6. Existing Perspectives¶
POOKA is not developed in isolation. It builds upon concepts that have emerged across multiple disciplines, including Information Architecture, Software Architecture, Knowledge Management, Identity & Access Management, Object-Oriented Design, graph-based modeling and Human–Computer Interaction.
Rather than replacing these disciplines, POOKA brings together selected principles within a single architectural perspective focused on sustainable Human–AI collaboration.
6.1 Information Architecture¶
Traditional Information Architecture focuses on organizing, structuring and presenting information so that it can be effectively understood and accessed by people.
POOKA extends this perspective by treating AI systems as architectural participants alongside human users. Information is therefore organized not only for human interpretation, but also for consistent interpretation by AI.
6.2 Knowledge Management¶
Knowledge Management addresses the creation, organization, sharing and preservation of organizational knowledge.
POOKA shares these objectives but focuses specifically on the architectural organization of knowledge as a persistent foundation for Human–AI collaboration, independent of specific organizational processes or technologies.
6.3 Object-Oriented Design¶
Object-Oriented Design introduced concepts such as encapsulation, composition and explicit relationships between objects.
POOKA adopts a similar architectural mindset by treating information as explicit architectural concepts with clearly defined responsibilities and relationships. However, these concepts represent information architecture rather than software components.
6.4 Domain-Driven Design¶
Domain-Driven Design emphasizes the importance of modeling complex domains through explicit language, bounded contexts and domain concepts.
POOKA adopts similar principles for organizing information but applies them beyond software design. Domains and Contexts become architectural constructs for organizing human and AI knowledge rather than software models alone.
6.5 Identity & Access Management¶
Identity & Access Management defines how identities are authenticated, authorized and represented within digital systems.
POOKA adopts the distinction between Identity and operational actors but extends it through explicit Delegation, allowing humans, AI systems and technical services to operate on behalf of an Identity while preserving governance and accountability.
6.6 Graph-Based Modeling¶
Graph-based models represent knowledge through nodes and explicit relationships.
POOKA shares the emphasis on explicit relationships but intentionally remains independent of graph technologies. Relations are architectural concepts that may be implemented using graph databases, relational databases, documents or other storage mechanisms.
6.7 Local-First Computing¶
Local-First Computing promotes user ownership, resilience and long-term control over information.
POOKA adopts similar principles by encouraging architectures in which knowledge remains under the control of the owning Identity whenever practical, while allowing cloud-based services and AI models to extend capabilities where appropriate.
6.8 Software Architecture¶
Software Architecture is concerned with the structure of software systems, including their components, responsibilities and the principles that guide their design and evolution.
POOKA adopts an architectural mindset but applies it to the organization of information rather than to software components. Its concepts describe an information architecture that remains independent of any particular software system.
6.9 Human–Computer Interaction¶
Human–Computer Interaction studies how people interact with computing systems and how those systems can be designed to support human understanding and control.
POOKA shares this concern for human understanding and control but addresses it at the level of information architecture. It focuses on how information is organized so that humans and AI can collaborate, rather than on the design of specific user interfaces.
6.10 Context-Aware Computing¶
Context-Aware Computing studies how systems can perceive and respond to the situation in which they are used. Within this field, context is commonly understood as any information that characterizes the situation of an entity.
POOKA shares the recognition that context determines how information should be interpreted, but treats Context as an explicit architectural construct rather than a condition sensed at runtime. Context in POOKA is represented, selected and governed as part of the information architecture rather than inferred from the environment.
6.11 Provenance Models¶
Provenance models, such as the W3C PROV family of specifications, describe how information came into existence by modeling entities, activities and agents, including agents acting on behalf of other agents.
Several POOKA concepts have close counterparts in these models: Artifacts resemble entities, Events resemble activities, Actors resemble agents and Delegation resembles acting on behalf of another. POOKA builds on this lineage but differs in scope: provenance models describe how information was produced, whereas POOKA organizes information, semantics, governance and behavior for ongoing Human–AI collaboration.
6.12 Personal Data Stores¶
Personal data store initiatives, such as Solid, place personal information under the control of an identity and allow applications and agents to access that information through explicit permissions.
POOKA shares this identity-centric perspective and its emphasis on explicit access, particularly in personal knowledge environments. It differs by defining a technology-independent architectural style rather than a platform or protocol: a personal data store may provide one possible implementation environment for POOKA concepts, but POOKA itself prescribes no specific infrastructure.
6.13 Multi-Agent Systems¶
Research on multi-agent systems addresses how autonomous software agents cooperate, including how norms and organizational rules constrain agent behavior.
POOKA addresses a related concern through Behavior, Delegation and Boundaries, which constrain how Actors, including AI systems, may operate. It differs in focus: multi-agent systems research primarily concerns interaction among autonomous software agents, whereas POOKA defines the information architecture within which humans and AI collaborate.
6.14 Positioning¶
POOKA should not be viewed as an alternative to these disciplines. Instead, it can be understood as an architectural synthesis that combines established concepts into a coherent model for Human–AI Information Architecture.
Its contribution lies not in replacing existing theories, but in connecting them within a common architectural language designed specifically for collaboration between humans and AI.