you
know,
throw
something
in
the
in
the
system:
logs,
hey
it's
time
to
start
thinking
about
new
flash
or
dcom
those
things
both
think
about
the
support
model,
as
well
as
shape
product
itself.
Very,
it's
a
it's
an
interesting
and
unique
place.
This
is
one
of
the
few
places
where
you
have
10-year
types
of
of
life.
We
have
some
customers
with
very
long.
A
In
the
thick
of
it
actually
maybe
about
halfway
through
yeah,
you
know.
B
People
on
these
they're
planetary
missions-
yeah,
oh
for
sure,.
A
B
Start
their
career
as
an
intern
and
they're
like
designing
the
thing,
and
then
you
know
one
of
one
of
my
friend's
dad
was
on
the
viking
lander
team,
and
that
was
basically
the
viking
lander
was
his
entire
career
at
nasa.
Basically,
because
of
the
time
span
of
often
this
that's
a
fun
one
too
yeah
so
kind
of
similar.
Here
we're
going
to
see
vehicles
that
are
running
this
for
a
really
long
time
and
the
underlying
like,
if
you're,
depending
on
ota,
they're,
about
to
decom
the
3g
system
10
years
from
now.
B
A
I'm
glad
you
brought
that
up
and
I
do
think
that's
I
mean
I
feel
like
that's
actually
why
something
that
is
drawn
off
of
what
we
already
do
with
rel.
Is
I
mean
this
guess?
That's
the
point
you
were
making
bruce
is
appealing
that
we,
that
was
the
thing
that
we
were
able
to
bring
early
on
into
the
linux
space,
was
a
slower,
update,
cadence
and
a
longer
more
traditional
support
life
cycle,
and
now
there
is
a
market
for
that
in
automotive.
A
She
was
highlighting
that
she
had
not
been
born
when
the
when
the
mission
launched
that's
how
long
that
that
platform
has
been
pretty
cool,
yeah
all
right
folks,
I'm.
I
would
value
my
participation,
drawing
my
participation
this
month
to
a
close,
because
I
need
a
little
bit
of
time
before
what
I'm
doing
next,
I'm
gonna
try
and
remember
to
send
out
an
invite.
I
don't
think
we'll
do
the
office
hours
portion
of
this,
because
it'll
be
right
before
the
new
year.
I
don't
think
anyone
will
be
around
for
it.
So
I'll.