A few months ago, we made a legislative attempt (zh_TW) to ask Taiwanese government in their procurement plans to prioritize machines supporting 0-keypress usb-key booting. It was not about forcing the government officials to use Linux on usb keys. It was not about forcing them to use Linux at all. It was a humble request that there be a way to set up the BIOS so that usb keys can be the first boot device without user intervention just like cdroms or floppies have always been, that these machines have a useful second life as donations to remote villages when they finish their services in the government. Microsoft adamantly objected (zh_TW), giving misleading arguments that distort the truth. Our attempt failed, but it revealed that MS is really afraid of usb-booting. We shouldn't really care about what MS thinks. We should just mind our own business and boldly go where no windows has gone before. But that happens to be exactly what MS is afraid of anyway :-)

Let's look away from the ICT business/community. Let's talk to people who care about digital divide. Labs with diskless computers can be set up in schools at remote villages. Linux-bootable usb keys loaded with scientific and educational software can be distributed to girls and boys. (No, we don't mean office apps, we mean apps offered by Freeduc-Science and the like.) School library is the place to restore the usb key if it gets messed up. Teachers will have much less hassle managing their labs. Newer versions of usb keys created by LUGs in the metropolitan can be delivered to the remote villages through snail mail. See full article in zh_TW. Hello, UNESCO?

Let's look away from the ICT business/community. Let's talk to organizers of international conferences or other international events such as sports and exhibitions. Diskless computers with internet connection can be set up at two or three sites such as the athletes' villages. Ask each foreign visitor to select her preferred language(s) and give her a linux-bootable usb key loaded with language(s) of her choice. Make them really feel at home in these events -- at least in terms of reading mails, instant messaging, and accessing the web. Let them marvel at the fact that the same computer that spoke Korean a few minutes ago now speaks Japanese. That's what I shall do with the EASTS academic conference in a few weeks.

Let's look away from the ICT business/community. Let's talk to businesses that help students to go abroad for degrees. Show them how a linux-bootable usb key (or a bunch of backup keys) can be heavenly helpful when the notebook computer an overseas student brings with him crashes. It won't be too hard to make a choice between having to use the "unfamiliar linux OS" in his own language to chat with his friends at the home country, and fretting over a sweetly familiar but dead windows. They will realize that non-techies, too, are entitled to a 80% computer-LOHAS life style.

Let's look away from the ICT business/community. Let's talk to the tourism business and specifically to the unions of hotels. Tell them how they could let their customers use their own languages to surf the net during their stay at the hotel. The hotels could lease diskless computers and sell usb keys. Let's ask the union to organize the hotels to register in some wiki or googlemaps, so that any foreign visitor can easily locate along her itinerary hotels that offer such service. See full article in zh_TW.

Do you see the cyan area? You get the idea. Let's talk to non-IT people and/or businesses who care about mobility and multilingual capability but who did not dare did not know they could ever ask MS for such privileges. Please share more possible applications in the comments. Please also let us know if you see interesting, immediately-perceivable aspects of FLOSS over the present combination of proprietary world other than mobility and multilingual capability. May I also stress that we are more likely to find ideas when we think about "combination" -- maybe combination with things beyond software, such as creative-commons-licensed digital contents. The proprietary world strangles themselves by failing to have a simple legal mechanism to cooperate freely. Tasks that require a combination of several digital resources more readily reveal the superiority of our free culture and the web 2.0 mashup characteristics.

Back to the main point: why emphasize diskless computers? Why are ordinary computers supporting 0-keypress usb-booting not enough? I see at least three reasons for it.

history of change of dominant players First, it's about lowering the barrier of entry. It's not about the purchase cost, but rather the maintenance/management cost. It is also about the psychological barrier of entry. A hotel manager wouldn't want to deal with upgrades, crashes, and viruses of a computer. (Yes, we surely will have loads of Linux viruses when it becomes main stream -- maybe not as severe as in windows, but annoying or detrimental nonetheless.) She might be more willing to deal with a broken TV set. And that's how we should emphasize the benefits of a diskless computer -- by comparing it to a TV set instead of a traditional computer. By lowering the barrier, we will have a much larger market for linux, thereby ignoring the dominant position of windows. That's what happened when Apple II ignored the dominant position of IBM mainframe (without fully replacing it). That's what happened when Wintel ignored the dominant position of Apple's products (without fully replacing it). That's also what's happening now: we probably won't see "the year of linux desktop" as a large-scale exodus of old timers ditching their windows in favor of linux. We are more likely to see an incredible number of new linux "computers" boldly go where no windows has gone before. OLPC, gphones, openmokos, and netbook are manifestation of this phenomenon. USB keys could be another example -- if only people realize how much less formidable a PC could be when it has no harddisks.

Secondly, innovation can be faster when the components are taken apart, the developments are made by separate people/teams, and the result can be freely combined by anyone. I have yet to finish reading The Future of the Internet and How to Stop it by Jonathan Zittrain to have a better argument for this. Still, you can already see that the innovation mechanism encouraged by the Bug Labs may prove to be helpful to accelerate the application of FLOSS into previously unimagined areas of life. More concretely, by separating the hardware vendors from the bootable usb-key vendors, many less well-known linux distributions may suddenly become visible to the ordinary people. Think for a moment about how it would ever be possible for HP or Acer to pre-load their PC's or notebooks with artistx specifically for the artists. Yet with diskless computers, this and a lot more analogous things can happen.

Finally, when hardware vendors start selling diskless computers, the debate as to which OS to pre-install will become moot. The shackle that MS places on the hardware vendors (though no one admits to its existence openly) will lose its anchoring point (zh_TW). We didn't know what MS did in order to get Windows into OLPC, eeepc, or aspire one, but we can be sure that MS will have to compete more fairly when it is not sufficient to secretly talk to a small number of hardware manufacturers.

Hilaire of the Dr. Geo fame showed me this encouraging product: the G-Key. The key by itself is no big deal. The exciting part is that their computers don't have disks -- not even ssd's. On the other half of this proposed formula is the Japanese product wizpy, which embeds a Turbolinux-bootable usb in a mobile phone. (Disclaimer: I am not associated with either manufacturers.) Imagine a day when you can go to an internet cafe in any big city of the world, plug your mobile phone into the usb port on a screen, and resume your work where you left off in another city. So join the request and let's promote similar things all over the world. The Tux fleets of usb keys are about to boldly go where no windows has gone before.