Summary

  • The Internet Standards Process gives a complainant two months from public knowledge of the challenged action or decision to file an appeal, while requiring the same dispute to be decided and communicated within "a reasonable period of time" — a section that expressly declines to set a fixed maximum.
  • The Board can direct that an IESG decision be annulled, may recommend action, may not pre-empt the IESG, and is final on whether procedure was followed; for appeals it does not process, the published promise is an acknowledgement of receipt plus a brief explanation of why.

A working-group participant who believes a decision was taken improperly faces a narrow door and a generous-sounding wait. The door is measured in months. The wait is measured in a phrase.

The ladder itself is documented. RFC 2026, the Internet Standards Process, sets escalation from a working-group chair to the Area Directors, to the IESG, and then to the Internet Architecture Board. RFC 2418 invents no parallel route: it sends readers to the standards-process document for the formal procedures for requesting review and for conducting appeals. The Board's charter, RFC 2850, describes it as serving "as an appeal board for complaints of improper execution of the standards process, with powers defined in [BCP 9]". The authority comes from the process document, not from the charter that names the role.

The door is measured to the month

Section 6.5.4 of RFC 2026 states: "All appeals must be initiated within two months of the public knowledge of the action or decision to be challenged." The interval runs from public knowledge, which is not always the date stamped on a decision, and the unit is two months rather than sixty days. A complainant who learns late is not necessarily early.

What the Board can do

The IAB's appeals page sets out the dispositions. Where a complainant is not satisfied with an IESG review, the Board examines the matter, attempts resolution "in a manner of its own choosing", and reports to the IETF on the outcome of its review. If circumstances warrant, it may direct that an IESG decision be annulled, restoring the situation as it stood before the decision was taken. It may recommend action to the IESG, or make other recommendations. There is a hard limit on that power: "The IAB may not, however, pre-empt the role of the IESG by issuing a decision which only the IESG is empowered to make." Finality is stated just as plainly: "The IAB decision is final with respect to the question of whether or not the Internet standards procedures have been followed." In working-group disputes the page makes the decision final on all questions of technical merit as well.

An annulment power plus a final word on procedure is not decoration. A body that could only sympathise would be a comment box. This one can undo, and what it undoes stays undone.

The other clock has no number

Section 6.5.4 requires the dispute to be decided and the decision communicated to the parties, including the complainant, within "a reasonable period of time". The same section carries a note that expressly declines to establish a fixed maximum. That is a duty with a shape and without a deadline: an appellant can rely on being answered, not on being answered by a date.

The record is where the difference between dispositions becomes visible. The IAB statement on submitting appeals, published on 4 February 2026, restates the two-month limit and provides that appeals failing its requirements will not be processed — with the public record carrying "an acknowledgement of receipt and a brief explanation of why the appeal cannot be processed". The appeals page carries the same commitment. The statement also permits a fresh compliant appeal until the later of the original deadline or 14 days after the Board's acknowledgement of the original one, and it bars revising an appeal once the Board has acknowledged it as compliant; withdrawal and a new filing remain open.

Read the two obligations side by side and the asymmetry is about paper, not candour. For an appeal the Board declines to process, the published promise is an acknowledgement and a reason. For an appeal it does process, the promise is to decide the dispute, to communicate that decision to the parties including the complainant, and to report the outcome of the review to the IETF. RFC 2026 section 8 separately lists the Board among the organisations required to announce standards-process activities publicly and to maintain publicly accessible records. The explanation is specified in one branch and left to the reporting duty in the other.

The paper trail differs by disposition; the obligation to answer is not absent from either.

What this review did not obtain

The Board's published appeal register, which the appeals page points to, was not obtained in this review: three retrieval attempts failed, first on provider capacity and then on two timeouts. Nothing here establishes whether any individual appeal was decided, how long a decision took, or whether reasons were published with it. That gap is a limit of this reading, not a finding about the Board. The appeals page itself displayed no publication date, so its currency is uncertain.

What would have to change

Two changes would turn a procedure with a quantified door and an unquantified exit into a remedy an appellant can plan around. One is a stated maximum for the decision, of the kind section 6.5.4 explicitly declined to write. The other is a published basis for each decision — a reasoned record findable by a reader who does not already know the parties. Either would make the second clock answerable the way the first one already is.

The Board's entry in the BTW directory tracks the institution; everything above rests on the documents cited.