►
From YouTube: ROS 2 Security Working Group (13 Sep 2022)
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
A
Kelly.
Please
go
ahead!
Actually,
sorry
before
we
start,
let's
approve
the
meeting
minutes
from
July
12th,
which
are
linked
on
the
agenda
alternate
agenda
on
the
chat
as
well.
So
let
me
get
that
out
of
the
way.
A
And
if
every
one
person
is
okay
with
that,
then
we
approved
the
meeting
minutes.
B
Yeah,
so
we
are
not
many
many
people
on
the
call,
so
maybe
I
will
bring
this
up
in
Matrix
or
or
in
a
next
meeting.
But
there
are
a
couple
of
couple
of
items
that
I
wanted
to
kind
of
bring
up
here.
One
one
is
that
we've
been
working
with
e
procima
the
past
half
a
year,
I
guess
with
some
additions
or
enhancements
for
the
fast
EDS
and
I
think
we
are
quite
close
to
finalize
it,
and
the
enhancement
is
pretty
much
to
kinda,
enable
a
high
router.
B
So
maybe
I
start
over
so
the
first
last
few
months
we've
been
working
with
eprasima
to
enhance
the
last
DDS
a
little
bit
and
the
idea
is
to
kind
of
add
support
for
multi-layer
networks.
When
Ross
security
features
are
enabled,
and
so
so
by
design
it
seems
that
the
Ross
and
and
DDS
hasn't
been
really
designed
to
work
in
multi-tier
network
or
or
multi-layer
network.
B
So
it's
it's
designed
to
pretty
much
work
on
a
flat
structure,
but
in
our
drone
use
case
that
I
have
explained
here
in
the
working
group
a
few
times.
We
have
like
a
very,
very
many
networks.
It's
like
a
network
of
networks
and
for
that
we've
been
working
with
the
procimer
to
kind
of
add
this
support
so
both
Discovery
and
and
the
actual
communication
and
rust
Tool
traffic.
So
to
work
when
we
have
multiple
different
networks,
so
we
have
the
local
virtual
land.
Then
we
have
Wi-Fi.
B
We
have
LTE
and
multiple
drones,
so
everything
is
Multiplied
and
everything
wants.
We
want
to
kind
of
communicate
seamlessly
between
those
different
networks
in
the
same
Rush,
2
Network
at
least
virtually
so
I
think
that
I
will
talk
with
the
iprosima
but
I
think
maybe
in
the
next
working
group
meeting
we
could
have
some
kind
of
presentation
about
it
or,
if
not
next
and
November,
but
I
think
we
are
doing
the
final
tests
with
it.
So
it's
it's
more
or
less
ready.
B
So
this
was
just
basically
to
give
some
kind
of
update
and
maybe
add,
item
to
the
next
next
call
I'd
send
then
the
other
thing
is
that
we
are
thinking
of
rewriting
all
our
Rost
two
notes
in
our
project
and
the
actual
idea
there
is
to
move
to
memory
safe
languages
and
I
was
just
wondering
if,
if
anybody
in
the
working
group
or
has
been
writing
Rush
2
notes
with
let's
say
rust
or
go
well
or
go
up
play
memory,
safe
languages
or
if
you
are
aware
of
any
kind
of
studies
or
research
about
using
Rush
tool
with
those
those
language
environments,
it
seems
that
most
of
the
Community
is
still
using
CAC,
plus,
plus
and
well
python.
B
C
C
Ahead,
just
to
say,
I
can't
speak
to
personal
experience
on
that
front,
but
one
of
the
guys
that
works
on
my
team,
Jacob
Hassell,
has
been
doing
a
lot
of
work
with
Ross
too,
and
rust
he's
been
on
an
actual
Ross
to
rust
working
group
for
a
while
now
so
I
think
he
probably
would
be
a
good
resource
to
get
in
touch
with,
and
you
know
be
able
to
bounce
ideas
around.
B
That
would
be
very
nice.
Could
you
maybe
write
his
name
in
a
chat
yeah.
C
I'm
gonna
see
if
I
can
get
some
links
or
something
where
he's
had
some
contributions
before
so
you
can
get
in
touch
with
him.
C
I'm
trying
to
get
to
his
actual
like
GitHub
profile
or
something
but
is
not
going
as
easily
as
I,
would
like
to
I'm
going
to
just
grab
this
first.
B
C
First
I'm
going
to
drop
over
here.
It's
just
a
GitHub
link
to
his
user
profile
that
is
Jacob
hassled
with
BCS.
C
C
There
is
some
email
information
so
that
you
can
reach
out
to
him,
but,
like
I
said
he,
he
has
been
a
very
big
proponent
of
rust
and
has
been
active
in
trying
to
move
Ross
to
rust
compatibility
forward.
So
you
know
I
think
that
could
be
a
great
connection
to
make.
He
definitely
has
some
opinions
on
things.
So,
if
you're
looking
for
opinions,
he
will
be
happy
to
share
them.
Okay,.
C
It
actually
was
interesting
because-
and
this
is
actually
one
thing
I
was
kind
of
thinking
would
be
interesting
to
talk
about.
With
this
group,
I've
heard
some
noise
from
the
Ross
II
steering
committee
about
possibly
looking
at
Xeno
as
being
a
new
middleware
option,
and
he
actually
was
looking
at
that,
because
I
believe
it
is
implemented
in
Rust,
and
he
had
a
couple
of
concerns
about
some
of
their
implementations
where
there
was
some
unsafe
memory
operations
happening,
but
I
know.
B
That's
interesting
so
it'd
be
one
of
the
one
of
the
middle
layers
to
be
added
to
kind
of
cycle
between
the
versions.
If
I
understood
correctly,
is
that
like
comparing
implementation
to
fast
EDS
and
so.
C
On
basically
yeah
now
you
know
I,
don't
know
a
whole
lot
about
it,
and
I
was
not
personally
part
of
those
discussions.
So
it's
just
kind
of
what
I'm
hearing
is
third
hand
but
yeah.
It
sounded
like
you
know,
some
of
the
senior
steering
committee
members
were
kind
of
saying
you
know
gee,
it
sure
seems
like
a
lot
of
people,
don't
really
like
DDS
and
it's
handbring
us
in
a
variety
of
ways,
especially
with
respect
to
Discovery.
C
So
maybe
we
should
consider
just
saying
you
know
what
never
mind
about
DDS,
let's
go
with
Xeno
and
maybe
that'll
make
everybody
happy,
but
then,
in
looking
at
the
actual,
you
know
implementation
and
documentation
for
Xeno.
There
was
very
little
in
the
way
of
security
support
in
there,
which
means
you
know,
which
would
be
a
big
concern
if
there
was
any
intent
to
have
Xeno
support
for
sros.
B
Okay,
yeah,
that's
that's
interesting!
It
security
perspective.
That's
very
interesting!
Change.
We've
been
quite
committed
in
our
project
to
work
with
fast
EDS.
Previously
we
proceed
my
Implement
that
this
big
ss11
support.
According
to
the
DDS
specification,
which
was
one
project
we
did
last
year
with
them,
and
now
we've
been
working
to
kind
of
I,
don't
know
if
we
are
facing
the
same
Discovery
issue,
what
you
refer
to,
but
that's
similar.
Similarly
like
we
have
had
a
problem
with
discovery
with
this
flat
Network
design
principle,
yeah.
C
C
Yeah,
like
I,
said
it's
not
a
third
hand
from
my
perspective,
I'm
hoping
there's
some
meeting
notes
out
there
that
might
have
a
little
bit
more
detail
about
what
was
actually
discussed,
but
I
haven't
dug
those
up
myself.
I've
just
been
listening
to
what
the
folks
who
were
involved
were
saying.
B
Yeah
roscon
is
going
to
happen
in
a
month,
so
let's
hope
that
there's
some
something
interesting
regarding
that.
B
Yeah
but
I
think
that
was
everything
that
I
had
I
had
to
kind
of
I
I
didn't
really
prepare
any
presentation
or
anything
or
nor
added
anything
to
that
to
end
up,
but
I
just
wanted
to
bring
these
two
two
things
to
the
table.
A
Okay,
great
and
if
we
do
update
us
on
Matrix
as
well,
if
you
learn
anything
else
about
this
initiative,
that's
definitely
interesting.
For
us,
I
mean
in
security
perspective.
C
I
did
find
the
TSC
minutes,
so
I'll
go
ahead
and
drop
those
over
there.
There
was
a
20-minute
discussion
on
what
would
it
take
to
build
a
Xeno
Ross
middleware
tldr
Xeno
has
a
lot
of
nice
properties
that
may
make
it
better
rmw
for
those
just
starting
to
learn.
Ross,
open,
Robotics
and
Zeta
scale
are
looking
for
contributions
and
funding
to
make
this
happen
to
do.
Update
a
September
meeting,
but
yeah
I
think
there's
a
very,
very
strong
question
about
what
the
security
capabilities
of
that
would
be.
C
Okay,
there's
very,
very
limited
stuff
from
what
I
looked
at
right
now,
let's
see
I,
don't
where
was
that
I'm
looking
if
you
have
not
seen
it
before
this
is
the
Project's
home
page.
C
And
I
believe
that,
basically,
all
it
said
was
that
they
had
support
for
the
client
verifying
that
the
server
was
who
it
claimed
to
be,
and
that
was
about
it.
C
C
That's
down
in
their
reference
manual.
B
B
Yeah
I
could
can
I
rate
this
back
and
see,
see
how
it
looks.
B
B
C
Let's
see
is
this
the
right
I'm
looking
I
am
looking
at
one
I'm,
not
sure.
If
it's
the
right
one
or
not,
though,
which
link
is
it
you're?
Looking
at.
C
C
That
was
the
only
thing
I
was
bringing
to
the
table
today.
A
Great
okay,
so
on
my
side,
what
I
added
to
the
agenda
is
a
follow-up
discussion
from
last
session
last
meeting
in
July
about
collaborating
on
a
very
short
paper
about
sros2,
initially
targeting
the
Journal
of
Open
Source
software.
A
Okay,
so
we're
gonna
have
a
lot
of
progress
yet,
but
just
wanted
to
bring
it
up
again,
and
so
maybe
we
can
distribute
some
from
the
work
and
sort
of
get
organized
and
push
it
through
yeah.
C
And
I
would
I'd
be
happy
to
help
out
with
that,
but
yeah
for
the
discussion
last
time
it
sounded
like
it
was
kind
of
a.
We
just
need
to
take
a
paper
that
someone
already
had
existed,
had
already
created,
cut
it
down
to
size
and
submit
so
I
didn't
know
if
there
was
any
good
place
to
jump
in
and
do
that.
C
I
thought
that
there
was
one
that
was
shared
during
that
meeting.
That
actually,
you
know,
was
like
hey.
We
could
use
this
as
a
basis
and
then,
whenever
we
looked
at
what
the
actual
Joss
you
know,
submission
criteria
looked
like
they
wanted
something
that
was,
you
know
simpler
and
smaller
than
what
had
already
been
developed.
C
A
A
C
Well,
I
guess
you
know
the
the
question
is,
you
know?
Is
there
something
in
the
existing
paper
there
that
would
be
appropriate
to
pull
over
here
because
I
don't
remember
exactly
what
they
were.
Looking
for,
I
thought
the
discussion
was,
it
seemed
like
it
was
a
fairly
close
match.
If
it's
not,
then
yeah
we'd
have
to
figure
out
what
would
be
an
appropriate
piece
to
pull
over
and
is
there
any
existing
material
to
work
from
or
are
we
having
to
create
something
from
scratch?.
A
A
I
guess
the
question
is:
is
there
an
interest
in
the
group
to
push
this
through
and
if
so,
what
the
timeline
should
be.
C
A
B
Yeah
sort
of
same
same
from
my
side,
I'm
I'm
happy
to
kind
of
be
involved,
but
yeah
I,
I,
guess
Victor
and
the
guys
are
proof
better
to
comment
on
that.
So
maybe
it's
better
to
take
it
to
Matrix.