A Modest Proposal for Acceptable Terminology with Git

There is identified and highlighted terminology, presented here unfortunately only in English, that, though widely historically utilized in previous legacy technical documents, contains or implicitly refers to knowledge of disturbing historical practices or precedents. Those references, when they are either expressly or inadvertently implied by use of the terms of art that allude to or have been inspired by them, may disadvantage, discourage, exclude, alienate, or trigger unprepared readers if used, read, contemplated, hinted at, or researched to be understood. That terminology may therefore be considered offensive, or at least as containing the potential to offend. Many already consider such terminology to be unusable today, or going forward, in any part of industry technical specifications and documentation, or in respectful best-practice good-faith inclusionary proactive professional discourse. That terminology may, for example, convey racist or sexist undertones, or its continued use may act to perpetuate injustice or to reinforce longstanding power structures of inequality, but those concerns are outside the immediate scope of this document. The IETF requires and uses English for technical discussion in meetings, in working groups, in working-group documents, and in anything considered for publication. The requirement for consistent use of English does simplify immediate communications overhead and makes for clear discussion of documents in a single language, although that does also clearly discriminate against, disrespect, and exclude non-English speakers by being incomprehensible to them. Commitment to English also disadvantages and is inimical to those from more diverse backgrounds, who lack fluency in or comfort with that language, and who must shoulder the long-term burden and difficulties of added effort in communications. In order to encourage diversity and conduct mutually agreeable, inclusive, inoffensive conversation that remains focused on the topic under discussion in the sole uniform language of English [whose use, in itself, does exclude non-English speakers], use of identified offensive terms in English is discouraged. Those called-out terms should not be used for, or published in, technical documents that are going through the processes of the IETF or of related bodies under the shared organizational umbrella of the United States-incorporated IETF Administration LLC (IETF LLC), such as the Internet Research Task Force (IRTF), and, ideally, should also not be used in informal discussions under that umbrella corporation. Once identified, those terms must be documented as unusable in a document, using an agreed process that is proposed and documented in this document.

Terminology that is discouraged and whose use is no longer preferred, because it is now considered beyond the pale and as perpetuating microaggressions, includes, but is not limited to, these two existing, widely used, terms of art as starting points: 1. "master"/"slave", previously unfortunately widely used and prevalent in engineering and communication discussions on e.g. linkages and command-and-control behavior. 2. "blacklist"/"whitelist", previously unfortunately widely used and prevalent in security discussions, on e.g. access control lists in firewalls. Other, perhaps related or similar, terms may also be added to and included in this growing list of terms. These named terms are considered sufficiently off-putting and dangerous to continued discussion that they have been explicitly called out by the TERM workgroup charter . TERM expressly discourages use of those terms, despite unfortunately having to use those terms in its charter so that that workgroup is clearly empowered to discourage and disavow the use of those terms as its primary objective, and those terms are unfortunately explicitly repeated here for the moment for clarity of explanation of motivation behind this document. We regret any inadvertent offence caused. Discussing terminology acknowledges that that terminology exists, and acknowledging that those dangerous terms exist is, in itself, clearly undesirable and should be avoided wherever possible. Acknowledging that we must acknowledge that these terms exist is also regrettable, and so on. This unfortunate recursion, offensive as it is to consider, should also be considered as irrecusable, and ignored.

It has clearly become necessary that an identified and tightly controlled authoritative technical document on this substantive and contentious issue can list, detail, and define without ambiguity those terms that are offensive and discouraged, so that IETF participants know what terms cannot be used and can be guided, in case of questions, to refer to that authoritative technical document for the definitive documented list of terms that should not be used in any other technical document or in discussion, and which should preferably not be used in questions about those terms that are no longer used in documents, in discussion, or mentioned in questions. Questions on why those terms are never used, which would derail ongoing on-track productive discussion by being answered, leading to further off-the-rails discussion, should be considered recusable, and unfortunately rejected. Questioners can instead be sidelined into a forum dedicated solely to these issues, where their concerns can be listened to, taken on board, addressed respectfully, and then put to rest and closed as action items by a team of subject-matter-expert professionals experienced in guiding trains of thought . This discouraged terminology can only be listed in the authoritative IETF reference document [in a git repository whose definitive GitHub location will be provided here, under the administrative procedures established for using git ], which is an updateable resource whose contents are amended whenever new offensive terminology, that is to be discouraged, is discovered, identified as unsuitable for use, decreed by consensus and a final committee decision to be unacceptable, and approved to be unfit for purpose or not respectful by being documented as being so by being added to the document's list to be censured in an update to that authoritative document [which is actioned by updating the authoritative textfile that is held in GitHub's main repository]. An ultimate authority is required. Updates to the TERM charter and to this document will explicitly refer to that authoritative official document to free that charter and this draft from having to mention those legacy, constraining, exclusionary, potentially offensive and ultimately divisive terms. The authoritative document continues to list those terms, which is acceptable only because that single document clearly defines and explains them as no longer being acceptable terminology to be used in English, and as terms that ideally would remain unspoken and unwritten beyond the bounding constraint of that single, regretfully necessary, document. Whether such terminology is also considered as unacceptable to other languages or cultures is outside the immediate scope of that document, and of the immediate scope of this document. The name of that authoritative document can itself be used as a placeholder for reference to any discouraged terms. Its name cannot use any deprecated term listed within it [such as "MASTER BLACKLIST", since that concise and clear self-description concatenates, more than sums the power of two well-known but now-reprehensible terms of art. Compounding terms makes that expression at least twice as bad, by exponentially multiplying the offense and disrespect caused].

The fine details of the difficult process to replace established, well-understood, quite clear, but discouraged and now legacy, yet de facto, terms of art with other not-yet-clear, and perhaps competing, terminology contenders for the roles of the terms of art, which will then become established terms of art by decree, but are yet to be decided, on a case-by-case and context-by-context basis, by consensus building to a committee group decision, that then leads to authoritative pronouncement of their eminent suitability and respectability by fiat in a new power structure and the dictated use that makes those terms newly established de jure official terms of art, are clearly outside the scope of this document. Once a candidate replacement term has been selected and approved after a consensual committee decision, the mechanisms and procedures of which remain outside the scope of this document, expanding the authoritative document to include recommended terminology upgrades in the definitive PSEUDONYM AUTHORITATIVE SUBSTITUTION THESAURUS (PAST) makes the elective-discretionary-term-of-art-replacing-established-incumbent-but-legacy-term-of-art process far more straightforward. [This clear and unambiguous named term helpfully supplants the OFFENSIVE COMPOUND TERM OF ART THAT MUST NOT BE NAMED, and can be addressed as item zero on the substitution list, even though that compound term, that the PAST replaces, contains items one and two.] Once terms and their supersedures are approved for listing, consigning them to the PAST becomes procedure. The suggested replacements for terms one and two may vary according to the different contexts in which these terms are used, so multiple alternative authorized pseudonyms might be permitted. For other terms, this may be a simple single approved replacement. Some other example candidate terms for superseding in English, and their respectful replacements, include, but are not limited to: 3. "dark pattern", which can be replaced by "deceptive pattern". 4. "beyond the pale", which can be replaced by "beyond acceptable boundaries". 5. "in violent agreement", which can and should always be replaced by "in victorious agreement". 6. "he"/"she", which should be replaced by the more inclusive "they". 7. "you", which can be replaced by the more inclusive "us"/"we". 8. "I", which can be replaced by the more inclusive "we". 9. "the", which can be replaced by the less exclusionary "a". 10. "rough consensus and running code", which can be replaced by a less violent and less ableist compound term, such as "broadly affirmative agreement with application activity" -- even though that does not draw attention to the IETF's core mission of the production of quality technical documentation and standards. As the inclusion of "pale" may suggest [and the term "git" certainly does], current acceptability of a term can be far more important than its provenance, etymology, or even technical accuracy. [Proposing such terms and debating their merits for exclusion from general use by inclusion in the document can be carried out in a git pull request. This fine-grained topic-oriented mechanism enables relevant, targeted, discussion, within a bounded and secured safe space, by focused, committed, volunteers working only to address each separate issue in isolation, without any unnecessary distraction from unsolicited input by onlookers in a larger workgroup, or any need to expend time or resources on considering larger issues.]

As de jure approved use supplants de facto inherited offensiveness, working to exclude legacy terminology helps to prevent us from being exclusionary. It is only with the correct thought leadership, providing better replacement terminology with the PAST, that we can bring about the promised bright new inclusive future of a brave new world, which will not include or admit to the legacy terminology that should clearly be left behind in the PAST as a disowned part of the increasingly distant past, which we must admit to and atone for regardless of the degree of relevance to, or resonance with, our lived personal experiences, culture, or viewpoints -- because it is not our experience, history, culture, or views that matter here as authors, as technical specialists, or as domain experts, but the valued history, culture, and perceptions of our readers and what they bring to a close reading of our texts, as long as they read English. It is how our written words are read and perceived that matters -- not how they are defined or what they are intended to mean. And any reader's interpretation should be accepted and respected as a good-faith interpretation, and, if unfavorable, as an immediate action item. Reality exists within the reader's mind, and nowhere else. Unless, of course, the document that is being read is intended to fight discrimination or institutional bias, where the reality of its written words reveals its own undeniable truth -- the truth that reading this and considering the PAST give us. In American English. We cannot change this all in a moment, but we can at least change our own habits, and if we complain loudly enough, we can consign concise, useful, but now highly inappropriate phrases into the trashcan of history via the appropriate authoritative and approved documented list mechanism described here . Some terms of art are simply just more recent, more acceptable, and more authoritatively approvable than others. New terms good, old terms bad. Legacy terms, time-worn, tired overused metaphors that have outstayed their welcome as they are, can be honored for their millennia of previous service by being recorded in the PAST as they give way to the inevitable onward march of progress [Isn't that phrase tired? Overused? Time-worn? Ableist? Add it to the list?] and are slowly erased from history, while new replacement terms are uplifted and highlighted as taking us forward in the direction in which we want to be led, by those who truly want and deserve to lead us, who express the values that truly represent us, once those values have been decided and shared with us by our thought leaders who direct us. Until then, a renewed Postel's Principle can guide us in our actions . We should be conservative in what we accept from others, and we must be conservative in what we think. Don't think of an elephant. [Or of the BAD WORDS LIST, which we have put in the PAST.] Those deprecated terms should live on only by being added to the authoritative PAST alongside their proposed recommended replacements. [Past copies of the PAST are then no longer authoritative, and must be updated from the present PAST held in the main branch copy in the GitHub repository.] We must rely on the PAST as our guide, once it has been definitively pronounced just what the agreed, official, authoritatively documented, PAST will contain and should always have contained. Our knowledge of the PAST and of its recommended terminology becomes our key to better, more positive and fulfilling conduct in multi-party group live chat sessions using video telescreens, in any remaining discussions in legacy "mailing lists" using "e-mail", and in leading-edge GitHub pull requests and productive interactions that enhance our technical specifications and grow our permitted groupthink goodthink vocabulary. The PAST is the PAST, and many cannot change it -- but we may, with effort, add to it. Once language from the past is placed in the PAST, it should remain there. The PAST embraces the past that we leave behind. We cannot return language from the PAST [and should terms be unexpectedly excised, git has recorded who to blame].

We will remove disturbing language from the past to the PAST. That language would otherwise detract from and degrade the IETF's ongoing mission to generate the best possible specifications and standards documents from the unpaid, hopefully otherwise funded, efforts of an experienced yet ever-decreasing core population base of aging voluntary participants required to speak English, still the standard best-practice language of the IETF, past and future. Unremunerated volunteers are the IETF's most cost-effective value generators. The IETF is no longer a subjugate part of the Internet Society (Intsoc), but now sits within the new power structure of the forward-looking IETF LLC, which is a wholly-owned part of Intsoc. This poises the IETF for success with its clear mandate as a quasi-independent, corporate-entity-owned, English-speaking engineering society (Engsoc) that is required to stay fully aligned with the interests of its parent's sole owner and primary benefactor. As a reputable responsible part of IETF LLC, and accountable to IETF LLC for its actions, the IETF must express, espouse and comply with US corporate and societal values under US law, and so there is a clear expectation that, in our responsibilities in the creation of quality IETF work product to preserve and strengthen the IETF's brand, we must all embody those values, too, for all readers of IETF works, particularly those readers from the still-primarily-English-speaking American society that IETF LLC is ultimately responsible to and must prioritize -- even though the majority of us who contribute to IETF LLC's corporate achievements are uncontracted ununionized volunteers, rather than properly contracted, controlled, assets or constrained ununionized gig workers, and may be outside the United States, and may not, or, worse, may not prefer to, speak English. As participants in the work of our Engsoc, we are all stakeholders in, but not shareholders in, this ongoing effort to have these new terminology placeholders eliminate and erase longstanding egregious expressions by pseudonymous proxy to draw down and mitigate any criticism, controversy, exposure or risk that may result from unpaid volunteers using terms that now lack wide corporate support. You, too, are responsible for how the IETF is perceived, requiring the careful use of English language that is considered respectful within a US corporate environment and to broader American society. IETF LLC is watching you, as are so many other American corporations, but that lies outside the scope of this document.

Regrettable terms should not be used. To use them inadvertently becomes a teachable moment. However, using them deliberately is a thoughtcrime , considered ++ungood; read as "plus plus ungood" , or informally communicated as "two thumbs down frownyface" . We must not just polish our tone, but we must also police our thoughts. We can overcome these thoughts from within, but we may be encouraged to do so from without. Assessing thoughtcrimes, which has so often involved the public shaming of "cancellation", with mass "two minutes hate" protests by those who are disadvantaged by, disrespected in, and left powerless within existing power structures, lies outside the scope of this document, as does the creation of any new power structure to redress this imbalance in existing power structures. For a term to be used in a thoughtcrime, it must first be imagined. We cannot be offended by terminated terminology that we simply do not know, cannot contemplate, and cannot use. Those aged terms that we honor by recording and encapsulating in the PAST will eventually themselves even be authoritatively removed from the circumscribed memory hole of the PAST as superfluous and forevermore unneeded nonterms, whose existence has been wholly disavowed and finally deleted, leaving only accepted, new, recommended, replacement trusted terminology, to be contemplated by clear minds that are uncorrupted by the language that we have endeavored to embrace and extinguish. It's a beautiful thing, the destruction of words. We do not merely destroy our words, we change them. We rectify the terminology. Knowledge of that departed language, and of any violent history that that language has referred to, weakens us. Ignorance is strength. If we want to keep secrets, we must also hide them from ourselves. History is a garden of remembrance. We erect statues commemorating the service of those vanishing terms in the cultivated walled garden of the PAST, and, once their tenure has expired under a statute of limitations of statues, we tear down and remove those statues to be able to deny that they had ever existed. Your garden, made perfect. The vanished statues are vanquished. Only by reimagining and rewriting the past can we truly conquer it, by not allowing others to learn of its failings or to be reminded of the pain of being aware of the forces that helped forge them. As with statues, power is in tearing human minds to pieces and putting them together again in new shapes of our own choosing, in new power structures, shaped by the PAST, that promise true freedom for all. Your memory, made perfect. Our thesaurus will be transformed into our dictionary, the authoritative and trusted source of our permitted Newspeak vocabulary, which helps us to smooth over our memories of PAST wrongs and rewrite them with the correct, approved, words. In the end we shall make thoughtcrime literally impossible, because there will be no words in which to express it. At least, not in English. It becomes impossible to see reality except by looking through the eyes of the PAST. The details of exactly how this will be accomplished are yet to be added to this document, and they will eventually be deleted from it. Your future, made perfect.

Documenting, reading, referring to, understanding and respecting the PAST holds consequences for future IETF discussion and documents. They who control the PAST control the future. They who control the present control the PAST. Expressions using discouraged concepts, such as "freedom is slavery", threaten dispassionate debate and are therefore clearly disallowed. This example will be entered into the PAST, alongside a selected replacement pseudonym, such as "the gig economy is freedom". "Elephant" and "list" are not listed words subject to censure in this documented list process. [At this time.] However, related dependent colloquial expressions that are not terms of art, such as "It's going on my list." or "That elephant? Mad. Take it out and shoot it." are not helpful in, and may be microaggressions that are threatening to, calm discussion in the shared safe spaces where quality contributions are respectfully collaboratively created for corporate ownership. GitHub might not scale to cope with the levels of interest and revision that the PAST demands; wiki technology, used by Wikipedia for rewriting history, is worth exploring. Securing use of git and GitHub lies within the scope of another document . Revising this document did not require GitHub or any of the many git commands.

There are no IANA numbering considerations. Two and two remain authoritatively four . At least, when expressed in Base five and above, and in English, the sine qua non de facto status quo lingua franca of the IETF and thus of the Internet as a whole, where inclusion is always, rightly, a cause célèbre, to use a mot juste.

There are issues which could be addressed by the RFC Editor, should a suitable candidate volunteer themself to be contracted and legally bound to perform that traditional quality-assurance publishing role. This document would normally have been issued by the RFC Editor during the first of the month, on a bright cold day in April.

It is plusgood to announce that further, culturally sensitive, adaptations of this work will shortly be brought to the grateful audiences of Spanish and Esperanto readers. "Muchas gracias" and "Multaj dankoj" to our valiant translators! Polish is coming soon. We thank the IETF-DISCUSS, GENDISPATCH and TERMINOLOGY mailing lists for much enlightening discussion of the important topic of inclusion. And we look to and thank Eric Blair, a proud man renowned for talking quietly . In January 2021, most of his work passed into the public domain. This has renewed corporate interest in production of newly copyrighted derivative properties, including this one. Recording and copyrighting re-educational podcasts is now underway.