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

Re: feedback wanted alternative Debian installation system

* Branden Robinson (branden@debian.org) wrote:
> The current version is, and I'd like to solicit feedback from
> Debian Developers to help us determine what needs to be done with PGI
> to make it worthy of a 1.0 version number.  PGI is in a late-beta stage
> and has been successfully used to install woody systems many times.


>      * PGI is flexible. Not only are many aspects of its behavior
>        customizable at build time (and even run time), but PGI supports
>        the simultaneous development of different "profiles", which are
>        referred to by code names that you select.

This is the killer feature that could take Debian into the next
generation, IMO.  If it were possible to "publish" these profiles, then
people could invest work into customizing Debian for a variety of
different purposes.  "DeMuDi," Debian Jr., and all the other specialized
Debian subprojects could all become special cases of the general ability
to publish installation profiles.

As it is, Debian provides excellent raw material.  Packages will work
properly when installed and come with reasonable defaults.  The problem
for non-developer users is that it's almost impossible to use Debian
without micromanaging.  There's no way to defer judgement about
configuration decisions to someone who knows what you want out of your

Profiles could alleviate the need to micromanage.  I could install
"DormRoom Debian" on my computer, and I would get X, a login manager, a
simple window manager, a media player, a filesharing client, a word
processor, and menus/buttons to make these particular apps easily
accessible.  Interested parties could invest time into polishing Debian
to provide a smooth desktop experience, and then share their work with

I think some of the more powerful customization would require that these
profiles have the ability to override default configuration files and/or
answer debconf questions.  I imagine this would be much more involved


Joshua Haberman  <joshua@haberman.com>

Reply to: