Re: Pkg-based base system.

From: Martin P. Hellwig <mhellwig_at_xs4all.nl>
Date: Wed, 17 Mar 2004 11:23:09 +0100
Scott Long wrote:
<cut>

> The trick here is to know when you start sliding too far down the
> slope.  It's hard to argue about sendmail, named, gcc, etc, but where do
> you stop?  Before long, you'll be chopping out nvi for the people who
> favor vim, and so on.  I'm actually more in favor of keeping FreeBSD as
> the 'reference implementation', and encouraging  the  to make
> derivatives off if it that satifies these kinds of needs.  But we will
> see where things head.  Above all, I support your work, but just ask you
> to be cautious and not this get carried away.

<cut>
As a _simple_ user/administrator not bothered with any knowledge about 
programming, I really do strongly agree with Scott about FreeBSD being a 
'reference implementation' (RI).
For me some one of the strongest point of FreeBSD is that you can start 
with little or no knowledge about unix with FreeBSD by just reading the 
handbook every step at a time. This is possible _because_ there are 
pseudo standard tools available and integrated with FreeBSD, and by that 
well documented in the handbook.
It makes my life allot easier and by this leaving some time to continue 
to learn about this system  so that maybe someday I can be more useful 
for the FreeBSD community.
But no matter how much I like this system and whatever more or less good 
arguments there are given, people would have other ideas about how it 
should be done, thats something IMO we should not only live with this 
but also encourage.  Colin his "nuke_a_part_of_world-o-matic" initiative 
is from a personal standpoint a good intermediate between a total 
package based system and a RI.
<dreammode> Maybe this can be placed in the port tree like being a kind 
of meta-port ,<fantasy> it might even be a distribution port so that 
when doing a "make iso" the system compiles it's own personalized binary 
distribution which then can be burned on CD-ROM  and installed on other 
systems.
 From a (personal) user perspective the advantage is that the RI is 
modified which avoids confusion about having another derivative.
Probably there are allot of (dis)advantages about this, although I might 
not understand all of the opinions, I really do like to read them 
</fantasy> </dreamode>.

MPH
Received on Wed Mar 17 2004 - 01:22:19 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:37:47 UTC