The Strengths and Weakness of Fedora/RHEL OS management

Nicolas Mailhot nicolas.mailhot at laposte.net
Tue Mar 28 07:52:47 UTC 2006


Le Lun 27 mars 2006 22:39, sean a écrit :

> No matter what you come up with though, it will be many years before
> you see wide spread adoption.   If anything, you might consider a
> project to create a system-wide config editor that knows all
> the different formats etc and provides a consistent CLI/GUI
> interface.

Yay, return of the linuxconf

It will break for the same reason that linuxconf failed - even a GUI needs
some config file consistency to work. It will fail like the current
printer setup fails. If you want it to work conf syntax and GUI/CLI tools
must be carefuly thought of and aligned, or you'll only produce brittle
GUI tools which eat  conf files at the first opportunity.

*If* a core set of apps could be moved to a modern consistent format then
one could try to evangelize it later. This is something the gconf people
botched big time by focusing on the GUI and generating files no one could
sanely modify in a text editor (and no xml is not the reason. Serializing
stuff without thinking is the reason. Abusing human-unfriendly UUIDs is
the reason)

My suggesting if you want this to happen is :

1. create a new gconf backend with files people can actually change in vi
of emacs
2. once you've proved you've a solid candidate both GUI and CLi cand work
with, try to sell it to everyone else

But if you're not able to switch gnome (which has a single point of entry
- gconf) you've got zip chances to get rid of the others legacy formats

-- 
Nicolas Mailhot




More information about the fedora-devel-list mailing list