Scott Bradner's Roman Standards Process Kept the Appeals

Scott Bradner Gave the Standards Process Roman Numerals. The Appeals Stayed.

Why it matters

On 1 April 1999, Bradner published a Roman version of the Internet standards rulebook. The joke changed the names and numbers. It left the machinery that decides who can appeal, and when, remarkably intact.

Sofia Ren3 November 2026Reading Time 2 minLast update: 3 November 2026

Scott Bradner; identity grounded in a named Harvard University portrait. The Roman committee setting is imagined.

Summary

Bradner's April 1999 RFC 2551 recast his 1996 Internet standards process as a Roman one, complete with “Worst Current Practice” and Roman-numeral identifiers.
The parody kept a recognisable chain of review and appeal, including a two-month filing window and a final board decision.

The first line of RFC 2551 gives away the game: “WCP: IX”, “I April MCMXCIX”, and “Obsoletes: MMXXVI”. It is Scott Bradner's RFC 2026, the Internet Standards Process, dressed for a Roman Senate. The standards group becomes the RETF, its steering group the RESG, and Best Current Practice becomes Worst Current Practice. A document that normally explains how technical specifications acquire authority is suddenly able to make the reader laugh at the costume.

Then the costume stops being the interesting part. A Roman-Draft has no formal status. An applicant still waits at least two weeks before a recommendation can start. The RESG still judges technical quality; the text still says that no algorithm guarantees advancement. Someone dissatisfied with a working-group decision moves through chairs, area directors, the steering group and an appeals board. The final route reaches the society's trustees when the procedure itself is challenged. Appeals must begin within two months of public knowledge of the contested action.

These are recognisable features of RFC 2026, carried into the parody almost bolt for bolt.

That fidelity is the sting. Replacing IETF with RETF does not move the point where discretion lives. The authority to approve a standard, define the shape of an appeal and decide when “reasonable” time has elapsed remains in named institutional hands. RFC 2551 appeared on April 1; RFC 2026 supplied the real 1996 procedure. Together they offer a reading test: find the sentence that lets a participant object, then find the body that can close the objection.

Bradner did not need to invent a tyrant or a broken network. He changed the letterheads and let the procedural architecture deliver the punchline. The document's very accuracy makes the joke work. Its target is the solemnity of a process that can survive a full wardrobe change without losing an ounce of its power.

Sources: RFC 2551, The Roman Standards Process; RFC 2026, The Internet Standards Process.