zhengj2007
commented
9 years ago From alanruttenberg@gmail.com on July 14, 2012 11:30:10
Proposed resolution: Eliminate talk of parts of dependent entities.
Larry> This makes me nervous, particularly in the context of information content entities. Aren't printed words part of a printed document copy, which specifically depends on the generic document?
How about: parts of dependent entities must exist in a one-to-one mapping with parts they depend on?
--
Alan> Doesn't work. Does the hue depend on a different part than the saturation?
Although we use parts for GDCs, I think that's probably a mistake and should be fixed in IAO.
In any case, not speaking of it (in this version) doesn't say it can't be done, but doesn't imply it can. If we say it can be done then I think a better explanation has to be given about what that means.
From alanruttenberg@gmail.com on July 14, 2012 14:15:37
Reference says: "As we shall see when we consider the parts of qualities such as color and tone below, the parts of a dependent entity may reciprocally s-depends_on each other. This idea has not hitherto been explicitly recognized in BFO, but it is documented at length in the literature on specific dependence".
FIrst, we haven't recognized parts of qualities. Second, how would we know whether such parts share parts, in which case we would arrive at a contraction because the definition of s-dependence says that it can't obtain between entities that share parts.
Proposed resolution: Eliminate talk of parts of dependent entities.
Original issue: http://code.google.com/p/bfo/issues/detail?id=111