, , , ,

The other day I was pondering about how to parse reference models, an exercise that readily folded my brain into a pretzel. One issue I often have is to judge what entities are derived from others, and what abstraction level is appropriate for a certain model. As long as the models are concrete I do fairly well, but when I assent into higher stratospheres of abstraction, I quickly feel dizzy. Let’s see: objects are entities. Categories are entities. Are processes also entities? Are processes categories? After all, both possess properties. Hrm …

You can probably appreciate that, in absence of a foundational ontology, one does not get very far, and concepts readily blur and turn out macerated. Since I am not a fan of lukewarm category soup, I foraged Wikipedia for elucidating entries, and after some clicks I found Hoehndorf’s dense but sustaining exegesis of upper-level ontologies. Exactly what I needed!

I perpetually need to exercise my UML skills, and translating text into a model significantly enhances my grasp of its content, so off I went and summarised Hoehndorf’s article in a UML class diagram: upper-layer ontologyThe diagram is rather off the cuff, so yes, there is probably one or the other glitch in this model. Anyway, I thought somebody might find it useful.