posted by Massimo Sandal on Thu 7th Jul 2005 20:14 UTC

"TuxMini, 2/3"

These are the people the GNU/Linux community should focus on now. To do this GNU/Linux does not need revolutions, it just needs to polish what it already has. Distributions like Mandriva or Xandros should be almost there. If it's perhaps real that today GNU/Linux is way too hard for grandma or the point-haired boss, it is perfectly within reach of the "Firefox target". It is useless to struggle to create a GNU/Linux desktop easy for all, at least at this stage. By aggressively targeting the "Firefox users" we could easily rise from the sad estimated 1-3% to a much more interesting 5-10% of desktops. This would be a major achievement, and it would really shift many things, preparing GNU/Linux for the rest of the computer users. This would be even much more important than it seems numerically, because that 5-10% of desktops would include the many of the most active and exigent Internet users. The Internet traffic would be even more biased towards GNU/Linux, helping the penguin to become a buzzword like Firefox did. Again, what we need is to educate the users: telling about this new alternative, about how and why it can fit their needs and warning them they need to learn a bit how it is different from Windows before actually using it, just because it's different, not because it's hard. These are the users that can be willing to do it happily, and we must and can attract them. We can do it just now; the system is practically already there. We must just make the right people aware of it.

II. Paving The Way To The Desktop With The TuxMini

There is still a big obstacle to the adoption of GNU/Linux on the desktop. This has nothing to do with GUI usability, hardware support or filesystem hierarchy. This is a much more simple and higher obstacle and, although its existence is obvious to anyone, it is often overlooked. The problem is: GNU/Linux practically never comes preinstalled.

Imagine a world where people have to manually install Windows to use it. They would face a cryptic and slow text-based install anytime, and all what they would get out of it would be a bare GUI without any software -apart from a bad browser, a media player and a few other goodies- and very limited hardware support. They would be lucky if their network card actually works. They would have to install all drivers by hand.

Do you think Windows would be very successful in such a world? Probably not. We all know why Windows is the winner: it comes preinstalled and preconfigured with 99% of PCs sold today. Trivial.

Now, imagine the user coming in front of a preinstalled, preconfigured GNU/Linux desktop. Hardware just works, the system boots into a login manager (just like WinXP) and after login a nice, slick desktop comes out, with a rational menu containing all apps and plug-ins needed for internet, multimedia and office use. Icons for drives like cdrom are on the desktop, with the familiar Trash and a new icon called Home. By the default behaviour of the applications the user will immediately learn this is the place where he can put his stuff. He finds a "Software Install" icon, with (let's say) a preconfigured Synaptic. Perhaps at the first time the user will be perplexed, but he will somehow figure out how it works and he will be soon happily installing and upgrading packages from the network. He will probably be happy to see virtual desktops. In a few hours he will have mastered all the basic things he will need to be somehow productive, and he's ready to routinely use the GNU/Linux desktop in a month or less. He will probably never see a shell in the meantime.

This scenario is not already truth, but GNU/Linux desktop is ready for it. What it needs is the preinstalled machine. But in my opinion a plain boring PC with GNU/Linux preinstalled wouldn't catch up. Given the user target we have defined before, we should look for something different. The native added values of GNU/Linux (cheapness, customization, plenty of preinstalled applications and security) should add up to a significant thing Windows lacks: the cool factor.

That's why I now hope some bold hardware producer is listening to me, because what we have to do now is to play on Apple's battleground. We must create a cheap substitute of the Mac Mini with GNU/Linux preinstalled and ready to go. We must create a powerful, flexible, small machine for the desktop of young students and workers that want something both cheap, functional and cool. Despite the hype, Mac Miniís are not cheap: the base configuration is too crippled to provide a pleasant experience, and upgrades are really costly. Macs also do not feature a plethora of really free applications out of the box, although the situation seems to be better than for Windows. iWork office apps are cheaper than MS Office, but they're surely not free. Moreover we can exploit the two opportunity windows that the Longhorn delays and the Apple switch to Intel are opening just now.

Table of contents
  1. "TuxMini, 1/3"
  2. "TuxMini, 2/3"
  3. "TuxMini, 3/3"
e p (0)    132 Comment(s)

Technology White Papers

See More