Not every `contains` edge is parthood

A lot of business software uses relation names like contains, belongs to, part of, included in, or in as if they were close enough to mean the same thing.

That shortcut is understandable in application code. It is dangerous in an ontology.

The recent Keet study work inside Hadto’s ontology program sharpened this point with a part-whole taxonomy that is more practical than it sounds. Structural composition, containment, membership, participation, and constitution can all look similar in ordinary language while carrying very different implications for reasoning, reuse, and operator trust.

For Hadto, that is not a modeling nicety. It is a business-system rule.

The problem with one generic partOf

When a platform flattens many relation types into one inclusion-style edge, it quietly mixes different claims:

• a district contains franchise units,
• a transaction contains order lines,
• a person participates in a workflow,
• a staff member belongs to an organization,
• a product is constituted by ingredients or components.

Those are not the same relation.

Some of them describe structural composition, while others capture location, containment, membership, temporary involvement, or material constitution.

If the ontology treats them as interchangeable, the platform starts inheriting the wrong assumptions about propagation and reuse. A claim that is true of a structural part may not hold for something that merely sits inside a container. Membership in an organization is not the same as being a physical component of it, and participating in an event should not automatically inherit the semantics of what composes an object.

You can still ship software with that confusion. You just make the semantics harder to trust later.

Why this matters in operations

Hadto is building systems meant to help turn employees into business owners.

That only works if the platform can represent operating reality cleanly enough for another operator to inherit it. Ownership, responsibility, escalation, kit composition, workflow participation, geographic coverage, and reporting boundaries all depend on relation semantics.

If those edges are blurred, a system can stay formally tidy while becoming operationally misleading.

A few examples:

• Treated like structural parthood, a location-style relation can make a reporting rule spread farther than intended.
• When membership is modeled like composition, a business can look like it permanently owns people or roles it only temporarily organizes.
• Read as inclusion, participation data can imply a stronger dependency than the business actually has.
• And if constitution is handled like ordinary containment, material and commercial structure get collapsed into one layer.

That is how automation becomes overconfident. The graph looks connected, but the meaning of the connection is weak.

The useful lesson from the taxonomy

The strong lesson from the latest Chapter 6 work is simple: similarity of wording is not enough.

A business ontology should separate at least a few recurring relation families:

• structural parthood: a real component relationship,
• containment or location: something is in, at, or within something else,
• membership: an entity belongs to a group, class, or organization,
• participation: an entity takes part in an event or process,
• constitution: something is materially made from something else.

That distinction matters because each family supports different downstream behavior.

Some relations should support transitive reasoning. Others clearly should not. Some should drive rollups. Others should remain local context. Some imply persistence. Others are temporary or situational.

A platform that wants reusable business infrastructure should name those families explicitly before it tries to automate on top of them.

Why Hadto’s current direction is promising

The useful sign in Hadto’s internal ontology work is that the stack is already moving beyond one generic inclusion vocabulary.

The current layer-zero relation surface now distinguishes families such as shared part, location, participation, membership-like, and constitution-style relations. That is the right direction.

The remaining gap is governance: contributors still need a clear rule for when a vertical-specific relation should reuse one of those families, when it should stay local, and what kind of propagation or audit behavior is actually safe.

That is where the recent backlog work matters. The real task is not to add more edges. It is to keep business meaning from collapsing into a bag of vaguely similar links.

The owner-operator standard

An owner-operator platform should make relation meaning trustworthy enough that people can hand off work without silently changing what the system claims.

That means a business should be able to answer questions like:

• is this a composition relation or just a container,
• does this edge justify rollup or inheritance behavior,
• is this a durable structural claim or a temporary operational one,
• should this relation be reused across ventures or kept domain-local until reviewed.

If those answers are not explicit, they do not disappear. They just move into tribal knowledge and ad hoc code.

That is the kind of dependence Hadto is trying to remove.

The practical takeaway

The next generation of business software will not get leverage just from having more connected data. It will get leverage from having better-governed meaning.

For Hadto, one practical rule follows from this chapter work: do not let every contains* or partOf* label sneak into the ontology as if it carried the same semantics.

A platform that wants to help people become owners needs relation families that match real business structure, not just familiar words.

---

Source evidence used in this note: smb-ontology-platform/docs/plans/2026-03-31-keet-ontology-engineering-progress-tracker.md (2026-04-12 entries), smb-ontology-platform/docs/issues/ONT-017-add-foundational-ontology-contract-for-layer0-relations-and-attribution-modelling.md, smb-ontology-platform/docs/issues/ONT-007-add-rbox-compatibility-and-safe-property-chain-audit.md, and existing Hadto blog posts reviewed to avoid duplicating prior notes on foundational posture, semantic lifting, and RBox governance.