COLLABORATION.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 Collaboration Specification¶
Purpose¶
Collaboration is a first-class capability of OntoCode. Teams should collaborate around semantic intent, not line-based text differences. Reviews, discussions, approvals, and change history should revolve around ontology concepts and their meaning.
Vision¶
Ontology engineering should feel like modern software engineering.
OntoCode should provide semantic equivalents of:
- Pull Requests
- Code Reviews
- Issue Discussions
- Pair Programming
- Design Reviews
- Architecture Reviews
- Continuous Integration
while understanding the ontology's semantic structure.
Design Principles¶
Semantic Over Textual¶
Review changes as:
- Classes added
- Classes removed
- Relationships changed
- Restrictions modified
- Inferred knowledge changed
instead of raw RDF/Turtle line diffs.
Context First¶
Every collaboration artifact is tied to semantic objects.
Comments attach to:
- Entities
- Relationships
- Restrictions
- Queries
- Graph regions
- Documentation
Reviewable Workflows¶
Every significant change should support:
- Preview
- Discussion
- Approval
- Merge
- Rollback
Collaboration Workspace¶
+--------------------------------------------------------------+
| Reviews | Discussions | Activity | Mentions | Notifications |
+--------------------------------------------------------------+
| Semantic Diff | Review Thread |
| | |
+--------------------------------------------------------------+
| AI Review | Graph | History | Checks | Approvals |
+--------------------------------------------------------------+
Semantic Pull Requests¶
Display summaries such as:
- 12 Classes Added
- 4 Classes Removed
- 18 Relationships Changed
- 6 New Restrictions
- 2 Unsatisfiable Classes Introduced
- Documentation Coverage +8%
Include semantic health before merge.
Semantic Diff¶
Show changes grouped by:
- Entities
- Relationships
- Constraints
- Documentation
- Imports
- Queries
- Reasoning
Allow filtering by change type.
Review Threads¶
Threads can attach to:
- Entity
- Axiom
- Graph node
- Graph edge
- Query
- Documentation section
Support:
- Resolve
- Reopen
- Mention users
- AI summary
Live Collaboration¶
Future capabilities:
- Presence indicators
- Shared cursors
- Live graph editing
- Follow collaborator
- Voice/video integration hooks
Edits synchronize through the Workspace Model.
Activity Feed¶
Track:
- Entity edits
- Refactorings
- Reasoning runs
- AI actions
- Reviews
- Merges
- Comments
Filter by user, ontology, or module.
Notifications¶
Notify users about:
- Mentions
- Review requests
- Failed checks
- AI review complete
- Merge conflicts
- Assigned work
Notifications should be actionable.
Merge Experience¶
Before merge:
- Run reasoning
- Validate constraints
- Execute configured checks
- Detect semantic conflicts
- Generate impact summary
Surface blocking issues clearly.
Semantic Conflicts¶
Detect conflicts beyond text.
Examples:
- Divergent class hierarchies
- Conflicting restrictions
- Duplicate entities
- Namespace collisions
- Reasoning regressions
Offer guided resolution workflows.
AI Collaboration¶
AI assists by:
- Summarizing changes
- Explaining reasoning impact
- Highlighting risky edits
- Suggesting reviewers
- Drafting review comments
- Identifying missing documentation
AI augments human review.
Governance¶
Support:
- Required reviewers
- Approval rules
- Protected ontologies
- Audit history
- Electronic sign-off
- Policy checks
Suitable for enterprise and regulated environments.
Integrations¶
Integrate with:
- GitHub
- GitLab
- Azure DevOps
- Jira
- Slack / Teams
- CI pipelines
Semantic review complements existing developer workflows.
Plugin Extension Points¶
Plugins may contribute:
- Review checks
- Policy validators
- Dashboards
- Approval workflows
- Notification providers
- Collaboration widgets
Accessibility¶
Support:
- Keyboard-first reviews
- Screen readers
- High contrast
- Reduced motion
- Accessible diff navigation
Success Criteria¶
Collaboration succeeds when teams discuss ontology meaning instead of serialization details. Reviews become faster, higher quality, and easier to understand because OntoCode presents semantic intent, reasoning impact, and contextual discussions as first-class concepts rather than forcing reviewers to interpret raw RDF diffs.