The Hurd project and DHCP/PPP/PPPoE
Hello fellow Hurd enthusiasts.
Let me start by saying that I am new to Hurd. I've been studying up on
the project and I must say that I am thoroughly impressed. It is an
older project, but when one looks at the history of open source
development, it's clear why the Hurd project has proceeded at a slower
rate. It's a very innovative project that is going where very few have
gone before. It's the wild wild west of programming. Never before has
an open source project attempted to tackle such an ambitious project.
Monolithic kernels are easy in comparison. I've got to tell you, I'm
hooked on Hurd. If there's anything that has the long term potential to
be a Win_ows killer, it's Hurd.
As a newbie, I do have the luxury of not being so involved in the
project that I can't see the obvious. I'm excited that there's a lot of
work going on right now. After studying the current status of Hurd and
observing phenomena such as the ever-growing popularity of GNU software
licensing, OSkit, and Hurd's association with Debian...I am
extraordinarily optimistic about the future of Hurd.
However, I've seen a lot of talk about making the Debian GNU/Hurd
operating system easier to install. I think that's a great goal, but
honestly it's not that hard to install for someone who finds the project
exciting and worthwhile. I'm not usually one who endorses the idea of
"Right of Passage" or creating a test for someone to be involved, but if
that happens due to the current state of affairs, so be it. In my
opinion, the moderately competent Linux user should be able to install
Hurd. Specifically, anyone who can create partitions using fdisk, mount
partitions, and extract tarballs should be able to install Hurd. Anyone
else should consider starting with Linux as an environment to upgrade
their knowledge.
Rather than focusing on a better install method using a cdrom, I suggest
getting DHCP,PPP, and PPPoE working. I have been going into #hurd on
irc.openprojects.net a lot lately due to my mission to learn all things
Hurd. After a week or so, curiousity overwhelmed me and I 'CTCP
Version' ed people to see what people they were running. Turns out it's
mostly Linux! I've asked why and found out that many run Hurd, but are
forced to use Linux as a router. Considering most major operating
systems use CVS or archives (like Debian Linux) for updating package
versions, I think it's a tactical mistake not to fully and properly
support DHCP, PPP, and PPPoE. Indeed, Hurd developers and users' NEED
to have a presence on the internet running Hurd to attract new
developers and have the tools required to move Hurd to the next phase,
"stable release". It's not practical to expect Hurd users and
developers to have another computer as a router or reboot to Linux to
send updated code to a CVS repository or access the internet.
I'm not trying to discourage people by pointing out Hurd is a bit
deficient in the networking category. That is not my intent. Hurd has
its strengths too. Hurd can run XFree86 ver4.0x. It has come a long
way! Hurd has a very flexible kernel with a lot of capabilities. My
point is that we should perhaps shift our priorities to overcome glaring
deficiencies that handicap development.
sincerely,
Kevin Werckman
P. S. I volunteer to help in any way I can. Unfortunately, I need to
upgrade my programming skills a bit. I'm doing the homework and will
get there. Meanwhile, I'd like to get a group of developers together
dedicated to fixing Pfinet to fully support DHCP, PPP, and PPPoE.
Anyone interested, email me. We can work on getting a special mailing
list and other resources needed for this sub-project.
Reply to: