Colombia has an academic network called Red CETCOL. The network was established in 1994 as a collaborative effort between COLCIENCIAS, the country's public science and technology foundation, and the Colombian Institute for Higher Education (ICFES). Modeled on the NSF Net concept, CETCOL was designed to foment the creation of interconnected databases throughout the country's universities and research institutions, with an emphasis on home-grown content and information-sharing. As with many of Latin America's academic networks, CETCOL was a pioneer in providing Internet access well ahead of commercial ISP services.

Despite an early start and government support, CETCOL's penetration has been largely unsuccessful. Only 20 percent of Colombia's 139 universities are CETCOL members. Even more significant, this participation has been inconsistent. In examining the demographics of Colombian university participation in CETCOL (and networking in general), a pattern emerges: networking is heavily concentrated within specific categories of institutions and notably absent in others. Geographically, academic networking is mostly confined to the two main political and economic centers of the country: Bogotá and Medellín. In fact, universities in these locations are twice as likely to belong to CETCOL as are those in the rest of the country. A second distinction exists between private and public universities, with the latter more likely to connect than the former. Finally, newer universities don't network as much as the country's traditional universities. The following chart breaks down CETCOL's membership by different categories of institutions.

Collectively, these three distinctions point to a disturbing trend: the sharing of research and knowledge is happening only within a select geographic, political, and economic group in Colombia. The country's "national" network is, in fact, a network of elites with little evidence of trickle-down of these efforts to the social and geographic periphery. Unless it adopts dramatically different strategies aimed at broader interconnectivity, Colombiaalong with most Latin American societieswill not develop into a "knowledge economy" along the lines of the newly emerging economies in Asia Pacific.

Since 1995 commercial ISP service has been increasingly available in Colombia. Interestingly, however, I found few university Web sites beyond the those provided through CETCOL, and these also show the same disparities between traditional elite institutions and the rural, private, and newer universities.

Why is CETCOL having problems?

What impedes the growth of academic networking beyond a concentrated group of institutions? I have identified three constraints: the overall political environment, especially as it pertains to telecom regulation; the deepening crisis in Latin American higher education; and the lack of an "information culture."

Colombia has attempted to liberalize its telecommunications state monopoly, TELECOM, with limited success. Vested interests, politics, institutional inertia, and the lack of a "regulatory culture" in Colombia have impeded modernization of the telecom sector and have kept prices for long-distance service at artificially high rates, thus limiting CETCOL's growth. Costs and quality of connectivity are especially problematic for the less-well-off institutions and those outside of major cities.

Early in its development, CETCOL sought a strategic partnership with TELECOM in an effort to join CETCOL's networking expertise and vanguard community of users with TELECOM's infrastructure. TELECOM, however, viewed CETCOL as a competitor for value-added services and refused to collaborate. Since legislation to liberalize the telecom sector began, TELECOM's unions, especially at the operational level, have fought the opening of competition and upgrading of technology which they perceive as a threat to workers' jobs. Strikes have shut down communications service several times in 1997 and, once installed, fiber optic cables are routinely cut, rendering them inoperable. With presidential elections coming up in 1998, the politics will only get further in the way of the lowered costs and improved quality of connectivity CETCOL needs in order to prosper.

A second major constraint to the development of academic networking is the marked decline in the quality of Latin American higher education. A 1996 report by the Inter-American Development Bank asserted that "higher-education performance in Latin America is low compared to most regions, low for the level of investment by society, and lower than in the past." The problems caused by diminishing resources available to public universities have been exacerbated by exploding student populations. Not only is overcrowding outstripping diminishing funds, but universities simply can't afford to hire the skills needed to develop and maintain networks.

The breakdown in public higher education was caused in large part by a disintegration of government regulation of education standards. One of the most notable symptoms of this decline in standards is the skyrocketing number of new private universities receiving government accreditation. Most of these institutions are unresponsive to the educational needs of students and unaccountable to regulation. These new universities usually do not produce (or consume) research, nor do they invest in information infrastructure, and they are far less likely to participate in academic networks. In the case of Colombia, although the number of these universities and the percentages of the total university student population they capture is increasing, only four of CETCOL's 27 members are universities established since 1980.

A successful academic network is based, in large part, on its content: the databases, library resources, and information flowing through it. In Latin America, the quantity and quality of research production has lagged far beyond the most of the world. For example, a recent study comparing Latin American research production with that of Asia found that research output, as measured by publications per R&D worker, is .14 in South Korea and .20 in Singapore, representing outputs three and four times Colombia's (.05).

In Colombia, along with most of the region, many individuals and institutions have resisted the decentralized, horizontal communications structures imposed by the global information revolution. Since the times of the Spanish crown, Colombia (and Spanish America) has been ruled from the center, as manifested by dictatorships and, until very recently, the complete absence of state and municipal elections (these local posts were appointed by the president sitting in the capital city). Institutions are usually very hierarchical, with policies and activities determined by the top individual, often in very personalized fashion. These traditions have clashed with the new network structures designed to facilitate communications across hierarchies, departments, and institutions. Often, for example, instead of LANs (Local Area Networks) at universities, one finds a lone dial-up connection in the university rector's office, although the rector has yet to learn how to use it! The absence of investment in LAN development and staff and faculty training at many (perhaps most) of Colombia's universities is a reflection of the failure to broadly promote information exchange, a sad irony for academic institutions. This has, in turn, prevented universities from producing the critical mass needed to create viable economies of scale for CETCOL and other academic networks.

CETCOL at a Crossroads: Service versus Content

The biggest dilemma facing CETCOL and other Latin American networks is how to remain relevant in the face of the continued growth of commercial ISPs. CETCOL's niche as a subsidized Internet access for universities has been completely outstripped by the advent over the past two years of commercial ISPs. Since these commercial ISPs do not have to limit their service to one group of institutions, they now have client bases large enough to give them substantial economies of scale, and they are not encumbered by the inter-institutional consensus-building inherent in trying to develop a national academic network. The one distinguishing feature left to CETCOL is content. However, the creation and maintenance of content on behalf of a relatively limited community of users is a relatively expensive undertaking, one which universities for the reasons discussed previously have yet to want to pay for.

CETCOL has staked its survival on Internet service, not content. Last April it re-inaugurated itself as one of Colombia's 13 ISPs and abandoned its original strategy of having select universities serve as regional operating centers on behalf of area universities, as inter-institutional rivalries and centralization of networking infrastructure impeded this model's success. This commercial service strategy is similar to the successful effort of REUNA, Chile's academic and research network. According to Larry Press (OnTheInternet March/April 97), REUNA is now 60% commercial, and the academic mission of the network has prospered as a result. However, there are fundamental reasons why REUNA's success in Chile can't be replicated by CETCOL in Colombia.

CETCOL's strategy is to capture enough of the commercial market to be able to subsidize development of content. This may be possible in Chile, where a strong commercial market, far better and cheaper infrastructure, and a relatively well-endowed university community, give REUNA a solid base for content development and a critical mass of trained and demanding users. In Colombia, however, where per capita GDP is less than half of Chile's ($1,720 versus $3,259), CETCOL must still seek to create the demand for content ("information culture") with income from a far smaller and more fragile commercial market.

CETCOL should get out of the ISP business altogether and focus its limited resources on the development of content and on training and education efforts aimed at expanding the currently deficient number of individuals and universities involved in academic networking. Connectivity service should be a priority only in areas of the country where the private sector doesn't have a commercially viable market. CETCOL and any other government-subsidized efforts should leverage LAN development within institutions so that networking ultimately reaches the broadest number of professors and students. The broader context of "information infrastructure" (telecom regulation, cost, and infrastructure) is beyond the control of CETCOL and the university community andgiven the grim reality of Colombian politicsmay take years to get resolved. In the interim, CETCOL can provide a valuable service to the academic community by helping to develop a badly needed "information culture" in Colombia -essential if the country is ever going to have the research, development, and educational capabilities to become a "knowledge economy" in the 21^st Century.

Notes, Acknowledgments, and References:

Much of the information in this article is based on a research paper which can be found in its entirety at Larry Press' Developing Nations website: http://155.135.37.1/fac/lpress/devnat/

Colombia's leading newspaper, El Tiempo, has a weekly section (available every Monday) on computers which includes telecom and Internet developments in the country. It is available on-line at: www.eltiempo.com

Ricardo Gomez of Cornell University studied the impact of networking on NGOs in Colombia. His analysis of the dilemma faced by COLNODO, the APC node which services Colombia's NGO community, points to the crisis in content versus (ISP) service also faced by CETCOL.

Cardoza, Guillermo. "Higher Education, Scientific Research and Sustainable Development in Latin America: A New Agenda." Paper presented at the Harvard University Conference on Higher Education Reform in Latin America. November, 1996.

Larry Press' review of REUNA and commercialized academic networking can be found at http://som.csudh.edu/fac/lpress/articles/commerc.htm or in the March/April issue of OnTheInternet (pg. 40, "Will Commercial Networks Prevail in Emerging Nations?")