Summary

  • In RFC 1527, Gordon Cook proposed a Corporation for Public Networking with fifteen constituency governors, nominated by the network community and approved by Congress. Its possible powers ranged from advice on federal spending to administering funds and network operations.
  • Public Law 102-194 supported a National Research and Education Network and named researchers, educators and students as access beneficiaries. It did not create Cook's corporation. The RFC records a contest over representation and subsidy choices, not a working institution.

A network budget needed a constituency

The High-Performance Computing Act of 1991 gave federal agencies a substantial assignment. Section 102 directed them to support a National Research and Education Network, with links among research and educational institutions, government and industry in every state. Agencies were to work with private providers, state and local agencies, libraries and educators to give researchers, educators and students appropriate access. The statute said what the program should seek and whom it should involve. It left plenty of room to fight over the definition of access.

Cook entered that fight with a peculiar document. RFC 1527, published in September 1993, was classified Informational. He explained that he had drafted a possible policy framework during an appointment at the congressional Office of Technology Assessment. OTA never officially delivered this framework for Congress. Cook released a shortened version to the Internet community. That provenance matters: the RFC is an original record of a policy proposal from a well-placed participant. It carries no legislative vote, agency charter or appropriation for the body he sketched.

His practical question was sharper than the usual promise to connect the country: whom would the network serve? Cook pointed to K-12 education, Native American communities, public libraries, and state and local networks. A research network could enlarge its backbone while these groups waited outside it. Even inside an institution with a link, students and faculty might lack documentation, training or usable tools. A subsidy for another institutional connection and a subsidy for easier use by people already connected would buy different things. Speed, reach and usability would compete for the same public attention and money.

Fifteen seats were a claim on the allocation desk

Cook's proposed Corporation for Public Networking gave those choices a physical shape: fifteen governors, nominated by members of the network community and subject to congressional approval. The proposed list had seats for NSF, the Department of Energy, NASA and the Advanced Research Projects Agency. It also included corporate users, K-12, higher education, public libraries with state and local networks, commercial network information providers, long-distance carriers, regional telephone companies, personal computer users, manufacturers, disabled users and university computing.

The list was more than a show of inclusiveness. A library representative would have a different answer from a carrier when asked whether the next public dollar should lower the price of access, extend a link, fund training or underwrite network capacity. Cook was trying to move that dispute from informal agency and supplier conversations into a named governing forum. The composition would make interests visible. It would also bring suppliers to the same table as prospective users.

A seat did not resolve the question of who could nominate its occupant, whose preferences that occupant carried, or how the body would handle a funding decision that favored one constituency over another.

Cook did not present a single fixed constitution. He offered Congress a ladder. At the first rung, a corporation could collect and express the views of NREN constituencies. It might recommend how NSF and other agencies spent their funds. Congress could then give it responsibility for distributing funds, ask it to press providers to remain interconnected, or enlarge its role in access planning. Further proposals included a backbone jointly run with NSF and a wider research and operational network. These powers are not interchangeable. A forum can complain about a grant; a fund distributor can deny it.

A body charged with a backbone would also make operational and procurement choices. Each step moves the allocation desk and changes who must answer for it.

The question of payment sat beside the question of voice. Cook discussed subsidies for additional connections, for greater use inside connected institutions, and for affordable access. He floated contributions from commercial participants in a more ambitious model. None of those options could be settled by drawing fifteen chairs. Congress would have had to decide the beneficiaries, financing rule, limits on the board and the division of authority among agencies, providers and the proposed corporation.

The proposal stayed on paper

Cook closed with four decisions he wanted Congress to make: the degree of network oversight; where to subsidize technology transfer; whether and how to subsidize a backbone; and whether to pay for more institutional connections or broader use inside institutions already connected. His order began with a prior question: who gets access? The proposed board was a way to give that answer a constituency and an accountable address.

The law already on the books offered a narrower record. Public Law 102-194 authorized a federal NREN program and directed collaboration with private and public partners. It did not name a Corporation for Public Networking, its fifteen governors or its proposed funding powers. RFC 1527 documented Cook's design, not a board in session. A later enacted statute or official charter creating that body would change the institutional story; the two original records here contain neither.

The striking part of Cook's plan is its refusal to let “the public” remain a decorative word. He itemized people and organizations that would demand access, then put them near the money. The same list exposed a problem he could not solve with a diagram: representation requires a rule for who appoints the representatives, and a governing board requires a mandate specifying what it may decide. Without that chain, fifteen seats are a seating chart for a power that has yet to be granted.

Sources