The Ultimate Distro

The name of Gaël Duval's new distro, Ulteo, with its hint of the word "ultimate", smacks of a certain ambition.  But Duval probably means it in the sense that it is the last distribution you will ever need to install, because thereafter it will "self-upgrade automatically," as the announcement of the alpha release put it.  Ease-of-use has been a constant theme in Duval's work.  When he launched his first distro, Mandrake, in July 1998, one of his stated goals was "to provide a working and easy-to-install linux-distribution to people who don't want to spend too much time in installing and configuring their Linux system : just install it and USE IT."

But if the vision has been steadfast, the path to achieving it has proved somewhat stony.  First Mandrake acquired Conectiva to form Mandriva, and then, in March 2006, Duval was "laid off", as the euphemism has it.  If you're interested, you can read Duval's comments on the whole affair, as well as those of François Bancilhon, CEO of Mandriva, and decide for yourself what really happened.  But looking at the bigger picture, what's interesting about the Mandrake/Ulteo saga is that it recapitulates so much of the recent history of free software, as new distros have continually been created in an attempt to resolve the perceived shortcomings of existing offerings.

In the beginning, Linus created two floppy discs, called "boot" and "root".  As Lars Wirzenius, Linus' Helsinki friend and someone who had the privilege of being present at the birth of Linux, explained to me a few years ago:

The boot disk had the kernel.  When that booted, it asked you to insert the other disk, and that had the whole file system for the Linux system.  All the stuff that these days would be put on a hard disc was on that floppy.  But it was a very, very small file system, very few programs, just enough to be called an independent Unix system.

Copies of these discs were placed on a server at Helsinki University.  They were soon mirrored around the world, for example at the Manchester Computing Centre (MCC), part of the University of Manchester, in the UK.  It was probably here that, in the nicest possibly way, the distro wars started.  The MCC decided it could do something a little better than Linus' basic two discs, and put together the MCC Interim distribution, which first appeared in February 1992, barely six months after Linus had revealed Linux to the world.  Shortly afterwards, other distros appeared: Dave Safford's TAMU (Texas A&M University) and Martin Junius' MJ collections, followed by Peter MacDonald's famous SLS release.

It was SLS that prompted a rather remarkable diatribe in the very first issue of Linux Journal, dated March 1994, that pinpoints the fundamental challenge facing any distro-maker:

Many distributions have started out as fairly good systems, but as time passes, attention to maintaining the distribution becomes a secondary concern.  A case-in-point is the Soft landing Linux System (better known as SLS).  It is quite possibly the most bug-ridden and badly maintained Linux distribution available; unfortunately, it is also quite possibly the most popular.

The author of these strong words was a young Ian Murdock, explaining what prompted him to create his own distribution, which he named "Debian" after his wife and himself - Deb+Ian.  As he told me in 2000: "I regret how harsh I was, because the guy was just trying to do something good."  They may have been typical young man's words, but they are also symptomatic of a feeling that seems to have welled up time and again within the free software community: that the current distros just don't do their job well enough - and that something better is possible.

There's a nice graphical representation of this constant sprouting and growth, and it's interesting to note that Murdock's Debian has proved a strong stock for new shoots of the distro tree.  But this shows only a tiny part of the total richness: the indispensable Distrowatch lists over 300 distributions in its main listing .

This is one of free software's greatest and least-appreciated strengths: the fact that it can continue to evolve in an almost organic fashion, untrammelled by constraints of economics, or even feasibility.  It is this fecundity that drives free software forward unstoppably, and that distinguishes it from the sterile code monster that is Windows, which, trapped within the carapace of its closed source, only slouches towards Redmond to be born every five years or so.  And it is precisely because of this ever-present, irrepressible urge to trump what has gone before, and to create the ultimate distro, that there will never be one.

Glyn Moody writes about free software at opendotdotdot

______________________

Comments

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

there's nothing like a little help from a friend...

AdyJ.'s picture

am just surfing to be able to use our open office and the new linux we had installed. if you have a difficult time installing have someone else do so...

It is very important to also

bryantan's picture

It is very important to also ensure that proper North America auto glass installation standards for windshield replacement are strictly observed in the car glass windshield replacement process. These standards should apply to all kinds of car glass windshield replacement, even in mobile windshield replacement. windshield replacement

Linux and radioprogramming

A hamradioenthusiast's picture

Hi, i am looking for a distro that is ONLY to program radios,
sutch as commercial and hamradios, of all brands and types.

It is a distro for hamradio allready at: http://www.datateknikk.no/AFU-Knoppix/knoppix_afu.iso
This is in GERMAN language.

SCO OpenServer

Ann Sutherland's picture

We use distro OS SCO OpenServer v5.0 and quite happy.

Quite

Glyn Moody's picture

I couldn't agree more.

Simply amazing how now after

Anonymous's picture

Simply amazing how now after all these years parts of the community still want to control. Its not yours! Its everyones to do as they please as long as they play and share nice. That is the beauty. The lack of control by those who have bad manners must be maintained at all costs. Freedom to build that 100,000th distro is yours. Oh how people really don't understand true freedom. Linux is moving just fine in all directions let it bloom the way it has.

SLS - yes

Morten Juhl Johansen's picture

I seem to remember SLS evolving into Slackware... which quite a few use - into Suse... which quite a few use. Oh, how I love open IT evolution.

Correction on the two floppies

Joe Klemmer's picture

The boot/root disks were made by HJ Lu. I remember playing with them back in November 1991. It was really something. And the timing was perfect as I was thinking seriously of buying Coherent at the time.

I also remember a very ingenious kid named Erik Ratcliffe who wrote what was possibly the very first "howto" when he chronicled his work of getting Linux to boot off of a HD. Imagine having to use a hex editor to hack the boot sector. Ah, the fun we had back then.

--
Indie Game Dev and Linux User
Contact Info: http://about.me/joeklemmer
"Running Linux since 1991"

Are you sure?

Anonymous's picture

Here's what Wirzenius wrote in January 1997:

Why are there so many Linux distributions?

The first Linux distribution, if you can call it that, was a pair of floppy disk images. One disk was used for booting, the other held a root filesystem, with enough tools to allow playing with the system or installing on a hard disk.

The floppy disk images were first made by Linus, but he delegated away the task as quickly as possible, since he wanted to concentrate on the kernel.

I have a copy of this old Web page, but I can't find the original at the moment. Anyone?

Whoops

Glyn Moody's picture

That's meant to be my comment - I forgot to log in first...

oh boy, here we go again...

chrisk's picture

After trying and re-trying many distros, I finally decided on ubuntu, and have been running it solid for a few months. Part ofa the motivation in choosing it is that I might as well jump on the bandwagon and go with the "popular" one in an attempt to popularize it.

In the year that I have been running Linux, I have realized that the success of M$ is in that they standardized the industry. When, oh when will all the Linux people decide on a single distro that they ALL get behind and make the "ultimate"? If they did this, I believe that OSx and Windowz would really be up the creek.

As long as there is someone that wants to rebuild the whole techie-wheel, there will be no peace. This shouldnt be as hard as reconciling all the major religions.... it is just software. :)

Diversity is the strength!

Anonymous's picture

If establishing great market share were the main objective, I would agree - standardize and be done with it. The somewhat chaotic mode of free and open source to try numerous different models has the effect of exploring many alternatives at a very moderate cost.

From a standpoint of only commercial offerings, sure, having no more than a few choices (three to five), similar to most of the major industries, tends to work well. Even there, though, if you look at it, there are a few dominant players, but lots of fringe smaller players.

I do not agree that cutting out all of the options and focusing only on one is the way to go. I do believe, however, that within the commercial community that standards and consensus are good things, particularly as size and scale emerge. But in that space, it is industry itself that ought to drive such things. Hobbyists and developers in the free space ought to do whatever works for them. If they get synergy with others, great, but if not, at least they get what meets their own needs - the beauty of software freedom - not found in proprietary solutions at all.

Computers are chameleons, not toasters.

Anonymous's picture

Why? Because the Linux you would want on your mother-in-law's desktop is NOT the same Linux would want running your bank's database server. Nor is it the same Linux you would want on your own desktop.

When all you have is a hammer...

Joe Klemmer's picture

The problem is that Linux, as an OS, is far to big to shoehorn into a little tiny hole. Imagine that an OS or Distribution is the equivalent of a tool (which, OC, it is). Microsoft makes hammers. They only make hammers. They may make various kinds of hammers, but they are all still hammers. Linux is the basis of the whole spectrum of tools. Hammer, saw, screwdriver, tweezers, awl, chainsaw, forklift, jackhammer, dump truck, space shuttle... It would be impossible to have one single distribution and have it encompass the broad range of work Linux does now. From embedded systems and cellphones to desktops and servers and even Big Iron. MS can not and will never be able to match this all encompassing reach. Yes, WinXX dominates the the desktop. But everywhere else, and that is a very large stage, it isn't even a thought.

--
Indie Game Dev and Linux User
Contact Info: http://about.me/joeklemmer
"Running Linux since 1991"

Good blog entry

PB's picture

Great blog entry. I started using GNU/Linux in 1998 and my first distro was Mandrake. Very easy install and the rest is history. I have used every major distro, and a few obscure ones, but have settled for the moment, on Debian. Again, thanks for the history lesson, much appreciated!

You know, I had forgotten

Anonymous's picture

You know, I had forgotten all about that history. I remember reading that USENET post sometime in September of '91. I remember thinking how neat it was that people were creating free operating systems (and not knowing what gnu even was).

It seems like such a long time ago, but I still have that vivd memory of reading that post on my little CGA monitor with my x386. (which was, btw, my first non-Apple computer) Back when everyone seemed to have a BBS, instead of a blog.

Those were the days.

I good little overview that

grazie's picture

A good little overview that for me answers a lot of previously unanswered questions.

The Trouble with 300 Distributions...

billg's picture

... is that people outside the community don't know which one to use.

Linux is hard enough to find and acquire, anyway, for people who aren't accustomed to locating and downloading large files. Those folks -- precisely the audience that Linux needs to appeal to in order to expand its desktop use -- rarely buy software, but when they do they probably buy it in a store.

If they do stumble onto a cite like Distrowatch, chances are they'll be flummoxed by the problem of choosing from 300 distros. It's exactly equivalent to MS releasing 300 versions of Windows, for which it would be be widely excoriated in the Linux press.

Diversity and fecundity may appeal if viewed from within, especially of you hold to the bastardized Darwinian notion that it all drives Linux forward. But, they won't sell Linux to people who've never heard of it.

Actually, here's how you handle that

Sum Yung Gai's picture

Ken "Helios" Starks has already shown us how to do it. Be warned, the below stories are a little long (Helios is quite detailed), but they are very enjoyable and most instructive. Oh, BTW, through some serendipitous karmic effect, it ultimately landed him a job as an IT Director where he converted an entire *company* to GNU/Linux.

Part I (PCLinuxOS):

http://blog.lobby4linux.com/index.php?/archives/64-Linux-Advocacy...A-Fo...

Part II (PCLinuxOS):

http://blog.lobby4linux.com/index.php?/archives/70-86-Year-Old-Great-Gra...

Part III (Fedora Core):

http://blog.lobby4linux.com/index.php?/archives/2006/10.html

Natural Selection

Glyn Moody's picture

I'd say that a different kind of natural selection kicks in here, in that "outsiders" are more likely to come across ads for Red Hat or Novell or Ubuntu etc. than Distrowatch. The latter is there for those who want it; for those who don't, it's probably invisible - as are the vast majority of distros.

All Linux Distros seems same to me..

AZLinux's picture

I have used Fedora, freeBSD, REL and our Local Language Linux, other than that i have never tried any distros.

Why cant everybody work on one Linux Distro and make it universal instead of making 101 useless packages...

Ads?

billg's picture

True, if any vendors were actually advertising a desktop Linux product. I know that Ubuntu has bought a few billboards in the Bay area, which is a bit like advertising the virtues of cheese in Wisconsin. Beyond that, is anyone buying adspace in mainstream media? I've never seen a Novell ad. Red Hat doesn't do consumer desktops.

Well

Glyn Moody's picture

It was also this kind of thing that I had in mind:

Besides the usual frippery that comes with a major stock exchange event, including Red Hat CEO Matthew Szulik ringing the bell to open the exchange, the NYSE will be spending a great deal of money on Red Hat advertising.

Some of it is just silly good fun. For example, you'll find a 10-foot-high red fedora on the NYSE building tomorrow morning, and street teams will be traveling around Manhattan on Red Hat-branded Segway scooters handing out 20,000 Red Hat baseball caps. The NYSE and Red Hat, though, will also be placing full-page Red Hat ads in all editions of the Wall Street Journal, and in the Investor Business Daily, and anyone who uses Bloomberg for financial news will be unable to avoid images of Red Hats.

Not Selling to Consumers

billg's picture

Those are quite likely institutional ads intended to convince investors that Red Hat is a smart place for their money, and, secondarily, to perhaps pick up a few corporate sales. It's all irrelevant to the consumer desktop market because Red Hat is not a player in that game. You can't buy Red Hat in the consumer retail chain.

No matter how good it might become, or how bad Windows might become, desktop Linux will not be in the conusmer desktop game as long as it is remains invisible in the consumer desktop market (ideologues aside). You've got to convince people who don't care about the software that runs their computer to care about Linux enough to make the effort to find it and risk installing it.

Have to agree with you on

Anonymous's picture

Have to agree with you on that note, Redhat's move away from the Desktop was not an unintelligent decision by any means. Linux on the average desktop computer probably is a ways off at this point. There is also the fact that most people who buy a computer in the retail chain, wouldn't care if the cpu was a squirrel in a cage, as long as it did what they want it to do, or at least what they think that they want. Most people, I say with much regret wouldn't know the difference between a Linux, Windows, BSD or Mac desktop, if you slapped them in the face with the install disk.
People have become so complacent that it doesn't matter what runs the computer, as long as it runs. Computers have become such a commodity, that people who can't get the spam and spyware off there computer go out and by themselves a nice new D**l with whatever the OS happens to be, and throw the "old" machine into the local landfill. They couldn't care less. (Although if I'm not mistaken, I think it's D**l, in france you can buy a laptop direct, with Mandriva preinstalled.) Thats one small step for Linux, but it's a step in the right direction.

How about some disks with those hats?

drew Roberts's picture

"Red Hat-branded Segway scooters handing out 20,000 Red Hat baseball caps."

Interesting but I do wonder why they didn't give out 20,000 Red Hat or Fedora distros along with the caps.

all the best,

drew

*grin* have you ever tried

Anonymous's picture

*grin* have you ever tried to install Fedora? They don't want to drive away people with little computer experience. Yes, there are people who can install Fedora all day long without incident, and then there are those who (I'm sure some of you remember this one) don't know the difference between there cup holder and the CD-ROM tray. Fedora is not an easy install, my personal experiences, anyway, have always been buggy. Save Fedora Core 5, had no problems whatsoever. As soon as someone makes a Distro that installs as quick and easy as a LiveCD can boot, Linux will move much easier into the mainstream.

errors with Fedora

Anonymous's picture

My biggest problem is I could never get an error free download. The best I could do was three of five disks. The other 2 disks were always errored.

Installing Linux

Anonymous's picture

Just a note. I have not had any errors or problems installing Linux. Now updating, and installing software is a different beast. Being a newbie to Linux it will take time. I have so far tried seven different flavors of Linux. The one I like the best is Debain, now the one with the best desktop interface in my opion is SimplyMEPIS. That interface on Debian would make it the go to Linux Distro.

White Paper
Linux Management with Red Hat Satellite: Measuring Business Impact and ROI

Linux has become a key foundation for supporting today's rapidly growing IT environments. Linux is being used to deploy business applications and databases, trading on its reputation as a low-cost operating environment. For many IT organizations, Linux is a mainstay for deploying Web servers and has evolved from handling basic file, print, and utility workloads to running mission-critical applications and databases, physically, virtually, and in the cloud. As Linux grows in importance in terms of value to the business, managing Linux environments to high standards of service quality — availability, security, and performance — becomes an essential requirement for business success.

Learn More

Sponsored by Red Hat

White Paper
Private PaaS for the Agile Enterprise

If you already use virtualized infrastructure, you are well on your way to leveraging the power of the cloud. Virtualization offers the promise of limitless resources, but how do you manage that scalability when your DevOps team doesn’t scale? In today’s hypercompetitive markets, fast results can make a difference between leading the pack vs. obsolescence. Organizations need more benefits from cloud computing than just raw resources. They need agility, flexibility, convenience, ROI, and control.

Stackato private Platform-as-a-Service technology from ActiveState extends your private cloud infrastructure by creating a private PaaS to provide on-demand availability, flexibility, control, and ultimately, faster time-to-market for your enterprise.

Learn More

Sponsored by ActiveState