[Top][All Lists]

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

Re: Broken dream of mine :(

From: Bas Wijnen
Subject: Re: Broken dream of mine :(
Date: Wed, 16 Sep 2009 21:09:46 +0200
User-agent: Mutt/1.5.18 (2008-05-17)

On Wed, Sep 16, 2009 at 08:42:58AM +0200, Damien Thébault wrote:
> Yeah, it seems to always end up this way, hurd architects think that
> one important feature is missing or is done wrong, and then wait for
> another set of years.

Nobody's waiting as far as I can see.  We're just not very many people,
and it happens to be very hard to get more people when the project is
just beginning, like the effort to move the Hurd to a new kernel.

> I'd like to see a GNU kernel with userspace drivers and maybe one or
> two more features, but I don't think it will come someday. Minix3
> seems better for this task because it has one advantage: it works
> right now.

The Hurd works right now as well.  It runs on top of the Mach kernel.
If all you want to do is play with userspace drivers, there are indeed
other systems which offer that.  Playing is not the goal of the Hurd.
It wants to be a usable base for production systems.  That doesn't mean
it's not usable for playing.  But if we write something which is only a
toy (like my previous kernel, which was designed as such), then we would
have failed.

The Hurd on Mach has disadvantages, which seem to be so bad that it's
better to take a new kernel altogether eventually.  That's why this list
exists: to move to a new kernel (it's called l4 because that was thought
to be the new kernel when this list was created).

We want to do it Right.  There's no real hurry; we already have a
suboptimal solution with Mach, so there's no point in creating another
one.  Of course it would be nice if things would go faster.  They would
if there would be more help.  However, I haven't seen anyone offering
that lately.  I admit we haven't been looking actively either.

> I know, this is just a rant, but I just wanted to share what I'm
> thinking from my external point of view.

No problem. :-)  Would you have an idea for how to improve things?  You
suggest that we should just take a random kernel which looks roughly ok
and go with that.  (I'm exaggerating it, but that is your idea, right?)
As I explained, we don't want to do that.  But perhaps you have some
other ideas?


Attachment: signature.asc
Description: Digital signature

reply via email to

[Prev in Thread] Current Thread [Next in Thread]