Mooers's law¶

The retrieval-system regularity that a system tends not to be used whenever having the information is more troublesome than not having it — adoption governed by the per-query cost-benefit ratio against a substitute, not the content's absolute value.

Core Idea¶

Mooers's law (Calvin Mooers, 1959) states that an information retrieval system will tend not to be used whenever it is more painful and troublesome for a user to have the information than not to have it. The structural claim is that use of a retrieval system is governed not by the absolute value of the content but by the per-query cost-benefit ratio: when the cost the user pays to retrieve and process information — time, cognitive load, query-formulation effort, result-list triage, social cost of disclosure, risk of receiving unwelcome findings — exceeds the cost of doing without, the rational response is non-use. The system sits idle not because its content is poor but because its use is too expensive relative to the substitute.

Mooers framed the law as a counter-intuitive warning to system designers: enriching content or expanding coverage can decrease use if those additions raise query cost faster than per-query value. The binding constraint is almost always the use side, not the content side. This reframes the design problem from "what should the system know?" to "what does it cost the user to extract what the system knows?" — and the intervention space follows: lower retrieval cost (improve query ergonomics, reduce result-set overload, embed the system in the user's existing workflow), raise per-query value (better answer quality, fewer false positives), or accept that substitutes whose cost is already low (asking a colleague, guessing, doing without) will dominate as long as the retrieval cost remains above them. The law is the earliest explicit statement in information retrieval of what the subsequent information-seeking-behaviour literature formalised as satisficing against friction: users do not maximise total value of information; they satisfice each information need against the marginal cost of satisfying it.

Structural Signature¶

Sig role-phrases:

the user with an information need — the agent making a use-versus-not-use decision per query
the retrieval system — the mechanism holding content of some per-query value, independent of how cheaply that value can be extracted
the cost of use — what the user pays to retrieve and process: login, query formulation, result triage, cognitive load, social cost of disclosure, risk of unwelcome findings
the cost of non-use — what the user pays to skip: decision-quality loss, error, ignorance
the named substitute — the cheapest comparator the system competes against (asking a colleague, guessing, doing without)
the per-query inequality — the governing comparison: use is abandoned wherever cost-of-use exceeds cost-of-non-use against that substitute
the satisficing rule — users meet each need against the marginal cost of satisfying it rather than maximising total information value, so the equilibrium use-rate is the fraction of queries where use wins the inequality
the three exhaustive levers — read straight off the ratio: lower the retrieval cost, raise the per-query value, or shift the substitute's cost (the last usually counter-productive); the use side, not the content side, is almost always binding

What It Is Not¶

Not a claim about content quality. A system can hold excellent, well-curated content and still go unused. The law relocates the explanation to the use side — the per-query cost of extraction — so "the content is poor" is precisely the diagnosis it rules out as almost always non-binding.
Not a verdict on user discipline. Non-use is the rational response to a cost-of-use exceeding the substitute, not a failure of will to be fixed by training. Framing it as undisciplined users keeps attention on the user's character and misses the binding constraint, which is the retrieval cost the designer controls.
Not a "more is better" coverage principle. Enriching content or broadening coverage can lower use, because new material inflates result-list triage and query-formulation effort faster than it raises per-query value. The law predicts this backfire as a sign change in the ratio, not as an anomaly.
Not a deterministic law of nature. It is an empirical tendency about a per-query cost-benefit comparison, not a physical law that always fires. It states which way adoption leans when use-cost crosses non-use-cost against a named substitute; particular users and needs can sit either side of the threshold.
Not about diffusion or adoption curves. The law concerns individual per-query rationality — each query weighed against its substitute — not population-level spread, network effects, or tipping points. It explains the equilibrium use-rate from marginal decisions, not the trajectory of uptake over a population.
Not the general friction prime. Mooers's law is the information-retrieval instantiation of friction-as-adoption-killer, carrying a specific per-query, information-need, cost-versus-value apparatus. The substrate-neutral shape ("when the cost of an action exceeds the cost of inaction, the action tends not to be taken") belongs to friction, cost_benefit_analysis, and satisficing, not to this named law.

Scope of Application¶

Mooers's law lives across information retrieval and its close neighbours — library science and knowledge management — and, by clear lineage, the developer-experience and documentation subfields it has been re-applied to; its reach is within that family, since the law needs a user with an information need choosing per query whether to pay a retrieval cost against a named substitute. The looser invocations (gym memberships, public transit) are analogy carried by the parent friction / cost_benefit_analysis / satisficing primes, not by this per-query, information-need machinery.

Information retrieval and library science — the home turf: a foundational reference in IR pedagogy, cited to explain why high-quality intranets, document-management systems, and knowledge bases are abandoned despite good content.
Knowledge management — the structural account of why most enterprise KM initiatives decay: contribution cost plus retrieval cost together exceed the per-query benefit.
Software-tooling and developer experience (DX) — internal-developer-platform design and the persistence of git grep over heavyweight enterprise code search as Mooers's-law instances at scale.
Help-desk and documentation systems — why users ask a colleague rather than consult well-written docs: the asking cost sits below the documentation-search cost.
EHR and clinical-decision-support adoption — alert fatigue as Mooers's law with safety consequences: physicians click through alerts costly to dismiss or act on.
Low-uptake disclosure and compliance regimes — filings, disclosures, and compliance systems users bypass because the per-query extraction cost exceeds the substitute of doing without.

Clarity¶

The law's clarifying force is that it redirects the designer's first question when a retrieval system goes unused. The instinctive diagnoses — the content is poor, or users lack discipline and should be trained — both keep attention on the content side and the user's character. Mooers relocates the explanation to the use side: a system can hold excellent, well-curated content and still sit idle because each query costs more than the user is willing to pay relative to the substitute. That separates two failures a designer otherwise conflates — unused because the content is bad (a coverage problem, fixed by adding or improving content) from unused because retrieval is too expensive per query (an ergonomics problem, fixed by lowering cost) — and it identifies the second as almost always the binding constraint.

This makes a counter-intuitive prediction legible and therefore checkable: enriching content or broadening coverage can lower use, because the new material inflates result-list triage and query-formulation effort faster than it raises per-query value. The sharper question the practitioner can now ask is not "what should the system know?" but "what does it cost the user to extract what it already knows, and is that cost above or below the substitute of asking a colleague, guessing, or doing without?" Phrasing adoption as a per-query marginal cost-benefit ratio against a named substitute is what turns vague laments about a "low-uptake system" into a locatable design defect.

Manages Complexity¶

Across information retrieval and its neighbours, low-uptake systems present as a heterogeneous catalogue of failures, each seemingly demanding its own story: the abandoned enterprise intranet, the well-written documentation users bypass to ask a colleague, the clinical-decision-support alert physicians click through, the document-management system nobody logs into, the low-uptake disclosure regime. Explained one at a time, each invites a bespoke diagnosis — this content is stale, those users are undisciplined, that interface is clunky — and the field accumulates an open-ended list of reasons systems go unused. Mooers's law compresses the whole catalogue to a single inequality evaluated per query: does the cost of using the system exceed the cost of doing without, measured against the cheapest available substitute? Every entry in the list is then one instantiation of the same ratio crossing the same threshold, and the analyst stops collecting failure-stories and starts reading adoption off two quantities — the per-query retrieval cost (login, query formulation, result triage, cognitive load, regret risk) and the per-query value — relative to a named comparator (ask a colleague, guess, do without).

That collapse does two things at once. It turns the design question from the high-dimensional "what should this system know, and why are users ignoring it?" into a low-dimensional cost-versus-value comparison, with the binding constraint pre-identified as almost always the use side rather than the content side — so the analyst tracks the marginal cost of a query, not the total value of the corpus. And it sorts the entire intervention space into exactly three levers that follow directly from the inequality: lower the retrieval cost, raise the per-query value, or shift the substitute's cost. Because the levers are exhaustive and read straight off the ratio, the practitioner reasons from the same small parameter set to both the diagnosis (which side of the threshold a given system sits on, and why) and the qualitative remedy — and the law's signature counter-intuitive prediction, that enriching content can lower use by inflating triage cost faster than value, drops out of the ratio as a sign change rather than arriving as a surprise needing separate explanation.

Abstract Reasoning¶

Mooers's law licenses reasoning moves that all evaluate one inequality per query — does the cost of using the retrieval system exceed the cost of doing without, measured against the cheapest available substitute — and read adoption off it. The most distinctive is a diagnostic relocation move that redirects the first question when a system goes unused. The instinctive explanations — the content is poor, or users lack discipline and need training — both keep attention on the content side and the user's character; the law relocates the explanation to the use side, reasoning that a system can hold excellent, well-curated content and still sit idle because each query costs more than the user will pay relative to the substitute. From that the analyst separates two failures that "low-uptake system" fuses — unused because the content is bad (a coverage problem) versus unused because retrieval is too expensive per query (an ergonomics problem) — and predicts that the second is almost always the binding constraint, so the search for a cause starts on the use side rather than the content side.

The signature move is predictive and yields a counter-intuitive result as a sign change rather than a surprise: enriching content or broadening coverage can lower use, because new material inflates result-list triage and query-formulation effort faster than it raises per-query value. The analyst reasons this directly off the ratio — a change that raises the numerator (retrieval cost) faster than it raises per-query value pushes the inequality across its threshold toward non-use — so "add more content" is predicted to backfire whenever the addition's triage cost outruns its value, and the prediction is checkable by asking whether coverage growth raised query cost faster than per-query value.

A boundary-drawing move forces every adoption question to be phrased as a marginal comparison against a named substitute, not a judgement of total corpus value. The analyst reasons that users do not maximise the total value of information; they satisfice each need against the marginal cost of satisfying it, so the quantity to track is the per-query retrieval cost (login, query formulation, result triage, cognitive load, regret risk) against the per-query value, relative to a concrete comparator — asking a colleague, guessing, doing without. This converts a vague lament about uptake into a locatable defect: the analyst identifies which side of the threshold a given system sits on and which substitute it is losing to, rather than reasoning about whether the content is "worth it" in the abstract.

The interventionist move reasons forward from the inequality to exactly three exhaustive levers and predicts each one's effect and its failure mode. Lower the retrieval cost (improve query ergonomics, reduce result-set overload, embed the system in the user's existing workflow) to move the inequality toward use without touching content; raise the per-query value (better answer quality, fewer false positives) to the same end; or shift the substitute's cost — but here the analyst reasons that raising the substitute's cost (banning colleague questions, hiding contact info) is usually counter-productive, because it raises costs without raising value and tends to run backwards from designer intent. Because the levers read straight off the ratio and are exhaustive, the analyst reasons from the same small parameter set to both the diagnosis and the qualitative remedy, and predicts that an intervention which adds content without lowering retrieval cost will not move use. The recurring habit the concept installs is to ask of any unused retrieval system not "what should it know?" but what it costs the user to extract what it already knows, and whether that cost sits above or below the named substitute of asking a colleague, guessing, or doing without.

Knowledge Transfer¶

Within information retrieval and its close neighbours — library science, knowledge management — the law transfers as mechanism, and the transfer is literal across every retrieval substrate the field has built. The same per-query inequality (does the cost of using the system exceed the cost of doing without, against the cheapest substitute?), the same use-side-is-binding diagnosis, the same counter-intuitive prediction (enriching content can lower use by inflating triage cost faster than value), and the same three exhaustive levers (lower retrieval cost, raise per-query value, shift the substitute's cost) carry without translation across paper card catalogues, online databases, enterprise search, code search, and EHR retrieval, and into the recurring cases the literature treats as Mooers instances — abandoned intranets and document-management systems, bypassed-for-a-colleague documentation, clicked-through clinical-decision-support alerts (alert fatigue), low-uptake disclosure regimes. The vocabulary does not strain across these because they share the one thing the mechanism needs: a user with an information need choosing per query whether to pay a retrieval cost against a named substitute. That information-need framing is the substrate.

Beyond information retrieval the honest account is twofold. The named law — Mooers's law, with its per-query cost-benefit ratio and its information-need satisficing apparatus — does not transfer non-metaphorically; the looser invocations one hears (a gym membership, public transit, any low-use system "because use-cost exceeds non-use-cost") borrow the shape and drop the specific information-retrieval-cost-versus-value framing that gives the law its diagnostic bite, so those are analogy, case (A). But the general shape the law instantiates is a genuine shared abstract mechanism — case (B): when the marginal cost of an action exceeds the marginal cost of inaction, measured against a substitute, the action tends not to be taken. That shape really recurs across substrates and carries a real lesson, and it is exactly why the "friction kills adoption" finding has been independently re-derived outside IR — in UX research and in behavioural economics (Sunstein's "sludge"). Crucially, that shape is already in the catalog as primes: it is the conjunction of friction (the cost that suppresses the action), cost_benefit_analysis (the marginal comparison), and satisficing (the decision rule by which users meet a need against seeking-cost rather than maximising). So the honest move is that the cross-domain reach belongs to those parents; Mooers's law is the information-retrieval instantiation of friction-as-adoption-killer, and its per-query, information-need machinery does not, as named, float free of the IR substrate. It also sits in a cluster of eponymous IR/HCI design laws (Hicks's, Fitts's, Tesler's, Postel's), which is part of why it reads as a domain-specific abstraction rather than a prime (see Structural Core vs. Domain Accent).

Examples¶

Canonical¶

Calvin Mooers, who coined the term "information retrieval" in 1950, stated his law in a 1959 editorial in American Documentation, drawing on his experience building early retrieval systems (including his Zatocoding punched-card index). He had observed a pattern that puzzled system builders: organizations installed capable retrieval systems, stocked them with genuinely useful material, and then watched them go unused. His diagnosis was that use is governed not by the content's value but by the trouble of extracting it. If a researcher must walk to the index room, learn a query notation, sift a pile of returned cards, and risk turning up an unwelcome result that creates new work, while a colleague two desks away can answer in thirty seconds, the researcher asks the colleague. The system's excellent holdings are irrelevant to that per-query decision.

Mapped back: The researcher is the user with an information need; the stocked index is the retrieval system whose content value is beside the point. Walking over, learning the notation, and sifting cards are the cost of use; the nearby colleague is the named substitute. Choosing the colleague whenever using the index costs more is the per-query inequality resolved by the satisficing rule — meeting the need against marginal cost, not maximizing information value.

Applied / In Practice¶

Clinical decision-support "alert fatigue" is Mooers's law with patient-safety stakes. Electronic health records fire pop-up alerts — drug-interaction warnings, allergy flags, dosing cautions — each time a physician orders a medication. Studies consistently find that clinicians override the large majority of these alerts, dismissing them with a click. The reason is a per-query cost-benefit calculation: most alerts are low-specificity false positives, so the cost of stopping to read and adjudicate each one exceeds the expected benefit, and the substitute (dismiss and proceed on clinical judgment) is nearly free. Piling on more alerts — enriching "coverage" — worsens uptake, because it raises triage cost faster than value, exactly the law's counter-intuitive backfire. The corrective is to lower use-cost and raise per-alert value: suppress low-value alerts, tier by severity, and tune specificity so a firing alert is worth the interruption.

Mapped back: The prescribing physician is the user with an information need; the CDS alert stream is the retrieval system. Stopping to read and adjudicate each alert is the cost of use; clicking through on clinical judgment is the named substitute. Overriding whenever attending costs more than ignoring is the per-query inequality. Suppressing low-value alerts and raising specificity are two of the three exhaustive levers — lower retrieval cost, raise per-query value.

Structural Tensions¶

T1: Use-side relocation versus genuine content failure (the corrective can blind). The law's signature move is to relocate the explanation for non-use from the content side (bad or missing material) to the use side (per-query extraction cost), and to declare the use side "almost always" binding. That corrective is valuable precisely because the instinctive diagnosis ("add more content," "train the users") so often misses. But "almost always" is not always: sometimes the corpus genuinely lacks what the user needs, and a designer who has fully internalized Mooers can reflexively attribute every non-use to ergonomics and ignore a real coverage gap. The tension is that the heuristic earns its power by deprioritizing the content-side explanation, and that same deprioritization is what makes it blind to the minority of cases where the content really is the problem. Diagnostic: Is this system unused because extraction is too costly (use-side, the usual case), or because it genuinely lacks the needed content (a coverage failure the relocation would wrongly dismiss)?

T2: Comprehensiveness versus usability (the counter-intuitive backfire is a standing opposition). The law's sharpest prediction is that enriching content can lower use, because new material inflates result-list triage and query-formulation effort faster than it raises per-query value. This is not an anomaly but a standing opposition: past a point, comprehensiveness and usability pull against each other, because a more complete corpus is by construction a higher-triage corpus. The natural improvement path — "make it better by making it more complete" — is therefore self-defeating beyond the point where added coverage's triage cost outruns its value. The tension is that the two things a designer most wants (a system that knows everything and a system that is cheap to query) are in direct conflict, so every act of enrichment must be weighed against the retrieval cost it imposes rather than assumed to help. Diagnostic: Does this addition raise per-query value faster than it raises triage and query-formulation cost — or does completeness here buy coverage at the expense of the usability that drives adoption?

T3: Rational non-use versus the outcome that mattered (local rationality, global cost). The law reframes non-use as the rational response to a cost-of-use exceeding the substitute, not a failure of user discipline — a genuine and humane correction. But per-query rationality can produce outcomes that are wrong at the level that actually matters: the clinician who rationally clicks through low-specificity alerts eventually clicks through the one that would have prevented harm, and the researcher who rationally asks a colleague propagates an error the system would have caught. The tension is that the same satisficing that is individually rational per query can be collectively or consequentially hazardous, and the law's non-judgmental framing ("it's rational, don't blame the user") can slide into excusing a non-use whose rare failures carry outsized cost. Rationality against the marginal substitute is not the same as safety or accuracy on the case that counts. Diagnostic: Is the rational per-query non-use acceptable given the cost of its rare failures — or does the tail outcome (a missed safety alert, a propagated error) make locally-rational skipping globally unacceptable here?

T4: Improving the system versus blocking the substitute (the tempting lever runs backwards). The inequality yields three exhaustive levers, but they are not equal: lowering retrieval cost and raising per-query value both move the ratio toward use by making the system better, while the third — shifting the substitute's cost by banning colleague-questions, hiding contact info, or mandating the system — raises cost without raising value and usually runs backwards from designer intent. The tension is that this counterproductive lever is often the most tempting, because coercing users away from the workaround looks cheaper and faster than the hard ergonomic work of lowering retrieval cost. Reaching for it treats the symptom (users prefer the substitute) while worsening the underlying ratio, and typically breeds a new, costlier workaround. The lever exists structurally yet is almost always the wrong one. Diagnostic: Does the proposed fix make the system cheaper or more valuable to use, or does it merely make the substitute harder — raising total cost without raising value and inviting a new workaround?

T5: Autonomy versus reduction (IR law or the instance of friction-as-adoption-killer). Mooers's law is a named information-retrieval regularity with proprietary apparatus — the per-query cost-benefit ratio, the information-need satisficing framing, the named-substitute comparator — and across IR, library science, knowledge management, and developer experience it transfers as mechanism, because each has a user with an information need choosing per query whether to pay a retrieval cost. But its portable shape is generic: when the marginal cost of an action exceeds the marginal cost of inaction against a substitute, the action tends not to be taken — carried by friction (the suppressing cost), cost_benefit_analysis (the marginal comparison), and satisficing (the decision rule). That shape has been independently re-derived outside IR (UX "friction," Sunstein's "sludge"), and looser invocations (a gym membership, public transit) borrow it while dropping the information-retrieval-cost machinery — analogy, not transfer. The tension is between a named IR law (one of a cluster with Hicks's, Fitts's, Tesler's) and the substrate-general friction-adoption pattern it instantiates. Diagnostic: Resolve toward friction / cost_benefit_analysis / satisficing for any non-retrieval adoption question; reserve Mooers's law for a user with an information need weighing a per-query retrieval cost against a named substitute.

Structural–Framed Character¶

Mooers's law is mixed on the structural–framed spectrum — an evaluatively neutral empirical regularity with a clean, portable friction-adoption skeleton, but one constituted by human information-seeking practice, stated in information-retrieval vocabulary, and standing as one of a cluster of eponymous design laws, so it holds the middle. The criteria: on evaluative weight it reads structural — the law renders no verdict, and it is careful to reclassify non-use as the rational response to a cost inequality rather than a failure of user discipline, explicitly declining to blame either the content or the user's character. But human-practice-bound points framed: the regularity exists only where there are users with information needs choosing per query whether to pay a retrieval cost against a substitute — designed retrieval systems, query effort, workflows; it dissolves without that information-seeking practice and has no referent in observer-free nature. Institutional origin leans framed: it is an empirical tendency about human-built retrieval systems (Mooers, 1959), named as a "law" and sitting in a cluster of IR/HCI design laws (Hicks's, Fitts's, Tesler's, Postel's) — a regularity of a designed practice, not a fact of nature. Vocab-travels is low: per-query retrieval cost, information need, result-list triage, named substitute are IR idiom. Import-vs-recognize is bimodal: within IR, library science, knowledge management, and developer experience it transfers as mechanism (a user, an information need, a retrieval cost, a substitute — the same across card catalogues, enterprise search, and EHR alerts); beyond that (gym memberships, transit) it is analogy carried by the parent, and the finding has been independently re-derived outside IR (UX "friction," Sunstein's "sludge"), which is itself the sign that the portable content is the parent, not the named law.

The portable structural skeleton is a single one: friction-as-adoption-killer — when the marginal cost of an action exceeds the marginal cost of inaction, measured against a substitute, the action tends not to be taken. That shape genuinely recurs, but it is exactly what Mooers's law instantiates from its umbrella primes — friction (the cost that suppresses the action), cost_benefit_analysis (the marginal comparison), and satisficing (the decision rule by which users meet a need against seeking-cost rather than maximizing total value) — not what makes "Mooers's law" itself portable: the cross-domain reach belongs to that friction/cost-benefit/satisficing conjunction, while the domain-accented cargo — the per-query information-need framing, the retrieval-cost catalog, the named-substitute comparator, the three-lever design apparatus, the counter-intuitive content-enrichment backfire — stays home in information retrieval. Its character: an evaluatively neutral, non-blaming empirical regularity, structural in the friction-adoption skeleton it borrows from friction / cost_benefit_analysis / satisficing, but constituted by human information-seeking practice and stated in IR vocabulary as one named design law among a cluster, leaving it mixed rather than a free-floating prime.

Structural Core vs. Domain Accent¶

This section settles why Mooers's law is a domain-specific abstraction and not a prime.

What is skeletal (could lift toward a cross-domain prime). Strip the retrieval system away and a thin relational structure survives: when the marginal cost of an action exceeds the marginal cost of inaction, measured against a substitute, the action tends not to be taken. The portable pieces are abstract — an agent facing a per-instance choice, a cost of acting and a cost of abstaining, a cheapest available substitute the action competes against, and a decision rule that meets the need against marginal cost rather than maximizing total value. That skeleton is genuinely substrate-portable — it is why the same "friction kills adoption" finding was independently re-derived outside information retrieval, in UX research and in behavioural economics ("sludge") — which is exactly why the entry houses it in the umbrella primes Mooers's law instantiates: friction (the cost that suppresses the action), cost_benefit_analysis (the marginal comparison), and satisficing (the decision rule by which an agent meets a need against seeking-cost). But this is the core Mooers's law shares, not what makes it Mooers's law.

What is domain-bound. Everything that individuates the concept is information-retrieval furniture that does not survive extraction: the per-query framing (adoption decided one information need at a time); the user with an information need as the deciding agent; the retrieval-cost catalog (login, query formulation, result-list triage, cognitive load, social cost of disclosure, regret risk of unwelcome findings); the named-substitute comparator specific to information-seeking (ask a colleague, guess, do without); the three-lever design apparatus read off the ratio; and the counter-intuitive content-enrichment backfire (adding coverage lowers use by inflating triage cost faster than value). The decisive test: remove the information-need-plus-retrieval-cost setting and Mooers's law is no longer Mooers's law but the bare friction-adoption pattern — a gym membership or a transit system "unused because use-cost exceeds non-use-cost" keeps the shape while dropping the retrieval-cost-versus-per-query-value machinery that gives the law its diagnostic bite. What remains is the parent, not the IR-specific child.

Why this does not clear the prime bar. A prime's vocabulary travels and its transfer is recognition of the same mechanism, not analogy; Mooers's law's transfer is bimodal. Within information retrieval and its close neighbours — library science, knowledge management, developer experience — the mechanism travels intact across card catalogues, enterprise search, code search, and EHR alerts, because each supplies the one thing the mechanism needs: a user with an information need choosing per query whether to pay a retrieval cost against a named substitute, so the use-side-is-binding diagnosis, the enrichment backfire, and the three levers all carry without translation. Beyond that substrate the named law does not travel non-metaphorically: the looser invocations (a gym membership, public transit) borrow the shape while dropping the information-retrieval-cost framing, and the fact that the same finding was re-derived outside IR under different names ("friction," "sludge") is itself the sign that the portable content is the parent, not the named law. When that cross-domain adoption lesson is actually needed, it is already carried, in more general form, by the friction / cost_benefit_analysis / satisficing conjunction. The cross-domain reach belongs to those parents; the per-query information-need framing, the retrieval-cost catalog, the named-substitute comparator, and the content-enrichment backfire are the domain-accented cargo that — like the eponymous design laws it clusters with (Hicks's, Fitts's, Tesler's) — should stay home.

Relationships to Other Abstractions¶

Current abstraction Mooers's law Domain-specific

Parents (2) — more general patterns this builds on

Mooers's law is part of Cost–Benefit Analysis Prime

Mooers's Law contains Cost–Benefit Analysis because use is governed by a per-query comparison between the marginal trouble of retrieving and acting on information and the marginal trouble of using a substitute or doing without.
Mooers's law is part of Satisficing Prime

Mooers's Law contains Satisficing because the user selects the least troublesome option that meets the immediate need rather than maximizing the total informational value available.

Hierarchy paths (18) — routes to 9 parentless roots

Mooers's law → Cost–Benefit Analysis → Commensurability

Show alternative paths (17)

Not to Be Confused With¶

Moore's law. The empirical regularity that transistor density on integrated circuits roughly doubles every couple of years. It is confused with Mooers's law almost entirely by the near-identical name (Gordon Moore versus Calvin Mooers); the two share nothing structurally — Moore's law is an exponential-growth trend in hardware capability, Mooers's a per-query cost-benefit account of why retrieval systems go unused. Tell: is the claim about doubling hardware capacity over time (Moore's law), or about adoption governed by retrieval cost versus a substitute (Mooers's law)?
Zipf's principle of least effort. The generalization that people minimize probable work, yielding regularities like the Zipfian frequency distribution of word use. It is Mooers's law's closest conceptual kin — both invoke effort-economizing agents — but Zipf's is a broad statistical principle about behaviour-at-large, while Mooers's is the specific per-query, information-need account with its retrieval-cost catalog and named-substitute comparator. Tell: is the subject a general least-effort tendency and its statistical signatures (Zipf), or the specific per-query use-versus-not-use inequality for a retrieval system (Mooers)?
The eponymous IR/HCI design-law cluster (Hick's, Fitts's, Tesler's, Postel's laws). The family of named design regularities Mooers's law sits among — Hick's (decision time grows with number of choices), Fitts's (pointing time scales with distance and target size), Tesler's (conservation of complexity), Postel's (be liberal in what you accept). Each is a distinct interface/usability regularity; Mooers's is specifically about whether a retrieval system is used at all, given extraction cost against a substitute. Tell: does the law concern choice time, pointing, irreducible complexity, or protocol tolerance (the cluster), or adoption decided by per-query cost versus non-use (Mooers)?
Diffusion of innovations / adoption curves (Rogers). The population-level account of how an innovation spreads through a community over time (innovators, early adopters, laggards; S-curves, network effects, tipping points). Mooers's law concerns individual per-query rationality — each query weighed against its substitute — and explains the equilibrium use-rate from marginal decisions, not the trajectory of uptake across a population. Tell: is the question how adoption spreads over a population over time (diffusion), or why a given user does or doesn't use the system on this query (Mooers)?
Sunstein's "sludge" / UX "friction." The behavioural-economics and design accounts of excessive process-cost that suppresses beneficial action. These are independent re-derivations of the same friction-kills-adoption shape outside information retrieval — which the entry treats as evidence that the portable content is the parent, not Mooers's named law. They share the mechanism but lack (and do not need) the per-query retrieval-cost-versus-information-value apparatus. Tell: does the analysis carry the information-retrieval cost catalog and named substitute (Mooers's law), or the general process-friction-suppresses-action pattern re-found in another field (sludge/UX friction)?
Friction / cost_benefit_analysis / satisficing (the umbrella). The substrate-general skeleton Mooers's law instantiates — when the marginal cost of an action exceeds the marginal cost of inaction against a substitute, the action tends not to be taken. Looser invocations (a gym membership, public transit) borrow this shape while dropping the retrieval-cost machinery, so they are analogy carried by these parents. Tell: the umbrella (treated in a later section) is what travels to any adoption question; "Mooers's law" as named applies only where a user with an information need weighs a per-query retrieval cost against a substitute.

Neighborhood in Abstraction Space¶

Mooers's law sits in a sparse region of the domain-specific corpus (69^th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.

Family — Surface Form & Underlying Structure (23 abstractions)

Nearest neighbors

Computed from structural-signature embeddings · 2026-07-12