Part of the WebMEM® Protocol
Fragment Class: FAQFragment
Location: /protocol/fragments/faqfragment/
Status: Current Draft
Last Updated: 2026-08-24
Overview
An FAQFragment is a WebMEM fragment class for publishing structured question-and-answer knowledge.
Each FAQFragment contains a clearly defined question and an answer intended to address that question within a known semantic or contextual scope.
The fragment may also identify the entity, subject, defined terms, source material, provenance, and relationships necessary to interpret the question and answer correctly.
Unlike a DefinedTermFragment, which establishes the canonical meaning of a term, an FAQFragment uses a natural-language question-and-answer structure to explain a concept, address a common uncertainty, clarify a decision, or communicate how a policy, product, process, or other subject applies within a particular context.
The semantic purpose of the class remains the question-and-answer relationship itself.
Semantic Purpose
The semantic purpose of an FAQFragment is to state:
For this question, within this defined context, this is the publisher’s answer.
An FAQFragment may establish:
- the question being addressed;
- the answer to that question;
- the entity or context to which the question applies;
- the subject or domain of the question;
- defined terms referenced by the question or answer;
- the source or sources supporting the answer;
- the provenance of the answer;
- relationships to other knowledge objects;
- and other contextual information required to interpret the answer correctly.
This allows question-and-answer knowledge to remain explicitly structured without reducing it to either an isolated factual value or an unstructured block of prose.
FAQFragment Within an SDT
An FAQFragment exists as one modular knowledge object within a Semantic Data Template (SDT).
The relationship is:
Web Resource → SDT → FAQFragment → Question + Answer
An SDT may contain one FAQFragment or multiple FAQFragments depending on the question-and-answer knowledge associated with the resource.
Multiple FAQFragments may address different questions about the same entity or context while sharing common identity, provenance, terminology, or source information established elsewhere within the SDT.
The FAQFragment does not replace those other knowledge objects. It represents the specific semantic relationship between a question and its answer.
FAQFragment Requirements
A conforming FAQFragment should establish enough information for the question, answer, and applicable context to be interpreted correctly.
| Element | Purpose |
|---|---|
| Fragment Class | Identifies the knowledge object as an FAQFragment. |
| Fragment ID | Provides a stable identifier for the FAQFragment within the published knowledge structure. |
| Question | States the question addressed by the fragment. |
| Answer | Provides the publisher’s answer to the question. |
| Context | Establishes the entity, subject, domain, location, audience, time period, or other scope required to interpret the question and answer when applicable. |
| Provenance | Identifies or references the source and lineage supporting the answer when evidentiary support is applicable. |
Additional information may be included when required by the subject, domain, entity model, terminology, relationship model, or applicable conformance profile.
Question
The question establishes the information need addressed by the FAQFragment.
A question should be sufficiently specific that the intended subject can be understood within the fragment’s declared context.
For example:
What does a $0 premium Medicare Advantage plan mean?
The question itself may contain terminology that is defined elsewhere in the SDT. Where explicit semantic alignment is useful, the FAQFragment may reference the applicable DefinedTermFragment or other knowledge object rather than embedding a separate definition into the question.
Answer
The answer provides the publisher’s response to the question.
An answer may contain:
- direct factual statements;
- explanatory language;
- qualifications or conditions;
- references to policies or procedures;
- references to defined terms;
- references to other entities or resources;
- or combinations of these forms of knowledge.
The answer should remain scoped to the question and should preserve any context necessary to prevent the response from being interpreted more broadly than intended.
When the answer depends on authoritative facts, source data, policies, or other evidence, provenance should identify or reference that evidentiary basis.
Entity and Context Scope
An FAQFragment may be general or scoped to a specific entity or context.
Context may include:
- a product or plan;
- a geographic area;
- an organization;
- a policy;
- a procedure;
- a defined term;
- a user or audience context;
- a time period;
- or another subject represented within the SDT.
For example, the question:
How much is the monthly premium?
cannot be interpreted correctly without knowing which plan, product, or other entity the question concerns.
WebMEM therefore treats applicable context as part of the question-and-answer knowledge object rather than assuming a machine will reconstruct that context from surrounding page content.
Example FAQFragment
The following example represents a question and answer associated with Medicare Advantage plan premiums using the current WebMEM HTML serialization.
<template
data-webmem-fragment
data-fragment-class="FAQFragment"
data-fragment-id="faq-zero-premium">
<section
data-entity-type="MedicareAdvantage"
data-subject="monthly-premium"
data-provenance-ref="#provenance-cms">
<h3 data-role="question">
What does a $0 premium Medicare Advantage plan mean?
</h3>
<div data-role="answer">
<p>
A $0 premium Medicare Advantage plan does not charge an
additional monthly plan premium. Members must continue to
pay any Medicare Part B premium they owe, and other plan
costs such as deductibles, copayments, or coinsurance may
still apply.
</p>
</div>
</section>
</template>
The fragment establishes that:
- the knowledge object is an
FAQFragment; - the fragment has a stable identity;
- the question concerns Medicare Advantage monthly premiums;
- the question is explicitly distinguished from the answer;
- the answer remains associated with the question and its declared context;
- and provenance supporting the answer can be resolved through the referenced provenance object.
FAQFragment and DefinedTermFragment
FAQFragment and DefinedTermFragment may address similar subjects but serve different semantic purposes.
| DefinedTermFragment | FAQFragment |
|---|---|
| Establishes the canonical meaning of a term. | Answers a specific natural-language question. |
| Answers: What does this term mean? | Answers: What is the response to this question? |
| Primarily establishes semantic definition. | Primarily establishes a question-and-answer relationship. |
| Designed to provide a stable definition that may be referenced elsewhere. | Designed to communicate an answer within the context established by the question. |
An FAQFragment may reference a DefinedTermFragment when terminology within the question or answer requires explicit semantic definition.
FAQFragment and ExplainerFragment
An FAQFragment and an ExplainerFragment may both contain explanatory language, but their structures serve different purposes.
An FAQFragment is organized around an explicit question-and-answer relationship.
An ExplainerFragment is organized around the explanation of a subject or concept without requiring a question to establish that structure.
For example:
- FAQFragment: “What does a $0 premium Medicare Advantage plan mean?” followed by the answer.
- ExplainerFragment: A structured explanation of how Medicare Advantage premiums work.
The appropriate fragment class should reflect the semantic form of the knowledge being published.
FAQFragment and Machine Fact Objects
An FAQ answer may contain factual assertions, but the FAQFragment itself remains a question-and-answer knowledge object.
When an answer depends on factual knowledge already represented elsewhere within the SDT, the FAQFragment may reference or align with those facts rather than redefining their semantic role.
For example, a DataFragment might publish a specific plan premium as a Machine Fact Object while an FAQFragment answers a question about what that premium means to the user.
This allows the factual representation and explanatory question-and-answer representation to remain distinct while participating in the same SDT.
Read the Machine Fact Objects specification →
Related Terms
An FAQFragment may identify defined terms associated with its question or answer.
Related-term references can help establish explicit semantic connections between natural-language explanations and the canonical terminology used elsewhere in the SDT.
For example, an FAQ concerning a $0 premium plan might reference defined terms for:
- monthly premium;
- Medicare Part B premium;
- deductible;
- copayment;
- coinsurance;
- or other concepts relevant to the answer.
Term relationships should use stable term identities when available.
Provenance
An FAQFragment should preserve the provenance appropriate to the answer it publishes.
The evidentiary basis may vary depending on the nature of the answer.
Provenance may identify:
- an authoritative dataset;
- a regulatory or policy source;
- an official publication;
- an editorial source;
- one or more WebMEM fragments containing supporting factual knowledge;
- the organization or author responsible for the answer;
- publication or revision dates;
- and other lineage information necessary to understand how the answer was produced.
When an answer synthesizes multiple sources, the provenance model should preserve the relevant source relationships rather than implying that the answer originated from a single source.
Read the Provenance specification →
FAQFragment and Relationships
An FAQFragment may participate in explicit relationships with other WebMEM knowledge objects.
For example, an FAQFragment may relate to:
- a DefinedTermFragment defining terminology used in the answer;
- a DataFragment containing supporting source-observed facts;
- a DerivedStatsFragment containing supporting calculated facts;
- a PolicyFragment establishing an applicable rule;
- a ProcedureFragment describing a process referenced by the answer;
- an ExplainerFragment providing a more complete explanation;
- or another FAQFragment addressing a related question.
Explicit relationships allow the FAQ to remain a focused question-and-answer object while connecting it to the broader knowledge represented by the SDT.
Read the Relationships specification →
FAQFragment and Resolution
An FAQFragment may participate in machine resolution when its question, answer, entities, terms, or relationships connect to more specific knowledge objects.
For example:
Question → FAQFragment → Answer → Supporting Fact → Source
or:
Question → FAQFragment → Defined Term → Canonical Definition
The FAQFragment does not itself define the resolution model. It participates in the resolution architecture through explicit identities, references, relationships, and provenance.
This preserves the original purpose of the class while allowing question-and-answer knowledge to participate in a larger resolvable information structure.
Read the WebMEM Resolution Model →
Conformance
A conforming FAQFragment must:
- identify itself as an
FAQFragment; - have an identifiable fragment scope;
- contain an explicit question;
- contain an answer corresponding to that question;
- establish entity or contextual scope when required to interpret the question or answer;
- identify or reference provenance when the answer depends on source-backed or authoritative knowledge;
- preserve the semantic distinction between the question and answer;
- and conform to the applicable WebMEM serialization requirements.
Defined-term alignment, related-term references, and relationships to supporting knowledge objects are optional unless required by an applicable domain or conformance profile.
Relationship to WebMEM 1.x
The FAQFragment class originated in the WebMEM 1.x Semantic Data Template specification and remains a defined fragment class in the current WebMEM Protocol.
Earlier specifications associated FAQFragments with TrustCast campaigns, memory-first publishing, retrieval conditioning, glossary co-citation, semantic reinforcement, trust scoring, and FAQ-specific retrieval behavior.
The current protocol preserves the semantic purpose of the class while separating that purpose from retired WebMEM 1.x retrieval, scoring, and memory-layer mechanisms.
FAQFragment continues to represent a structured question-and-answer knowledge object within an SDT.