0:00
apologize to the simultaneous
0:01
translators because I have no idea how
0:04
you would translate this into Japanese
0:07
um so one of the things we've talked a
0:09
lot about over the last few kernel
0:11
releases is the introduction of rust
0:13
into the kernel and um I think it has
0:17
been relatively steady and quiet what's
0:21
your perception well the rust still is
0:23
at the point where we don't we have the
0:27
initial infrastructure got merged last
0:29
year
0:30
um it's been growing but we don't have
0:34
any part of the konel that really
0:36
depends on Rust yet to me rust
0:40
was one of those things that a made
0:42
technical sense but to me personally
0:45
even more important was that we need to
0:48
not stagnate as as a kernel and as
0:52
developers and um so I am
0:56
always excited by trying something new
0:59
and not getting too comfortable doing
1:01
the same thing I mean I've been working
1:03
on on the colonel now for 32
1:06
years uh yeah 32 years and that's a long
1:10
time to work on one single thing but
1:13
it's still interesting because it's not
1:15
the same single thing I mean Linux 32
1:18
years ago was very different from what
1:20
Linux is today obviously and and I
1:23
actually often look for things where
1:27
where we can do new things and we can do
1:30
things differently because it's so easy
1:32
to get stuck in a rut and say this is
1:35
working just fine right and and
1:39
rust has not really shown itself as the
1:44
next great big thing but I think during
1:46
next year we'll actually be starting to
1:51
integrate drivers and
1:54
some even major subsystems that are
1:57
starting to actively use it so it's one
1:59
of those those things it's going to take
2:01
years before it's a big part of the
2:03
kernel but it's uh it's certainly
2:05
shaping up to be one of those so you're
2:08
writing rust code yourself you're
2:09
reviewing rust
2:11
code I have been reading rust code a bit
2:14
just so that I can make some kind of
2:17
judgment calls on on when something is
2:20
too horrendous to be included in the
2:22
colel but I have to admit No I um I mean
2:26
the colonel we rely on literally
2:29
thousand of people every single release
2:32
we have a thousand people involved and
2:34
they're not the same thousand people
2:37
quite often we have people in fact for
2:40
the longest time we've had the
2:41
statistics be roughly that every release
2:45
about half the people involved send just
2:48
one patch and a lot of them never show
2:51
up again they may have something small
2:54
they wanted to fix that they cared about
2:56
and they were not really kernel people
2:58
they found it for some other reason and
3:01
they sent their small patch to the colel
3:05
and they were never interested in doing
3:07
anything more but
3:10
then the other half keeps coming back
3:14
and and when it comes to rust I'm not
3:17
going to be the one who manages the rust
3:20
Cod because that's not my expertise as
3:22
is true of so many other parts of the
3:25
kernel uh I'm
3:27
honestly I'm less of a programmer these
3:31
days than I am I call myself a technical
3:33
lead because I'm not I'm not a manager I
3:35
don't manage people I manage code so I
3:38
call myself a technical lead I'm not I'm
3:41
not my day-to-day work is not
3:43
programming it is um merging other
3:46
people's codes and rust will be one of
3:49
those things