Remaining Primary Directions Frontiers

Purpose

This document defines the remaining frontier zones in Moira's primary-directions program now that the most recoverable geometry, relation, and target families have been admitted narrowly and validated.

It exists for one reason:

• to stop further expansion from drifting into doctrine that is not yet trusted

This is a research-governance document, not an implementation checklist.

Current Position

Moira now has a substantial primary-directions core:

• major recoverable geometry families
• explicit direction spaces
• explicit time keys
• explicit relation doctrine
• explicit narrow target families
• branch presets
• worked-example and fixture-backed validation on the most recoverable surfaces

That means the next work is no longer "add another branch because software has it."

The next work is:

• frontier ranking
• recoverability judgment
• doctrinal risk management

Ranking Rule

Every remaining frontier should be judged on four axes:

1. source_quality
   • are there explicit, formula-grade sources?
2. mathematical_recoverability
   • can Moira derive a real governing law?
3. implementation_risk
   • how likely is false unification or doctrinal drift?
4. layer_fit
   • should this live in the engine, a service layer, or remain research-only?

Frontier Matrix

Frontier	Source Quality	Mathematical Recoverability	Implementation Risk	Layer Fit	Current Judgment
wider non-Ptolemaic reflected doctrine	low-to-medium	partial	high	engine only if branch law appears	defer
midpoints in primary directions	low-to-medium	partial	high	likely engine only for narrow explicit branch	research_only
mundane aspects as a family	medium in places, but uneven	partial and method-bound	very high	engine only by method-specific branch	defer
wider fixed-star doctrine beyond conjunction	medium	low-to-partial	medium-to-high	engine only if conjunction-first discipline is preserved	defer
wider parallel families beyond current closures	medium in narrow branches, weak globally	partial and method-bound	high	engine only by method-specific branch	defer
neo-converse	medium	partial	medium	engine if governing law is recovered cleanly	research_only
field_plane	low as a single unified doctrine	low	very high	research-only until decomposed law is explicit	last_and_defer

Frontier Notes

1. Wider Non-Ptolemaic Reflected Doctrine

What is known:

• Moira now has a first narrow reflected branch:
  • Ptolemaic zodiacal antiscia / contra-antiscia
• the mathematical substrate for reflection is explicit

What is not known:

• whether other method families admit reflected points by the same directional law
• whether reflected doctrine should remain zodiacal-only in some families

Risk:

• easy to overgeneralize from one narrow branch

Current policy:

• defer until a branch-specific law appears

2. Midpoints in Primary Directions

What is known:

• midpoint mathematics in the repo is strong
• midpoint doctrine in primary directions is historically real in some later software and derived families

What is not known:

• one source-safe governing primary-direction law that makes midpoint targets admissible without ambiguity

Risk:

• very high risk of importing midpoint substrate into primary directions without a real directional doctrine

Current policy:

• research only

The governing research packet now lives in:

• primary_directions_midpoint_family_matrix.md

3. Mundane Aspects as a Family

What is known:

• several geometry families become genuinely distinct in wider mundane aspect doctrine
• Campanus / Regiomontanus and Morinus especially point toward this

What is not known:

• one global mundane-aspect doctrine

Risk:

• extremely high
• this is where many method families stop looking interchangeable and start demanding distinct branch laws

Current policy:

• do not admit generically
• only admit method-specific mundane-aspect branches if their laws are explicit

4. Wider Fixed-Star Doctrine

What is known:

• conjunction to angles and planets is now explicit and validated
• opposition does not currently deserve admission

What is not known:

• whether aspects, wider mundane doctrine, or other star relations should be admitted at all

Risk:

• moderate if the conjunction-first discipline is preserved
• high if widened by folklore pressure

Current policy:

• fixed stars are closed for now on the current recoverable surface

5. Wider Parallel Families

What is known:

• Ptolemaic zodiacal parallels / contra-parallels are real
• Placidian direct and converse rapt parallels are real

What is not known:

• whether additional families deserve admission without fresh governing laws

Risk:

• high risk of pretending the family is more uniform than it is

Current policy:

• closed for now

6. Neo-Converse

What is known:

• the label is live in modern software
• converse doctrine is already an explicit axis in Moira

What is not known:

• a sufficiently explicit and source-safe governing law for admission

Risk:

• medium
• lower than field_plane, but still high enough to require a dedicated research pass first

Current policy:

• research only

The governing research packet now lives in:

• primary_directions_neo_converse_research.md

7. Field Plane

What is known:

• it is a composite ambiguity zone
• it likely decomposes into:
  • space doctrine
  • latitude doctrine
  • projected-plane doctrine
  • naming drift

What is not known:

• whether it is one thing at all

Risk:

• highest in the entire subsystem

Current policy:

• leave for last
• do not implement as a single opaque switch

Engine vs Service-Layer Guidance

Engine Candidates

These may eventually belong in the engine if a governing law becomes explicit:

• method-specific mundane aspects
• a narrow midpoint branch
• a narrow non-Ptolemaic reflected branch
• neo-converse

Research-Only for Now

These should remain research-first, not implementation-first:

• field plane
• any global widening of reflected doctrine
• any global widening of parallels

Not a Service-Layer Escape Hatch

Moira should not move a frontier to the service layer merely because the engine law is unclear.

If the law is unclear, the honest status is:

• research
• deferred
• or rejected

not:

• "hide it one layer up"

Recommended Next Order

If work continues after the current pause, the clean order is:

1. documentation and validation consolidation
2. dedicated research packet for neo-converse
3. dedicated research packet for midpoints in primary directions
4. only then reassess whether any method-specific mundane-aspect branch is recoverable
5. leave field_plane last

Present Declaration

Moira is now past the most recoverable primary-directions families.

The remaining frontier is not "more branches."

It is:

• careful research
• branch-specific recoverability judgment
• and refusal to widen doctrine faster than mathematics can support it