Protocol Designers Were Not the Creators of the Internet

RFC 1122 defines the Internet as a network of networks, a network of networks exists only when independent networks are built, interconnected, routed, and operated as a utility, and the designers of the protocols that ride those networks did not create the Internet.

Preamble: A spoke did not create the wheel. Air does not create a tire. And semantics are not a Cisco BFR. The definitions governing every term on this page are published at The Governing Definitions and Controlling Facts of marknichols.com, and those definitions control. The memo this page was built around, MN-UTIL-01 of January 2026, is now the posted Internet-Draft draft-mnichols-protocols-not-the-internet-01 of September 25, 2026, reproduced section for section at Protocols Are Not the Internet: The IETF Internet-Draft by Mark Nichols, and this page states the rule the memo rests on and the conclusion the rule forces.

0. Prologue: this page is written in deposition form, its sections and paragraphs are numbered for citation, and any claim on it can be cited, challenged, or defended by its locus, section number, paragraph number, and sentence ordinal.

0.1 Section 1 states the rule in the IETF's own words, section 2 states what the rule forces any reader to admit, section 3 maps the dependency stack as a wheel, section 4 states the category statement and the activation gap that separate interoperability from utility-grade operation, section 5 records the memo and its posting as an Internet-Draft, section 6 preserves the public record of the LinkedIn post of February 3, 2026, section 7 states the documented-challenge terms, and section 8 records the renumbering of September 25, 2026.

1. The rule is the IETF's, not this site's: RFC 1122 of October 1989 states that the Internet is a network of networks, and RFC 1983 of August 1996 states that the Internet is a multiprotocol internet.

1.1 RFC 1122, Requirements for Internet Hosts, edited by Robert Braden and published in October 1989, prints at Section 1.1.2, Architectural Assumptions, as assumption (a): "The Internet is a network of networks." The document's next printed line reads: "Each host is directly connected to some particular network(s); its connection to the Internet is only conceptual."

1.2 RFC 1983, the Internet Users' Glossary, edited by Gary Malkin and published in August 1996, defines an internet as "a collection of networks interconnected with routers" and states: "The Internet is a multiprotocol internet." A glossary that calls the Internet multiprotocol is not defining the Internet as one protocol suite, and a host-requirements document that calls a host's connection to the Internet conceptual is placing the Internet in the interconnection of networks, not in the host's software.

1.3 The Internet therefore exists because networks interconnect, not because a protocol was published. That sentence is not philosophy and not this site's opinion; that sentence is the architectural assumption stated in the host requirements RFC of the Internet's own standards body, and every claim on this page is measured against that assumption.

2. What the rule forces any reader to admit is stated in four denials, each on its own subject, because if the Internet is a network of networks, then the Internet is not a protocol, not a paper, and not a laboratory's output, and the Internet exists only when independent networks are physically and operationally interconnected into a working fabric.

2.1 The Internet is not a protocol. TCP, the Transmission Control Protocol of RFC 793, and IP, the Internet Protocol of RFC 791, are utilities that ride networks, and RFC 1983 states the Internet is a multiprotocol internet, so no one protocol, and no suite of protocols, is the network of networks that RFC 1122 names.

2.2 The Internet is not a paper. A Protocol for Packet Network Intercommunication, published by Vinton G. Cerf and Robert E. Kahn in IEEE Transactions on Communications in May 1974, describes itself as a protocol presented for sharing resources across packet switching networks that already existed, and a paper that presupposes networks cannot be the network of networks the paper rides on.

2.3 The Internet is not a university laboratory's output. The Stanford University plaque dedicated on July 28, 2005 casts the headline BIRTH OF THE INTERNET over a body that claims conception during 1973, specification in December 1974, and testing during 1975, and conception, specification, and testing are design work in a laboratory, not the interconnection and operation of networks, as Stanford University Is Petitioned to Retract the Headline of Its BIRTH OF THE INTERNET Plaque and to Keep Every Name in Its Body states on the bronze's own text.

2.4 The Internet exists only when independent networks are physically and operationally interconnected into a working fabric. Internet protocols enable interoperability, the World Wide Web enables publishing and retrieval, and neither interoperability nor publishing is the fabric; the fabric is provisioned transport, interconnection, routing policy, redundancy, locality, and continuous operation, the work this page's section 4 names infrastructure activation.

2.5 Any claim of creating the Internet therefore requires infrastructure activation across networks, and a claim that rests on protocol authorship alone rests on the wrong category: the designers of TCP and IP designed the rules by which unlike networks exchange data, and the people who built, interconnected, routed, and operated the networks made the Internet.

2.6 The four denials of this section are the four denials the site states on the names themselves at Article 27 of The Governing Definitions and Controlling Facts of marknichols.com, and the IETF's own process document states the same ground: RFC 2026 of October 1996 defines the Internet as a collaboration of autonomous, interconnected networks and describes adherence to the protocols as voluntary, as section 8.1 of the posted Internet-Draft records at Protocols Are Not the Internet: The IETF Internet-Draft by Mark Nichols.

3. The dependency stack is mapped as a wheel, from the air in the tire at the top to the axle at the bottom, and TCP sits at layer 4, the spokes, because no spoke created the wheel and no protocol created the Internet.

3.1 The wheel maps the dependency stack that RFC 1122's sentence, "The Internet is a network of networks," implies: the wheel runs from the air in the tire at the top to the axle at the bottom, with the protocols and applications, the code, in one column, and the infrastructure activation that turns that code into a utility, the build and the operation, in the column beside it, so that a reader can trace what each layer depends on and what depends on it.

3.2 TCP sits at layer 4, the spokes. A spoke carries load between the rim and the hub, and a spoke carries nothing unless a rim, a hub, a tire, and an axle exist around it; no spoke gets to claim it created the wheel, and no protocol or application gets to claim it created the Internet. The Internet becomes real only when networks are built, interconnected, routed, and operated as a utility, and that work is the column beside the code, not the code.

3.3 The OSI model is a teaching model, and the mapping on the wheel illustrates dependency, not strict boundaries: the layers are drawn to show what rests on what, and the site's account of the OSI model itself, the scheduled successor that never shipped, stands at OSI: The Coulda, Woulda, Shoulda Protocol of Broken Dreams.

4. Interoperability and utility-grade operation are different categories of contribution, the memo's terminology names the difference, and the activation gap is the space between a standards-compliant endpoint and a corridor that carries a secure transaction across borders.

4.1 Public headlines and institutional profiles frequently use "created the Internet" as shorthand for protocol authorship, and some extend the same shorthand to the World Wide Web, its HTTP, HTML, and URIs. The memo this page records separates two categories of contribution: interoperability standards, which define the protocol semantics that let heterogeneous systems and networks exchange data, and utility-grade Internet operation, which is repeatable, measurable, and enforceable service behavior suitable for commerce and institutional dependence across borders.

4.2 The category statement is one sentence: designing protocols, leading protocol development, or leading the development of the Web's software is not equivalent to creating utility-grade Internet operation. Protocol correctness and adoption do not imply utility-grade operation, and a whole-system creation claim requires whole-system evidence. The statement does not diminish protocol invention credit; the statement clarifies a different claim, the creation of utility-grade, enforceable service behavior, which public narratives compress into a single creator story, and the compression has reached educational materials, where a school textbook presented a single-person "created the Internet" claim as literal technical history.

4.3 Interoperability is the ability of heterogeneous systems and networks to exchange data using shared protocol semantics.

4.4 An internet, lowercase, is a network of networks in which packets can be forwarded between administrative domains, and the Internet, capital I, is the publicly joinable network of networks of the registry lineage defined at Article 1 of The Governing Definitions and Controlling Facts of marknichols.com.

4.5 Internet utility, or utility-grade Internet operation, is a service environment in which end-to-end outcomes are sufficiently predictable and accountable to support commerce and institutional dependence across borders, and its defining attributes are measurability, repeatability, enforceability, and operational accountability.

4.6 A corridor is the practical end-to-end path experienced by a flow, shaped by transport provisioning, interconnection capacity, routing policy, and queue treatment across administrative domains.

4.7 Infrastructure activation is the work required to turn interoperable networking into a dependable utility, including provisioned transport, interconnection strategy, routing policy control, redundancy, locality, and continuous operations.

4.8 The activation gap is the difference between standards-compliant endpoint behavior and the corridor properties required for predictable, long-lived sessions and secure transactions across borders. In an interoperable internetwork, reachability may exist while utility-grade behavior does not, and the activation gap exists when endpoints remain standards-compliant while corridor properties fall outside the envelope that applications and transactions require.

4.9 Six observable conditions can occur while every protocol remains correct: high round-trip-time variance and multi-second spikes; loss and jitter that collapse long-lived sessions; congested interconnects and chokepoints; policy-driven queue treatment and class-of-service degradation; application and middlebox timeouts that expire before recovery converges; and unstable routing policy interactions under load or failure. These failures are utility failures, not protocol definition failures, and every one of them is the failure Digital Island's Tier-0 network was engineered in 1996 to remove, as The Tier-0 Architecture: Merchant Transport and the Missing Telecommunications Layer for Internetworking and the Internet documents.