Background and Context
This maxim was collected October 26, 2018, one of three numbered "Rules of Good Instructional Design" filed within about half an hour of one another — by creation timestamp it is actually the earliest of the three, preceding the First Rule by roughly a minute and the Third by nearly half an hour, despite carrying the "Second" label. Together the set reads as a single sitting's worth of accumulated professional irritation, likely provoked by a concrete run-in with a course or job aid where the same instructions lived in two places and only one of them got fixed. Where the First Rule targets the correctness of the sentence itself, the Second Rule targets something one level up: the architecture of the document — where information lives, and what happens when it has to live in more than one place at once.
Interpretation
The maxim has two clauses doing different jobs. The instruction — "don't repeat assignment instructions in more than one place" — is a structural rule about where information should live. The consequence clause — "you may look like an idiot when you forget to update one of them" — names the specific failure mechanism with unusual precision for a maxim this short. Compare it to the First Rule's flatter "otherwise, you may look like an idiot": that one leaves the mechanism implicit, trusting the reader to infer how a typo damages credibility. This one spells it out. The danger isn't duplication itself — two correct, matching copies of an instruction cause no harm the moment they're written. The danger is what duplication becomes over time: a document that will, almost certainly, need to be revised at some point after it ships, at which point every copy has to be found and updated in lockstep, and the maxim's whole force rests on the empirical observation that this basically never happens perfectly. Someone updates the instructions in the syllabus and forgets the copy pasted into the assignment prompt; the two documents now disagree, and disagreement is worse than either version alone, because the reader can no longer tell which one is authoritative.
This is the instructional-design cousin of a principle software engineers gave a name to independently: DRY, "don't repeat yourself," and its companion notion of a single source of truth — the idea that any given fact should have exactly one authoritative home, with every other place that needs it referencing that home rather than copying its contents. Mike's maxim arrives at the identical structural insight from an entirely different discipline, which is itself worth noting: this isn't a borrowed software engineering slogan retrofitted onto instructional design, it's a working teacher/designer rediscovering, from lived experience with course materials, the same failure mode that version-control and technical-writing disciplines built entire tooling ecosystems to prevent.
The maxim's real subject, then, is drift — the slow, silent divergence between two things that started out identical. A single wrong instruction is a bounded problem: it's wrong, someone eventually notices, it gets fixed. Two contradictory instructions are a compounding problem: now the reader has to figure out not just that something is wrong, but which of two competing claims to trust, and worse, they may not even notice the contradiction exists until they've already acted on the wrong one — a student who follows the syllabus version of a due date that the slide deck quietly changed doesn't find out they're wrong until the assignment is late. The credibility cost the First Rule describes as instantaneous — the reader's trust swerves the moment they spot an error — here becomes retroactive: the reader trusted a version of the instructions that turned out to have been superseded somewhere else, and the designer's authority is damaged not by a visible mistake in the moment, but by an invisible one that surfaces only after someone has already relied on it.
What makes this a design rule rather than merely a proofreading habit is that it's preventable at the point of authoring, not just catchable at the point of review. The First Rule's fix is a better second pass over existing prose. This rule's fix is a different structural choice made before the prose is even duplicated: write the instruction once, in one canonical location, and link or reference it everywhere else it's needed, rather than copying its text. The maxim is, in effect, an argument for information architecture as a form of care — not "write it correctly" but "build the document so that a future correction only has to happen in one place."
Current Relevance
The maxim's underlying risk has, if anything, multiplied with the number of channels a single piece of instructional content now typically lives in. A 2018 course might have duplicated instructions across a syllabus and a handout; a 2026 course routinely spreads them across a learning management system, a shared slide deck, a pinned Slack or Teams message, a recorded video's on-screen text, and an AI-generated study guide derived from all of the above — five or six places where "the same" instructions can drift out of sync, each maintained by a different tool with no shared awareness that the others exist. Distributed and asynchronous teams compound the problem further: when a change is made by someone in one time zone, there is often no single moment where everyone who touches a downstream copy is in the room to be told to update theirs.
Generative AI tooling introduces a version of this risk the maxim couldn't have anticipated in 2018 but describes with uncanny precision anyway: an AI system prompted to "summarize the assignment instructions" or "draft a reminder email" will happily produce a new, fluent-sounding copy of instructions pulled from whatever source document it was given — instantly creating one more place those instructions live, one more place that will silently fall out of date the next time the canonical version changes, and one that looks exactly as authoritative as the original because AI-generated prose carries no visible marker of its own staleness. The maxim's discipline — one canonical source, everything else a pointer to it — is arguably more load-bearing now than in 2018, precisely because the tooling that makes duplication effortless has outpaced any tooling that would flag the duplication as a liability.
The same dynamic plays out beyond the classroom, in any workplace that documents a process more than once. A company handbook that restates a vacation policy already covered in the HR system, a project wiki that copies an API contract also defined in the code itself, a customer-facing FAQ that repeats a return policy the terms-of-service page states independently — each is a version of the identical structural bet: that someone will remember to update every copy when the underlying rule changes, and will remember it correctly every time, indefinitely, across however many years the document survives and however many different people take over maintaining it. The maxim's warning scales with the number of hands the document passes through; a rule two people maintain from memory is fragile in a way a rule maintained by one canonical source and a set of references is not.
Impact and Legacy
As a private aphorism rather than a published statement, the Second Rule's legacy lives entirely inside the small numbered set it belongs to. Read alongside the First Rule (correctness of the sentence) and the Third Rule ("good context is key"), it occupies the middle ground between the two: more structural than proofreading, more concrete than the abstraction of "context." Where the First Rule assumes a document is otherwise sound and asks only that its prose be checked, the Second Rule assumes the prose might be perfectly correct at the moment of writing and asks a harder question: will it stay correct once it exists in more than one place? That question has no equivalent inside the First Rule at all — a lone, well-proofread document can't drift from itself. The Second Rule is the one member of the set that names a failure mode specific to documents that get revised after they ship, which in instructional design is essentially all of them. Its influence, like its siblings, is scoped to the practical standard Mike applied to his own materials and whatever he reviewed, rather than to any wider public discourse — but the independent convergence with software engineering's DRY principle suggests the underlying insight is not idiosyncratic; it is a structural truth about maintained documents that different disciplines keep rediscovering under different names.
Contrasting Views or Controversies
The maxim's clearest limitation is that it doesn't distinguish administrative repetition from pedagogical repetition, and instructional design has a long, well-supported tradition of the latter. Retrieval practice and spaced repetition — repeating a core concept across a lesson, a review activity, and a later assessment — are among the most robustly evidence-backed techniques for improving retention; a designer who read this maxim too literally and stripped all repetition from a course would be discarding a genuinely effective teaching tool along with the genuine liability. The maxim, read carefully, targets a narrower target than "repetition" in general: procedural or administrative instructions — due dates, submission steps, formatting requirements — where two copies create only risk and no learning benefit, as opposed to conceptual content, where a certain amount of restatement is the mechanism by which learning actually happens. The maxim's brevity doesn't make that distinction explicit, which leaves room for a well-intentioned but overcorrected reading.
A second objection concerns resilience. Sometimes deliberate redundancy is a defensible design choice rather than an oversight — posting critical instructions in both the LMS and a follow-up email, for instance, so a learner who misses one channel still encounters the information through another. The maxim's fix should arguably be "one authoritative source of truth, referenced or mirrored elsewhere" rather than "delete every copy but one" — a redundant channel for reaching the learner is different from a redundant, independently-editable copy of the content, but the maxim's short form collapses that distinction, and a designer applying it too rigidly could remove a genuinely useful safety net along with the genuine liability of drift.
A third, more structural point: the maxim locates the fix at the level of individual authoring discipline — remember not to copy-paste — when the more durable fix is tooling. Version-controlled single-source documents, transclusion, and content-management systems that let one canonical block populate multiple destinations automatically solve the problem the maxim describes without relying on any individual's memory to "update all the copies." Read as a personal habit rather than a systems argument, the maxim risks placing the burden of a structural failure mode onto individual vigilance, in exactly the way Deming's quality-management thinking warns against — the same critique the Third Rule's sibling maxim on incompetence draws elsewhere in this folder.
Practical Application
- Individual (designer/writer): Before pasting a block of instructions into a second document, ask whether a link or cross-reference to the canonical copy would serve the same purpose — treat every duplicated paragraph as a future maintenance debt, not a convenience.
- Individual (reviewer): When editing a set of instructions, actively search for other places the same information might live — a slide deck, an email template, a pinned message — rather than assuming the document in front of you is the only copy.
- Organizational (learning teams): Designate one canonical location for procedural information (a syllabus, an LMS page) and treat every other surface — slides, emails, chat pins — as a pointer to it, not an independent restatement, so a single edit propagates everywhere it needs to.
- Organizational (tooling): Where the platform allows it, use transclusion, shared content blocks, or a single-source content-management approach so "updating all the copies" is a system property, not a task that depends on someone remembering to do it by hand — the same instinct that led technical-writing and software teams to build docs-as-code and version-control workflows around exactly this failure mode.
- Civic/professional (any procedural documentation): Apply the same discipline to policy handbooks, onboarding packets, and compliance documents — anywhere a rule, deadline, or requirement is stated in more than one official document, a later amendment that updates only one of them creates exactly the kind of silent contradiction this maxim warns against, with higher stakes than a missed homework deadline.
- Individual (auditing your own materials): Periodically search your own body of work — old slide decks, saved email templates, a personal wiki — for a fact, date, or figure you know you've since revised, and check whether an earlier, uncorrected copy is still circulating somewhere you forgot about; the maxim's failure mode is silent precisely because the outdated copy never announces itself.
Background on the Author
Mike Hower's collected maxims read as accumulated field notes from a career spent in instructional design, and this one in particular reads as the kind of lesson that gets written down only after living through its failure mode directly — watching, or being told about, a learner who did exactly what one copy of the instructions said and was marked wrong against a different, updated copy elsewhere. That the Second Rule is, by creation timestamp, the very first of the three "Rules" set down that day gives it a plausible claim to being the freshest irritation of the sitting — the maxim someone reaches for first when something has just gone wrong in exactly this way. Filed alongside a First Rule about surface correctness and a Third Rule about context, it supplies the set's structural middle term: not just "get the words right" and not just "give enough context," but "make sure the words you got right stay right everywhere they live" — a distinction that reads as earned through direct experience with materials that had to be revised after they shipped, rather than derived from abstract design theory.
The maxim's authority rests less on formal credential than on the accumulated evidence of the other entries in the same folder — "Mike on Incompetence," "Mike on Courseware Design," "Mike on Getting Things Done" — which together sketch someone who has spent enough time building, reviewing, and revising instructional materials to have developed strong, specific opinions about exactly where the work quietly breaks. A designer who has only ever authored a document once, and never had to go back into it months later to correct a due date or a rubric line, has no particular reason to have noticed this failure mode; the maxim reads as the kind of rule that gets written down only after living through the version where it wasn't followed.