An idea I like is to take the diagram into a third axis, along these lines:
First: Reposition the two arrays so that they're loosly stacked (shifting to a 3-d perspective drawing in that animation. Then, take the addition operator duplicate into (*/ @ $) copies that spread from the original into a matrix the same size as the nouns. This verb matrix would be produced between the two noun layers described above. (They were described as "loosly" stacked so that there is natural room for this third layer to be drawn.) The diagramming of the verb as existing for each atom of each array seems to me a natural visual representation of the rank-zero relationship. (Having a visual representation of rank relationships does not require that they be explained when first shown, yet will provide a visual mnemonic for eventual discussion.) Resolution to the result might be shown by drawing the results in a new (fourth) parallel panel. Or the animation could show a convergence of the two noun panels onto the infixed verb panel, with all of them replaced by the result noun at collision. (The latter appeals to me.) This technique would allow the addition to be shown as conceptually parallel, avoiding inaccurate implications of sequencing. It would also take advantage of the multi-axial thinking that J involves. -- Tracy On Tue, Feb 23, 2010 at 11:59 AM, bob therriault <[email protected]>wrote: > Thanks Tracy, > > I wrestled with this as well. The options I explored were: > > 1) ... > ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
