[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: GNU/kFreeBSD talk at DebConf15



I'd love to do this, but heading to Germany in August isn't really practical for me right now (I also don't have any meaningful credentials :P). Some key points I'd recommend:
- Why kFreeBSD is good for the ecosystem (competition fuels open source projects, encourages writing of platform-independent glibc code)
- glibc + GNU userspace = familiar environment for the Linux user, with the benefit of some cool BSD kernel features
 * pf vs iptables (readability, speed+statefulness)
 * ZFS vs LVM (ZFS send/receive and upgrades, RAID effectiveness)
 * jails vs linux solutions like linux containers and UML (one tool included in the kernel vs a conglomeration of many)

Some other cool kFreeBSD tricks:
- Linux compat layer - Maybe include a demo of a Squeeze i386 Linux system running in a jail and what this could mean for the future
- Static FreeBSD binaries can run in a kFreeBSD system
 
>This would necessarily include a demo / showcase of what jessie-kfreebsd
can do already.
- minetest, openarena, and sauerbraten all in the repos and functioning
- WebGL stuff works on IceWeasel (from sid): (http://pirateradiotheater.org/files/m.grl/demos/06_models/)

On Sun, Jun 14, 2015 at 10:15 AM, Steven Chamberlain <steven@pyro.eu.org> wrote:
Hi,

We should give a talk on GNU/kFreeBSD at DebConf15.  I'm happy to
volunteer for this, but also welcome others to participate, or just
suggest some additional topics to cover.

The deadline for submissions is 15th June, so I'm putting in a
request now for a 20-minute talk.  If we have even more things
to cover, we could maybe still try to schedule a lightning talk.

Ideal outcomes of a talk from my POV would be:
  * convince more people that Debian should support multiple kernels;
  * encourage more people to get involved, explaining how;
  * tempt some people to try it for the first time, or see how it's
   improved since they tried it last;
  * start thinking about how kfreebsd, hurd or other ports should fit
   alongside official testing/stable suites.

This would necessarily include a demo / showcase of what jessie-kfreebsd
can do already.  I'd also like to throw in:
  * a bit of history;
  * how the jessie-kfreebsd release was (will be) done, give credit to
   those responsible;
  * some principles about porting - as I see it - how we do/don't want
   to do things.

Regards,
--
Steven Chamberlain
steven@pyro.eu.org


Reply to: