Henry Hardy’s “History of the Net” — version 8.5, dated September 28, 1993, prepared for Grand Valley State University — preserved the full text of a message Brian Reid sent on April 3, 1988, headers and all.

The To: field lists three addresses, backbone@purdue.edu first. The Subject: is “Re: soc.sex final results.” The In-Reply-To: names Gene Spafford. The body runs four sentences, followed by a name and a notation:

To end the suspense, I have just created alt.sex. That meant that the alt network now carried alt.sex and alt.drugs. It was therefore artistically necessary to create alt.rock-n-roll, which I have also done. I have no idea what sort of traffic it will carry. If the bizzarroids take it over I will rmgroup it or moderate it; otherwise I will let it be.

Brian Reid T5 (5th thoracic)

What backbone@purdue.edu was, what Reid’s standing with it was, and what “T5 (5th thoracic)” meant to the people receiving that message — those three things make it worth finding.

No power de jure

In 1988, no institution formally governed Usenet. There were no bylaws, no elected leadership, no legal structure with authority over what the network carried. What there was instead was a de facto arrangement: the administrators at universities and research institutions operating the backbone infrastructure made practical decisions about which newsgroups their machines propagated. If the backbone sites declined to carry a group, the group had effectively no existence for most of the network.

Steven Bellovin — himself one of those administrators, and a backbone cabal member — described the arrangement precisely in a 2019 retrospective: “The Backbone Cabal had no power de jure; in practice, though, any newsgroups excluded by the entire Cabal would have seen very little distribution outside of the originating region.”

Between July 1986 and early 1987, this informal body undertook a systematic reorganization of the Usenet newsgroup namespace — the Great Renaming. Before it, newsgroup names had accumulated without consistent hierarchy or taxonomy. The reorganization introduced a tiered structure: comp.*, rec.*, sci.*, soc.*, talk.*, news.*, misc.*. These became the established hierarchies, referred to as the seven main hierarchies.

The renaming came with procedures. Creating a new newsgroup within any of the established hierarchies now required following a defined process: a period of discussion, a community vote, and approval from the backbone. Groups that didn’t clear this process — regardless of community interest — didn’t get created, and if the backbone declined to carry them, they didn’t propagate.

The Great Renaming solved the organizational problem it was designed to solve. In doing so, it also encoded the backbone cabal’s practical, de facto authority as something closer to formal procedure — with the cabal as the approving body.

Brian Reid was a voting member of that body.

G.T.’s Sunset Barbecue

The alt.* hierarchy was founded at a barbecue on May 7, 1987.

Reid, John Gilmore, and Gordon Moffett of Amdahl met at G.T.’s Sunset Barbecue in Mountain View, California — two months after the Great Renaming concluded. Hardy’s history records what they decided: “We set up a link between us, and each of us set up a link to amdahl, and we vowed to pass all alt traffic to each other and to nurse the net along.”

Gordon Moffett, Hardy notes, “had no specific beef or goal, but he wanted to help.”

The structure they were building differed from the renaming’s organized hierarchies in one key respect. In the seven main hierarchies, a newsgroup could be killed: the backbone could decline to propagate it, and without distribution, it would fade. In alt.*, there was no administrative kill mechanism. Groups would persist or die entirely on the basis of whether readers kept reading them.

Reid, who was helping design this, was a voting member of the backbone cabal — the body whose decisions about the seven main hierarchies he was effectively routing around. He didn’t leave the cabal to build the alternative. He built the alternative from inside the cabal, while remaining in it.

Re: soc.sex final results

The proximate cause for alt.sex was a rejected vote.

At some point before April 1988, the Usenet community conducted a vote on creating soc.sex. Hardy’s history records the outcome: the vote passed, but Gene Spafford, who was administering the group creation process, refused to create it. The community had voted; the institution had overridden it.

On April 3, 1988, Spafford sent his announcement of the final results to backbone@purdue.edu. Reid replied to the list.

The message’s tone is worth attending to. “To end the suspense” frames the preceding institutional deliberation — will the community get what it voted for — as a suspense that Reid is now ending by doing the thing himself. The framing implies that what the cabal had treated as a decision was from Reid’s perspective merely a delay.

“It was therefore artistically necessary to create alt.rock-n-roll” is a joke, but the joke is also an argument. Reid has just created alt.sex; the alt network now carries alt.drugs. The claim is that artistic completion — the canonical trio — logically requires alt.rock-n-roll. The point underneath the joke: groups in alt.* exist because someone creates them and readers read them, not because the backbone approved them. The cabal’s logic simply doesn’t apply to the question of whether to create alt.sex.

Richard Sexton, writing in 1995, described the act as follows: “Brian Reid simply, quietly and quickly created alt.sex…” Sexton’s framing is accurate as far as the act itself goes, but it misses the institutional architecture of the message — the recipient list, the subject line, the timing in relation to Spafford’s announcement. It wasn’t sent to a neutral mailing list. It was filed in the cabal’s own inbox, in reply to the cabal’s own decision, by one of the cabal’s own voting members.

(Hardy and Reid’s April 3 message subject line both name soc.sex as the immediate cause; the Sexton account appears to describe an earlier episode rather than the same one.)

T5 (5th thoracic)

Reid explained the signature in a later communication to Hardy, preserved in the 1993 document:

“‘T5’ is the name of a vertebra (the 5th thoracic vertebra). This was my attempt to remind these people that I was an official voting member of the backbone.”

“These people” were the administrators reading backbone@purdue.edu. The reminder was built into the message’s closing notation, appended below his name, legible to everyone on the list. What it said was: I am one of you. I am doing this anyway.

The received account of alt.* — a community rebellion by people the cabal excluded — is not quite wrong, but it’s not quite right either. Reid was not outside the cabal. He was a voting member of it, working at Digital Equipment Corporation, operating infrastructure the network depended on. His standing in the institution was intact.

The T5 signature made the standing explicit. Reid was not just announcing alt.sex to the backbone cabal’s mailing list; he was signing the announcement as a cabal member. The act and the institutional credential were in the same message, addressed to the same audience.

The backbone cabal’s authority was practical — it rested on controlling which groups the backbone propagated. There was no mechanism to address a voting member who used a parallel infrastructure, one outside the backbone’s propagation decisions, to create a group the backbone had declined to create. The cabal could still decline to carry alt.sex on the backbone. What it couldn’t do was prevent alt.sex from existing and spreading through the alt.* infrastructure that Reid had helped establish nearly eleven months earlier.

No artificial death, only natural death

Five years after the April 3 message, Reid wrote to Hardy in a communication dated September 23, 1993. Hardy cites it as [Reid, 1993b]. In it, Reid reflected on what the alt.* design had turned out to mean:

“In retrospect, this is the joy of the alt network: you create a group, and nobody can kill it. It can only die, when people stop reading it. No artificial death, only natural death.”

“In retrospect” is doing real work in that sentence. Hardy’s text is explicit about Reid’s understanding at the time: “At the time I sent that message I didn’t yet realize that alt groups were immortal and couldn’t be killed by anyone.”

The “no artificial death” principle is cited often as though it were a founding design choice — the thing the alt.* hierarchy was built to achieve. Reid’s own account says he arrived at it retroactively, as a description of something the design had produced rather than intended. He built a structure with no kill mechanism; years later, he named what that had turned out to mean.

What it produced: the backbone cabal retained authority over the seven main hierarchies after April 1988. New groups there still required the process, the vote, the approval. What the cabal could not reach was a parallel structure that didn’t require its participation to function. In alt.*, a group that readers kept reading stayed alive. The backbone had no administrative lever.

The authority the cabal held was procedural — it rested on controlling the process of newsgroup creation and propagation within the established hierarchies. Outside those procedures, the authority didn’t extend. Reid, who had voted within those procedures and understood how they worked, had built a structure that operated without them. What he named in 1993 — “no artificial death, only natural death” — was not a principle he had enacted. It was a consequence he had discovered, one that turned out to be permanent in exactly the way the cabal’s authority was not: not by decree, not by procedure, but by the absence of any mechanism to end it.