Linux Doesn’t Need a Unified Distro

Every now and then, someone on a Linux forum will suggest that—in order to make inroads into the desktop market—Linux should make a unified distribution. While that may, in fact, be one model for success in the desktop market, it’s also a pointless suggestion for several reasons:

  1. The current model is just as likely to be successful as a “unified distro” model
  2. A unified distro simply won’t happen
  3. Linux standards are already in the works (and posting about them in a forum doesn’t make them get adopted any more quickly)
  4. Not every Linux distribution aims to “conquer” the desktop market

The irony is that a lot of times when this suggestion is made, the person making the suggestion objects to the high number of distros out there and then wants to create a new “unified” distro. If X number of distros is too many, why would you want X+1 distros?

There’s an implicit assumption in the suggestion that Linux lacks direction and that all of the developers developing these different distributions are, in fact, wasting manpower. However, because Linux and most of the projects associated with the kernel are open source, energy spent developing one distro or parts of one distro often comes back to help the rest of the community. For example, Ubuntu bases itself on the unstable branch of Debian, but Ubuntu developers give back to the Debian community the improvements they make. Linspire, though reviled by many in the Linux community, actually makes a lot of its apps open source and sponsors many open source projects.

The other side of open source is doing whatever interests you, not some company. This is why a “unified” distro will never happen. If that “unified” distro aims at imitating Windows in order to woo Windows users, then someone (either who is working on this “unified” distro or who isn’t working on it) will say, “Hey, I don’t want to imitate Windows. I want some other desktop,” and a new distro will appear. Then, someone else will say, “I want a distro that specializes in older hardware,” and a new distro will appear. Someone else will say, “But this ‘unified’ distro doesn’t suit the needs of schools,” and a new distro is born. Then, a new one will appear for servers, for business desktops, etc. Pretty soon, even if a unified distro happened for one week, you’d end up with as many distros as you have now.

A unified distro will not last because there’s nothing tying this unified distro together (open source means anyone can fork at any time—no one company owns the Linux kernel) and because people have different interests and needs.

More importantly, as any evolutionist can tell you, diversity leads to survival and growth. Sure, Microsoft dominates the desktop market now, but the dinosaurs once ruled the earth, too. In fact, that lack of diversity is a major contributing factor in the prevalence of viruses and malware in the computing world.

And, in fact, if Linux had a unified desktop distro, that distro would bring all of desktop Linux down with it, should it fail—as much as it would bring all of desktop Linux up with it, should it succeed.

There’s a false assumption in the unified distro idea, though, that competing efforts will not lead to mass adoption. That’s like saying people will never eat pizza because there are too many kinds (deep dish, New York-style) or too many topping choices (anchovies, garlic, mushrooms) or too many sizes (small, medium, large). People like choice, as long as they know what the choices are.

That’s the real problem—people don’t know what the choices are. In fact, they get overwhelmed by the number of Linux distros because they aren’t used to having a choice and because they know next to nothing about the choices. The names of major Linux distributions just sound like gobbledygook to the unitiated (Blag, Mepis, Ubuntu, SuSE, Gentoo, Mandriva, Fedora). Perhaps two of the few descriptive Linux names out there are Damn Small Linux and Linux from Scratch—and neither of those is going to seriously take the Linux desktop up above the 10% mark.

What will help Linux succeed on the desktop are the following (not a unified distro):

  1. More publicity for Linux desktop choices. Linspire, SuSE, and Red Hat make it into the news every so often. A few commercials might help Linspire. SuSE and Red Hat target more of the corporate arena.
  2. Mainstream vendors like HP, Dell, etc. selling Linux preloaded. Presumably, HP is already planning to sell Ubuntu computers.
  3. Experienced users helping/teaching new users to install and configure Linux on their already existing computers (I heard that Linux User Groups already have installation parties that happen every so often).

As a closing note, I often hear from some purists that “Linux doesn’t need more users” or “Linux doesn’t need more market share.” As long as we’re using the word need, I agree. Linux doesn’t need anything, but Linux could certainly benefit from a greater desktop market share. It doesn’t have to dominate the desktop—no operating system should—but it could use at least 10% of the market.

If Linux were to gain a substantial amount of the desktop market, hardware vendors would be more likely to make Linux drivers or release driver code to Linux developers, and commercial software and gaming companies would be more likely to port their wares to Linux. The percentage of Linux users who dual-boot to Windows just to play computer games is staggering. Cedega can’t handle everything.

So can we please stop with the unified distro talk? It isn’t a good thing. It isn’t going to happen. And, if it did happen, it would be only momentarily until a split happened. It’s also just not needed.

One thought on “Linux Doesn’t Need a Unified Distro”

  1. An interesting post– with many valid statements. But also full of generalizations and assumptions that are debatable. Saying something will “never” happen takes more the flavor of strong opinion than market trend.

    No one is presenting the idea of adding “yet another distro” to make a unified package. The concept is that of removing the “distro” concept entirely– and replacing it with a universally agreed upon and marketed Linux package with a consistent GUI (versatile, but consistent), core end-user programs (word processing, basic game package, browser, etc), and in a format / name / logo designed to gain universal market recognition.

    Currently the new Linux user IS faced with a plethora of “distros”… all claiming to be the best on the planet. An army divided against itself cannot stand. All that those in favor of a unified Linux present is that the Linux community needs to present a unified concept to the desktop community. Not three concepts, not two dozen concepts: one. What additional packages people wish to install from there is up to them… just as with any operating system.

    What is standing in the way of this is the general attitude of “it can’t be done”… and the ever-present resistance to change, insisting on personal agendas, and refusal to cooperate for the greater good of the community and world as a whole. Some Linux users have a tendency (as is common in the computer field) toward an “elitist” attitude that insists new customers adapt to the platform– rather than the platform adapting to the needs of the customers. That is an old and tired philosophy that needs to change.

    I disagree that unification of the Linux community will never happen. In truth, it is almost inevitable. The question is: does it happen in the near future… or does it take another decade or two? That will largely depend on how willing individuals in the Linux community are willing to cooperate– to drop personal agendas– to compromise for the greater good of the end goal: to create the best, most widely-accepted operating system on the planet.

    That will never happen so long as our community remains fragmented among numerous and unnecessary “distros” rather than a basic core desktop package with additional “add-on” packages. Fragmentation creates chaos and hurts the community. For Linux to go mainstream– regardless of the kickers and screamers– the negativity will need to give way to a more end-user-friendly, desktop-friendly attitude of cooperation.

Leave a Reply

Your email address will not be published. Required fields are marked *