Re: YotH 2010 -- a Year of the Hurd 2010

[olafBuddenhagen]

Hi,

On Mon, Jan 24, 2011 at 09:02:40AM +0100, Thomas Schwinge wrote:
> On Wed, Jan 12, 2011 at 04:04:29PM +0100, Arne Babenhauserheide wrote:

> > But though I like the idea, I think Olaf is right with the point, that the 
> > Yoth is not positive enough.
> 
> But we have to be careful to not exaggerate; to not raise false hopes.

How could listing specific advances raise false hopes?

> > - GSoC 2010: 
> >     - dde linux26 (device drivers!)
> 
> Not yet ready.

Actually, the network part is pretty much ready, apart from some minor
issues and cleanups... But then, that's true for other parts of the Hurd
too.

> (And this wasn't a GSoC project, even.)

True.

> A (stable / usable / ...) procfs implementation is very nice of course,
> but it's not interesting per se -- it's only procfs.

It is one less missing piece for a usable system. I don't see how it is
any less interesting than other improvements.

I agree though that it might not be among the most important bits to
mention in the digest... Feel free to drop it.

> Please send changes against these sources:

As I have rewritten almost all of the text, there is probably no point
in sending a diff... Also, I'd prefer not to bother with texinfo.

Hope you like my version...


GNU Hurd (<http://www.gnu.org/software/hurd/>)

Originally started as the GNU project's take at replacing proprietary
UNIX kernels, the Hurd is nowadays pursued mostly because of it's unique
multi-server microkernel architecture -- bringing advanced operating
system research to the mainstream. According to our mission statement,
the goal is creating "a general-purpose kernel suitable for the GNU
operating system, which is viable for everyday use, and gives users and
programs as much control over their computing environment as possible."

The Hurd doesn't fully deliver on the "everyday usability" goal yet; but
it is seeing continuous improvement -- and 2010 has been no exception.
So let's take a look at the progress throughout the year.

  * In addition to various general stability and portability fixes,
    several people (including Samuel Thibault, Pino Toscano, Emilio
    Pozuelo Monfort and others) have been working on fixing issues with
    specific Debian packages. 68% of all Debian packages are also
    available for Debian GNU/Hurd now.

  * As a participant in the Google Summer of Code program, Jérémie
    Koenig ported the modern Debian Installer to work with the Hurd. CD
    images using the new installer (available from
    <http://people.debian.org/~sthibault/hurd-i386/installer/cdimage/> )
    are replacing the previous CD images, which were using an installer
    based on the old Debian Bootfloppies (and running under Linux) --
    Philip Charles has been maintaining these single-handedly for almost
    ten years!

  * Meanwhile, Michael Walker started the Arch Hurd distribution, and
    together with a bunch of other enthusiastic Arch developers (Allan
    McRae, Matthias Lanzinger, Alexander Preisinger, Stephen Gilles,
    Diego Nieto Cid) they got it working in an amazingly short amount of
    time. Now you have a choice between two well-featured distributions
    for the Hurd. These new people of course also help forwarding Hurd
    development in general -- Diego in particular contributed various
    patches to the Hurd console and other components.

  * Jérémie Koenig created a new implementation of the procfs
    translator, which is considerably more robust and efficient than the
    previous one. Tools such as "top" can be used without problems now.

  * Some other translators (gopherfs, netio, tarfs), which were created
    by external contributors in the past, have been fixed up by Manuel
    Menal, and packaged in Debian. Thus some of the results of Hurd's
    extensibility are now easier to access.

  * Also as a Summer of Code participant, Emilio Pozuelo Monfort was
    investigating specific compatibility problems exposed by the
    extensive testsuites coming with some software packages. Emilio's
    analysis uncovered a bunch of bugs in the Hurd, and he fixed several
    of them. As these bugs usually affect other programs too, this
    should improve stability and compatibility in general.

  * After several years of development, Samuel Thibault finally merged
    his Xen domU support to GNU Mach, which enables Hurd to run as a Xen
    guest. It was now deemed mature enough, after being in heavy use on
    the Debian GNU/Hurd package build daemons for a long time, and later
    also the Hurd web server and some of the public Hurd systems listed
    on <http://www.gnu.org/software/hurd/public_hurd_boxen.html>.

  * Zheng Da worked on a new hardware driver framework, which is based
    on the Dresden (Fiasco) L4 group's DDE project, and allows running
    modern Linux drivers as userspace server processes. Most network
    cards already work perfectly with this new framework. (It's not
    integrated in the mainstream Hurd yet, so needs to be compiled and
    set up by hand.) Other driver classes, such as hard disk
    controllers, will require further work.

  * Carl Fredrik Hammar finished and presented his thesis, ``Generalizing
    mobility for the Hurd'',
    <http://lists.gnu.org/archive/html/bug-hurd/2010-01/msg00078.html>,
    and passed with distinction.

That's only a short digest of what happened during the last
year.  You can read our regular ``Month of the Hurd'' news blurbs on
<http://www.gnu.org/software/hurd/news.html>, or by subscribing to our RSS
feed at <http://www.gnu.org/software/hurd/index.rss>.

If you are interested in doing a university project on a multi-server
microkernel-based operating system for example; or if you just want to
contribute to Hurd development in general: you can find some pointers on
<http://www.gnu.org/software/hurd/contributing.html>.  Of course you can
contact the Hurd developers about anything on the <bug-hurd@gnu.org>
mailing list, or the #hurd IRC channel on freenode.net -- see
<http://www.gnu.org/software/hurd/community/communication.html> for
details.