| Internet-Draft | Author Assignment | June 2026 |
| Carpenter | Expires 7 December 2026 | [Page] |
This document describes ethical guidelines for assigning authorship in RFC documents, including guidelines for the use of artificial intelligence during document preparation, and for inclusion of material from other documents. It also discusses the related issues of acknowledgements, editors and contributors. The various RFC streams may apply these guidelines, or set their own guidelines, which will have priority.¶
This note is to be removed before publishing as an RFC.¶
Status information for this document may be found at https://datatracker.ietf.org/doc/draft-carpenter-rswg-authoring-ethics/.¶
Discussion of this document takes place on the RSWG Working Group mailing list (mailto:rswg@rfc-editor.org), which is archived at https://mailarchive.ietf.org/arch/browse/rswg/.¶
This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79.¶
Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet-Drafts is at https://datatracker.ietf.org/drafts/current/.¶
Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress."¶
This Internet-Draft will expire on 7 December 2026.¶
Copyright (c) 2026 IETF Trust and the persons identified as the document authors. All rights reserved.¶
This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (https://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document.¶
Ethical questions sometimes come up about who should be listed as the author(s) of an RFC, who should be listed as editors or contributors, and what acknowledgements are appropriate. Additionally, questions have arisen about the use of artificial intelligence tools during the drafting of future RFCs.¶
The policy guidelines below address these questions and may be applied by all RFC streams as defined in [RFC7841] and [RFC9920], and by any streams defined in future. Each stream has its own approving body [RFC8729], and may explicitly accept these guidelines and define exceptions to and variations from them. In case of discrepancies or gaps, the stream's own policy and the preference of its approving body will prevail over this document. In particular, the final list of authors and editors will be determined by the RFC stream concerned according to its own procedures.¶
The guidelines are intended to be compatible with the RFC Editor's style guide, including [RFC7322], and with an earlier RFC Editor authorship policy [RFCED-policy]. Intellectual property issues are out of scope.¶
The Editorial stream will apply these guidelines from the date of publication of this document as an RFC. Other streams which wish to adopt these guidelines should do so explicitly, also defining any exceptions and variations.¶
For the IETF stream, there is an existing IESG statement on Internet-Draft Authorship: [IESG-policy]. For the IAB stream, see Section 1 of [RFC4845]. For the IRTF stream, Section 4 of [RFC9775] covers this topic.¶
Appendix A covers some general aspects of authorship ethics as background information.¶
Aspects not covered by this document are left as operational choices for the streams and for the RFC Production Centre (RPC).¶
Contributors are people who made smaller contributions to the document than the authors, for example providing initial ideas that others have transformed into publishable text, or drafting only a few paragraphs. People who did not make any such contribution should not be listed as contributors.¶
Listing someone as a contributor is a factual statement that does not imply responsibility for the document as a whole.¶
People should not normally be listed as contributors without their explicit permission. If a person objects to such a listing, and especially if they do not support the document as posted, it may be appropriate to include them in the list of acknowledgements with a suitable disclaimer (Section 4).¶
The dividing line between contributors and authors is a matter of judgement and cannot be rigidly defined. It may vary between the various RFC streams. However, the RPC's practice is to query any document that has more than five listed authors (including editors). Any list of more than five authors must be approved by the relevant RFC stream's approving body, bearing in mind that listed authors and editors bear long-term responsibility for the contents.¶
Acknowledgements should be given to people who have made significant creative contributions smaller than those from the authors and contributors, or to people who have made useful comments, provided critical reviews, or otherwise contributed significantly to the development of the document. The dividing line between people who are acknowledged and those listed as contributors is a matter of judgement and cannot be rigidly defined.¶
Acknowledgements may also be given to people or organizations that have given material support and assistance, but this should not include the authors' regular employers unless there are exceptional circumstances.¶
An acknowledgement should be written as a description of a fact. It does not and should not signify that the person acknowledged agrees with or supports the document. In general, people who do not wish to be listed as an author or a contributor, but have in fact made a significant contribution, should be given an acknowledgement. In unusual circumstances, acknowledgements of contributions have specifically indicated that the contributor does not support the document as posted. A disclaimer such as the following might be used:¶
Thanks to <insert names> for their valuable comments and help during the development of this document, even though they did not fully agree with the WG's conclusion.¶
When in doubt, it is usually better to include an acknowledgement than to omit it.¶
A common occurrence is that an RFC from some years ago requires updating. This is often done by people who were not the original authors. The question then arises of whether to list the original authors on the "bis" draft, even if they are long gone from active participation.¶
When an RFC is drafted by one or more new people but reuses significant amounts of text from one or more earlier RFCs, a situation arises that often requires thought and careful handling. The criteria above suggest that the authors of the original documents should continue to be listed as authors. After all, there is rarely any question that the earlier publications constitute "a substantial creative contribution" to the revised document. However, there are no guarantees that the prior authors will want to be listed as authors of the new draft and take on whatever responsibilities that implies. Ideally, those assembling the newer version will consult with the authors of the previous ones and make mutually acceptable arrangements, but, especially when that is not feasible, sensitivity to all possible issues will be needed.¶
It goes without saying that normally nobody should be listed as an author, contributor or editor against their will. Ideally, the parties involved will agree among themselves, or defer to the preference of the relevant RFC stream approving body. However, we need flexibility to deal with unusual cases, such as these:¶
If an author or editor wishes to withdraw, for example because they no longer agree with the premise of the document, this should be honoured, although the person may then be listed as a contributor or be mentioned in the acknowledgements.¶
As noted above, an acknowledgement is a statement of fact (the person contributed to the discussion). In some cases it may be included even if the person acknowledged objects, for example if they made a suggestion that might later be viewed as prior art.¶
Generalising the point made in Section 5, an earlier author or contributor may deserve to be listed, even if they cannot be contacted when a document is updated after a long interval. Each such case needs to be considered on its merits.¶
In particular, an author or contributor might be deceased.¶
Disputes about authorship, editorship, contributors and acknowledgements for future RFCs will not be settled by the RPC and must be resolved by the relevant RFC stream according to its own procedures. This includes any cases where an author or editor is asked to withdraw.¶
If significant amounts of text are copied from other RFCs or Internet-Drafts, this should be suitably acknowledged. Unauthorised or unacknowledged copying from any other documents constitutes plagiarism and is not allowed. Authors and editors are expected to take reasonable steps to avoid accidental plagiarism.¶
Authors will use various editing programs and other tools for document preparation, and in general these do not raise any ethical concerns. For example, if tables, graphs or diagrams are generated using a specialized software program, this is of no concern. If formal notation is verified by specialized software, this is also of no concern.¶
If an AI tool is used for document preparation, the following guidelines apply:¶
The authors or editors remain entirely responsible for any content generated by AI.¶
The authors or editors remain entirely responsible for all intellectual property matters.¶
An AI tool must not be credited as an author.¶
If AI usage has been limited to improving English grammar, translating from a draft in another language, or other purely editorial uses, this is no different in principle from older tools like spelling checkers.¶
OPEN ISSUE whether to include this: If, however, a substantial part of the document was created by AI, this must be disclosed, typically in the Acknowledgements section. This requirement is to avoid any confusion about the authorship of the document and to ensure that its readers are not misled.¶
This document does not discuss intellectual property rights (IPR) and in no way preempts or alters the various RFC streams' rules and requirements concerning IPR. All authors and editors are strongly advised to be familiar with the applicable rules, e.g. [BCP78],[BCP79].¶
It is worth noting that if a document includes complete acknowledgements and references, it will be simpler to clarify its status as possible prior art in years to come.¶
Copyright in RFCs is governed by the IETF document [BCP78], the IETF Trust/IPMC's Legal Provisions, and applicable national and international law.¶
The word "contributor" used in this document might not mean the same thing as the word "Contributor" used in the IETF document [BCP78].¶
None, really.¶
This memo includes no request to IANA.¶
[RFC Editor: please remove.]¶
draft-carpenter-rswg-authoring-ethics-00, 2026-04-11:¶
Original version (derived from draft-carpenter-whats-an-author-03).¶
draft-carpenter-rswg-authoring-ethics-01, 2026-04-23:¶
Many small changes after first round of comments.¶
Underline that each stream can make its own rules.¶
Added very short section on dispute resolution.¶
draft-carpenter-rswg-authoring-ethics-02, 2026-05-26:¶
Further clarified that each stream may establish its own guidelines.¶
Replaced "stream manager" by "approving body".¶
Moved background material to Appendix, and trimmed it.¶
Removed reference to academia.¶
Removed reference to order of author list.¶
Removed some redundancy.¶
Numerous minor edits.¶
draft-carpenter-rswg-authoring-ethics-03, 2026-06-05:¶
Removed the word "ethics" from the document title¶
Adjusted scope such that each stream may explicitly accept these guidelines and define exceptions and variations¶
Stated that IPR is out of scope.¶
Added statement that the Editorial stream will apply these guidelines¶
Stated that the list of authors and editors will be determined by the RFC stream¶
Reordered several sections for clarity¶
Moved text about author withdrawal to the Exceptions section¶
Mentioned removal of authors in the Disputes section¶
Made author/editor responsibility more explicit¶
Made lack of contributor responsibility more explicit¶
Generalised requirement to acknowledge copying from other RFCs (previously only stated for AI, which was silly)¶
Generalised plagiarism rule (ditto)¶
Re-ordered bullets about AI usage¶
Added brief justification for AI disclosure, and marked it as an open issue¶
The Internet technical community that contributes to the RFC series has some peculiarities. Perhaps the most important is that we generally encourage the free flow of ideas and their re-use in fresh documents. In other words, internal plagiarism between RFCs is normal. Sometimes that means that small or large sections of text are copied from one document into another, and subsequently changed as the discussion evolves. Within the RFC series, we consider this to be normal procedure as long as due acknowledgement is given. Indeed, when technical text has been carefully verified in a previous RFC, reuse of existing text is an important tool to avoid restating a specification or concept, and possibly introducing new unintended interpretations which might cause interoperability issues.¶
The only exception to this is an RFC that carries a "no derivative works" legend according to [BCP78].¶
Valuable comments on this document and its 2015 predecessor [I-D.carpenter-whats-an-author] were received from Loa Andersson, Andy Bierman, Carsten Bormann, Scott Bradner, Dave Crocker, Jay Daley, Martin Dürst, David Farmer, Stephen Farrell, Joel Halpern, Bob Hinden, Russ Housley, John Klensin (who also contributed some text), Larry Kreeger, Mirja Kuehlewind, Watson Ladd, Eliot Lear, John Levine, Jean Mahoney, S. Moonesamy, Lucas Pardue, Craig Partridge, Colin Perkins, Tom Petch, Alexandru Petrescu, Pete Resnick, Eric Rescorla, Michael Richardson, Nathanael Ritz, Rich Salz, Rob Sayre, Yaron Sheffer, Martin Thomson, and Joe Touch.¶
Especially given the topic of this draft, the author apologises for any accidental omissions.¶