Archaic wrote these words on 09/14/05 00:41 CST:

> Stepping in even later than you... :) While I didn't expect the first
> round proposal would be seen with much favor, I must point out the true
> impetus of this undertaking. None of the criteria Matt listed gives the
> "why" behind this, only the criteria that was settled on. The "why" can
> be stated (hopefully) simply and succinctly:

[snip one of the best posts in this forum I've read]

First of all Archaic, I would like to point out that your message
was so perfectly stated that it really made me think about the big
picture here. Well done, sir.

The crux of the issue seems to be Gerard's desire to eliminate any
dependency, or for that matter, even any thought to building on top
of LFS. He's the boss, if this is his desires, then we have to do
what is necessary.

Wrong is it may be.

You've explained things perfectly. To satisfy Gerard's request,
there is no other choice but to eliminate the groups from /etc/group
and the associated rules definitions from the Udev rules file.

However, I'll be on record as saying this is a mistake. Though the
goal of LFS is to build a minimal system capable of building upon,
I believe this is carrying things a bit too far.

There is no harm in creating groups which have no users assigned to
the group. There's no security, nor any other implications I can
think of. To not anticipate that LFS will be build upon, in whatever
fashion, be it a server or desktop capacity, and not having the
groups on the system with appropriate Udev rules to create device
nodes appropriate to the use of those nodes, is simply wrong.

Here is the important thing that we must consider:

Device nodes are going to be created upon the kernel's discovery
of the hardware. Do we want these device nodes to be created
correctly, or not?

If the decision is that LFS can live with improperly created device
nodes then go forth with the plan. If it is decided that the device
nodes should be created with the proper group assignment, then *LFS*
should do this.

-- 
Randy

rmlscsi: [GNU ld version 2.15.94.0.2 20041220] [gcc (GCC) 3.4.3]
[GNU C Library stable release version 2.3.4] [Linux 2.6.10 i686]
00:51:00 up 165 days, 24 min, 3 users, load average: 0.00, 0.00, 0.07
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-dev
FAQ: http://www.linuxfromscratch.org/faq/
Unsubscribe: See the above information page

Reply via email to