On Jan 23, 2009, at 6:44 AM, Ian Bailey wrote: Hi Andreas, ... The big challenge in ontology is to figure out what’s different about a system to any other physical item. ...
Ahem. Allow me to suggest that this is an excellent example of how not to approach ontology engineering. Words like 'system' (and 'organization' and 'context' and many others) which are used in informal language as loose, vague descriptors but have been co-opted by more formal uses as classifiers, are not going to be useful ontology categories. Setting out to decide where their precise edges are is like trying to nail down the ocean. It will never succeed, because there is no fact of the matter to decide what the truth is. (At best, maybe a PatC-ish process of formal voting in a well-behaved body of highly motivated people, all following Roberts rules of order, might eventually come to a decision, but it will be one which satisfies nobody.)
I have in the past made lists of alternative "definitions" of the word 'system', and there were dozens. My favorite is the one given in what was at the time (maybe15 years ago) widely cited as a foundational text in general systems theory: "a system is a set, S, with a distinguished subset, T". In other words, anything at all can be thought of as a system; and this is what one should expect, just as anything at all can the thought of as a "thing" or an "entity" or a "context" or ... These are not natural kind words: they do not partition a meaningful subset out of the universe of entities. Rather, they are words that indicate a kind of intention, what Dennett calls a "stance", towards a thing: it is being treated as a system. But anything can, under appropriate circumstances, be so treated. Even subatomic particles, it turns out, are made of quarks.
Now of course, various organizations use the word in particular ways, with more exclusive intended meanings whose boundaries can be investigated. But different organizations will draw their meaning boundaries in different places, sometimes wildly different. Again, not a good start for creating a useful ontology.
Pat Hayes
------------------------------------------------------------ IHMC (850)434 8903 or (650)494 3973 40 South Alcaniz St. (850)202 4416 office Pensacola (850)202 4440 fax FL 32502 (850)291 0667 mobile
|