Many years ago I remember someone, possibly Bill Joy, pointing out that the issue of UNIX market penetration is not converting Microsoft users to UNIX because the majority of the people in the world were not yet computer users.

Today, it is hard for the average American to comprehend this. While not everyone who lives in the United States has their own personal computer, few will be free from having some interaction with a computer of some sort every day. Many will start their day with a computerized alarm clock or coffee brewer, prepare their breakfast using a computerized stove or microwave and then get in their computerized car to go to a job that more times than not will have a computer involved in it.

If you don't work, it is likely you will still get to deal with a computerized entertainment center, ATM or store checkout system. You will likely use some computerized communications device such as a cellular phone as well.

Let me contrast that picture with an average person in Nicaragua. While there are computers, Internet access and even Internet cafes, most people do not own a cell phone, computerized alarm clock, personal computer or even a car. Most stores don't have a computer. And, outside the capital, ATMs are few and far between.

The most basic reason for this is money. With an average monthly income of less than $100, it is not likely that these people will be buying a lot of computers much less computerized alarm clocks. Food and housing is clearly a higher priority.

But, there is more to this vision than they are too poor to buy a computer. Looking at it another way, they have better things to do with their money. Writing receipts by hand and counting inventory by hand is just plain cheaper than installing computers to perform these tasks.

Before you say, "but their productivity is lower", think again. They are writing receipts which means they are employed and the cost of doing the job manually is lower than the cost of automation. Sounds to me like they have done a good job maximizing their return on investment.

Things Are Changing

Now that I have set the picture for what is a little piece of what is so commonly called The Third World, let me point out that things are changing. What made it possible for most Americans to have a computer was having enough disposable income to be able to buy one. As the volume has increased, the costs have continued to drop.

Today, for around $500 you can buy a computer that is hundreds of times more powerful than what would have cost you $15,000 just 20 years ago. (For a real data point, SSC's first computer was a Codata 3300. It had an 8MHz 68000 processor, 750KB of RAM and a 27MB hard disk. With no console, Ethernet it cost $12,600. We added a 1200 baud modem for $700 and a dot matrix printer for $500.)

This hardware trend continues. The $100 computer will happen. Broadband connectivity is becoming very inexpensive. Soon, a computer will be the cheapest way to get access to books and other information--even for the people in the third world.

There is, however, one roadblock. As the price of hardware and infrastructure goes down, software costs have continued to become a larger percentage of total system cost. Today, it is easy for someone to spend more in proprietary software than on hardware. While more people in Nicaragua will be able to spend a month's salary on a computer, most will not be able to spend a year's salary on typical software.

Take a look at this WorldWatch article, for example. The People's Republic of China has designed their own hardware specifically for running Linux and is embracing Linux and Open Source software. While this acceptance of Linux isn't unique to China, it is certainly the biggest example. What this means is that it will quickly be possible for low-cost computer technology to get into the hands of billions--yes, I said billions--of people.

Going back to the Nicaragua example, let's see what kind of difference cheap computers could make. Much like roads made it possible for transportation systems to connect isolated communities, Nicaragua could transition from day-long trips to the city to a 10 minute trip to a local Internet cafe in order to order supplies, find a book or just communicate with another community.

This goes way beyond personal convenience. For example, if government information and access to records is made available on-line, the government no longer has to field requests by phone or mail, process them manually and return printed documents. As I pointed out earlier, labor costs are not the big issue but there would also be significant savings in terms of physical resources--less use of paper, printing and physical delivery as well as the likelihood of smaller facilities.

Society could change much more quickly in Nicaragua because of this than it did in the U.S. Why? Because Nicaragua hasn't made a bunch of the mistakes the U.S. has already made. Instead, they get to preview the choices and pick the best path.

For example, in the U.S., most people own a personal car. What this means is that mass transit tends to be poor and up to 50% of the landscape of a city is taken up with support for vehicles. This makes cities larger, increasing the need for more roads, parking lots and support facilities.

In Nicaragua the situation is quite different. Public transit is the most common way to get around beyond the local area (where feet, bicycles and horses are more common). Thus, someone living in a rural location in Nicaragua can choose between the asphalt highway or the information highway for commuting to work, seeking information or buying products without losing a current investment.

For the government, once again the choice is simple. The costs associated with offering computer connectivity is much lower than the cost of building highways. So, in addition to the savings in paper, printing and delivery of services, highway costs are reduced. Thus, there is another savings in resources making it easier to justify government subsidies of computing resources.

In conclusion, I see this as a time for third world countries to learn from the mistakes of the first world. As we watch cities such as Los Angeles try to reconstruct a mass transit system that they destroyed over the last 50 years, the third world can use computers and free software technology to skip over those costly mistakes.

Phil Hughes, reporting from Costa Rica