l4-hurd
[Top][All Lists]
Advanced

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

Re: Status of project?


From: Bas Wijnen
Subject: Re: Status of project?
Date: Wed, 02 Feb 2005 10:14:15 +0100
User-agent: Debian Thunderbird 1.0 (X11/20050116)

B. Douglas Hilton wrote:
I have been following Hurd development for more than a few years, though,
and I continue be interested in L4-Hurd from an engineering standpoint.
As an engineer, I desire to have a "kernel in my pocket" as it were,
such that
I could concievably hack in support to control industrial equipment and
write
a simplified userland tailored to the process.

I really like that idea as well, indeed.

The Linux kernel would
appear
a good choice for this, and it probably is the best choice right now,
but its
sheer bulk of code is highly unwieldy. Microkernel seems to imply small,

While all the pieces are smaller, you should not expect the result to
be.  Linux uses modules to allow you to remove most parts you don't use
at run time, so it shouldn't make much of a difference.

Of course things like libcrypt in the kernel (so non-swappable) whenever
you have an encrypted filesystem mounted is a problem that doesn't exist
for microkernels, but they have their own problems.  I wouldn't expect
the result to be significantly smaller (or larger).

and
hopefully understandable source code.

It should indeed be more understandable.

The GPL license is incredibly relevent to industrial projects, because
if the
modified software is never intended to be re-published "for profit" or
otherwise,
then one is pretty much free to do what one wants with it.

I still don't see the problems for companies of releasing their changes
when they redistribute software, as long as they're main business is not
software selling.  So I would expect it to be just as useful if they do
redistribute it.  But most managers seem to think otherwise.

The kernel level understanding allows you to
write simple
device drivers for the equipment, and the GNU tools allow you to run the
userland however you want on whaterver hardware you want to use. This is
the
ultimate: fast (to implement), cheap (non-proprietary), and good (does
what you want).

While in practise usually true, non-propriatary strictly does not imply
cheap. ;-)

What I like most about the approach is the power it gives to the user,
combined with the security it gives to the system.  Any user can do
whatever he or she likes, which is great.  Still if one user messes
things up, the other users (or even the other processes of that same
user) will not have any problem.  My experience with device driver
writing for Linux is that I often need to reboot.  That should not be
needed with a microkernel (which does not include device drivers).  I
look forward to installing GNU/L4-Hurd on my Yopy PDA. :-)

I'm probably due to pitch in again some more myself on this project. I
switched
my workstation over to Gentoo a while back so thats my new development
platform,

I tried Gentoo some time ago and really liked it, but I went back to
Debian, mostly because I don't want to run non-free software, and Debian
seperates that very well.  AFAIK gentoo doesn't mind (or warn) if you
install pine, for example, while it isn't free software at all.

Anyway, this is getting less on topic all the time. :-)

I'm sure I will figure out how to
build L4-Hurd
on Gentoo easily enough the next time I try it.

Speaking of which...  that sounds like a great idea. I admit, I'm bored.
I have no
exams coming up. I haven't been active here for a while. All you
students hit those
books, and meanwhile I'll recompile it all and update the Wiki if I can.

Great. :-)

Thanks,
Bas

--
I encourage people to send encrypted e-mail (see http://www.gnupg.org).
If you have problems reading my e-mail, use a better reader.
Please send the central message of e-mails as plain text
   in the message body, not as HTML and definitely not as MS Word.
Please do not use the MS Word format for attachments either.
For more information, see http://129.125.47.90/e-mail.html

Attachment: signature.asc
Description: OpenPGP digital signature


reply via email to

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