un

guest
1 / ?
back to lessons

Two Solos, One Gap, One Engine

The ML Triangle has four components: your specification, their specification, an ML engine, a gap to close.

Unturf provides infrastructure for certain gaps: education, code execution, data distribution. Curriculum authors hold one specification. Students hold the other. The ML engine sits between them, classifying responses & generating adaptive feedback. unhomeschool is one lesson-shaped instance of this pattern.

But the pattern predates unturf. It applies anywhere two solos carry complementary domain depth & a real gap separates them. A nutritionist & a patient navigator. A machinist & a design engineer. A village elder & an oral historian. A bookkeeper & a cash-strapped cooperative. Any of these pairs can form a triangle.

The question to ask is not: does unturf have a child entity for my domain? The question is: who holds the specification on the other side of my gap?

The four-part check

Before forming any triangle, verify all four components exist:

1. Your specification: you hold domain knowledge deep enough to articulate what a correct answer looks like, what a partial answer looks like, & where the boundary falls between them.

2. Their specification: someone on the other side of the gap also holds domain knowledge. They know what they need. They can recognize when they get it.

3. A real gap: the two specifications do not currently meet. Distance, language, cost, format, or access prevents them from connecting.

4. An ML bridge: a classifiable operation exists between the two specifications. The engine can receive input from one side, process it against the other side's specification, & return something useful.

If any component is missing, the triangle does not form. If all four are present, it forms the same way regardless of which capital form the domain produces.

Identify the Triangle

Apply the four-part check to a concrete scenario.

A solo builds curriculum for home electricians. Another solo runs a tool inventory system for contractors. Describe the gap between them & sketch the ML bridge that would close it.

Tacit Knowledge in Bodies & Land

Living capital holders, nutritionists, farmers, herbalists, midwives, physical therapists, carry specifications built from direct observation: bodies, soil, plants, birth outcomes, recovery patterns. Their knowledge accumulates in handwritten notes, memory, & pattern recognition sharpened over thousands of cases.

Material capital holders, machinists, fabricators, land stewards, builders, carry specifications built from physics & craft. A machinist who has run 10,000 tool paths knows tolerances that no manual covers. A builder who has framed 200 houses knows which stud bays will fail before an inspector does.

Both forms resist digitization. You cannot export a midwife's 20 years of birth outcome patterns to a spreadsheet without losing most of what makes the knowledge useful. You cannot encode a machinist's feel for chatter into a tolerance spec. The tacit component is real & it is large.

This resistance to digitization is precisely why ML bridges are valuable here. The gap is not information asymmetry: both solos know things. The gap is translation: the midwife's pattern recognition & the postpartum care navigator's matching logic exist in different formats. An ML bridge does not replace either specification. It translates between them.

ML Triangle Capital Map

Typical solos by capital form

Living: nutritionist, farmer, herbalist, midwife, physical therapist, community health worker

Material: machinist, fabricator, land steward, builder, repair technician, tool-maker

The complementary solo on the other side of the gap varies by domain. A midwife's complementary solo might be a postpartum care coordinator, a home visiting nurse scheduler, or a doula training program. A machinist's complementary solo might be a design engineer, a materials sourcing specialist, or a tolerance database maintainer.

The gap between them is always the same shape: deep domain knowledge on each side, no shared format in the middle.

Midwife's Specification

A concrete scenario for living capital.

A midwife has 20 years of birth outcome patterns stored in memory & handwritten notes. A solo builds postpartum care matching tools. What would the midwife's specification contain: & what would the ML bridge do between them?

Theory Meets Practice

Intellectual capital holders, researchers, librarians, open source developers, domain writers, hold organized, documented knowledge. It exists in artifacts: papers, codebases, curricula, databases, reference collections. The knowledge is real & often deep. The problem: it exists in a format disconnected from the practitioner who needs it.

Experiential capital holders, chefs, craftspeople, tradespeople, athletes, musicians, hold the inverse: deep practice with minimal documentation. A chef who has made 5,000 sauces holds a specification for flavor balance that no recipe book fully captures. An Olympic rower holds a specification for stroke efficiency built from thousands of hours of feedback on water.

The ML Triangle between an intellectual solo & an experiential solo closes the gap between theory & practice. The intellectual solo's specification describes what correct form looks like in the abstract. The experiential solo's specification describes what correct form feels like in practice. The ML engine classifies concrete performance against both.

This is the unhomeschool structure

A curriculum author (intellectual capital) writes a lesson. A student (experiential capital) attempts an activity. The ML engine classifies the student's response against the curriculum author's specification & returns adaptive feedback. The author never sees the student. The student never sees the author's raw specification. The engine translates.

The same structure works for any domain where theory & practice exist on separate sides of a gap. The ML engine does not need to understand the domain: it needs the two specifications to be precise enough that it can classify inputs against them.

Rowing Triangle

A scenario from experiential capital.

An Olympic rowing coach can identify stroke inefficiency in 2 seconds of video. A sports analytics startup has 10,000 hours of race footage. Describe the ML Triangle that closes the gap between them.

Hardest Specifications to Write

Social capital holders, connectors, community organizers, trust-holders, carry specifications built from networks. They know who knows whom, who follows through, who carries which kind of credibility in which room. This knowledge lives in memory & in the texture of long relationships. It does not appear in any database.

Cultural capital holders, storytellers, tradition keepers, historians, artists, carry shared meaning. They hold the living archive of a community: the stories that encode land use patterns, conflict resolution methods, seasonal practices, intergenerational contracts. This knowledge survives only through telling & retelling.

Spiritual capital holders, contemplatives, chaplains, meaning-makers, carry presence. They know how to hold a person in difficulty, how to create conditions for reflection, how to surface purpose when it has gone quiet. Their specification is relational & contextual in ways that resist explicit articulation.

These are the hardest capital forms to write a specification for. Social capital feels like 'knowing people.' Cultural capital feels like 'knowing stories.' Spiritual capital feels like 'being present.' None of those descriptions is a specification.

What makes these specifications writable

A social capital specification names the trust-structure of a network: who vouches for whom, on what basis, in what domain. The ML bridge does not replace the trust: it routes introductions to the right vouching path.

A cultural capital specification names the pattern vocabulary of a tradition: the recurring themes, the canonical examples, the stories that encode which values. The ML bridge does not replace the storytelling: it surfaces relevant stories for a given question.

A spiritual capital specification names the conditions & practices that reliably generate reflection, meaning-recognition, or intention-setting. The ML bridge does not replace the contemplative: it scales the scaffolding.

The articulation work is hard. But a solo who can write these specifications becomes a bridge for anyone trying to serve that community: without extracting the community's capital in the process.

Village Elder's Specification

A scenario from cultural capital.

A village elder holds oral histories spanning three generations: stories that encode land use patterns, seasonal practices, & conflict resolution methods. What would a specification for that knowledge look like? Who is the complementary solo on the other side of that gap?

Three Parts, One Structure

Every specification has the same three-part structure regardless of which capital form the domain produces:

Part 1: What you hold. Your domain, your capital form, your tacit knowledge. Not 'I know a lot about X', name the specific vocabulary you use to classify what you observe. A midwife names her risk categories. A rowing coach names her stroke inefficiency taxonomy. A village elder names her pattern vocabulary for land use & conflict. If you cannot name the categories, you do not yet have a specification, you have expertise. Expertise becomes a specification when it names its own distinctions.

Part 2: Who needs it & cannot reach it. The other solo, the gap, the distance. Not 'anyone interested in X': a named role with a concrete task that requires your classification vocabulary. The postpartum coordinator who needs birth outcome routing. The sports analytics startup that cannot scale without the coach's taxonomy. The land restoration practitioner who needs historical precedent before intervening. The gap must be real: if the two solos can already exchange specifications directly, no bridge is needed.

Part 3: What ML does between. The bridge operation: input format, classification logic, output format. Not 'ML connects them': name the function. Take a narrative birth record, classify it against the midwife's risk taxonomy, return a care-matching signal. Take a video clip, score it against the coach's stroke vocabulary, return an efficiency rating. Take a practitioner's question, match it against the elder's pattern vocabulary, surface relevant stories. The bridge is a function, not a relationship.

ML Triangle Specification Structure

Capital form changes the vocabulary: not the structure

Living capital specifications use biological & observational vocabulary: health indicators, risk markers, care thresholds.

Material capital specifications use physical & technical vocabulary: tolerances, material properties, process parameters.

Financial capital specifications use flow vocabulary: runway, burn rate, invoice categories, payment timing.

Intellectual capital specifications use knowledge-structure vocabulary: curriculum scope, topic dependencies, assessment criteria.

Experiential capital specifications use performance vocabulary: form categories, error classifications, mastery thresholds.

Social capital specifications use trust-network vocabulary: vouching relationships, reputation domains, introduction paths.

Cultural capital specifications use pattern vocabulary: recurring themes, canonical examples, encoded values.

Spiritual capital specifications use conditions vocabulary: practices that generate reflection, intention-setting triggers, meaning-recognition patterns.

The vocabulary changes. The three-part structure holds.

Write Your Own Specification

Apply the structure to your own domain.

Write a draft specification for your own domain using the three-part structure: what you hold, who needs it, what ML does between. Be specific: name the capital form, name the gap, name the bridge operation.