Form Facet¶

Reserve one classification axis in a faceted catalog for genre, document type, or format — assigned independently of topic — so 'what kind of object is this' and 'what is it about' become separate, combinable filters.

Core Idea¶

A form facet is the classification axis in a faceted catalog reserved for genre, document type, or physical format — treated as a dimension of description that is assigned independently of topic, so that "what kind of object is this" varies orthogonally to "what is it about." A catalog that uses it can answer "show me everything about X" and "show me everything in the form of Y" as separate, combinable filters; without the form facet, those two questions must be merged into a single subject string. The structural commitments are a controlled form vocabulary (the closed list of genre and format values), a rule that form is assigned without reference to topical classification, and interface support for filtering by form either alone or jointly with other facets.

In LCSH and MARC practice, genre and form terms are carried in field 655 and in the LCSH genre/form headings vocabulary; OCLC FAST exposes them as a separate fast:genre/form facet. RDA distinguishes content type, media type, and carrier type as intersecting form-family axes. In digital library interfaces — HathiTrust, Europeana, DPLA — "format" appears as a top-level facet alongside subject, date, and language. The same design recurs in research data archives (ICPSR, Dryad, Figshare filter by dataset, codebook, software, preprint independently of topic) and in e-commerce (binding — hardcover, paperback, ebook — and media type — vinyl, CD, digital download — as facets independent of genre or artist). The canonical demonstration: a Library of Congress search for "civil war" filtered by the form facet "motion pictures" returns films about the Civil War; filtered by "diaries" it returns Civil War-era diaries; the topical axis is constant while the form facet selects the kind of object, a combination the precoordinated subject string cannot achieve without explicit enumeration of every topic-form pairing.

Structural Signature¶

Sig role-phrases:

the form axis — the single classification dimension reserved for what kind of object this is (genre, document type, carrier), held apart from what it is about
the controlled form vocabulary — a closed list of genre and format values (motion pictures, diaries, correspondence, datasets, treaties)
the independence-from-topic rule — form is assigned without reference to the topical classification, the orthogonality the joint filter depends on
the per-item assignment — every resource receives a form value from the vocabulary
the joint filterability — form is exposed as a search axis usable alone or in combination, so the topical axis can be held constant while the form facet selects the object kind
the topic×form coverage — separating the axes trades the precoordinated topic-form product (every pairing enumerated) for a finite form vocabulary times a finite topic vocabulary
the flattening locus — the form axis is precisely where an object's native genre can be forced into an ill-fitting foreign type

What It Is Not¶

Not a subject heading. The form facet records what kind of object this is (genre, document type, carrier), not what it is about. When "diaries" or "motion pictures" gets smuggled into a topical string, kind-of-thing is masquerading as aboutness; the facet exists precisely to pry those two apart so each can be filtered on its own.
Not the general form/content relation. form_and_content is the substrate-spanning relation between a work's shape and its substance; the form facet is the narrow operationalization of the form half as a controlled, jointly filterable axis inside a retrieval system. The facet presupposes a form vocabulary, an independence-from-topic rule, and a faceted interface — machinery the general relation does not carry.
Not the whole faceted vocabulary. It is a single axis (genre/format) within a faceted catalog, not the faceting design itself. Treating "form facet" as the entire orthogonal-description apparatus overstates it; its distinctive content is just that the axis pried loose is always kind of object.
Not a value derived from topic. Form is assigned without reference to topical classification — that independence is the orthogonality the joint filter depends on. If the form value is read off or implied by the subject, the axes are not orthogonal and "everything in the form of diaries, on any topic" stops being answerable.
Not a catalog that enumerates every topic-form pairing. Separating the axes trades the precoordinated topic×form product (every kind-of-object against every topic, enumerated) for a finite form vocabulary times a finite topic vocabulary; "films about the Civil War" is satisfied by holding topic constant while the form facet selects the object kind, with no composite built in advance.

Scope of Application¶

The form facet lives across the cataloging family — one substrate, faceted description of catalog resources, in which the axis pried loose is always kind of object; the general form/content relation and the orthogonal-axes pattern that recur further afield belong to the parents (form_and_content, paradigmatic_vs_syntagmatic_relations), not to this LIS axis.

Library cataloging — the LCSH genre/form vocabulary carried in MARC field 655 and OCLC FAST's fast:genre/form axis; RDA splits the family further into content type, media type, and carrier type.
Digital libraries — HathiTrust, Europeana, and DPLA expose "format" as a top-level facet beside subject, date, and language.
Research data archives — ICPSR, Dryad, and Figshare let users filter by resource type (dataset, codebook, software, preprint) independently of topic.
Museum and archives cataloging — object-type vocabularies (the Getty AAT object-type facet, ICA archival levels) separate the type-of-thing from what it depicts or documents.
E-commerce — binding (hardcover / paperback / ebook) and media type (vinyl / CD / digital) are facets assigned independently of book topic or music genre.

Clarity¶

Naming the form facet makes legible a distinction that a subject-only catalog silently fuses: what kind of object this is (genre, document type, carrier) versus what it is about. When both are forced into a single precoordinated subject string, "civil war" and "diaries" must be enumerated together as a compound heading, and the two questions a user actually has — everything about a topic, everything in a form — cannot be posed apart. Reserving an independent axis for form dissolves that conflation: the cataloger can now ask whether genre and format are being assigned without reference to topical classification, which is exactly the orthogonality the joint filter depends on, and the searcher gains the ability to hold one axis constant while sweeping the other. The confusion it clears is the one where "kind of thing" masquerades as a topic; the question it sharpens is "is form a separate, controlled axis in this catalog, or smuggled into the subject heading?"

The form facet also localizes a specific cost of the monolithic alternative. Without it, covering "primary-source video lectures on the French Revolution" requires the subject vocabulary to pre-enumerate every topic-form pairing — an explosion the facet trades for a finite form vocabulary times a finite topic vocabulary. Seeing the axis as its own dimension also clarifies where a foreign-category imposition can go wrong (form is one of the axes along which a catalog may flatten an object's native genre into an ill-fitting type), and reframes "how do we file this kind of object?" as the prior question "is the kind-of-object dimension orthogonal to subject here, and does it draw from a closed, navigable form vocabulary?"

Manages Complexity¶

The complexity a form facet tames is the cross-product of two descriptive questions a catalog otherwise has to answer together: what is this about and what kind of object is it. A subject-only catalog, lacking a separate axis for genre and format, can express "primary-source video lectures on the French Revolution" only by precoordinating the pairing into a compound subject string — and doing that for every kind-of-object against every topic means the heading vocabulary must enumerate the full topic×form product, an explosion that grows unmanageable exactly as either vocabulary grows. The form facet collapses that by factoring the product into its two terms: the cataloger maintains a finite, closed form vocabulary (motion pictures, diaries, correspondence, datasets, treaties) on its own axis and assigns it without reference to topical classification, so the expressible description space stays the full topic×form product while the artifact maintained is the sum — a topic vocabulary plus a form vocabulary, not their multiplication. What had to be enumerated is now read off as one selection per axis.

The qualitative behaviors the cataloger and searcher then read straight off the facet structure, rather than re-deriving per query, all follow from the single orthogonality commitment — that form is assigned independently of topic. Because the two axes vary independently, the answerability of the searcher's two distinct questions ("everything about the Civil War, in any form" and "everything in the form of diaries, on any topic") follows immediately, and the joint query ("films about the Civil War") is satisfied by holding the topical axis constant while the form facet selects the object kind — no topic-form composite need have been built in advance. The designer reads "can a user hold one axis constant and sweep the other?" directly off "is form drawn from a closed vocabulary and assigned without reference to subject?" The same structure localizes a characteristic failure without inspecting records one by one: when an object's native genre is forced into an ill-fitting type, the form facet is precisely the axis along which that flattening occurs, so the diagnosis ("the kind-of-object dimension is being collapsed") is read off the facet rather than hunted for in the data. The sprawling "how do we file every kind of object on every topic?" reduces to "is the kind-of-object dimension orthogonal to subject, drawn from a closed form vocabulary, and jointly filterable?"

Abstract Reasoning¶

The form facet's defining move is a separation diagnostic aimed at one specific conflation: what kind of object this is (genre, document type, carrier) masquerading as what it is about. The analyst reasons from the catalog's behavior to whether form is its own controlled axis or smuggled into the subject string — if "civil war" and "diaries" can only be enumerated together as a compound heading, form has been fused into topic; if they filter apart, form is a separate dimension. The operative question becomes "is the kind-of-object dimension orthogonal to subject here, drawn from a closed form vocabulary, or hidden in the subject heading?" The inference targets the genre/format axis in particular, not orthogonal description in general: the thing being pried loose is always "kind of thing" from "aboutness."

That separation grounds a predictive move about joint retrieval. Because form is assigned without reference to topical classification, the analyst predicts that the searcher's two distinct questions are independently answerable — "everything about the Civil War, in any form" and "everything in the form of diaries, on any topic" — and that the joint query is satisfied by holding the topical axis constant while the form facet selects the object kind, so "films about the Civil War" returns from filtering "civil war" by "motion pictures" with no topic-form composite built in advance. The reasoning runs from the orthogonality of form-to-topic to the existence of the cross-section, and it reads "can a user hold one axis constant and sweep the other?" directly off "is form drawn from a closed vocabulary and assigned independently of subject?" The cost-side corollary is a prediction about the monolithic alternative: a subject-only catalog must pre-enumerate every topic-form pairing (every kind-of-object against every topic), an explosion the facet trades for a finite form vocabulary times a finite topic vocabulary.

A failure-localizing diagnostic names where a specific pathology surfaces. When an object's native genre is forced into an ill-fitting type, the form facet is precisely the axis along which that flattening occurs, so from a misfiled object the analyst infers "the kind-of-object dimension is being collapsed into a foreign category" and locates the fault on the form axis rather than hunting through records — the form facet is one of the axes along which native-category flattening shows up. A boundary-drawing move fixes the concept's scope: the form facet is the single axis for resource genre/format within faceted catalogs, so its inferences presuppose a controlled form vocabulary, an independence-from-topic rule, and an interface exposing form as a joint filter — they apply across the cataloging family (library genre/form headings, digital-library format facets, e-commerce binding and media-type, research-data resource-type) and are distinct from the general orthogonal-description reasoning that belongs to the broader faceted family; here the axis is specifically what kind of object, held apart from what it is about.

Knowledge Transfer¶

Within the cataloging family the form facet transfers as mechanism: the separation diagnostic (is genre/format its own axis or smuggled into the subject?), the joint-retrieval prediction, and the closed-form-vocabulary requirement carry intact across every subfield that indexes resources by kind. In library cataloging it is the LCSH genre/form vocabulary in MARC 655 and OCLC FAST's fast:genre/form axis; RDA splits it further into content type, media type, and carrier type as an intersecting form-family. In digital libraries (HathiTrust, Europeana, DPLA) "format" is a top-level facet beside subject, date, and language. In research data archives (ICPSR, Dryad, Figshare) resource type — dataset, codebook, software, preprint — filters independently of topic. In e-commerce binding (hardcover/paperback/ebook) and media type (vinyl/CD/digital) are facets independent of genre or artist. A cataloger who learns to ask "is the kind-of-object dimension orthogonal to aboutness, and does it draw from a closed vocabulary?" carries that question and the native-category-flattening diagnosis without translation across all of these, because they are one substrate — faceted description of catalog resources.

Beyond that substrate the case is (B): the entry adds no portable mechanism of its own; what travels is its parent. The form facet is faceted classification applied to one specific axis (genre/format), and faceted classification in turn instantiates paradigmatic_vs_syntagmatic_relations (selection-axis versus combination-axis) and orthogonal decomposition. Those parents recur across substrates — separation of variables, basis selection, normalized schema design — but "form facet" itself does not float free of the catalog. Its near-neighbor form_and_content is the general structural relation between a work's shape and its substance, and that relation does travel across domains; the form facet is the narrow operationalization of the form half as a controlled, jointly filterable axis in a retrieval system, and that operationalization presupposes a form vocabulary, an independence-from-topic rule, and a faceted-search interface. Invoking "a form facet" for a non-catalog kind-versus-aboutness split renames the components and borrows the orthogonality idea while dropping the cataloging machinery — illuminating by resemblance, but the cross-domain lesson belongs to the form/content and orthogonal-axes parents, not to the LIS axis (see Structural Core vs. Domain Accent).

Examples¶

Canonical¶

The Library of Congress maintains Genre/Form Terms (LCGFT), a controlled vocabulary kept deliberately separate from its subject headings and recorded in MARC field 655 rather than the topical 650 field. Catalog a documentary about the American Civil War and it receives a topical heading — "United States—History—Civil War, 1861-1865" in a 650 field — and, independently, a genre/form term — "Documentary films" in a 655 field. A patron can then search "civil war" and filter by the form value "Motion pictures" to get films about the war, or filter the same topic by "Diaries" to get Civil War-era diaries. The topic stays fixed while the form axis selects the kind of object — a combination a single precoordinated subject string cannot produce without enumerating every topic-form pairing in advance.

Mapped back: The 655 genre/form field, held apart from the 650 topical field, is the form axis, and LCGFT is the controlled form vocabulary — a closed list. Assigning "Documentary films" without reference to the Civil War subject is the independence-from-topic rule. Searching a topic and filtering by form is the joint filterability, and avoiding a heading for every topic-form pair is exactly the topic×form coverage the factoring buys.

Applied / In Practice¶

Europeana, the European Union's aggregator of digitized cultural-heritage objects from thousands of libraries, museums, and archives, exposes a top-level "Type" facet — TEXT, IMAGE, SOUND, VIDEO, 3D — alongside facets for subject, provider, and date. A researcher studying the French Revolution can hold that topic constant and pull only the IMAGE objects (engravings, paintings) in one pass, then switch the same query to TEXT (pamphlets, letters) in another, without the aggregator having to precoordinate "French Revolution engravings" and "French Revolution pamphlets" as distinct subject strings. Because heterogeneous institutions contribute objects described in different local schemes, the independent media-type axis is precisely what lets a single interface filter by kind-of-object across all of them uniformly.

Mapped back: Europeana's "Type" facet is the form axis and its fixed value set (TEXT/IMAGE/SOUND/VIDEO/3D) is the controlled form vocabulary, assigned separately from subject — the independence-from-topic rule. Holding "French Revolution" fixed while sweeping media type is the joint filterability, and serving that combination without pre-enumerated topic-form strings is the topic×form coverage at aggregation scale.

Structural Tensions¶

T1: Orthogonality stipulated versus genres that carry topic (independence is imposed, not found). The joint filter depends on form being assigned without reference to topical classification — that orthogonality is what makes "everything in the form of diaries, on any topic" answerable. But independence is a design rule laid over a domain where kind and aboutness sometimes genuinely co-vary: a "field guide," a "case report," a "war diary," or a "screenplay" carries topical or content commitments in its very genre, so the stipulation that form says nothing about subject can misdescribe objects whose kind and aboutness are entangled. The tension is that the economy of the facet requires an independence the material does not always honor, and forcing the split cleanly can file a partly-topical genre as if it were topic-neutral. Diagnostic: For this genre value, is the kind-of-object assignment truly independent of the subject, or does the genre already imply part of what the object is about?

T2: Sum-not-product economy versus lost compound meaning (factoring cannot say a pairing is itself a category). Factoring the topic×form product into a topic vocabulary plus a form vocabulary is the facet's central bargain: maintain the sum, express the product, and never pre-enumerate every pairing. But precoordination could assert that a specific topic-form compound — "war diaries," "nature documentary," "field recording" — is itself an established genre with its own warrant and readership, a meaning the factored version can only reconstruct as a mechanical intersection of two independent selections. The tension is that the economy which frees the catalog from enumerating pairings also strips it of the ability to mark certain pairings as meaningful units, so a combination that a community treats as a single named kind is recoverable only as a query, never as a category. Diagnostic: Is the topic-form combination at issue a mere intersection the searcher assembles, or a recognized compound genre whose specific meaning the factored axes cannot record?

T3: Filterable closed vocabulary versus native-genre flattening (the same closedness cuts both ways). A controlled, closed form vocabulary is what makes the axis navigable and jointly filterable — a finite list the searcher can sweep. But that closed list is precisely the flattening locus the entry names: an object whose native genre is not in the vocabulary gets forced into an ill-fitting foreign type, and the coarser the list, the harder the misfit. Europeana's five buckets (TEXT/IMAGE/SOUND/VIDEO/3D) let thousands of heterogeneous institutions be searched uniformly by discarding the fine genre distinctions each local catalog maintains. The tension is intrinsic: the finiteness that makes the axis a usable filter is the finiteness that guarantees some objects' real kinds are flattened into it, so navigability and fidelity to native genre trade directly against each other. Diagnostic: Does the closed form vocabulary have a value that fits this object's native genre, or is it being forced into the nearest available type to stay on the axis?

T4: One clean axis versus form as an internally fractured family (the binary hides a bundle). The concept sells a clean binary — kind-of-object held apart from what-it-is-about — and insists the axis pried loose is always form. But form is not one axis: RDA already splits it into content type, media type, and carrier type as intersecting form-family axes, and a documentary on DVD is at once "moving image" (content), "video" (media), and "disc" (carrier), which come apart for a text read aloud or a film issued on several carriers. The tension is that treating form as a single facet under-describes objects whose content, media, and carrier diverge, while splitting it into its sub-axes reintroduces the multi-dimensional complexity the clean form-versus-topic story was meant to have dissolved. The privileged binary is itself a facet that needs faceting. Diagnostic: Does a single form value describe this object adequately, or do its content, media, and carrier diverge enough that one form axis collapses distinctions that matter?

T5: Autonomy versus reduction (its own LIS axis or the catalog instance of its parents). "Form facet" is a library-and-information-science construct with its own cargo — MARC 655, LCGFT, OCLC FAST's fast:genre/form, RDA's content/media/carrier split, the faceted-search interface. Yet it adds no portable mechanism of its own: it is faceted classification applied to one axis, and faceting instantiates paradigmatic_vs_syntagmatic_relations (selection-axis versus combination-axis) and orthogonal decomposition, while its near-neighbor form_and_content is the general, substrate-spanning relation between a work's shape and its substance. Those parents recur far afield — separation of variables, basis selection, normalized schema design — but "form facet" does not float free of the catalog; it presupposes a form vocabulary, an independence-from-topic rule, and a faceted interface. Invoking "a form facet" for a non-catalog kind-versus-aboutness split borrows the orthogonality idea while dropping that machinery. The tension is between a distinctive cataloging axis and the recognition that its portable content already belongs to its parents. Diagnostic: Resolve toward the parents (form/content, paradigmatic vs syntagmatic relations, orthogonal decomposition) when asking what travels beyond the catalog; toward the form facet when designing kind-of-object retrieval over catalog resources in situ.

Structural–Framed Character¶

The form facet sits at the framed-leaning position on the structural–framed spectrum: it is a cataloging design construct through and through — an axis inside a retrieval system, constituted by controlled vocabularies and a faceted interface — but one that names a description dimension rather than a verdict, which keeps it off the framed pole occupied by evaluative labels like ad hominem. Assessed criterion by criterion, the framing marks dominate. On evaluative_weight it points mildly structural: to call an axis a "form facet" is not to convict or approve anything — the term classifies a retrieval dimension (kind-of-object held apart from aboutness), a design choice about how a catalog is built, not a normative finding. On human_practice_bound it is emphatically framed in the strongest sense: the concept is constituted by a cataloging practice and dissolves the instant that practice is removed — the entry states plainly that it "presupposes a form vocabulary, an independence-from-topic rule, and a faceted-search interface," so strip away the catalog, the controlled vocabulary, and the filter interface and there is no form facet left, only objects that happen to differ in kind; nothing about it runs observer-free in nature. Institutional_origin is equally pronounced: the entry is an artifact of a specific disciplinary lineage — MARC field 655, the LCGFT vocabulary, OCLC FAST's fast:genre/form axis, RDA's content/media/carrier split, Europeana's Type facet — machinery invented inside library and information science, not a fact of nature someone merely named. On vocab_travels it scores framed: its operative vocabulary — genre/form heading, controlled form vocabulary, independence-from-topic rule, joint filterability, topic×form coverage — is pinned to the retrieval-system substrate and loses its referents off it. And on import_vs_recognize it patterns framed: within the cataloging family the mechanism is recognized intact across library headings, digital-library format facets, research-data resource types, and e-commerce binding, but beyond it the entry's own verdict is the strongest available — case (B), it "adds no portable mechanism of its own," so a non-catalog invocation "renames the components and borrows the orthogonality idea while dropping the cataloging machinery."

The genuinely portable structural skeleton is orthogonal decomposition — factoring a description space into independent axes so a product is expressed from a sum of vocabularies, one axis held constant while the other is swept. That skeleton travels as mechanism (separation of variables, basis selection, normalized schema design), which is what tempts a structural reading. But it does not lift the facet off the framed side, because orthogonal decomposition is precisely what the form facet instantiates from its parents — paradigmatic_vs_syntagmatic_relations (selection-axis versus combination-axis) and the general form_and_content relation between a work's shape and its substance — not what makes "form facet" itself travel: the cross-domain reach belongs entirely to those parents, and the entry is explicit that the facet contributes no portable content beyond them, only the LIS-specific specialization (the axis pried loose is always kind-of-object) plus the retrieval machinery that stays home. Its character: an evaluatively neutral but thoroughly cataloging-constituted description axis, structural only in the orthogonal-decomposition skeleton it wholly borrows from its form/content and paradigmatic-vs-syntagmatic parents, framed by the controlled-vocabulary-and-faceted-interface apparatus that pins it to the catalog.

Structural Core vs. Domain Accent¶

This section decides why the form facet is a domain-specific abstraction and not a prime — an unusually clean case, because by the entry's own verdict the facet adds no portable mechanism of its own; its entire cross-domain payload is its parents'.

What is skeletal (could lift toward a cross-domain prime). Strip the catalog and a thin relational structure survives: a description space is factored into independent axes assigned without reference to one another, so a product of descriptions is expressed from a sum of vocabularies, and one axis can be held constant while another is swept. The portable pieces are abstract — two orthogonal dimensions, per-item assignment on each, and joint selection across them. That skeleton is genuinely substrate-portable — it is separation of variables, basis selection, normalized schema design — which is why it recurs in the catalog as the parents the form facet instantiates: paradigmatic_vs_syntagmatic_relations (selection-axis versus combination-axis) and the general form_and_content relation between a work's shape and its substance. It is the core the form facet shares; it is not what makes the form facet distinctive.

What is domain-bound. Everything that makes it a form facet in particular is library-and-information-science furniture and none of it survives extraction: the form axis reserved specifically for kind of object (genre, document type, carrier); the controlled form vocabulary (motion pictures, diaries, correspondence, datasets, treaties); the independence-from-topic rule; the joint-filterability requirement; the topic×form coverage economy; and the concrete apparatus — MARC field 655, LCGFT, OCLC FAST's fast:genre/form, RDA's content/media/carrier split, Europeana's Type facet. The decisive test: remove the catalog, the controlled vocabulary, and the faceted-search interface and there is no form facet left, only objects that happen to differ in kind. The facet is constituted by the very retrieval-system machinery the prime bar asks it to shed, and its one specialization over the bare skeleton — that the axis pried loose is always kind-of-object rather than any orthogonal pair — is itself LIS-specific, not portable.

Why this does not clear the prime bar. A prime's vocabulary travels and its transfer is recognition of the same mechanism, not analogy. The form facet's transfer is bimodal. Within the cataloging family it travels intact — the separation diagnostic (is genre/format its own axis or smuggled into the subject?), the joint-retrieval prediction, the closed-form-vocabulary requirement, and the native-genre-flattening diagnosis all carry without translation across library genre/form headings, digital-library format facets, research-data resource types, and e-commerce binding, because they are one substrate: faceted description of catalog resources; that is recognition. Beyond the catalog the named axis does not travel: invoking "a form facet" for a non-catalog kind-versus-aboutness split renames the components and borrows the orthogonality idea while dropping the cataloging machinery — analogy, not the mechanism recognized anew. And — decisively — when the bare structural lesson is needed cross-domain, it is already carried, in more general and substrate-neutral form, by the parents: the shape/substance relation by form_and_content, the axis-independence by paradigmatic_vs_syntagmatic_relations and orthogonal decomposition, none of which require a controlled vocabulary or a faceted interface. The cross-domain reach belongs entirely to those parents; the named entry contributes only the LIS specialization plus the retrieval machinery, both of which should stay home. It is a genuine, useful cataloging axis in situ, but it has no substrate-spanning content of its own — which is exactly what keeps it below the prime bar.

Relationships to Other Abstractions¶

Current abstraction Form Facet Domain-specific

Parents (2) — more general patterns this builds on

Form Facet is a kind of Retrieval Facet Domain-specific

Form Facet is the genre, media, carrier, or kind-of-object species of Retrieval Facet.
Form Facet is a decomposition of Form and Content Prime

Form Facet operationalizes the form side of the general form-content relation as a controlled, independently filterable catalog axis.

Children (1) — more specific cases that build on this

Faceted Vocabulary Domain-specific is part of, conditional Form Facet

A Form Facet is an internal axis of a faceted vocabulary when object kind, genre, media, or carrier is independently represented.

Not to Be Confused With¶

Subject heading / topical facet. The axis that records what a resource is about — its topic. The form facet records what kind of object it is (genre, document type, carrier), held deliberately apart. When a genre term like "diaries" is smuggled into a topical string, kind-of-thing is masquerading as aboutness, which is exactly the fusion the form facet exists to undo. Tell: does the value answer "what is this about" (subject heading), or "what kind of object is this" (form facet)?
The general form/content relation (form_and_content). The substrate-spanning relation between a work's shape and its substance, which recurs across domains and is the form facet's near-neighbor and parent. The form facet is the narrow operationalization of the form half as a controlled, jointly filterable retrieval axis, presupposing a form vocabulary, an independence-from-topic rule, and a faceted interface. Tell: is it the general shape/substance relation that travels cross-domain (form_and_content), or the catalog axis bound to retrieval machinery (form facet)?
The faceted classification scheme itself. The whole orthogonal-description apparatus — several independent axes (subject, date, language, format) exposed as combinable filters. The form facet is a single axis within that scheme, not the faceting design as a whole. Its distinctive content is only that the axis pried loose is always kind-of-object. Tell: is the object the entire multi-axis faceted design (faceted classification), or the one axis reserved for genre/format (form facet)?
Content type / media type / carrier type (RDA's form-family sub-axes). The internal splits within the form family — a documentary on DVD is at once moving-image content, video media, and disc carrier, and these come apart for a text read aloud or a film reissued on several carriers. They are sub-axes that a single "form" value bundles together. Tell: does one kind-of-object value describe the item adequately (form facet as one axis), or do content, media, and carrier diverge enough to need separate axes (the RDA sub-axes)?
The orthogonal-decomposition parents (paradigmatic_vs_syntagmatic_relations, orthogonal decomposition). The substrate-neutral patterns the form facet wholly instantiates — factoring a description space into independent selection axes so a product is expressed from a sum of vocabularies, one axis held constant while another is swept. These recur as separation of variables, basis selection, and normalized schema design; the form facet adds only the LIS specialization (the axis is kind-of-object) plus retrieval machinery. Tell: strip away the catalog and what remains — independent axes jointly selectable — is carried by these parents, not by "form facet." (Treated fully in a later section.)

Neighborhood in Abstraction Space¶

Form Facet sits in a crowded region of the domain-specific corpus (33^rd percentile for distinctiveness): several abstractions share nearly its structure, so a description that fits it tends to fit its neighbors too.

Family — Faceted Classification & Metadata (6 abstractions)

Nearest neighbors

Topic Facet — 0.90
Geographic Facet — 0.88
Faceted Vocabulary — 0.86
Method-as-Metadata — 0.86
Preferred label — 0.84

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