Summary

  • By October 1992, Ecuanex exchanged email and conference traffic with APC through a twice-daily UUCP connection, while Ecuanet offered live Internet access through Banco del Pacífico’s satellite link to Miami.
  • The January 1993 Sprint/ICM peering gave Ecuanet a distinct 64 kbps path and began the routing process for 29 Ecuadorian networks; a separately proposed university backbone was not evidence of a working route.

One country, two different meanings of “online”

An October 1992 message from Ecuanex’s technical administration did not describe a single Ecuadorian network. It said that two services with similar names belonged to different organizations. Ecuanex offered email and electronic conferences to about 100 individuals and organizations, including universities, non-governmental groups and social and ecological organizations. Its node exchanged messages with the Association for Progressive Communications (APC) in San Francisco twice a day over UUCP. Most users connected with an offline reader. Telephone-line quality and availability were among the main technical problems.

The same correspondence described a newer Internet link opened by Banco del Pacífico in the preceding months. The bank supplied a portion of its satellite capacity to Miami and dedicated domestic circuits; universities and larger users were about to use the service. A later October note from Ecuanex made the distinction explicit: Ecuanex connected through San Francisco using UUCP, while Ecuanet had recently begun a “real live Internet service” through a satellite connection to Miami. That phrase belongs to the correspondent.

The transport details explain the practical difference: store-and-forward mail was useful international communication, but it was not the same as a continuously available interactive IP path.

The distinction matters because a connection date can conceal different services. Ecuanex’s users could exchange messages and conference posts without each user’s computer reaching an Internet host directly. Ecuanet’s users could reach Internet resources through a bank-backed gateway, but only where an institution or dial-up customer had a usable path to an Ecuanet node. Neither arrangement establishes that every Ecuadorian was online.

A bank’s communications network became an Internet route

Banco del Pacífico’s role was more than sponsorship. A U.S. National Science Foundation official who visited Ecuador in October 1992 described Ecuanet as a nonprofit association with free access for universities and similar nonprofit institutions. It received bandwidth on the bank’s domestic network, which served automatic teller machines and inter-branch voice and data, as well as on the bank’s satellite link to Miami. The Internet service therefore sat on a communications system built for other uses.

That shared infrastructure expanded what Ecuanet could offer without first building a separate national and international network. It also shaped capacity and access. A report on the Galápagos described a research station only a short distance from the bank branch in Puerto Ayora, but the station did not yet have an Internet connection. A terminal server was planned. In a test from the mainland, the link was operating at 2,400 bits per second; the correspondent said that available satellite speeds could vary with banking use and time of day. The report records a test and a plan, not a completed Galápagos service or a nationwide speed profile.

EcuaNet’s own presentation at a national data conference in March 1993 described a broader offer. The proceedings, published later that year, listed nodes in Guayaquil, Quito, Cuenca, Ambato, Machala, Manta and Galápagos. It described dedicated lines for institutions and dial-up connections for smaller organizations and professionals. Education and research institutions, along with medical centres, were offered membership without a fee; professionals were quoted US$10 per month, while other organizations contributed according to their size.

These are the operator’s published terms and advertised footprint, not an independent census of active subscribers or performance.

The membership model made the access boundary visible. A university could receive different treatment from a professional or a private organization, and the connection still depended on equipment, a local line and access to an Ecuanet node. “Free” membership did not eliminate the cost of a modem, a dedicated circuit or the bank’s shared capacity.

The route that went live—and the backbone that was only proposed

A third network appears in the contemporary record, but in a different tense. A November 1992 description of the Red Ecuatoriana de Información Científica y Técnica (REICYT) laid out a proposed university and research network. It planned to connect six universities, the national science council and a Central Bank information service. The design called for 9.6 kbps leased lines among Quito, Guayaquil and Cuenca, a 64 kbps satellite connection from Quito to Princeton’s JVNCnet, and 25 dial-up lines. The feasibility study had been completed, and the project was expected to begin within months.

Its estimated investment was about US$1.5 million, divided into 18-month and six-month phases.

That plan should not be folded into Ecuanet’s live Miami route. They had different sponsors, endpoints and status. The REICYT document is valuable precisely because it states the intended footprint and the work still to be done. It does not confirm that the proposed circuits, satellite antenna or dial-up lines were operating.

On 20 January 1993, Sprint’s Internet engineering team announced that a new Sprint/ICM–Ecuanet link had become active. The Sprint/ICM router at the PanAmSat teleport in Homestead, Florida, peered with Ecuanet, AS 2277, over a 64 kbps channel. Sprint also submitted network-administration-change requests for 29 Ecuadorian networks to start routing them on NSFNET. The message called this the first Latin American backbone-network connection served by that new Homestead router. It did not call it the first Internet connection in Latin America: Ecuanex and Ecuanet’s earlier link were already documented.

The activation and the routing requests are separate evidence. The 64 kbps channel was active; the submissions initiated routing for 29 networks. They do not prove that all 29 were immediately reachable, that every institution used them or that Ecuanet’s bandwidth was unshared.

Two organizations were still negotiating their boundary

The distinction did not disappear after the Sprint link. In December 1994, Intercom/Ecuanex described plans to join university campus networks into a wider network, expand conferences through its APC affiliation and let local users host information services. It said that it had proposed a direct, interactive Internet connection through Banco del Pacífico’s Ecuanet and was waiting for a response. The note is a snapshot of a pending proposal, not proof that the connection was later accepted.

The request makes the earlier architecture legible. Ecuanex had a functioning community service and an international UUCP exchange; Ecuanet had a packet-routed Internet path on bank communications infrastructure. Ecuanex still saw value in direct interactive access. The two services were not interchangeable simply because both exchanged messages across borders or both used the word “Internet.”

Reading the access record

The practical unit of history here is not a founding date or a national label. It is the path a user could actually use: an offline reader and a scheduled UUCP exchange, a modem or dedicated line to an Ecuanet node, then a shared satellite route and, from January 1993, a named peering and routing process. The record also shows a planned university network whose topology was specific but whose implementation was still prospective.

That view keeps three questions apart: whether a message could cross a border, whether an institution had an interactive IP route, and whether a nationwide backbone had been built. Ecuador had evidence for each question at different stages, but not a single event that answered all three. Counting a proposal as a service overstates deployment; counting UUCP mail as a direct IP circuit erases the useful but different service Ecuanex had already made available.

The most consequential dependencies were local lines, shared satellite capacity, node availability and the terms for connecting organizations. They determined who could use a route and when. By 1994, Ecuanex’s request for a direct connection through Ecuanet showed that the operational boundary between the services still shaped access.

Sources