Pre-installed Linux: Forget PC's. Why not USB-Disks?

As a 11-year GNU/Linux desktop user (and a gnu-tools user 5 years longer than that) I am time and again excited by the prospect of seeing major vendors preloading their PC and/or notebook products with Linux. But none of the news seemed to have lived up to the expectation of us the FS users. The most recent concerns Dell's reluctant interest in doing this.

But then it occurs to me: why not put most of our efforts into easier and more effective tasks?

blindspot Let me give a few examples about this generic idea first. It is more difficult to convince people to switch from MS Office to OO.o; it is much easier to convince people that they should at least allow their children and/or students the exposure to the collection of wonderful education software such as Dr. Geo, gcompris, celestia,... because this is several FS projects fighting collectively against (not-united) proprietary software vendors much smaller than Microsoft. It is more difficult to push free software in developed countries; it is much easier to push it to countries not yet addicted to Microsoft products, as evidenced by the OLPC project. In my frequent FLOSS talks across Taiwan, I try to demonstrate "unusual" (from the Windows users' point of view) things using my Linux desktop, and point out that their lack of exposure to Linux prevents them from even asking any question in the shaded area in the above picture.

Now consider the external USB harddisk makers. Today in Taiwan, one can get a 120GB disk at the price of about $100 US dollars. What risk do they run at all if these makers chop off 5GB to preload some version of Linux on their products and make sure it is bootable?

  1. Microsoft pressure? I don't see any.
  2. Support issues? The customers will fully understand that they get this gratis and extra bonus without any warranty or support.
  3. Consumers are unfamiliar with Linux? But Linux is irrelevant. Many consumers are already familiar with the Windows version of some cross-platform applications that allow personal tweaking just like firefox with its plugins and extensions. The alternative is not Windows, but rather a USB harddisk that is completely passive, which forces me to use other people's desktop that may not have my own plugins, a truely unfamiliar environment!
live technologies one linux machine driving an entire classroom one usb disk saving an entire infected classroom

I do see consumer interest if they advertize it as "a 80-gram notebook without the CPU and the peripherals" because in my talks I see the majority of the population amazed by the concept of live-CD, let alone the possibility of booting your own environment from a usb thumb drive or external usb harddisk. People just don't even know that such possibilities as Live CD, Live USB, and DRBL, exist at all. Much less can they realize what these things can do when combined. Imagine their surprise when you bring a DRBL-Live usb disk + edubuntu packages to a Windows classroom heavily infected by virus, boot the server machine with it, and instantly bring all machines not only back to life but also fully endowed with educational software! Why not put more of our efforts in promoting such things against ... nothing! The latest slogan in my advocacy strategy is: "Save an entire infected classroom with one usb external harddisk." and I can't imagin any Windows follower object to this idea.

I have written an article in Chinese about this idea, separately addressing three different groups of people: the end users, the potential beneficiaries whose business may not be computer-related, and the FS advocates. I am also exercising my whatever tiny influence over the minor computer hardware vendors to convince them to do this. If the world would help by creating such demands, we may see USB harddisks (and even USB thumb drives when the practical issues are resolved) pre-installed with Linux long before the major PC vendors do so with their internal harddisks. At some point they will follow suit, but it won't matter too much to us by that time, because the usb-boot proposal is in better accordance with the idea of a more mobile future, except that we already have the necessary docking stations now, wherever there is a usb-bootable PC.

And the more general and more important point is: Let's focus on what the FS can easily do and the proprietary software cannot. These will be the places where we can make most progress in FS adoption. This is not to diminish the importance of efforts such as OO.o, firefox, wine, and samba that bravely face the challenge of the Goliath. It's just that we a flock of penguins will have more fun to boldly go where no (or only very weak) proprietary software has gone before.

Comments? Let's discuss it at linuxtoday.com .