Summary

  • The retained rendering of RFC 9280 mentions errata in one place, in Status of This Memo, and the sentence points the reader to a live web page.
  • The RFC Editor's errata page is described to this review as recording an erratum as an annotation attached to the RFC rather than as a change to its text, in two categories and four states.
  • Those states are described as managed and decided by stream leads or relevant Area Directors, which keeps the judgment layer outside the administrative contract.

There is a sentence that nearly every reader of an IETF standard skips, and it is the one that tells them where the errors are kept.

RFC 9280 constitutes the RFC Editor function of the IETF; the rendering this review retained describes it as Informational and dated June 2022. Read that rendering for the word errata and the trail is short — one hit, in Status of This Memo: "Information about the current status of this document, any errata, and how to provide feedback on it may be obtained at https://www.rfc-editor.org/info/rfc9280".

The sentence does three small things at once. It asserts that status information exists. It names errata as part of what a reader might need. And it locates both somewhere else, at a page rather than in the document you are holding. A reading of the same rendering found the word errata in that section only: no list of error categories, no named deciding role, no procedure. That is a finding about this text as retained on 18 November 2026, not a claim about the RFC series at large, and the difference matters for everything that follows.

What the page is described as recording

The errata page's own wording was not obtained in this review, so this article describes it rather than quoting it. The description returned states that an erratum does not modify the original RFC text; it provides an annotation associated with the document. Two categories organise the reports: Technical and Editorial. Four states track them: Reported, Verified, Held for Document Update and Rejected.

That single design choice settles what a correction does to the artefact. It does not rewrite the paragraph you were reading. It adds a record beside it, and the record carries the label. Of the four, Verified is the strongest, and even that one, as described, annotates. Nothing in the description says a verified report republishes the RFC or amends its sentences. The third label is the more revealing one: Held for Document Update promises a future document, not a corrected present one. A reader who wants the fix now is told that the fix is scheduled for somewhere else.

Where the deciding pen actually sits

The description places management and decision of those states with the respective stream leads or relevant Area Directors, who review a report for accuracy and relevance to the RFC's stream. Not the author. Not the working group that produced the text. And not whoever holds the contract to run the editing function — here, IETF Administration LLC, the entity whose contractual role this article examines.

That allocation is worth stating precisely because it is easy to flatten. It is not "the administrators decided". It is a division of labour across two layers. The contracted administrator sits in the administrative layer, running a function under contract. The judgment layer — is this claim about the standard correct, and does it bear on this stream — is described as held by volunteers in stream leadership and Area Director roles. Read the two together and the arrangement looks deliberate rather than accidental: the party with the budget and the service obligation does not hold the pen that declares a published standard wrong.

Naming that division is a description of where authority is documented and exercised. It is not an allegation about anyone who runs a queue, and this article makes none.

A courtesy, or a chore

Which leaves the question the pointer raises and does not answer. When a document instructs you to go and look elsewhere, has it extended a courtesy or handed you a chore?

The honest answer is conditional, and the condition is not rhetorical. If the page is maintained, authoritative and precise, the pointer is a courtesy: the text stays stable, the status stays fresh, and each does the job it is good at. If the pointer is the only practical route and the page is unclear or unavailable, the same sentence becomes a burden, because the authoritative state of the standard then lives where the standard cannot show it to you. Nothing in what this review obtained settles which reading applies. The page's own sentences are exactly what was not obtained.

What this review did not obtain

Four items, so that no reader mistakes a gap for a finding. The errata page's own wording. Any per-RFC errata record. Any instrument stating that verification obliges republication or amendment. And the instrument that names the errata deciding authority, which is known here only through description. Both source documents are published by the same organisation, rfc-editor.org, which is a limitation of this evidence set and not a hidden one.

What would change the reading is specific. A per-RFC errata record showing verification triggering republication or a textual amendment, or a provision inside RFC 9280 itself that set out the procedure. Either would move correction from beside the standard into it, and this article would need rewriting rather than patching.

Until then the artefact is half of a pointer pair. The standard carries the text. The status travels elsewhere, on a page the standard does not contain, under a pen the standard does not hold.