Summary

  • Jean Armour Polly made early Internet access intelligible and repeatable for public libraries through a nontechnical guide, a peer discussion list and practical experiments in rural institutions.
  • Project GAIN showed that a working connection was only the first layer: effective public access also required trained staff, physical-terminal rules, continuing support, affordable recurring service and a mandate to serve people without their own equipment.

The most famous object in Jean Armour Polly’s Internet story is a surfboard. The more consequential object is a library desk.

Her March 1992 article, Surfing the INTERNET: an Introduction, opened with a Macintosh, communications software, a modem and an ordinary phone line. From that modest stack, Polly promised trips to distant computers. Yet the guide was not selling effortless motion. It patiently explained electronic mail, discussion lists, remote catalogues, Telnet, anonymous FTP and emerging discovery systems. A person needed host names, commands, account arrangements and enough confidence to recover when the route was not obvious.

That is why Polly’s enduring contribution is larger than a phrase whose first use remains debatable. She helped define the human operating layer between a network and a public that could not be expected to think like network engineers.

A friendly metaphor for an unfriendly interface

Polly wrote for librarians, not for protocol designers. The distinction shaped the work. Her article described useful destinations, how to reach them and what access might cost. It treated online catalogues, electronic journals and discussion groups as parts of library service rather than curiosities attached to a research network.

Her later account of the title is revealing. “Surfing” was meant to combine pleasure with skill, stamina, uncertainty and risk. In other words, the metaphor did not claim that the Internet was easy. It gave beginners permission to try a system that was still difficult.

The guide’s second version made that permission operational. It could be copied for nonprofit educational use, retrieved through anonymous FTP, corrected by email and replaced with a newer file. Distribution was part of the design. The guide was not a sealed manual delivered from an expert centre; it was a portable, revisable interface that one librarian could hand to another.

The scale of its influence cannot be derived from the document itself. Download counts, changes in patron behaviour and the number of libraries that acted because of it are not supplied. But the design is visible: plain language lowered the entry cost, free circulation widened the potential audience, and a correction address made the document maintainable.

A community for the people running the terminals

Instructions solved only the solitary beginner’s first problem. Operating a public terminal created questions that did not fit neatly inside a manual: who could use it, for how long, under which policy, with what assistance, and at whose expense.

In December 1992, Polly and John Iliff founded PUBLIB, an electronic discussion list for public librarians. OCLC’s surviving description shows the breadth of the operating surface: public and staff Internet access sat alongside intellectual freedom, facilities, administration, policy and practical “how-to” questions. The network connection was becoming an institutional issue.

PUBLIB mattered because a public library could be locally isolated even after it joined the Internet. A librarian facing an unfamiliar modem failure, a patron-access dispute or an emerging online resource needed peers who had encountered comparable choices. The list did not create binding policy, and it did not eliminate differences among libraries. It created a place where operating knowledge could travel faster than formal guidance.

Project GAIN tested the layers around the line

Project GAIN turned that intuition into a bounded field experiment. The full report describes five rural New York public libraries and one Indian nation school, not a national sample. Beginning in December 1992, the programme paired Internet connectivity with equipment, software, training, evaluation and continuing support for roughly a year.

The project’s preparation is as important as its outcomes. The team developed selection criteria, interviewed candidate sites, assessed training needs, surveyed the baseline, evaluated access options, designed connectivity and signed an agreement allocating responsibilities. Each site received a SLIP account. The line arrived inside an operating plan.

Funding also arrived in layers. A $65,000 foundation grant paid for project management, evaluation, training, help-desk work and some purchases. Apple supplied a common computer package, while other sponsors contributed materials and services. Sites that had to make long-distance calls received an $800 subsidy. Those figures belong to 1993–94 and cannot be carried into present budgets, but they expose a durable problem: the nominal network fee was not the whole cost of access.

The report says long-distance charges at some sites could exceed the access-service charge. The feeling that a meter was running could discourage exploration. A technically available connection could therefore remain behaviourally scarce.

Public access was an operating commitment

The programme identified a cluster of success conditions: an enthusiastic person at each site, standardized equipment, a usable connection method, common initial training, local user groups, sharing among sites, telecommunications support and persistent help. Project manager Wally Babcock answered questions and contacted sites that went quiet. Support was not an emergency add-on; it was part of the access architecture.

When libraries opened terminals to patrons, a second set of costs appeared. The room had to be arranged. Time on the machine had to be allocated. Patrons needed instruction. Equipment and software had to be protected. Staff had to choose between supporting the new service and performing other library work. Some sites were overwhelmed by demand.

This is the point at which a connection became a public institution. A modem could establish a session, but it could not decide how a scarce terminal should be shared. A provider could sell network time, but it could not supply the local trust that allowed a hesitant resident to ask for help. A guide could describe FTP, but it could not watch the desk while other services waited.

Project GAIN’s report therefore reached a sharper conclusion than a simple connectivity success story: connecting the building might not be the hardest problem. Who received access, how recurring costs were paid, how librarians and patrons were trained, which applications were worth learning, and which policies governed use could matter more after the link worked.

The librarian was not a workaround

It is tempting to treat the intermediary as a temporary feature of an immature Internet. Better software, cheaper connections and personal devices did remove many command-line barriers. They did not abolish the decisions that Polly’s work made visible.

Every public-access system still has an admission rule, a support burden, a cost allocation and a theory of who may be left outside. When those choices are hidden inside an app or outsourced to a platform, they do not disappear; they become harder to contest.

Polly’s story restores the library to Internet history as more than a room containing a terminal. The library supplied legitimacy for shared access. Its staff translated remote systems into local use, balanced one patron’s session against another’s opportunity, and kept a non-commercial door open for people who did not own the necessary machine.

The line was indispensable. So was the person on the public side of it.

Sources