Re: Improving Hurd

[Marcus Brinkmann]

On Sun, Apr 21, 2002 at 08:43:52PM +0200, Jan Atle Ramsli wrote:
> As far as I'm concerned, your mail above is a significant piece of
> documentation - I have a good mind to rip it off and paste into a
> 'Programmers guide' (with your permission, of course :-)

If it's a free document (eg under a free license) feel welcome.

> This is probably one Hurdle (sorry) overcome: Who do I talk to, who does
> this & that, & what 2 do to be most productive, etc - meta-questions.
> I have never been part of a multi-programmer-free-software project.

Ah, ok, you probably could have improved the situation by letting us know
about your background.  Most of us hack free softare for years now, and we
sort of take for granted that anybody who ends up on this mailing list has a
similar history.  Especially as the Hurd is still a niche operating system.

> Non of us are administrators, there is nobody 'in charge' - I guess
> somebody who could be in charge hestitates to _take_ charge for fear of
> flames & stuff.

Well, in fact, Thomas and Roland are in charge for the Hurd source.  Well,
maybe me, too, by now, to some extent, and it seems Neal gets there, now he
has CVS access.  That much is pretty clear simply by the fact that we are
the ones who are technically permitted to make changes to the source code.

But, that is mostly just infrastructure.  We are not shy to give away
responsibility (maybe keeping the last word on issues).  If you turn out to
become our ueber-hacker for device drivers, we will happily let you do that
and become in charge for device driver issues.

However, at least here, this is a very dynamic process.  You won't get a
medal or a hat that let's you know you are in charge.  You become in charge
simply because you are there and do a good job (assuming you do ;), and by
us trusting you to do the right thing.

Likewise, if it turns out you don't have the time anymore or run out of
speed, all we can do is thank you for your contribution and look for
something else to take up what's left to do.
 
> There is no 'System Specification' - if you want one, make one.
> (Then, what if it turns out that the design is wrong according to
> specification? Fix it!)

Well, this is certainly problematic, but we have specifications for some
parts of the system (for example, we say we are POSIX compatible, so we can
use the POSIX standard to verify against).  But for the internal stuff, we
have specifications in our heads, and the source code.  Comments, the header
files etc.  Writing specifications takes a lot of time, so this is a
resource problem again.
 
> Also, and this I think is as important: There is no 'support
> infrastructure' - there's a bunch of guys, _just_like_you_, somewhere,
> who may, may not, or may partly know what you ask, what you thought you
> asked ... and the answer may even be wrong. 

Yep.  Again a resource problem, you won't find many people who like to be
called at night for Hurd installation problems without getting paid :)
 
Thanks,
Marcus

-- 
`Rhubarb is no Egyptian god.' Debian http://www.debian.org brinkmd@debian.org
Marcus Brinkmann              GNU    http://www.gnu.org    marcus@gnu.org
Marcus.Brinkmann@ruhr-uni-bochum.de
http://www.marcus-brinkmann.de