Terms

The terms explorer is the Concepts view of the Context hub. You use it to keep vocabulary consistent across locales. Each concept is one meaning, with terms in several languages, a lifecycle status for each term, and a domain.

The server keeps terms and content memory for every project in the workspace, so the team has one approved record of its wording.

In the Context hub, each concept also carries typed relations to other concepts (what replaced it, what to use instead, which competitor it stands against), per-market validity, observations, comments and a story of its changes. The sections below cover everyday term management. Context covers relations and governance.

Outdated recording

The interface has changed since this was recorded. The workflow it demonstrates still applies, but some screens and labels differ from the current release. Concepts and Content memory show a list beside a detail pane.

Opening the Concepts page

Click Context in the main sidebar, then Concepts. The Terminology button in a project's source view opens the same page, because every project shares the workspace's terms. A link to one concept, from Activity, the dashboard or a review, opens the list with that concept selected.

The list

The page is a list of concepts beside a detail pane. Each row shows the term the concept is headed by, that term's status, and how many languages the concept has terms in. A concept is headed by its term in the language most of your projects are written in.

• Search, at the top of the list, matches term text as you type.
• Status narrows the list to concepts with a term in that status.
• Market narrows it to one market. It appears once the workspace defines markets.
• Show more, at the foot of the list, reads the next page.

The server returns each search a page at a time, so a workspace with thousands of concepts searches as quickly as one with ten. While the list has focus, the arrow keys, j and k move the selection.

A change-set waiting for review sits above the list, with Review to open it. Proposed concepts join the list once their change is approved.

Term lifecycle statuses

Each term carries a lifecycle status that shows its approval level:

Status	Meaning
Preferred	The recommended term; use this one
Approved	Accepted for use
Admitted	Acceptable but not preferred
Deprecated	Retired; replace it with the term to write instead
Proposed	Under review, not yet approved
Forbidden	Must not be used

The detail pane

Select a concept to see it beside the list:

• its definition;
• Write instead: the deprecated and forbidden forms, each with the term to use in its place;
• By language: its terms in each language, the source language first, each with its status and its other forms, and the markets that cover the language;
• Where it came from: its domain, voice profile and coordinates, whether it belongs to one project or the whole workspace, and who added it and changed it last;
• Relations to other concepts; select one to open it;
• History, Discussion and Evidence: the concept's story, the comments on it, and the observations recorded for it.

Editing concepts

The actions for the selected concept are in the pane's toolbar:

• Approve sets the term the concept is headed by from proposed to approved. It appears while that term is proposed.
• Edit sets the status of each of the concept's terms.
• Relate adds a relation to another concept. The cross on a relation removes it.

How an edit applies depends on whether it is governed:

• Ordinary edits, such as adding or correcting a translation, adjusting a domain, or moving a term to a non-governed status, apply directly.
• Governed edits, such as setting or clearing a term's forbidden or preferred status, setting or clearing a concept's do-not-translate flag, recording a replaced_by relation, or deleting a concept, are bundled into a reviewable change-set whose blast radius is measured before anyone approves it. When the server refuses a governed edit, the page offers to open Changes, where the edit goes into a change-set.

A concept created with its do-not-translate flag set is governed too. Promoting an entity marked do-not-translate, approving a context-scan term marked that way, or adding such a term through the MCP term_add tool proposes the concept in a change-set of its own, and the concept appears once that change-set merges.

Adding concepts

Concepts arrive from your projects and from review:

• kapi push sends a project's terms to the workspace. Ordinary terms apply directly; governed ones arrive as a change-set waiting for review.
• Mark as term in a review session, or approving a term a context scan proposed, creates the concept.

To load terms from a spreadsheet or a TBX file, import them into a project with kapi terms import, then push the project.

Deleting concepts

Deleting a concept is a governed action. The delete goes into a change-set, and the concept is removed when that change-set is approved and merged. A concept that other content references is therefore always reviewed before it goes.

Terms in the editor

Terms are also used within the editor:

• Context panel: shows per-block term matches with source terms, target suggestions, status badges, and domain labels
• Term checks: validate that a target uses the terms in force at its point

When a block is selected in the editor, the context panel looks up matching terms and displays them alongside memory matches. See The editor for details on the context panel.

Concept model

The terms system is concept-oriented, in the manner of the TBX (TermBase eXchange) standard:

• A concept represents a single meaning or notion
• Each concept has terms in one or more languages
• Terms have metadata: lifecycle status, domain, and notes
• Concepts are scoped to the workspace and shared across all projects

The framework owns this concept and term model; for its architecture, see the framework terms reference.

On the platform, the same concepts are the nodes of the context graph, with typed relations, per-market validity, observations, comments and governed change-sets. To model how concepts relate, start at Context.