OK. I thought you were suggesting an alternative approach with the same power, rather than a “just what I need for this problem” approach.
I agree that it is about pulling tools out of a bag, and pragmatic utility is the ultimate arbiter.
+44 750 338 5279
From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx [mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx] On Behalf Of Edward Barkmeyer
Sent: 17 March 2015 23:44
Subject: Re: [ontolog-forum] Endurantism and Perdurantism - Re: Some Comments on Descriptive vs. Prescriptive Ontologies
That is exactly the point. If I am modeling the world that results (by inference) from a plan, I can take that to world to be the (actual) universe of discourse for that reasoning exercise. I don’t need it to be “certain that the plan would not change …”; I mean to analyze the effects of that plan. And I can examine multiple plans, each in its own world. To go beyond that, I am in a Plantinga universe in which some states are Actual and others are imported into that universe as explicitly named ‘possible worlds’ along with their content. If you have a reasoning tool that can deal with that kind of situation, then you are dealing with possibilia in a particular engineering approach to achieving some analytical result. But it has nothing specifically to do with ‘branching time’. I can make similar models of possible current worlds based on limited knowledge and some specific assumptions for each such world. (And OBTW, that is a common phenomenon in ‘intelligence operations’, where the ‘intelligence’ is of a different kind from AI. ;-))
The whole problem, as others have said, is that certain ontologists become wedded to particular philosophies of ‘time’ and ‘possibilia’, and they tend to view alternative models as “lesser” or outright “incorrect”. My point was only that you make a model for a purpose, and you can choose any model of time that is adequate to solve the problem(s) for which the ontology/model was intended. These things are not epistemological. “All models are wrong; some are useful.”
Another of my pet peeves is the insistence of certain persons that any reference to future states requires some kind of modality, or ‘branching time’. There are situations in which you want to examine possibilia, yes. But for much knowledge engineering, I prefer the view that there is a world you take to be ‘actual’ and if there is time in that world, then there is, and some of that time can be in the future.
[MW>] Yes. This works fine if there is only one plan you are considering, and it was certain that the plan would not change and would be implemented on time. How would you handle multiple plans for the way a project might be implemented? Or how would you handle dealing with developing a strategy that worked for multiple scenarios?
+44 750 338 5279