Re: GSoC application deadline passed

[Arne Babenhauserheide]

El Tuesday, 18 de March de 2008 15:46:25 olafBuddenhagen@gmx.net escribió:
> Hi,
>
> On Mon, Mar 17, 2008 at 12:26:30PM +0100, Arne Babenhauserheide wrote:
> > El Sunday, 16 de March de 2008 20:55:08 olafBuddenhagen@gmx.net
> >
> > escribió:
> > > > When we had the HURD in my informatics class, one task was to find
> > > > out information about the HURD, and as I searched for information
> > > > on the current state of the HURD, it looked quite dead.
> > >
> > > That's interesting. When was that class, and where?
> >
> > It was in 2006 in Heidelberg: Informatics 3: Operating Systems and
> > Networking. ->
> > http://pvs.informatik.uni-heidelberg.de/Teaching/bsnw-0607.html
>
> Nice to hear that someone gets interested in the Hurd this way :-)

How else do people come here? 

> Well, I assumed so, because I don't remember having seen you speak up
> before :-)

I'm mostly quiet, that's true. 

I did some things in the wiki and was in contact with Thomas Schwinge, but I'm 
simply not that proficient at OS design... I just learned coding in Python 
last year - before that I wouldn't have said I could write anything, and this 
is still true for anything but Python (I study physics, informatics was just 
a minor part of my studies). 

> > because I realized that the ideas behind the Hurd fit with what I
> > believe: It should be possible for users to add system services
> > dynamically, because a computer should restrict its users as little as
> > possible - instead it should be a product increasing our freedom - and
> > I am not free in changing my computer as i want it, if I have to hope
> > for some kernel hackers to integrate a new piece of software into the
> > kernel, so I can use it - or for someone to patch the kernel for each
> > new release to add the feature I need.
>
> Pretty nice description of the ideas behind the Hurd :-)

:-)

Maybe I should put it up somewhere. 

> > Maybe it would be possible to get the best of two worlds: "Avoiding a
> > really late and still incomplete release" and "having regular
> > releases" by just visibly featuring the Debian distribution along with
> > livecds, qemu images and news on the Hurd pages.
>
> That's not quite the same. It probably makes no difference technically,
> but it does make a major difference psychologically.

And maybe that difference is what hinders the Hurd releases at the moment. 

> Also note that having both Debian releases *and* official Hurd releases,
> we get more exposure... :-)
> 
> > It preserves the current status: "The Hurd doesn't do releases until
> > it is mostly feature complete" while putting out news and promoting
> > the work of the Debian people as well as the advances inside the Hurd.
>
> But do we *want* to preserve that terrible state?...

I don't, but I'm no coder. 

If the people who do most of the coding don't want to do Hurd-only releases, 
then it may be necessary to just say "OK, you're doing work noone else here 
could, and if you don't want releases of the hurd itself, so be it."

(And I know that it is rather awkward to write this in the public list, where 
they read in... so be it, I can't change that without excluding others)

My goal for the Hurd is to get more visibility, and if Hurd-only releases are 
a no go to some key people, other ways to reach that goal may be necessary. 

> Sooner or later we will *have* to release -- and the longer we wait, the
> stronger the pressure to come up with something ground breaking. It
> get's worse and worse. We need to break this evil circle; the sooner we
> do that, the less painful it will be.
>
> The strategy I suggest is to create a 0.3 release very soon; put up the
> tarball and some release notes on the server, but never announce it
> anywhere. A bit later, do that again with 0.3.1. Maybe with 0.3.2 or so
> we could already announce it in a very careful manner. With 0.3.3, more
> openly. And so on.
>
> This way, by the time most people realize that we are doing releases
> (again), it will be business as usual. "Yeah, right, another development
> release. We have been doing these for some time now, haven't you
> noticed?..."

That sounds nice to me. 

With this, the Hurd would also have release notes again at last, so *I* could 
see, what changed (instead of trying to understand it from commit 
messages). :-)

The problem is: Will the major coders go with that? 

> No, the draft isn't linked. I had to paste it directly into the
> application form. We can change the draft as much as we like (for next
> year perhaps...), but I don't think we can still influence the version
> that the students will see...
> 
> Well, maybe I can still edit it as organisation administrator. Let's
> see.

Just take a look at the mentoring organizations... 

http://code.google.com/soc/2008/hurd/about.html

The link to the ideas goes directly to the wiki page, so the students see the 
most recent version. 

Or is that already your work? 

Best wishes, 
Arne
-- 
Unpolitisch sein
Heißt politisch sein
Ohne es zu merken. 
- Arne Babenhauserheide ( http://draketo.de )
-- Weblog: http://blog.draketo.de

-- Mein öffentlicher Schlüssel (PGP/GnuPG): 
http://draketo.de/inhalt/ich/pubkey.txt

signature.asc
Description: This is a digitally signed message part.