GNU/Linux Horizons: How OLPC May Impact the Future

by Mario Miyojim for the opinion section

Readers of this article may also be interested in: Will the Empire Last Long?

The future: OLPC project and LAMP repercussions

The GNU Manifesto and the Linux kernel development are two important and unpredictable events that took place within the past quarter century, tightly coupled with the internet explosion. However, I propose that we are beginning a new phase of the Open-Source saga, a particularly bright phase, with the hardware development of the OLPC project.

Recently, Nicholas Negroponte announced that the OLPC now can run Windows on it, whereby the initial market may have doubled. If 1/4 of the population of the world (assumed as 5 billion) are not adults, 1/3 of those are older than 5 and younger than 12 years old, and 1/2 of those are the ones to benefit from the OLPC, we are looking at market of 5 billion / 24, or about 200 million children. That market will probably be reached in about three years. Over these years, children who are two or older will get into the age bracket considered, so that number tends do grow.

There will be spinoffs from the OLPC project, making use of the display and energy savings features of its current hardware. I predict that a laptop similar to the OLPC will be used by writers, journalists, scriptwriters and other people who rely on text, due to its low-power reflective daylight fairly large display, which is the closest an electronic gadget can get to a book. Many people still today can only concentrate on documents printed on paper, due to this reflective property. There will be other surprising, unpredictable applications for the OLPC features, I believe.

In the developing countries business environments, the LAMP middleware will be a major force, especially for ad hoc point-of-sale applications (pharmacies, car service stations, restaurants, etc.) because the small businesses cannot afford the big ones (Oracle, SAS, Lotus Notes, etc.). With LAMP becoming influential, technical schools are going to teach PHP programming, and there will be crash training courses for the current Windows-only programmers. There will be specialists setting up Windows-only applications running under Wine or a virtual machine on a case by case basis as an antidote to the Windows Vista influence, as well as to protect them against software piracy charges.

Hardware prices are a great barrier to the growth of computer usage in developing countries. For instance, in Brazil, a 4 GB RAM, dual CPU, 250 GB hard drive server from Dell costs $2200, while the final price for a lean PC sold locally is around $1200. How powerful the movement and the influence on society if there was a start to at least one native manufacturer (to assemble affordable hardware) in each country. Ofcourse costs and overhead would be kept low precisely because of the cost of software today being greater than the cost of hardware. Therefore, these systems would be preinstalled with GNU/Linux and open-source applications for professional use (as distinct from entertainment, multimedia use). I propose that the big assemblers like Dell, HP, Toshiba have their market restricted to wealthy countries, because they maintain the current price levels. The lean and mean assemblers from Asia, Africa or South America may come to invade the U.S. and Europe, like Toyota and Honda did with their cars.

The present: GNU/Linux preinstalls

Microsoft has grown powerful due to Windows preinstalls on OEM computers (see this article). GNU/Linux is poised to take a slice of that market on merit only. A major computer OEM, Dell, is preparing to start a new sales segment: PCs with GNU/Linux preinstalled. If buyers require the ability to run Windows-only applications that they depend on, Dell should preinstall a virtualization framework on Ubuntu, so that the user can install any version of Windows by oneself according tosupplied instructions. That would be a big plus that will attract buyers who are willing to migrate to GNU/Linux but cannot due to this one hurdle: vital applications.

Dell played an unpleasant trick on potential customers in the past, by announcing OEM GNU/Linux computers, then hiding the ad links and charging a higher price than for Windows computers. But that was when Dell was afraid to lose the benefits from helping Microsoft's business plan. This time, the circumstances are quite different: Dell is afraid of losing sales further and not afraid of Microsoft scare tactics; GNU/Linux is more polished and ready than ever to be accepted by common users. If Dell succeeds, then its competitors will have to do the same.

A profession that is now taking shape is that of the family consultant: a professional who maintains family and friends' computers remotely by means of remote access tools. In emergency situations, the family consultant will visit the patient's home. A family consultant will be a professional akin to the family doctor of times past, but for the computers and appliances of the home. These will be more helpful than the corporate support technicians, more reliable, and less expensive, too. For a couple of years, their major task will probably be to install on some computers a GNU/Linux + virtualizer + Windows (3.1, 95, 98, NT4, 2k, XP) combination, to prolong their functional environment and prevent migration to Windows Vista, which is costly and problematic. By doing so, they will save each home a lot of money and hassles. Some of the family consultants will interact with domestic hardware assemblers who, together, may end up becoming the local competition for Dell.

The past: Silent Knowledgeable Adopters in Europe

My opinions in the web forums facing a vociferous pro-Windows minority of posters brought me in contact with a representative of a silent and knowledgeable open-source adopter community. Dr. Alain Empain has a degree in Botany and a Ph.D. in Ecology, and became a Bryologist at the National Botanic Garden of Belgium in 1980.

He lives in the country among cows, sheep, horses, chicken, forests, caves, cliffs. He grows his own vegetables, and has a pond he dug himself. In his property there is a small stream, and a passive solar house he built himself with family help.

Dr. Empain found the Botanic Garden without any Informatics, so it took him the period from 1980 to 85 to introduce computers into the research activity. In 1986 he connected a few PC terminals to a Unix minicomputer. Thereafter, he dedicated full time to software development, hardware maintenance, UNIX server deployment, databases for 12 years. By the end of 1993, he began to play with Linux, and everything changed. He installed the first Linux web server in the Federal Government domain and probably fully switched the first government department to Linux by 1998.

The evolution of their IT structure until 1999 is summarized on the Belgium Botanic Garden web site.

In 2000, he left his occupation for 6 years to recycle into Bioinformatics in Molecular Genetics (something like a sabatical) in his University. He put together a Grid-Computer with all his Linux servers, on which a 6-month computational job for a single CPU--comparison between the full human and bovine genomes--took less than a week.

Now he is back at the Botanic Garden, to supervise the servers and work on his collected samples from tropical Africa.

During his absence, the servers have received a great boost in Linux-only NAS etc. However, most new users have been trained earlier on Windows, so the Linux workstations are now diluted.

Dr. Empain plans to run the Botanic Garden Library client under Wine or to run Windows under VMWare, temporarily.

A newly acquired, expensive camera mounted atop new Olympus microscopes only works under Windows, and he sees that as a problem. As he succeeded to use a new high-resolution [1.3 Mpixel] webcam to take pictures from a microscope, he is now developing a gui for it. In this manner, he will be able to replace an expensive camera with a Windows-only interface by a much cheaper hi-resolution webcam, with an open-source interface. In Dr. Empain's Taxbench web page, there is phrase that stands out as the prime motivation to many open-source projects: It would be a pity to switch to Microsoft just to use a camera; it is something that everyone who reads this should deeply think about.

This camera project reminds me of a patchwork of open-source applications I did for a simulator to print color graphs and tables on a $150 inkjet printer instead of a Windows-driven $6000 color laser printer in 1997, which gave me the belief that open-source software can always provide an affordable solution for any problem.

Despite the dominance of Windows workstations in the Botanic Garden, OpenOffice is now the de-facto standard, as well as Firefox and Thunderbird. This represents a significant victory of the FOSS movement, in the circumstances.

You could say that Dr. Empain is a unique person, and that there must be just a few or none like him in each country, regarding the dedication to practical applications of GNU/Linux and open-source software in each one's field of activity. However, such individuals exert a strong influence on younger people, with a life-long, cascading effect; that is what will bring an ever wider adoption of open-source software in the world.

What I am doing while I wait for the future

I am practicing with virtualization now, using VMWare, and am pleased with the flexibility one would achieve by running a Windows virtual machine together with a regular GNU/Linux distribution. It can serve the high purpose of allowing a painless migration process from Windows to GNU/Linux for the next 2 to 3 years.

I did not yet have a chance to play with the KVM project or the Xen project, which also look interesting.

A group of people and I are also preparing to start up an industry that will provide affordable hardware, software and services for third-world countries, that can be a driving force of that future.

This article published by permission from Mario Miyojim, with special thanks to Dr. Alain Empain. For further information and reading regarding Dr. Empain's projects.

This opinion piece should not be construed as factual information. It contains the opinions and personal experiences of the author at the time of publication and where applicable includes the references to other texts for further research. However, the opinions and personal experiences that have been posted do not necessarily express the opinions of Linux is a registered trademark of Linus Torvalds. IBM is the registered trademark of International Business Machines. Microsoft, Microsoft Windows are trademarks or registered trademarks of Microsoft Corporation both in the United States and Internationally. MS is used in certain cases to further denote Microsoft Corporation. RedHat is the registered trademark for RedHat Inc. All other trademarks or registered trademarks in this opinion piece belong to their respective owners.

Linux is a registered trademark of Linus Torvalds. Microsoft, Windows Vista, Vista are trademarks or registered trademarks of Microsoft Corporation in the United Statest and Internationally. All other trademarks or registered trademarks in this opinion piece belong to their respective owners.