Skip to content

HUMAN_INTERFACE_GUIDELINES.md

Document type: Product design specification (target state). Not a shipped feature list. See ROADMAP_MAPPING.md for release mapping and SHIPPED.md for what works today.

OntoCode Human Interface Guidelines (HIG)

Purpose

These guidelines define the user experience standards for every OntoCode interface. Every screen, component, workflow, and interaction should follow these principles to create a cohesive, modern ontology engineering environment.


Design Goals

The interface should feel:

  • Modern
  • Calm
  • Fast
  • Intelligent
  • Predictable
  • Discoverable
  • Professional

Users should spend their mental energy understanding ontologies---not learning the UI.


Core Principles

Context First

The interface revolves around a single Current Focus. Every major view reacts automatically to the selected entity.

Never require users to manually synchronize multiple windows.

Progressive Disclosure

Start simple.

Only reveal advanced semantic concepts when users need them.

Novice users should never feel overwhelmed by OWL terminology.

Workflow First

Optimize for tasks instead of tools.

Primary workflows include:

  • Explore
  • Understand
  • Edit
  • Validate
  • Refactor
  • Review
  • Document
  • Publish

Layout

Global Structure

Top - Universal Search / Command Palette

Left - Explorer - Favorites - Recent Items

Center - Workspace

Right - Inspector

Bottom - Problems - Graph - Query - AI - Git - Output

No dedicated "Graph Mode" or "Reasoner Mode."


Navigation

Always provide:

  • Breadcrumbs
  • Back/Forward navigation
  • Recently viewed entities
  • Favorites
  • Command palette
  • Keyboard shortcuts

Navigation should never trap users.


Visual Hierarchy

Emphasize information in this order:

  1. Entity name
  2. Human-readable description
  3. Semantic relationships
  4. Diagnostics
  5. Technical metadata
  6. Internal identifiers

IRIs should never dominate the interface.


Editing

Prefer inline editing over modal dialogs.

Changes should:

  • update immediately
  • validate continuously
  • support undo/redo
  • provide non-blocking feedback

Search

Search is the primary navigation mechanism.

Search should index:

  • entities
  • relationships
  • annotations
  • documentation
  • queries
  • diagnostics
  • git history
  • AI actions

Results should be grouped by type.


Panels

Panels must:

  • synchronize automatically
  • remember layout
  • support keyboard navigation
  • avoid duplicate information

Each panel should answer one question well.


Feedback

Use subtle feedback.

Prefer:

  • inline validation
  • status chips
  • badges
  • lightweight notifications

Avoid disruptive modal dialogs whenever possible.


Motion

Animations should communicate state changes.

Use motion for:

  • expanding trees
  • graph transitions
  • docking panels
  • loading

Animations should be fast (150--250 ms) and never delay interaction.


Accessibility

Support:

  • WCAG 2.2 AA
  • full keyboard navigation
  • screen readers
  • high contrast
  • reduced motion
  • scalable typography

Accessibility is a core requirement, not an enhancement.


AI

AI should appear as contextual actions rather than a separate application.

Examples:

  • Explain
  • Suggest improvements
  • Generate documentation
  • Repair issues
  • Refactor
  • Translate syntax

AI should always explain why it made a recommendation.


Error Handling

Errors should:

  • identify the affected entity
  • explain the problem in plain language
  • provide suggested fixes
  • link directly to the source

Never display cryptic parser errors without context.


Performance

Target interactions:

  • Navigation: \<100 ms
  • Search: \<100 ms
  • Inspector updates: \<50 ms
  • Workspace switches: \<150 ms

Perceived performance is as important as measured performance.


Extensibility

Plugins should inherit the same interaction patterns.

Extension authors should be able to create native-feeling experiences without reimplementing navigation, layout, or styling.


Success Metric

A new user should be able to install OntoCode and begin exploring an ontology within minutes, while experienced ontology engineers should feel they have the power and efficiency of a first-class modern IDE.