A controlled vocabulary is a managed set of terms or labels used consistently to describe, organise and retrieve information. Instead of letting every contributor invent a different subject name for the same idea, the system defines which labels should be used, which alternatives should lead to them, and what each concept includes or excludes. An education library might group the search expressions Sec 1 Math, Secondary One Mathematics and Secondary 1 Maths under a carefully defined concept for the subject and stage—without claiming that every informal variation is an official curriculum term.
Why should vocabulary learners care? Because words do more than decorate essays or fill dictionaries. They function as addresses in a knowledge system. A consistent term can help a parent find the right lesson, a student discover a relevant explanation, a tutor compare exercises correctly and a research reader distinguish nearby ideas. When those labels are poorly managed, a search may miss useful material, combine unrelated topics or make two different things appear identical.
This guide explains what a controlled vocabulary is, how concepts differ from labels, why preferred and alternative terms matter, and how broader, narrower and related relationships improve discovery. It uses original school, library and study examples, practical worksheets and a small vocabulary-governance procedure. The reference foundation includes W3C SKOS and Dublin Core, both of which show how carefully labelled concepts can support interoperable information. A controlled vocabulary can improve the organisation of knowledge; it does not automatically make a source accurate or a search system intelligent.
Editorial note: All school catalogues, example search results and concept identifiers are fictional illustrations of information organisation. No invented label is represented as an official education authority’s terminology.
Contents: from word variations to searchable concepts
Open all 54 chapters, examples and tests
- The simplest example: one idea, several search phrases
- From loose words to managed concepts
- A concept is not identical to its label
- Preferred labels make display consistent
- Alternative labels preserve how people actually search
- Hidden labels are a different design choice
- One preferred label per language is a SKOS integrity constraint
- Scope notes prevent false mergers
- The canonical label does not make a claim true
- A controlled vocabulary is not a dictionary
- A thesaurus is a structured kind of knowledge organisation system
- A taxonomy is not the whole story
- Broader does not always mean identical to superclass
- Related terms are not necessarily siblings
- The danger of false synonymy
- Same spelling can require two concepts
- Different spelling can be the same local concept
- Official names and informal aliases need different status
- Controlled vocabulary versus free tagging
- User language should still inform the scheme
- A vocabulary entry needs a stable identifier
- Stable identity and changing labels are not contradictions
- Version control matters for subject indexes
- Governance needs an owner and a correction path
- How to propose a new concept
- How to identify a collision
- A hub is a navigation service, not a synonym pile
- Internal links should have clear destinations
- A thesaurus relation is not a search ranking guarantee
- A small educational scheme in five fields
- Worked case: Primary and Secondary vocabulary routes
- Worked case: a word used as two subject labels
- Worked case: two apparently similar Science topics
- Worked case: the wrong synonym
- Worked case: a missing alias
- Worked case: conflicting official and informal terms
- A search test must include successes and failures
- Precision and recall are a useful trade-off
- A result count is not proof of search quality
- Controlled vocabulary and faceted classification
- Controlled vocabulary and knowledge graphs
- Cross-language labels need careful checking
- Mapping between two vocabularies is a separate task
- The difference between label mapping and curriculum equivalence
- Who should change the meaning of a term?
- How a vocabulary becomes outdated
- Preserve historical names without pretending they are current
- Editorial review is a form of quality control
- A minimal maintenance log
- Learning activity: build ten concepts
- Diagnostic: seven common vocabulary-management errors
- Twelve questions with short answers
- Frequently asked questions
- Primary references and one useful next action
1. The simplest example: one idea, several search phrases
Imagine two pupils searching a fictional class library. One enters ‘Sec 1 Maths’ and another enters ‘Secondary 1 Mathematics’. Both want the same course-level subject collection under the library’s chosen scope.
If the system treats those spellings as unrelated subject tags, one pupil might miss half the material. A controlled vocabulary can identify one preferred concept label and record common alternatives as retrieval routes.
The important distinction is that the alternatives are connected to the same deliberately scoped concept. A casual synonym list without an agreed concept boundary would not by itself guarantee good search results.
2. From loose words to managed concepts
A vocabulary list contains words. A controlled vocabulary adds rules about what the terms mean for a particular domain, how they are selected and how their relationships are maintained.
The W3C SKOS Reference provides a standard model for concept schemes with preferred labels, alternative labels, scope notes and semantic relations. It is especially relevant to thesauri, classification schemes, taxonomies and subject headings.
A school can use those ideas informally without claiming that a spreadsheet is a full SKOS implementation. The practical aim is stable meaning that several users can understand.
3. A concept is not identical to its label
A concept is the idea or category that a scheme intends to represent. A label is a human-readable expression associated with it. Several labels can identify the same concept in different languages or common forms.
Consider a class concept about fractions. The English word ‘fractions’ is a label; a Mandarin translation could be another language label, and ‘fraction numbers’ might be a local search synonym if readers use it, even if the preferred educational term remains fractions.
The scheme should not treat every spelling difference as a different topic. Nor should it merge different concepts just because their labels look similar.
4. Preferred labels make display consistent
A preferred label is the expression the system uses as the main display name for a concept. In SKOS, skos:prefLabel represents a preferred lexical label.
For an invented internal collection, the preferred label could be ‘Secondary 1 Mathematics’ because that is the agreed public-facing form for the library’s purpose. Users can search with familiar variations and still arrive at that entry.
This is a local display convention, not a declaration about current official nomenclature. A curriculum or assessment claim should separately use the relevant authoritative current source.
5. Alternative labels preserve how people actually search
People type abbreviations, local spellings, variants and familiar names. If a library only recognises the preferred label, it may hide material from readers who use ordinary alternatives.
SKOS includes skos:altLabel for alternative labels. A search interface can choose to use these labels for finding a concept while displaying the preferred one.
Alternative labels are not unlimited synonyms. Each should be reviewed to ensure it reliably points to the intended meaning in the specified context.
6. Hidden labels are a different design choice
Some spellings are useful for search but unsuitable for public display, perhaps because they are common misspellings or machine-oriented variants. SKOS includes skos:hiddenLabel for such lexical forms.
A hidden label can help a reader recover from a typing error without promoting that error as the recommended term. The specific search behaviour depends on how the site implements the vocabulary.
Do not assume that adding a hidden label to metadata automatically changes a website’s search engine. The indexing and query system must actually make use of it.
7. One preferred label per language is a SKOS integrity constraint
In a SKOS representation, the model specifies at most one preferred label for a given language tag on a resource. Two competing English preferred labels for the same concept would conflict with that integrity condition.
An editor can keep one preferred English display label and place other English alternatives under alternative labels, while also recording appropriate labels in other languages.
Language tags and policy need care: en, en-SG and other regional distinctions may be represented differently depending on the intended application. Consistency is a design responsibility, not merely an automatic outcome.
8. Scope notes prevent false mergers
The same expression can refer to different ideas. ‘Primary’ may mean a school stage, a primary source or something of first importance. A library should not merge these meanings under one concept just because the spelling matches.
A scope note states what a concept means within the scheme and, where helpful, what it excludes. SKOS offers skos:scopeNote as a documentation property.
Write a one-sentence scope note with a positive example and a plausible exclusion. This can be more valuable than adding dozens of alternative spellings to an ambiguous entry.
9. The canonical label does not make a claim true
A science article can have the correct tag ‘evaporation’ and still explain the process inaccurately. The tag helps a reader locate the article; factual validity depends on the content and evidence.
Similarly, a local label ‘Approved’ does not itself establish that an authorised review occurred. Labels and status records should agree with real events.
Teach two questions: ‘Is this record classified under the right concept?’ and ‘Is the claim within the record justified?’ They need different checks.
10. A controlled vocabulary is not a dictionary
A dictionary primarily describes words and senses, their forms and patterns of use. A controlled vocabulary manages which terms represent concepts within a specific information system.
The dictionary may record several meanings of ‘bank’. A subject vocabulary might maintain separate concepts for financial institutions and riverbanks so that a search does not mix unrelated records.
The two tools complement each other. A good dictionary can inform scope notes, but the library still chooses which categories it needs.
11. A thesaurus is a structured kind of knowledge organisation system
In information science, a thesaurus can express relations among subject concepts, including preferred terms, equivalent search expressions and hierarchical or associative relationships.
This differs from a casual writer’s thesaurus that mainly offers possible synonyms. An indexing thesaurus needs carefully scoped concepts and dependable relations for retrieval.
SKOS can represent controlled vocabularies of different types; it does not require every small scheme to have the same level of complexity.
12. A taxonomy is not the whole story
A taxonomy often emphasises hierarchical organisation, such as a general subject with more specific subjects beneath it. A controlled vocabulary can also include preferred labels, synonyms, scope notes and related concepts that do not form a simple tree.
A school library could organise Mathematics, Algebra and Linear Equations hierarchically under an appropriate policy, while also relating Algebra to other subjects without pretending all relationships are parent–child links.
Do not use the word taxonomy as if it automatically supplies every facet, synonym, evidence source or mapping rule.
13. Broader does not always mean identical to superclass
SKOS offers skos:broader and skos:narrower relationships. They indicate hierarchical links in a knowledge organisation scheme, but a broader concept relation should not casually be interpreted as formal class subsumption in every application.
An everyday hierarchy such as ‘Mathematics → Algebra’ may support browsing, whereas a more formal ontology can express stricter logical constraints. The meaning of each relation must be appropriate to the scheme.
The practical question is ‘Will this parent label help a reader browse to the intended narrower topic?’ rather than ‘Can this arrow be drawn?’
14. Related terms are not necessarily siblings
Two concepts can be related without one being a subtype or part of the other. A school scheme might relate ‘Vocabulary Learning’ to ‘Reading Comprehension’ because the activities often interact.
SKOS offers skos:related for associative concept relations. The link should have a reason, not merely arise because the topics appear in the same classroom.
Label each relationship carefully: synonym, broader, narrower, related or part of an external mapping. A generic ‘connected to’ edge can become unhelpful when it hides the nature of the connection.
15. The danger of false synonymy
Suppose a staff member adds ‘percent’ as a synonym of ‘fraction’ because both appear in mathematics lessons. The concepts overlap in use but are not interchangeable.
A search for percent may legitimately return material about converting fractions to percentages, but it should not necessarily be treated as the same concept as a generic fraction.
Use an associative relationship or a mapping to a specific conversion lesson instead of collapsing distinct ideas into one entry.
16. Same spelling can require two concepts
The word ‘draft’ can mean an early version of writing or a current of air, among other senses. A controlled vocabulary serving mixed subjects may need separate scoped concepts.
A term’s spelling does not tell the search system which sense a reader intends. Context from the subject, page or question can help.
An index should avoid using one undifferentiated concept merely because its label appears in two domains.
17. Different spelling can be the same local concept
A parent searches ‘PSLE vocab’ and a teacher uses ‘PSLE vocabulary’. A local vocabulary-learning library may choose to route both expressions to a common topical entry if the intended scope is the same.
The scheme should preserve the formal meaning of PSLE and avoid merging all Primary 6 English material automatically under an examination concept.
An alias is safe when it directs a user to the intended topic without erasing the distinctions the curriculum or task requires.
18. Official names and informal aliases need different status
An organisation’s published qualification or school subject should be named according to its current official wording when presenting factual information. Common informal search forms can be retained as alternative labels.
An alias is a reader-access route, not an endorsement that the informal term is officially defined or equivalent in every policy context.
Review official definitions before using a local index to answer questions about eligibility, examination rules or course names.
19. Controlled vocabulary versus free tagging
Free tagging lets contributors create terms as they work. This can capture emerging language quickly, but it also produces inconsistent spellings and categories.
Controlled tagging reduces uncontrolled variation by offering a managed set of concepts. It can slow down recording new ideas unless there is a reasonable request and review process.
A practical library may combine both: a trusted subject field for navigation and an optional free note for details that do not yet fit.
20. User language should still inform the scheme
A controlled vocabulary does not mean administrators should ignore the words readers actually type. Search logs, questions and feedback can reveal missing aliases or confusing labels.
Any analysis of user queries should follow appropriate privacy and data-handling requirements. Patterns, not private personal details, are usually sufficient to improve labels.
A useful addition is evidence-led: readers repeatedly search for a legitimate variant, and the editor confirms it belongs to the existing concept.
21. A vocabulary entry needs a stable identifier
A label can change over time: a department may rename a subject collection, or a clearer name may replace an ambiguous one. A stable internal identifier can let systems keep referring to the same concept across a label change.
SKOS concepts are resources with identifiers, often IRIs in web applications. The human-readable name is a property of the concept rather than its sole identity.
A spreadsheet can begin with a local concept ID, but a deployed cross-system scheme needs deliberate identifier policy. Do not reuse retired identifiers for unrelated meanings.
22. Stable identity and changing labels are not contradictions
A concept can retain its underlying scope while the preferred display label changes to improve clarity. The old label can sometimes remain an alternative for search.
But if the definition itself changes significantly, a new concept or carefully documented version may be needed. Silent changes can break historical comparisons.
Keep identifier, current label, past label, scope note, review date and change reason distinguishable in the governance record.
23. Version control matters for subject indexes
Imagine a school resource index at the beginning of the year. Later, the school changes the name of one local learning pathway. Historic records should not be retroactively rewritten as though the newer label always applied.
A reviewed change note can say when the preferred label changed and whether the underlying concept remained the same. The public system can redirect an obsolete alias appropriately.
The State Transition guide explains why Draft, Reviewed and Published revisions should remain distinct events.
24. Governance needs an owner and a correction path
A controlled vocabulary is not truly controlled if any editor can silently change meanings that many articles rely upon. Someone should be responsible for checking collisions, reviewing proposals and recording approved updates.
That governance can be lightweight for a small tuition resource library. A simple entry may record Proposed Term, Existing Concept, Evidence, Decision and Effective Date.
The point is traceability and consistency, not unnecessary bureaucracy. A vocabulary becomes more useful when users know how to report a mistaken alias.
25. How to propose a new concept
Start with the learner’s question and the intended scope. Search existing entries before inventing a new category. If an existing concept covers the need, consider adding an alternative label or a narrower guide.
If the concept is genuinely distinct, write a plain-language definition, two examples, one exclusion and the intended broader or related links.
Review the new entry against near neighbours. A new title without a meaningful new scope can create an SEO collision and an indexing problem.
26. How to identify a collision
Two entries may compete for the same search intent even when their titles differ. For example, ‘What Is an Equation?’ and ‘Understanding Equations: What They Are’ could duplicate content for readers if they answer the same question at the same level.
Conversely, ‘Solving Linear Equations’ and ‘Why a Linear Equation Can Have No Solution’ can be distinct when their scope and exercises differ.
Compare the intended user question, central concept, level and unique teaching outcome. Do not decide collisions solely by checking whether the exact slug exists.
27. A hub is a navigation service, not a synonym pile
A learning hub can introduce the main pathways and point to detailed articles. The page’s job differs from the detailed article’s job: help readers choose the right topic before they read the full explanation.
A controlled vocabulary can supply the labels used by the hub, while the content architecture adds useful links and summaries.
Do not make every page repeat the entire hub. Stable concise navigation and distinct article purpose often serve readers better.
28. Internal links should have clear destinations
A link labelled ‘Mathematics vocabulary’ should lead to an appropriate topical route, not a random unrelated page containing that phrase once.
The displayed anchor text can be reader-friendly, while the concept identity is managed behind the scenes. When a link’s destination changes, the editor should verify it.
Internal-link quality is a practical expression of controlled meaning. A correct label attached to the wrong URL is a navigation error.
29. A thesaurus relation is not a search ranking guarantee
Adding an alternative label can help a properly configured system match more queries. It does not automatically improve visibility in every search engine or guarantee a particular ranking.
Algorithms, indexing rules, page quality and competing results still determine discovery. A vocabulary project should measure actual retrieval behaviour rather than promise traffic from metadata alone.
Treat better terminology as one part of making a collection understandable and findable.
30. A small educational scheme in five fields
Begin with Concept ID, Preferred Label, Alternative Labels, Scope Note and Relationships. Add a review date and status if the scheme is shared or frequently changed.
Example: a local concept labelled ‘Fractions’ may include an alternative familiar phrase, a note distinguishing fractions as numbers or representations, and a narrower link to ‘Equivalent Fractions’ if the scheme uses that hierarchy.
The number of fields is a teaching starting point, not a universal standard. A multilingual or public dataset may require language tags, identifiers, mapping history and editorial attribution.
31. Worked case: Primary and Secondary vocabulary routes
A fictional library has four documents: one about Primary 5 descriptive writing, one about Primary 6 comprehension, one about Secondary 1 algebra and one about vocabulary learning strategies for adults.
An editor labels all four simply ‘vocabulary’. The label is not false as a broad topic, but it is too coarse to help a parent find the relevant school stage and task.
A controlled vocabulary can preserve ‘vocabulary’ as a broad hub concept while distinguishing level, subject and specific learning outcome. It should not invent curriculum rules or mix unrelated age and assessment categories.
32. Worked case: a word used as two subject labels
A museum-style educational library stores ‘scale’ under Mathematics, Geography and Music. The word can refer to ratios, map relationships or musical organisation.
If the index merges the senses, a student may open irrelevant records. Split or scope the concepts appropriately, then add context-sensitive labels or subject facets.
The word form is identical, but the reader’s intended concept is not. Controlling the vocabulary improves precision, not merely consistency of spelling.
33. Worked case: two apparently similar Science topics
A school index includes ‘Evaporation’ and ‘Boiling’. Both involve liquid changing to gas, but their processes and conditions are not identical.
An editor should not combine them under one preferred concept solely to reduce the number of labels. Instead use distinct concepts and an appropriate broader relation to change of state or a related connection.
The categorisation should preserve scientifically important differences rather than flatten the subject into a single everyday phrase.
34. Worked case: the wrong synonym
An editor adds ‘must’ as an alternative label of ‘recommended’ because both appear in assignment instructions. This creates a dangerous conceptual collapse.
An instruction that recommends extra reading differs from an obligation to complete it. The scheme should keep those concepts distinct and record a semantic relation only if useful to the domain.
See Semantic Features for how a small difference in meaning changes the correct action.
35. Worked case: a missing alias
A parent searches a common abbreviation for a study level; the collection uses only the full display phrase. The search returns nothing although relevant articles exist.
An editor checks that the abbreviation has the intended meaning in the current scope, records it as an alternative label and tests that the query retrieves the expected concept.
The change improves access without duplicating content or inventing another nearly identical article.
36. Worked case: conflicting official and informal terms
A public course description uses an official current name, while older enquiries use a former or informal label. The system can keep the official name preferred and record a historical alias with suitable time or scope notes.
The editor must not claim that requirements remain unchanged just because the old term redirects to the current page. Qualification and syllabus rules require separate verification.
A controlled vocabulary protects both accessibility and the integrity of a real-world name.
37. A search test must include successes and failures
Test with the preferred label, one common alternative, one misspelling, one neighbouring concept and one ambiguous homonym.
The successful tests should retrieve the intended content; the negative tests should show whether the system wrongly merges different subjects. A vocabulary that retrieves everything is not necessarily good at finding the right thing.
Record what the actual search system does. Do not claim the index works simply because the labels look organised on a spreadsheet.
38. Precision and recall are a useful trade-off
In information retrieval, precision concerns how many retrieved results are relevant, while recall concerns how many relevant items were retrieved. A broad ambiguous synonym can increase the number of results while worsening relevance.
A narrow preferred label may miss common reader queries. A controlled vocabulary can help navigate the trade-off through careful alternative labels and context.
These are conceptual retrieval measures; a precise numerical evaluation requires a defined corpus and relevance judgements, not a single anecdotal search.
39. A result count is not proof of search quality
An index returns a hundred pages for ‘algebra’. The large number alone does not tell us whether the first ten answer the student’s particular question.
The learner may need subtopic labels, level filters and a clear definition of what each page covers. Quantity without useful distinctions can create another obstacle.
A controlled vocabulary should lead to answers that match the reader’s intent, not merely a higher displayed count.
40. Controlled vocabulary and faceted classification
A term scheme manages concept labels and relations. Faceted classification adds independent dimensions such as subject, age or stage, resource type and intended activity.
These work together but are not identical. A document can be about Fractions, intended for a particular learning stage and presented as a worked example without forcing all those properties into one long compound category.
The companion article on faceted classification will explore that multi-axis approach. Keeping these functions separate prevents a fragile hierarchy with thousands of mixed labels.
41. Controlled vocabulary and knowledge graphs
A vocabulary can supply stable concepts and predicates for a structured graph. A graph might connect one educational article to the concept Fractions and identify a prerequisite concept.
The graph adds machine-readable relationships between entities; the controlled vocabulary supplies agreed meanings for labels and links.
A label can be controlled without creating a knowledge graph, and a graph can exist without being well governed. Both require appropriate identity and semantic rules.
42. Cross-language labels need careful checking
A concept may have labels in English, Mandarin, Tamil or another language, but languages do not necessarily divide meaning into exactly the same categories.
Do not automatically declare two words exact equivalents because a machine translation suggests them. Check the intended concept, specialist usage and any important lexical gap.
The Lexical Gap guide explains why meaning transfer sometimes needs descriptive phrases.
43. Mapping between two vocabularies is a separate task
Two institutions may use different preferred subject terms. SKOS provides mapping properties, such as exact match, close match, broad match and narrow match, to express connections between concepts in different schemes.
An exact match should not be asserted casually merely because two labels look alike. A close match is not a promise that every record has identical scope, and mapping should be reviewed when the sources change.
Maintaining a mapping is different from silently replacing another institution’s official terminology.
44. The difference between label mapping and curriculum equivalence
Two educational systems may teach similar concepts at different stages or assess them differently. Linking related Mathematics topics for navigation does not establish official grade or qualification equivalence.
The mapping should state what is shared and what remains different. A reader seeking progression or admissions advice needs authoritative current requirements.
Use a controlled concept scheme to find the material, then separately validate claims about educational policy and eligibility.
45. Who should change the meaning of a term?
A local school project may legitimately choose its own label for a revision stage. A national assessment term belongs to a different authority. The editor should distinguish internal governance from official nomenclature.
Metalinguistic negotiation can help groups discuss which local label would work better, but not override an external standard by mutual preference.
See Metalinguistic Negotiation for that distinction.
46. How a vocabulary becomes outdated
Over time, readers may adopt new terms, curricula may change, and some subject labels may become ambiguous or misleading.
A useful governance schedule reviews high-impact entries and checks whether their scope notes and mappings remain accurate. It should prioritise terms affecting action or official requirements.
Do not automatically delete old entries. Preserve appropriate aliases and change notes where they help users interpret historic material.
47. Preserve historical names without pretending they are current
A past article may use the subject’s historical name correctly for its period. Replacing every occurrence with today’s label could obscure the record.
A controlled vocabulary can route historical search expressions to an entry that explains the name change and applicable dates.
The editor should distinguish historical terminology, current preferred terminology and genuine changes in curriculum content.
48. Editorial review is a form of quality control
When an editor proposes merging two terms, review the definitions, examples, exclusions and potential consequences for current articles.
If the merge would put unrelated content on one search results page or distort a subject meaning, reject it or choose a related-link relationship instead.
The review need not be elaborate, but it should be documented when other pages depend on the concepts.
49. A minimal maintenance log
Keep the concept identifier, old value, proposed value, evidence, approval decision, editor and effective date. Record whether the change affects labels, scope or relationships.
An internal log makes it possible to investigate why a reader suddenly sees different results for the same query. It also protects earlier content from accidental reinterpretation.
This is one kind of provenance information. The forthcoming provenance guide will show how sources, transformations and versions can be traced through a larger workflow.
50. Learning activity: build ten concepts
Choose a small topic, such as reading-comprehension skills or basic Mathematics representations. Write ten concepts with clear preferred labels and a sentence defining each.
Add one common alternative for two concepts, two narrower links, one related link and a plausible near-neighbour that must remain separate. Then invite a second reader to find a resource using informal wording.
Measure the observed navigation result. If the wrong concept is chosen, revise the labels or scope note and repeat. The exercise demonstrates practical information organisation rather than only word memorisation.
51. Diagnostic: seven common vocabulary-management errors
Error one: treating every spelling as a new concept. Error two: treating related concepts as synonyms. Error three: missing official-name changes. Error four: assigning an ambiguous label without a scope note.
Error five: giving one concept competing preferred labels in the same language under a SKOS policy. Error six: treating a correct tag as proof that the article is true. Error seven: claiming search improvements without checking the actual retrieval implementation.
Each calls for a different repair. An accurate diagnosis avoids mass deletion or careless rewriting of established entries.
52. Twelve questions with short answers
1. Is a controlled vocabulary simply a long word list? No. 2. Can one concept have alternative labels? Yes. 3. Does a preferred label have to match every reader’s search phrase? No. 4. Are related topics automatically synonyms? No.
5. Does SKOS include scope notes? Yes. 6. Do hierarchy and association mean the same thing? No. 7. Is a controlled tag proof of factual truth? No. 8. Can a historical alias remain useful? Yes.
9. Does a term change always need a new concept ID? Not necessarily; it depends on scope and identity. 10. Can two schemes be mapped? Yes, carefully. 11. Will labels automatically change search ranking? No. 12. What is the practical goal? Consistent, meaningful and inspectable routes to information.
53. Frequently asked questions
Is a controlled vocabulary different from a glossary? Yes. A glossary explains terms; a controlled vocabulary governs concept labels and their use in information organisation. Do small schools need formal SKOS files? Not necessarily; a simple reviewed table can begin the work.
Can teachers still use natural language? Yes. Alternate labels and plain definitions can preserve familiar wording while keeping the index consistent. What if a term is missing? Review whether it is an alias, a distinct concept or a misleading request.
Why is this part of vocabulary learning? It shows how words become stable addresses in a knowledge system that people and software can use to locate the right explanation.
54. Primary references and one useful next action
The W3C SKOS Reference defines concept labels, notes and relations in a standard knowledge-organisation model. Dublin Core Metadata Terms recommends controlled vocabularies for subject descriptions and defines the broader metadata fields used to identify resources. The Library of Congress Subject Headings linked-data service provides a public example of maintained subject authority concepts.
All class libraries, example identifiers and title decisions in this article are invented instructional cases. The standards do not guarantee any particular site’s search ranking or editorial correctness.
Choose a topic where learners use three names for the same idea. Select one preferred label, document appropriate alternatives, write a scope note and test whether an unfamiliar reader can find the intended resource without confusing a near-neighbour. That is vocabulary operating as information infrastructure.