Concept — the main addressable unit of knowledge
Concept — the main addressable unit of knowledge
Concept is an independent semantic object that is useful for the system to have its own identity. It has a name, URI, short description, current content, and links to other Concepts.
A Concept does not have to be just an encyclopedic article, a fact, a task, a case, or a conclusion. These distinctions should not be turned into separate entities in advance unless they provide distinct behavior or a useful system query. The practical principle is: do not multiply entities without necessity.
What a Concept provides
- stable addressability and human-readable URLs ( ЧПУ );
- an independent SEO/GEO page;
- reuse of meaning in different contexts;
- the ability to constantly improve current content instead of accumulating conflicting versions;
- links to other semantic objects;
- a convenient reading and updating unit for AI.
A Concept is a living document: over time, its content can be refined, supplemented with work results, new links, and more precise conclusions.
Types
The type field is used only where there is a practical need for group filtering or specialized behavior. A colon sets the filtering hierarchy: for example, project:default allows finding both a specific type and all project:%.
At the start, there is no need to pre-create case:*, problem:*, solution:*, and other assumed categories. First, meaning accumulates, and then stable families become visible from the actual data array.
Related ideas: