[Rule] Whither RULE?
Richard Kweskin
rkwesk at hellug.gr
Thu Aug 23 20:51:47 EEST 2007
On Thu, 23 Aug 2007 19:27:14 +0800
"C David Rigby" <c.david.rigby at gmail.com> wrote:
> Greetings to one & all,
>
> I'm going to throw my $0.02 into the conversation, for what it is worth. A
> bit long for an email, perhaps, so I apologize in advance if it is a bit
> long-winded. Errors and misconceptions are my own, and I'm happy to have
> them corrected by others.
I for one am glad to read what you have in mind.
snip
> Observations:
>
> Historically, the goal of our project has had two primary facets:
>
> 1. Support older hardware rendered obsolete by the "creeping
> featuritis" or bloat of most recent distributions
> 2. Use components from an existing distribution, Fedora in our case,
> rather than create yet another distro from the sources
I still like these but am open to other distros.
snip
> We were also pretty excited at one point about the idea of recycled
> computers being useful in developing countries. Ingo & friends did a great
> job with Vum:box
> (http://www.vum.at/index.html?/content/projekte/vumbox/vumbox.html),
> for example.
I tried to make a useable (to total non-users of computers but requiring everything in the local language and character set) system with a light gui and am still struggling with that. The hardware I've got (lots of ibm p2 desktop boxes) is about on par with my old vaio laptop. I would be delighted if more such projects wanted to use our stuff as Vum did.
> In response to some remarks from Richard and James, I note that:
>
> 1. Debian can still be installed from floppy diskette. I just
> installed the latest version on my ancient dual PII-333 system (which we
> used for a while to host some RULE-project files - that was some years ago).
> It required three floppies - boot.img, root.img and net-drivers-1.img.
I have these floppies but couldn't get my cdrom using the pcmcia interface recognised at that early stage. Still, I like the idea of our script(s.) I think Franz did an excellent job of polishing it.
snip
> Suggestions:
>
> OK: the obvious one, which is the reality check: RULE/slinky has outlived
> its usefulness on all fronts, and should be retired. NOTE: this is the point
> where everyone is supposed to vehemently insist that I am quite mistaken.
>
> The RULE/slinky project IS functionally dead if there is no one to continue
> development of slinky in some form. That does not mean that RULE is dead,
> though. How so? Well, as much as RULE has been about slinky in the past,
> there is no reason to constrain ourselves to just slinky.
>
> Ways forward (and I am open to other suggestions):
>
> 1. Regardless of the future of slinky, RULE as clearing house for
> information on using FOSS operating systems on older hardware still makes a
> lot of sense. Because, despite what I said earlier about efficiency versus
> continued use, there remains plenty of operational old hardware out there,
> and a lot of it is still being used. Worse, a lot of it is
> Internet-connected and running old, corrupted, bootlegged, unsupported
> versions of trojan, worm and virus infested Windows. If we can help get even
> a fraction of that installed base of diseased systems off the Internet, or
> help to cure them, so much the better!
> 2. Relax our goals. We can continue to develop slinky for Fedora, but
> we should abandon the goal of supporting any hardware that does not at least
> meet the CPU minimums of Fedora. To my mind, that means P-II or better CPU.
> RAM is negotiable, but requiring 128 MB or better would not be unreasonable.
> This assumes that someone can continue slinky development.
> 3. Change our goals for slinky. Make it truly an alternate installer
> to Anaconda that has as it's goal an installation of Fedora that is even
> more minimal than the standard minimal Fedora installation. Or simply a
> lighter-weight installer for a full Fedora installation, which would look a
> lot like the Fedora Core floppy project described at
> http://www.thisiscool.com/fcfloppy.htm.
> 4. Focus on a different distribution. I think Richard is right on the
> money with his remarks on Debian. For example, Debian still supports i386,
> and the installer will run in 48MB of RAM (see this link:
> http://www.debian.org/releases/stable/i386/ch02s05.html.en). My basic
> installation on the IBM PC 365 required about 480MB for the filesystem plus
> swap space. So, the 500 MB minimum shown on this page seems correct. A link
> on that page leads to a fuller discussion of the issue. I suppose it could
> be argued that Debian's attention to low-end systems means that it does not
> really need much trimming, but if we relax Debian's policy requirements, we
> could produce a lighter system.
I went for Debian since so many linux users in Greece were choosing it. I developed a sort of system of installing, using a text mode standard Debian installer and purging a few packages afterwards before upgrading from the net. I am happy to elaborate if this list wishes.
> Other suggestions:
>
> With regard to the website, I would be happy to continue with the task of
> reinstalling the historic material. I think Chris is right though - this is
> the lowest priority activity, if our goal is to revitalize the RULE. If we
> decide, collectively, that having the material from the SPIP site and
> earlier available to us is important, I suggest to Marco that we activate
> and give ourselves access to the Book module of Drupal. Eventually we will
> have a book of the historic site. In the interim, make the raw tarball of
> prior content available for download.
>
> Starting a Drupal Book about slinky is also a good idea. We can build it
> collaboratively, and a Drupal Book can be exported to Docbook, from which
> conversion to straight XHTML, PDF and so forth makes republishing it in
> other formats easier.
>
> Marco, please consider making comments and other contribution formats
> (Blogs, for example) available to register users. A forum would be useful as
> well. All of these interactive forms of contribution help to build a
> community once people find it easy to contribute. I suggest three tiers of
> user privilege as a start: yourself as Administrator, trusted managers that
> can manage content and remove offensive/off topic material, and registered
> contributors that have as much access as you can possibly tolerate.
>
> For the near term, at least, you can count on my help in policing for
> inappropriate content and helping with setup. Obviously, do not make
> yourself crazy with over work. If you're too busy, you're too busy. As well,
> I will eventually return to the hectic megapolis lifestyle, at which point
> my available time for contribution will be greatly diminished.
>
> I am also willing to take a look at updating slinky for Fedora 7 if Franz is
> unable to continue. I have the Fedora 7 iso available, though I have not yet
> committed it to DVD+R. With the new laptop computer and the old desktop
> systems here, I've got a good testing environment. (Maybe too good - the
> laptop is REALLY new, and even the Ubuntu alpha for Gutsy Gibbon, the next
> release, cannot cope with the sound card.) I doubt I could get it ready in
> time for the conference in Crete that Richard mentioned, which is two weeks
> away.
I accept that. However, you have inspired me to make an attempt with the last slinky, just changing the base rpm names to match what I have of fedora 7.
> Future work on slinky or other installation tools would be helped by a
> collaborative development environment such as CVS or Subversion. I do not
> know much about these things, admittedly, but I'm happy to learn. Richard,
> would Athens LUG be willing to host such resources?
Sure, if that is what is wanted.
> Anyway, those are my admittedly extensive thoughts on various topics.
> Possibly more than two cents worth. I look forward to more discussion!
>
> Kindest Regards,
> C David Rigby
Well written and well thought out!
Richard
This full static mirror of the Run Up to Date Linux Everywhere Project mailing list, originally hosted at http://lists.hellug.gr/mailman/listinfo/rule-list, is kept online by Free Software popularizer, researcher and trainer Marco Fioretti. To know how you can support this archive, and Marco's work in general, please click here