ATMS may be a useful concept here. The Knoweledge Engineering Environment (KEE) used them. They were also called viewpoints in the Automated Reasoning Tool. http://aaaipress.org/Papers/AAAI/1986/AAAI86-003.pdf
Date: Fri, 14 Dec 2012 05:59:09 -0500 Subject: Re: [agi] Internal Representation From: [email protected] To: [email protected] I did not express that very clearly. I feel that when we think about something -it is as if- we tend to put boundaries around the ideas that we are using at that moment. But those boundaries are not solid and they are not static. When we realize that some idea-thing has repercussions outside of those boundaries we can include the additional insight into our thinking by either implicitly changing the boundaries or by associating the central ideas with a note about how outside objects of thought may impact or interact with the central ideas. I originally derived this idea by examining how we use logic and how logic might be used if we were expressing our ideas about things (idea things) logically. I realized that we always think about things as if they could be examined ex-situ so to speak but when we need to examine additional ideas that can have an effect or be related to the central ideas we can change the parameters of our consideration without problem or formal declaration. Jim Bromer On Thu, Dec 13, 2012 at 6:00 PM, Jim Bromer <[email protected]> wrote: The use of overlapping transcendent boundaries or viewpoints is way to describe the local relations between ideas that I think we use in when we are thinking. The complexity is that it is a way of ignoring the complexity of actually fitting one group of ideas into the greater collection of ideas. Instead, the transcendent or momentary connection which led to a particular way of looking to a problem will dissipate as we consider the idea from that vantage. We may have a sense of the main paths that led to that vantage but we forget the complexity of possible routes that those paths can lead to even in cases where they can lead to strongly related nodes of knowledge. So we can simplify a problem in the near term but we leave many difficult problems until we start to examine the problem from another vantage. It was just my way of understanding how we use logic (in tight local sets) even when that logic may be related to other sets of logic. If we want to make the initial logical analysis a better model we start looking at the problem from related sets of logical propositions and once we get that worked out then we will start looking to see how the two groups might be related in other ways by looking for other propositions that may relate the parts in a more sophisticated model. Jim On Thu, Dec 13, 2012 at 12:13 PM, Piaget Modeler <[email protected]> wrote: Getting back to "overlapping transcendent boundaries" or "viewpoints". Jim, what do you see as the complexity of using viewpoints for inference and reasoning? ~PM ------------------------------------------- AGI Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-f452e424 Modify Your Subscription: https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-58d57657 Powered by Listbox: http://www.listbox.com
