I see. Well, would those who are reviewing the project proposals be
willing to consider a proposal which is only a partial completion of a
project with a promise to continue working on it? I was looking at the
POSIX standards,
http://www.opengroup.org/onlinepubs/000095399/idx/realtime.html
(that's them right?) and they seemed to have broken Real time into a
few Topics. Might it be possible to pick a few, maybe Asynchronous I/O
and Process Scheduling for example, and then propose working on them
for SoC and then continue the rest once SoC is over? I don't want to
be difficult or anything but I'd really like to work on this project
if at all possible.

As for filesystem experience, mine is pretty limited, see Phase #2 at
http://www-inst.eecs.berkeley.edu/~cs162/sp06/Nachos/index.html but
again, if there is some possible break down of the journalling project
that somebody could help me define that would be nice. I should also
mention that I think that they implemented Journaling for UFS in
FreeBSD in last years summer of code, see=20
http://www.freebsd.org/projects/summerofcode-2005.html, so depending
on how different the systems are, and how successful the project was,
that could be a potential time saver couldn't it?

But anyways, any comments/suggestions would be much appreciated. If
the feedback on this is negative perhaps I'll look more closely at
Improving/Extending Filesystem Resizer or Defragmentation for FFS, but
in any case is there anybody who wouldn't mind reviewing a proposal
once I've gotten some work done?

Thanks Again,
Raymond Chetty

On 4/30/06, Thor Lancelot Simon <tls@rek.tjls.com> wrote:
> On Sun, Apr 30, 2006 at 08:38:05PM +0930, Brett Lymn wrote:
> > On Sat, Apr 29, 2006 at 07:22:23PM -0700, Raymond Chetty wrote:
> > >
> > > I was thinking about trying my hand working on ether Real time suppor=
t
> > > or Journaling for FFS but the descriptions on the projects website ar=
e
> > > pretty vague and don't really provide much in the way of direction fo=
r
> > > somebody who is new to NetBSD, or kernal programming in genreal.
> >
> > Both are fairly large projects and are probably not good candidates
> > for SoC especially if you are not familiar with kernel programming.
>
> I don't think journalling is out of reach for Summer of Code, if the
> person doing the project already has some filesystem experience.
>
> Thor
>