My guess is that it would help if we could imagine that we had a primitive called 'frame' in the same way as we have one called 'shape' i.e. $ Then one way to get to the precise meaning of frame is to have a J definition of that ficticious primitive. At a guess, this primitive requires the 'rank' of the cells in order to then return the appropriate frame. If we have an array of shape (x,y,z), and we stipulate cells of rank 3, then the frame is presumably empty? Not sure of this... but anyhow, just a thought...
Enjoy ken -----Original Message----- From: Programming [mailto:[email protected]] On Behalf Of Henry Rich Sent: 17 January 2016 23:59 To: [email protected] Subject: Re: [Jprogramming] Definition: Frame of an argument The terminology I use is an (x by y by z) array of cells, or an array of cells with frame (x,y,z), emphasizing that the frame is a (part of the) shape rather than an array. Henry Rich On 1/17/2016 6:16 PM, Raul Miller wrote: > Hmm... ok, reviewing > http://www.jsoftware.com/help/primer/frame_and_cell.htm 'frame' does > get used that way. > > I was thinking of the frame as having a shape rather than being the shape. > > Then again, since you can think of an array as being (for example) an > (x,y,z) frame of cells, I do not think that my interpretation was > entirely incorrect, either. So I suppose I have gotten myself into a > "much ado about nothing" sort of issue. > > Thanks, > ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
