Summary

  • In August 1992, UWI Mona’s CUNet mail route depended on one or two daily calls to Puerto Rico and a person who manually distributed the messages on campus.
  • The August 1994 JAMNet announcement described a different operating arrangement: a 64 kb/s satellite link, an FDDI ring for 400 UWI hosts, and a separate CAST connection, while the old dial-up remained active.

A message had a departure time

The Internet often enters history as a line on a map: a country was connected, a gateway opened, a university came online. A report from Jamaica in August 1992 describes a smaller unit of access. Sam Lanfranco, after visiting the Mona campus of the University of the West Indies and speaking with planning officer Keith Manison, told a mailing list that the Jamaica–Puerto Rico link consisted of Manison calling the CUNet system in Puerto Rico once or twice a day. He uploaded and downloaded email. Then he had to deliver messages manually to anyone else on campus.

That is a specific account of one university’s operation, not a census of Jamaican communications. Its value is the detail of the handoff. The international route could carry messages, but it did not deliver them to every recipient. The schedule, the person who placed the call, and the campus distribution step were all part of the service. In practice, a mailbox was not a continuously reachable endpoint; it was a place in a queue that moved when the call happened and when someone carried the mail onward.

The connection sat within CUNet, an Organisation of American States initiative for Caribbean universities. A September 1993 public exchange about whether an individual with a modem could communicate with Jamaica identified Archie Marshall at the College of Arts, Science and Technology (CAST) as sysop of the Jamaican CUNet node. It described UUCP links to the University of Puerto Rico and its computing facility, CRACIN. The answer pointed the questioner to a university node and a named operator; it did not promise a general consumer dial-up service.

An institutional path was being built, but the record does not show that any Jamaican household could simply dial in.

Two institutions, two paths

The transition announced in August 1994 was not merely the old email run made faster. Saul Hahn, coordinating RedHUCyT and CUNet for the OAS, wrote that a 64-kilobit-per-second satellite link had connected Jamaica to NSFNet on 15 August. Through an FDDI ring, 400 hosts at UWI would have full Internet connectivity. CAST had a direct IP link of its own, which the announcement said enabled connectivity to many outside users.

Those figures describe institutional infrastructure. “400 hosts” means hosts at UWI, not 400 people or public subscribers. The two named campuses also mattered differently: UWI’s ring joined hosts across the university, while CAST’s direct link extended connectivity to outside users. The OAS announcement does not tell us how many people used either route, what service terms they faced, or how far access reached beyond those institutions.

There was also no clean switch-off between eras. Hahn said the CUNet dial-up connection to UPR in San Juan, established during the project’s first phase, was still operating after the new connection. The older route remained alongside the satellite link. The article of faith that a single “arrival date” replaced every earlier practice does not fit the record. A new backbone and a familiar fallback coexisted.

The announcement itself preserves some of the work behind that change. It credits Jamaican coordinators, technical visitors from Costa Rica, NSF contacts, Sprint, and others; it mentions an emergency addressing problem that led to overnight troubleshooting of the routers and modems. An attached ping test sent 14 packets and received 12, reporting 14 percent loss and round trips from 625 to 699 milliseconds. That is one setup-era observation to one destination, not a long-run service measurement. It does show why an announcement of connectivity and evidence of a working path are different kinds of record.

What the timetable tells us

A 2007 Commonwealth of Learning report dates UWI Mona’s first dial-up email access through Puerto Rico to 1991 and says the link later became saturated. That retrospective may be right about the start, but the 1992 correspondence is the contemporary snapshot used here: by then, calls occurred once or twice daily and mail required human delivery. Keeping those claims separate avoids turning a later chronology into a precision the earlier source does not offer.

Heng Lu’s notes are an editorial lens, not evidence about Jamaica. Their useful question is who controlled the practical reach of shared infrastructure. In this case, the answer changed by layer. CUNet and the Puerto Rico node set an international exchange path; a UWI coordinator set the call in motion and handled local delivery; domestic and campus facilities bounded who could get to the message. In 1994, a satellite circuit and separate campus arrangements widened the institutional path, but the sources still do not establish universal access.

The shift from a daily mail run to IP connectivity was a change in the service’s operating surface: where hosts could attach, how messages moved, and which institutions could offer a route outward. The human timetable was not incidental to the early network. It was part of how the network worked.

Sources

  • Sam Lanfranco, “Links to Jamaica,” 19 August 1992, NSRC archive.
  • Percy van Kanten, “Internet access in Jamaica,” 8 September 1993, NSRC archive.
  • Saul Hahn, OAS RedHUCyT/CUNet, “Jamaica on Internet,” 16 August 1994, NSRC archive.
  • Commonwealth of Learning, “Jamaica Connectivity Report 2007,” report PDF, used only for the retrospective 1991 date.
  • NSRC, Jamaica archive index, used only to verify catalog titles and dates.
  • OAS/RedHUCyT, 1995 regional overview, contextual reference only; no additional claim is drawn from it.