I just had the pleasure of meeting with Guy F. de Téramond, who is the Minister of Science and Technology of Costa Rica. Don Guy, as he is referred to by his staff and others, is a serious Linux and free software advocate.

For those not familiar with the structure of Costa Rica's government, government ministers are appointed by the President. For those in the US, this is much like a cabinet-level position. This means a pro-Linux person is very high up in the government.

I originally met Guy when I was down here in June and spoke at the Costa Rica Linux Users Group (GULCR). Guy also spoke, and he really impressed me with his interest in Linux, and his serious desire to bring good internet connectivity to the general public. This meeting was a chance to fill in the blanks in my knowledge of his interests and commitments.

First off I learned that Guy is actually a research physicist who received his degrees in Paris between 1968 and 1977. He was a Full Professor of Physics at the University of Costa Rica from 1982 until 2000, when he was appointed to his current position.

On the computer side of things, he was responsible for interconnecting Costa Rica to BITNET in 1990 and to the Internet in 1993.

Convinced of the potential this connectivity would offer the country, he was responsible for the creation of CRNet, a fiber router-based backbone linking all major academic and research institutions in Costa Rica. Going beyond the borders, he was involved with the interconnection of Nicaragua, Panama, Jamaica, Honduras and Guatemala to the Internet.

When he came to the Ministry he wanted to continue this connectivity project at the national level.

Costa Rica received a $1.2 million grant to start a DSL pilot project, which offers DSL connectivity in five of Costa Rica's 240 phone districts.

Guy explains, "We are just using the infrastructure that is there; just putting the logical elements on top of the fiber and then using the copper line. So DSL is fantastic technology because you are using what is [already] there....This second phase has two crucial parts. We are going to deploy 100,000 DSL lines in all 240 of the nation's phone districts."

To support all this connectivity, there needs to be increased bandwidth to the Internet outside of Costa Rica as well. That brought the discussion to Arcos. Arcos is a ring-based system in the Caribbean that covers Mexico, the other countries of Central America, the Caribbean islands and Miami. It has a bandwidth of about one TB/sec, and its structure is such that all the repeaters will be located on land, making an upgrade much less expensive than with typical sea-based repeaters.

The final project is now on a fast track to completion, and the goal is to start deployment in January 2002.

Knowing that Guy is totally into Linux, I figured it was time to discuss the future of free software in Costa Rica.

Phil: Your connectivity project makes it possible for someone who lives in Quepos to decide they want to become a computer consultant. Is there something else that you have planned to get Linux into the community?

Guy: This is not a simple issue; the one best technology does not exist. The best technology is the integration of technologies. It is generally what is the most simple--simple and powerful. You don't need a crystal ball to know what is going to be the dominant technology.

In the last decade we have always bet on the best technology. Not by knowing every detail of the technology, but by common sense. At the start of the decade, there was discussion about what was going to be the dominant telecom data technology. At the time you had NetBUI, NetBIOS, the IBM protocols and DecNet, but at that point it was obvious that one universal technology was platform independent--TCP/IP--because the protocols were above the hardware.

There were some discussions to advise the universities as to what would be the dominant technology. One day I got really annoyed with those discussions and told the people, "Well, look, TCP/IP is going to be the technology, and I'm not going to come to the meetings anymore. When you come to the same conclusion, please call me. If you come to another conclusion, don't call me."

At the University [of Costa Rica], there was a very integrated system of different technologies. At the end-user level, every user or department was free to put whatever OS or application they wanted on their systems, but most picked Windows or Mac--mostly Windows at the user level. A few crazy fellows in the Math and Engineering departments picked Linux.

In the center we have about 100 servers. They were all running Linux except the databases that were, at that time, running under Solaris on Sparcs.

I think that Linux, in the last two or three years, has really made some progress--more towards the end user. Basically, what we see again, if you go to layers, is at the bottom you have the common infrastructure and at the top you have the user. The more you go toward common infrastructure, the more it is open source, and the more you go toward the end user it is proprietary software. This is how it is right now.

In the coming years, we will see this line going up or going down. That is going to depend a lot on what the Linux and Open Source community does in developing applications for the end user.

Phil Hughes is the Publisher of Linux Journal.