Proposal: Add Category "System Inception Patterns"

[cpoepke]

Summary

This proposes adding a new category:

System Inception Patterns

Motivation

There is currently no clear category for patterns describing:

• how systems are started
• how architecture is validated early
• how risk is reduced before scaling

These patterns are often confused with:

• Architecture patterns (which describe structural organization, not temporal phase)
• Agile process patterns (which describe team/workflow concerns)

Definition

System Inception Patterns describe:

How a system is incrementally started, validated, and evolved — with a focus on early feedback, end-to-end viability, and risk reduction before scaling.

Scope

Includes:

• Early system validation techniques
• Incremental system shaping at project start
• Technical and product risk reduction in inception phase

Excludes:

• Architecture structure patterns (e.g., Vertical Slice Architecture, Hexagonal Architecture)
• Team/workflow processes (e.g., Scrum, Kanban)
• Deployment and operations concerns

Proposed Patterns

• Walking Skeleton
• Tracer Bullet
• Thin Vertical Slice (also used post-inception; primary use case is inception phase)
• Spike Solution
• Minimum Viable Product (MVP) (framed as inception-phase risk reduction, not general product strategy)

Overlap Analysis

Pattern	Closest Risk	Clear Separator
Walking Skeleton vs Tracer Bullet	Both end-to-end	Skeleton = production-capable; Bullet = exploratory
Thin Vertical Slice vs Vertical Slice Architecture	Both "vertical"	Slice = delivery technique; Architecture = structural style
MVP vs Walking Skeleton	Both "minimum"	MVP = market demand validation; Skeleton = architecture validation

Open Questions

• Should MVP be included here or separated into a Business/Product strategy category?
• Should Thin Vertical Slice be a dual-category entry (inception + development practices)?

References

• Alistair Cockburn — Agile Software Development (2001): https://www.pearson.com/en-us/subject-catalog/p/agile-software-development/P200000007027
• Andrew Hunt & David Thomas — The Pragmatic Programmer (20th Anniversary Ed.): https://pragprog.com/titles/tpp20/the-pragmatic-programmer-20th-anniversary-edition/
• Kent Beck — Extreme Programming Explained (2nd ed.): https://www.pearson.com/en-us/subject-catalog/p/extreme-programming-explained-embrace-change/P200000007477
• Eric Ries — The Lean Startup: https://theleanstartup.com/

[raifdmueller]

Thanks @cpoepke for the very thorough analysis with the overlap matrix — this is quality work. The 5 proposed anchors (#420–#424) are all valid, well-established, and definitely belong in the catalog.

However, I'd like to decline the new category "System Inception Patterns" and instead sort the individual anchors into existing categories. My reasoning:

1. Our categories are structured by concern, not by project phase — we have Testing, Documentation, Architecture, Development Workflow, etc. "Inception" describes when you apply something, not what the concern is. A category based on temporal phase would be an orthogonal axis.

2. 5 anchors is thin for a top-level category — our existing categories typically hold 8–15 anchors. A category with 5 entries risks fragmenting the taxonomy.

3. You identified the overlap yourself in your Open Questions: MVP sits in product strategy, Thin Vertical Slice is dual-purpose. When half the proposed patterns don't cleanly fit, that's a signal the category boundary isn't stable.

4. The anchors already fit existing categories well:

   • Walking Skeleton, Tracer Bullet → software-architecture (architectural validation technique)
   • Thin Vertical Slice, Spike Solution → development-workflow
   • MVP → strategic-planning (product hypothesis validation)

This gives each anchor a stable home without needing a phase-based axis that would then logically require "Maintenance Patterns", "Scaling Patterns", "Sunset Patterns", etc.

I'll comment on #420–#424 individually and proceed with PRs placing them in the existing categories. Happy to discuss if you disagree — the anchors themselves are great additions regardless of where they land.

[raifdmueller]

All 5 proposed anchors are now live in the catalog via #425, sorted into existing categories by primary concern:

• Walking Skeleton → software-architecture ([Anchor Proposal]: Walking Skeleton #420)
• Tracer Bullet → software-architecture ([Anchor Proposal]: Tracer Bullet #421)
• Thin Vertical Slice → development-workflow ([Anchor Proposal]: Thin Vertical Slice #422)
• Spike Solution → development-workflow ([Anchor Proposal]: Spike Solution #423)
• MVP → strategic-planning ([Anchor Proposal]: Minimum Viable Product (MVP) #424)

The proposed System Inception Patterns category itself is declined (full rationale in the earlier comment). Thanks @cpoepke for the thoughtful proposal and the thorough overlap analysis — the content is great, we just disagree about the taxonomy axis. (Ready to close as 'not planned'.)