Formal Question: Where Does "Portion of Reality" Live in BFO, Especially for Relational and Compound Targets?

[Alican Tüzün]

I am writing to raise a concern that emerged while working with the Information Artifact Ontology (IAO), specifically the paper "Towards Foundations for the Information Artifact Ontology" (Smith & Ceusters).

Problem 1: POR Cannot Be Placed in BFO

When the term 'target' refers to a portion of reality (POR), I could not categorize it under any BFO categories. The fact is that POR is not just an object aggregate, but also includes relationships, time, and space. As it is a portion of reality.  Is POR simply informal shorthand for BFO: Entity? Or is it pointing to a genuine gap?

Problem 2: The RU → POR Relationship

Setting aside the POR categorization problem momentarily, a second problem emerges at the level of Representational Units (RU). The paper defines:

RRU (Referring RU)  ──is_about──→  POR
NRU (Non-Referring) ──fails──→     ???

I am struggling with such issues in the business domain. E.g., a compound term "business model", or some other very vague and ambiguous terms that usually denotes some POR.

But what is the formal nature of the is_about arrow itself? Is it a relation between two BFO entities? If POR is not a BFO entity, the relation is formally ungrounded.

Problem 3: Compound Terms in Applied Domains

This creates an impractical situation in domains like business. Consider the term "business model." Its target POR is not a single entity; it is a configuration of material entities, processes, qualities, and ICEs at a specific time for a specific firm.

Therefore, this configuration is:
Not a Material Entity
Not a Quality
Not a Process in isolation
Not an Object Aggregate in any straightforward sense

Yet practitioners use this term to refer to something real. Under the current BFO/IAO framework, where does this POR live? And if it has no formal place, does that mean "business model" is permanently an NRU?

I would very welcome the group's thoughts on the status of POR and formal introduction, or whether existing categories can be shown to be sufficient with appropriate argumentation.

LG,
Alican

 

[Werner Ceusters]

Alican,

Good questions, but I don't see any problem.

P1) A POR is a part of reality, carved out in whatever way. PORs that are occurrent (op) or continuants particulars (pc) fall under BFO. PORs that don't, do not fall under BFO. But the op's and oc's that are components of such PORs can be described in BFO terms. If you need things of the POR sort, you can define them in a separate ontology (say 'PORO') and create bridging axioms with BFO.

P2) In the paper cited, is_about is defined as a relation between a representation (i.e. a quality) and a POR. So you can use it in BFO for any POR that you can represent in BFO. For other PORs, you can use the relation in PORO.

NRUs are representations that are not veridical, so there is no is_about that can have them in domain position. 

If ''business model' is a vague term, it is a terminological problem, not a problem of BFO.

P3) If you hold that 'business model' is not vague, but designates a POR that cannot be represented in BFO, represent it in PORO. It seems however that the term is mostly used to designate plans, i.e. a subtype of representation. And that such plans tend to be concretized in processes of certain sorts and in parts of which continuants such as the company, its employees, its equipment, ... participate, while thus realizing the company's disposition to do business.

I hope this helps.

W