The use of ‘is boundary of’ in several definitions.

[cameronmore]

In several definitions, the phrase ‘is a boundary of’ is used, which is not an object property.

One-Dimensional Geospatial Boundary = A Fiat Line that is a boundary of some Geospatial Region.

Government Domain Border = A Geospatial Boundary that is a boundary of some Government Domain.

Rather than change the phrase to be ‘is a spatial part of’, I am thinking this is a good candidate for a new term.

My first attempt:

x is boundary of y = x is an instance of a Fiat Line or Fiat Surface && y is an instance of a Site && x is is a tangential part of y.

I like the direction the direction this definition goes, but wonder whether it’s possible to add a clause to the effect that ‘any thing on one side of x is located in a spatial part of y and on the other side is not a spatial part of y.’

@alanruttenberg might have some thoughts on this

[alanruttenberg]

Boundaries don't always seem to belong(be part of) to a single thing. For example, consider a sea wall. Does the boundary belong to the wall or the sea. Or consider a fiat boundary that divides an apple into two fiat parts. Which half is the boundary part of.

So one direction you could go to address this would be to say a boundary could be part of two things. In the cases I give, the boundary belongs to both and they overlap.

The examples also push against your definition, which only allows boundaries delimiting sites. Maybe specialize and define site boundary so as not to rule out those other cases. But even with sites we get the same problems as I list above. I drill a hole in a piece of wood. Is the boundary part of the wood or the hole?

Another direction would be to say that there's always two boundaries involved where it seems one boundary is, with each a part of the the thing on one side or the other.

For temporal regions, we have that the first time point can be a part of a temporal region or not, leaving that to the user. first and last instant are boundaries in spirit.

When I've tried to sort this out I've leaned towards a definition that talks about separating two things - that a boundary is the sort of thing, that anywhere along it's length, there are two different things one on each side of the boundary. Even that gets strained if you have a three or more way boundary. E.g. at Four Corners, the states of Utah, Colorado, Arizona, and New Mexico meet.

In the ontology I built for work I have

'has immaterial entity host': Relates a site or boundary to the entity relative to which it is defined
Maybe it should say to "a entity" rather than "the entity" to allow for there to be more than one thing relative to which the immaterial entity is defined. The range is independent continuant, so for a looser sense of your site boundary you could use a relation like this relating the boundary to the site.

[arazdow]

> Does the boundary belong to the wall or the sea.

Poetic, but one wonders how this affects intended uses of the CCO. We have open and closed intervals in number theory to stand on where the boundary is in a substrate mapping to the reals. From: Alan Ruttenberg ***@***.***> Sent: Monday, January 19, 2026 7:41 PM To: CommonCoreOntology/CommonCoreOntologies ***@***.***> Cc: Subscribed ***@***.***> Subject: Re: [CommonCoreOntology/CommonCoreOntologies] The use of ‘is boundary of’ in several definitions. (Issue #746) alanruttenberg left a comment (CommonCoreOntology/CommonCoreOntologies#746) <#746 (comment)> Boundaries don't always seem to belong(be part of) to a single thing. For example, consider a sea wall. Does the boundary belong to the wall or the sea. Or consider a fiat boundary that divides an apple into two fiat parts. Which half is the boundary part of. So one direction you could go to address this would be to say a boundary could be part of two things. In the cases I give, the boundary belongs to both and they overlap. The examples also push against your definition, which only allows boundaries delimiting sites. Maybe specialize and define site boundary so as not to rule out those other cases. But even with sites we get the same problems as I list above. I drill a hole in a piece of wood. Is the boundary part of the wood or the hole? Another direction would be to say that there's always two boundaries involved where it seems one boundary is, with each a part of the the thing on one side or the other. For temporal regions, we have that the first time point can be a part of a temporal region or not, leaving that to the user. first and last instant are boundaries in spirit. When I've tried to sort this out I've leaned towards a definition that talks about separating two things - that a boundary is the sort of thing, that anywhere along it's length, there are two different things one on each side of the boundary. Even that gets strained if you have a three or more way boundary. E.g. at Four Corners, the states of Utah, Colorado, Arizona, and New Mexico meet. In the ontology I built for work I have 'has immaterial entity host': Relates a site or boundary to the entity relative to which it is defined Maybe it should say to "a entity" rather than "the entity" to allow for there to be more than one thing relative to which the immaterial entity is defined. The range is independent continuant, so for a looser sense of your site boundary you could use a relation like this relating the boundary to the site. — Reply to this email directly, view it on GitHub <#746 (comment)> , or unsubscribe <https://github.com/notifications/unsubscribe-auth/AABY3JSTHSAYLNDNLN4JNB34HV2RJAVCNFSM6AAAAACGB4VX4GVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTONZQGU2TONRZGU> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>