To translate a CASE 1.1 competency framework quickly, translate the human learning statements and explanatory text, but keep the authoritative competency identifiers, coding scheme and alignment relationships under explicit control. In 1EdTech’s Competencies and Academic Standards Exchange, a standard is not just a sentence. It is an item in a structured framework that other systems may use to connect lessons, assessments, reports and resources. A fluent translation that changes which outcome a resource points to is a failed translation.
People searching for CASE competency translation, multilingual curriculum standards, CFItem translation QA or CASE 1.1 localization usually have a practical integration problem. Their original academic standards are written in one language, while teachers, families or course designers need another. They may have a curriculum database, a learning platform and assessment items that already refer to unique standard identifiers. The safest process turns the translated statement into a reviewed representation with traceable provenance rather than treating it as a new unlinked sentence.
The short answer is to freeze the source package, map every source CFItem to its intended translated item, preserve alignment links where equivalence is justified, and use CASE 1.1’s isTranslationOf association where the implementation and publisher’s model support that relationship. Check the exact scope of each verb, grade, condition and evidence requirement. Test the translated framework with the consuming system and keep a documented distinction between translation, adaptation and a new curricular standard. This guide explains that workflow with a structured example, diagnostics and an exercise.
A competency is a learning promise, not just a sentence
Consider an outcome that says, “Explain how changing one variable affects the result of a fair investigation.” A student, teacher and assessment developer can read the sentence as ordinary English. But a system might use a persistent identifier to associate that outcome with ten lessons, three quiz items and a performance rubric. The reference is valuable precisely because it remains stable when the sentence appears in different screens. A translation must serve these readers without creating a different goal behind their backs.
The statement contains several important ideas: the student should explain, the explanation concerns the effect of changing one variable, and the investigation is fair. A translation that reduces “explain” to “name” lowers the cognitive demand. One that omits fairness removes an experimental-design condition. A beautiful sentence may sound educational while describing a weaker performance. The translator needs a clear theory of learning evidence, not only a bilingual dictionary.
CASE is designed for exchanging academic standards, skills and competencies across software systems. Its framework objects, items and associations allow consumers to link educational content to a structured definition. This makes it especially useful for a school or institution that wants different platforms to agree on what a lesson is trying to teach. The exchange format does not, by itself, decide whether two language versions are semantically equivalent; that remains an editorial and curriculum-governance judgement.
An imported framework often becomes invisible infrastructure. A teacher clicks an assessment objective and sees a friendly label, while the platform stores a CFItem identifier. If that label is mistranslated, the screen misleads the teacher even though the data link remains technically intact. If the identifier is changed during localization, the teacher may see an accurate statement but lose alignment to previous records. The quality target is both semantic correspondence and traceable structural identity.
For Singapore readers, distinguish a local classroom translation from an official syllabus statement. A translated school resource may help bilingual learners understand a learning outcome, but it should never be presented as an official MOE or SEAB wording unless it is actually an authorized publication. A public article can teach how to preserve competency meaning without claiming to set the curriculum or grant equivalence on behalf of an awarding body.
What CASE 1.1 actually stores
The framework document
A CFDocument represents the standards framework or document. It carries metadata such as title, identifier, language, versioning information and status. The framework may be adopted, updated or deprecated according to publisher governance. A translated edition should not be invented by changing only the title inside the original record. Its ownership, language, authorization and relationship to the source must be decided before it is distributed to other systems.
The CFDocument.language metadata identifies the language of the framework document, but implementers must follow the exact field conventions required by their CASE 1.1 system. The 1EdTech implementation guidance discusses browser-locale conventions that may differ in punctuation from ordinary BCP 47 presentation. Do not blindly normalize en_US to en-US or vice versa in an existing integration. First inspect the specification, provider behavior and consumer tests, then record the accepted representation for that deployment.
The competency item
A CFItem carries a unique identifier, a resolvable uri, a fullStatement, and potentially fields such as abbreviatedStatement, humanCodingScheme, educationLevel, CFItemType, notes and subject metadata. Some fields express natural language; others express identity, classification or references. The human coding scheme might resemble SCI.P6.FI.2, which is useful to readers but should not be translated as prose. Changing the code may break reporting conventions even if it looks like a harmless editorial improvement.
The word “full” in fullStatement matters. It is the principal declaration of what the learner should be able to know or do. A summary for navigation is not a substitute. Translators should compare the complete outcome, including conditions, quantifiers, mathematical expressions and performance verbs, before approving a short label. The abbreviatedStatement may be edited for concision, but it needs independent review because a shorter phrase is more likely to omit a critical qualification.
CASE 1.1 also permits Markdown and mathematics-oriented LaTeX in several human-text fields, including CFItem.fullStatement and notes. That increases expressive power for science and mathematics, but changes how translation must be checked. A translator might move natural-language words around an equation while preserving symbols. Rendering the original and translated statements is essential because syntactically present backslashes or delimiters may not be displayed as intended by the consuming platform.
Associations and alignment
A CFAssociation connects framework elements and supplies relationship semantics. It can represent parent-child structure, alignment or other supported relationships. CASE 1.1 also introduced isTranslationOf, which is particularly relevant to language editions. This relation makes the translation relationship explicit instead of inferring it from titles. The exact direction, endpoint typing and supported association behavior must be checked against the official information model and the provider’s implementation.
Do not confuse isTranslationOf with a promise that all associated text is perfectly equivalent. A system can store a relationship that an editorial process has not properly justified. The association is evidence that the publisher expressed a link; accuracy still depends on reviewing the statements and governing changes. A consumer might use that link to navigate between languages, but the institution must decide what to do when one edition is updated before another.
Association sequence and display order are also distinct from semantic alignment. The implementation guide explains that sequenceNumber on associations is relevant to display ordering under a parent, while listEnumeration on the item is informational. Translating or re-sorting the visible numbering should not silently alter hierarchy. A teacher selecting the fourth outcome in a topic expects its position to reflect the publisher’s intended sequence, not an accidental lexical sort of translated titles.
The first important policy decision: one item or linked items?
There are two different patterns a school may be considering. In a presentation-only approach, the authoritative source item keeps its identifier and a separate interface or content layer supplies localized wording. In a framework-publication approach, a publisher may issue a translated framework whose items have their own managed identifiers and explicit isTranslationOf links to the authoritative original. The right choice depends on publisher control, provider support, consumer capability and whether the translated material is an official edition.
The first pattern is appropriate when a teaching interface merely helps readers understand an unchanged official outcome. The English CFItem still owns the reference and the app decides which explanatory translation to display. The second pattern is useful when the translated text needs a discoverable presence in a CASE ecosystem. The original and translated records then must be linked and versioned carefully. Neither approach is a universal rule; the dangerous approach is drifting between them without a documented identity policy.
Do not reuse a GUID just because a translated sentence feels equivalent. A persistent identifier is a published reference, not an ordinary database row number that editors may recycle freely. At the same time, do not fabricate a whole new unlinked competency if a published isTranslationOf relationship should explain the editorial relationship. The publisher and integration owner must specify whether identities are shared across representation layers or newly minted for language-specific CASE items.
The translation process must also distinguish an adaptation. If a source outcome asks students to interpret a local historical document and the target course replaces it with a different historical event, the target may serve a similar pedagogical role but it is not a literal translation. If grade expectations, evidence standards or scope change, the edit may require a new competency and a different relationship, not isTranslationOf by default. Mislabeling adaptation as translation contaminates later curriculum mapping.
A useful decision record therefore answers four questions for every target item: Who authorizes the target wording? What source item and source version does it represent? Is the performance demand materially unchanged? What stable link will consumers use when the source changes? This small governance step prevents months of confusion after translations have spread to lesson banks, teacher dashboards and assessment reports.
A structured example of a reviewed translation
The following JSON is an illustrative editorial comparison record, not a complete valid CASE service response. It deliberately keeps source and target representations inside a local review fixture, so that no one mistakes example data for a certified CASE 1.1 payload. The structure can be validated as JSON; production CASE objects and CFAssociation payloads must be built from the official schema and checked with the provider’s conformance tools.
{
"source": {
"identifier": "d0c0c9f2-ec69-4d54-aad5-44a80dbeae01",
"uri": "https://curriculum.example.org/ims/case/v1p1/CFItems/d0c0c9f2-ec69-4d54-aad5-44a80dbeae01",
"humanCodingScheme": "SCI.P6.FI.2",
"fullStatement": "Explain how changing one variable affects the result of a fair investigation."
},
"translationReview": {
"language": "zh-Hans",
"statement": "解释在公平实验中改变一个变量会如何影响实验结果。",
"relationshipIntent": "isTranslationOf",
"sourceVersion": "2026-08",
"reviewStatus": "subject-specialist-approved"
},
"protected": ["identifier", "uri", "humanCodingScheme"],
"releaseDecision": "pending CASE provider association validation"
}Notice that the example’s source.identifier and uri are frozen. The review record does not claim a target identifier already exists, because the publisher has not yet completed that decision. relationshipIntent records a plan rather than pretending that an officially serialized CFAssociation has been created. This distinction lets a curriculum editor approve the meaning while a systems engineer later checks the exact provider model, association endpoint and permissions.
The Chinese sentence conveys explaining how changing one variable affects the outcome of a fair investigation. A bilingual science educator should still review whether the accepted local term for fair investigation is appropriate and whether the phrase preserves the one-variable condition. A computer can check that the identifier remained unchanged; it cannot reliably decide whether the wording reflects the intended scientific method in a particular curriculum.
A weaker translation might say, “State the result after changing a variable.” That loses the explanatory demand and may imply that changing multiple variables is permitted. Another weak version might add “predict,” which changes the required evidence from explanation to forecasting. Neither should be approved just because it contains the familiar terms for experiment and variable. The review should compare the task a student would be asked to perform under each sentence.
The fixture’s sourceVersion is an editorial trace, not a standard CASE field inserted into the source CFItem. For a real project, the publisher may use defined version and date metadata, together with an internal manifest, to connect the target to the exact source release. Use existing standard fields correctly before creating custom extensions. The CASE 1.1 extension mechanism is available in specified entities, but a consumer may ignore extensions it does not understand.
Source preparation and extraction workflow
Begin with a package inventory
List the source CFDocument, all source items, associations, relevant rubrics, language metadata and the provider’s service version. Download or retrieve the source package in a controlled way, then record a hash and extraction date. Identify whether the publisher declares the document adopted, draft or deprecated. Translating a deprecated framework without clearly labeling it risks making an obsolete outcome look current in a target-language search result.
Capture the framework hierarchy before selecting sentences. Many competencies depend on their parent topic. A short outcome like “Use this relationship to solve problems” is uninterpretable when removed from its node context. The translator needs the parent heading, preceding item and associated subject, not only the fullStatement string. Preserve this context with each extracted segment so that language decisions do not require repeated searches through a large PDF.
Build a field-permission table
Mark fullStatement, approved notes and descriptions as candidates for translation. Treat identifier, uri, humanCodingScheme, association types, endpoint references, timestamps and other structural attributes as protected unless a separate publisher-approved migration is underway. Classification labels such as CFItemType may be controlled vocabulary values rather than free prose. Do not translate a machine classification because the user interface could render its code as English.
For each item, log its source value, path, context, term decisions and permitted target length. The 1EdTech implementation guidance discusses best-practice field lengths, including roughly four thousand characters for fullStatement in common storage designs, but that guidance should not be confused with a universal normative hard cap. A consuming product might impose its own stricter limits. Check the real system before shortening a sentence or making it misleading just to fit a database column.
Clarify instructional verbs and conditions
Create a small performance-verb glossary with examples: identify, describe, compare, infer, justify, evaluate, design and explain. These verbs signal different evidence demands. Ask a subject specialist which distinction matters in the local syllabus rather than assuming a single bilingual equivalent works in every topic. A learner who must justify a method needs to give reasons; a learner who only identifies a method may need to name it. Translating both into the same target verb collapses the assessment design.
Conditions can be equally important. “Using measurements,” “with reference to a given graph,” “without a calculator,” and “for familiar situations” restrict what counts as meeting an outcome. If the target edition drops one condition, educators may design easier or harder tasks than the source permits. Reviewers should annotate the source sentence as action, object, context and performance condition, then check that those elements survive in the target formulation even if word order changes.
Preserve mathematical and technical notation
CASE 1.1 permits Markdown and math markup in designated fields. Preserve the formula’s actual mathematical meaning while translating prose around it. If a source uses \(a^2+b^2=c^2\), the target can move the explanatory words, but should not change the variables’ roles or accidentally transform an exponent into ordinary text. Use a renderer that supports the project’s selected Markdown and LaTeX conventions, and review the visible output in the consuming platform rather than only the raw string.
An equation may use Latin variables that are conventional symbols; do not transliterate them simply because target-language words use another script. If a source uses a culturally familiar example or units, identify whether the item is about the underlying mathematical relation or about a locally specific context. A genuine adaptation may be justified, but it must be approved and labeled as such. Translation without authorization should not silently change quantities, signs, diagrams or evidence requirements.
A practical release sequence
First pass: translate only the approved readable fields with the source context visible. Second pass: review terminology across sibling items so that the same concept does not gain three incompatible target names. Third pass: check cognitive demand and conditions against the source. Fourth pass: ensure text, Markdown and LaTeX render. Fifth pass: compare all protected machine fields and relationships. Sixth pass: import into a staging consumer and verify retrieval and alignment, with one curriculum reviewer signing off on content and one technical reviewer on integration.
This sequence is fast because it separates failures by owner. A linguist can repair a verb choice without asking a developer to rebuild an API. A developer can repair a broken URI without second-guessing curriculum terminology. A curriculum specialist can reject a changed performance expectation without editing JSON by hand. When responsibilities are mixed, every defect produces a meeting; when they are separated, defects move directly to the person who can solve them.
A terminology memory should include source phrase, target phrase, item type, age group, domain, evidence demand and approval status. A general translation memory entry containing only the two sentences can be dangerous because the same English verb may require a different level of specificity in different curricular contexts. Reuse should be a suggestion anchored to meaning, not a command to substitute words mechanically. Reviews are especially important when language editions serve different regions.
For each target-language item, maintain an alignment ledger with source identifier, target identifier if issued, translation relationship, source revision and content approval date. Mark a target stale when the source changes substantively. If a source item’s statement now says “compare and evaluate” instead of “compare,” the old translation no longer necessarily carries its full scope. The ledger provides a precise list of items that need renewed review rather than forcing the school to retranslate the whole framework.
Diagnostic checks for integration and meaning
The GUID changed because the sentence changed language
If an editor duplicates the framework and replaces all source identifiers with randomly generated ones, previous lesson and assessment associations may no longer resolve. That may be appropriate for a new publisher-managed target framework, but it is not a harmless localization detail. The repair is to document the identity model and publish explicit mappings, potentially using isTranslationOf in a supported CASE 1.1 relationship. Verify consumer behavior before migration rather than assuming every platform understands language equivalents.
The GUID stayed the same while the outcome changed
The opposite problem is more subtle. Reusing the original identity for a target statement that changes grade demand or removes an assessment condition makes the same identifier represent different learning promises. Reports appear connected but may compare unlike evidence. Repair begins with a semantic comparison and an ownership decision: correct the target text if it should be a translation, or publish a distinct curriculum item with the proper relationship if it is an adaptation.
A hierarchy was translated as an outline
A visible outline number may look like the source of item order, but CASE associations govern relationships, and recommended display sequencing can be carried on associations rather than inferred from a label. If the translator renumbers items or rearranges alphabetical titles, the consumer may show siblings in a different pedagogical order. Compare parent links, sequence metadata and displayed outline. Repair the presentation without inventing a new hierarchy unless the publisher authorizes a substantive curricular restructuring.
Grade language shifted
English “Primary 6,” “Grade 6” and “Year 6” are not automatically identical educational levels across jurisdictions. A target-language label may need a contextual explanation while preserving the original educationLevel metadata as the source defines it. A learner’s age or schooling stage cannot be inferred from a number alone. If the audience is in another national system, let the curriculum owner decide whether to provide an equivalence note or publish an adaptation rather than silently replacing the level.
Mathematical formatting disappeared
An escaped backslash, unsupported delimiter or stripped Markdown mark can turn a correct formula into confusing plain text. Compare the rendered statements in each consuming platform. Some systems may expose raw LaTeX instead of typeset mathematics; the CASE specification recommends rendering support, but real capability varies. If a client cannot handle the approved source notation, treat that as an interoperability problem and design an accessible fallback, not as permission to remove mathematical content.
Long statements were truncated
A database import may accept a record yet store only the first portion of its text. An outcome that loses “without using a calculator” changes the task while appearing syntactically valid. Measure exact target length, inspect the retrieved value and compare the full display. A field-length warning is more useful than pretending the value is complete. Repair may require a provider configuration change, shorter genuinely equivalent wording, or a supported long-text representation, followed by renewed subject review.
A deprecated outcome appeared as current
When source frameworks are replaced, an old translation can remain visible in search tools and learning-object tags. A teacher may then assign content aligned to outdated expectations. Keep version and adoption-state metadata visible in the publication workflow. Where replacedBy or equivalent supported associations exist, preserve them. A localization team should not reclassify a deprecated item as current simply because target-language readers have never seen its replacement.
Translation alignment was assumed from similar titles
Two outcomes can share the words “understand fractions” but differ in whether they require modeling, calculation or explanation. Text similarity is not a reliable identity key. Use authorized identifiers and published associations, then test actual semantic equivalence. A target item should not be connected merely because a machine embedding reports a high cosine score. The repair is to route uncertain pairs to a subject expert and record the unresolved status rather than auto-link them.
The source language metadata was normalized incorrectly
A consumer may expect a particular locale notation while another integration uses BCP 47 tags. A transformation that looks like harmless punctuation can break filtering or language selection. Read the 1EdTech field guidance and inspect the provider’s actual values. Define an explicit mapping between external locale identifiers and UI locale tags if necessary, and test round trips. Do not write one dialect tag into every layer without checking each layer’s requirements.
An association now points outside the CASE ecosystem
CASE 1.1 permits association references involving non-CASE frameworks, with target-type metadata to help consumers interpret them. If a translation pipeline assumes every target is an internal CASE item, it may drop or misresolve legitimate external links. Keep source and destination references stable and respect targetType where used. For language equivalence across systems, verify both the relationship type and the kind of target rather than relying on a string that happens to look like a URL.
Worked classroom scenarios
Mathematics learning progression
Suppose a school publishes a progression from recognizing equivalent fractions to justifying fraction operations. A translation that renders every outcome as “calculate fractions” hides the increasing cognitive demand. Group items by learning progression during review and compare the required evidence in adjacent outcomes. If one item asks for a visual representation and another asks for symbolic reasoning, target wording should preserve that contrast even when the source phrases share much vocabulary.
Science practical skills
In science, “measure accurately,” “estimate uncertainty” and “evaluate reliability” describe related but distinct tasks. A generic translation for “quality of data” may blur them. Review the associated rubrics and experiments before finalizing terms. Where a source refers to variables, controls or fair testing, preserve the underlying investigation method rather than adopting an attractive but technically different phrase. The target needs to guide assessment design, not merely sound scholarly.
Language-learning standards
An English proficiency item may refer to inferential reading, spoken interaction or audience-aware writing. Translating such terms into another language requires a decision about whether the target is explaining an English-language skill or requiring the student to perform in the target language. Those are different learning outcomes. A bilingual explanatory edition should make the language of performance explicit so that families do not misread translation as permission to change the assessment language.
Technical and vocational competencies
A vocational outcome might say a learner must isolate electrical power before maintenance. The translation must preserve temporal order and safety conditions. “Identify the switch” is not equivalent to “verify isolation.” The standards owner may need to involve a safety specialist. A school can add explanatory notes for learners, but a non-authorized translation should not replace a technical or regulatory requirement with a simplified suggestion.
Cross-national curriculum comparison
An education researcher might align two countries’ frameworks by translating their outcomes into a common language. Such comparison is useful, but it is not the same as producing a publisher-approved language edition of either framework. Keep the countries’ original identifiers and a separate research mapping with confidence and assumptions. The isTranslationOf relationship should not be used to assert that two independently authored national standards are translations of one another merely because they cover similar mathematics.
Multilingual teacher dashboards
A school platform may render source and target statements side by side. The interface should retain the code and source version while allowing a teacher to choose a preferred display language. A useful design permits easy access to the authoritative original and highlights translations awaiting review. It should not show a green check merely because the target text exists; the check must mean a defined QA state, such as curriculum approval against a specific source revision.
Assessment item tags
A test question tagged to an authoritative CFItem identifier should continue to point to that item after the dashboard is localized. If a separate translated framework is issued, the mapping needs explicit governance before analytics can aggregate across language editions. Check whether a quiz’s cognitive demand matches both the source and approved translated wording. Translation errors can propagate to recommendations and mastery charts, making incorrect alignment appear as student weakness.
Resource discovery and search
A bilingual resource library can index translated competency wording to help teachers find materials in their preferred language. Search expansion is a presentation feature; do not overwrite canonical skill identifiers to gain keyword reach. Store alternate labels with provenance, and ensure search results show which framework and year are being referenced. When results are ranked by lexical similarity, let teachers inspect the original outcome and relationship before assigning high-stakes teaching or assessment content.
Historical revisions
An outcome may be split into two items in a later framework edition. Translating the old single statement into two target sentences does not automatically create the two new official items. Preserve the historical structure and document the revision relationship defined by the publisher. If students’ past records use the old identifier, keep its meaning recoverable. A migration script should never erase the history that makes earlier reports interpretable.
Community translation versus official adoption
Volunteer teachers may prepare helpful bilingual explanations long before a formal publisher authorizes a translation. Such contributions should be visibly marked as unofficial and separated from authoritative framework records. Use a review workflow that records translator, subject reviewer, source version and license. If an official language edition later appears, compare the two rather than silently treating the volunteer wording as canonical. Ownership matters as much as linguistic polish.
Changed-context transfer exercise: translating an algebra standard
A curriculum team now works on an algebra outcome: “Represent a linear relationship using a table, a graph and an equation, and explain how the representations correspond.” The proposed target sentence says, in effect, “Draw a graph of an equation.” The translator keeps the original humanCodingScheme but generates a new random identifier, removes the parent association and changes the item type. The editor argues that a new language must have a new record. Identify each defect and decide what a responsible next step would be.
The semantic defect is severe. The target has removed the table, the explanation of correspondence and the relationship among three representations. Restore the entire performance demand before discussing JSON. The structural defects are separate: the identifier policy has not been approved, the parent association was lost and a classification was changed without curricular justification. A new target identifier may eventually be correct for a published translation edition, but it must be issued and linked under the provider’s governance, not fabricated during sentence translation.
The repaired workflow begins with a bilingual subject review of the algebra statement, including a worked linear function shown as a table, plotted points and an equation. Then the publisher chooses whether to provide a presentation-only translation or a distinct CASE item with an isTranslationOf relationship. Finally, the team tests the chosen record in a consumer and confirms that its parent, ordering, code, source provenance and assessment tags behave as intended. The exercise shows why meaning, identity and association need separate approval.
Editorial checklist and release evidence
Before starting, identify the framework owner, license, adoption state, source version, expected target language, grade level and receiving systems. Freeze the original payload and create an item inventory. For each translation, preserve the source identifier and context in a ledger. Decide in writing whether the project creates alternate display copy or publisher-managed target records. Do not mix both models invisibly, and do not imply that the translation is official without authorization.
During review, compare each learning verb, object, quantity, constraint, representation and evidence requirement with the source. Verify consistency among sibling and parent outcomes. Render Markdown and mathematics. Check labels and field lengths in the actual consuming UI. Maintain separate findings for language, curriculum, data structure, accessibility and licensing. An editor who cannot explain what a particular outcome asks the learner to demonstrate should hold it for clarification.
At release, validate the JSON and the provider’s CASE conformance path where applicable, inspect imported values, verify associations and their direction, and test resource and assessment links. Archive changed-field diffs, translation approval, source hashes, outcome version mappings and rollback decisions. Mark unresolved items clearly. A clean import is necessary but does not prove valid pedagogy; curriculum owners and technical owners must each sign off on the portion they understand.
Limits and conclusion
CASE 1.1 is an exchange specification, not a machine that certifies educational equivalence. It can express framework items and relationships, including a translation relationship, but it cannot decide that a target phrase demands the same student performance. Implementations may also vary in support for optional fields and display behavior. Always check the exact official specification and the receiving platform’s version; a research fixture that parses as JSON is not a production-conformant CASE response.
The quickest trustworthy way to translate curricula is therefore not to translate everything indiscriminately. Extract the right text with enough context, maintain an exact source-to-target identity plan, review cognitive demand, protect associations, and test the resulting framework where teachers and learners will use it. A smaller set of approved equivalent outcomes is more valuable than a large database of attractive statements that no longer connect to the right learning evidence.
A translation succeeds when a teacher reading the target language makes essentially the same instructional decision as a teacher reading the authoritative source, and both can still trace the outcome to its owner, revision and aligned resources. That is the durable meaning of speed in curriculum translation: fewer repeated guesses, less broken alignment and a reliable path from a sentence to learning.
Frequently asked questions
Does every translated CASE item need a new identifier?
No universal answer applies. It depends on whether the translation is presentation-only or a publisher-managed target item and how the provider governs identity.
What is `isTranslationOf` in CASE 1.1?
It is a supported association type for expressing a translation relationship. It does not replace subject-matter review of semantic equivalence.
Can I translate `humanCodingScheme`?
Treat it as a protected reference code unless the authoritative publisher explicitly defines a controlled migration. Use a separate readable label for natural-language presentation.
What if the target language needs more words?
Preserve the learning demand first, then test the actual consumer’s field and display limits. Do not delete a qualification merely to meet an arbitrary length.
Are Markdown and LaTeX allowed in CASE 1.1?
Specified human-text fields may use them. Validate rendering and mathematical meaning in the target platform.
Should I translate `CFItemType` values?
Not by default. A classification value may be used as data. Localize its presentation separately where the system supports that.
Is a translated outcome automatically an official syllabus statement?
No. Authorization and ownership must be clear; a community explanation cannot impersonate the authoritative publisher.
How can I detect a changed cognitive demand?
Break the sentence into performance verb, knowledge object, condition and evidence requirement, then compare every element with the target version.
What happens when the original framework is revised?
Use source-to-target version mapping to identify stale translations, review substantive changes and maintain historical references for older student records.
Can AI perform the entire alignment automatically?
It can suggest mappings and draft language, but identifiers, publisher authority, curriculum equivalence and integration behavior require validated human oversight.
Authoritative sources
- 1EdTech, CASE 1.1 Specification (final).
- 1EdTech, CASE 1.1 Best Practice and Implementation Guide.
- 1EdTech, CASE REST/JSON Binding.
- 1EdTech, CASE 1.1 Certification Guide.
- 1EdTech, CASE standards portal.