On 11 May 1995, Barry Mallis wrote:
> At my company we have incorporated a writing system in document generation
> for our ISO 9001 Quality Management System. It's called Information
> Mapping.
>
> One of the principles of the writing technique is based upon studies which
> show that seven, plus or minus two, is the number of items the brain can
> retain in a mesage. While there are always exceptions on either side of
> the bell curve, we have found that the use of 7 + or - 2 bullets, or
> points, or items or whatever in the body of a document works effectively.
> I wonder if this is in any way related to the algorithms you mention.
>
> Regards,
> Barry Mallis
> Total Quality Resource Manager
> MARKEM Corporation
> Keene, NH 03431
> bmallis@markem.com
Barry, I'm glad you brought up the George Miller concept of the "magical
number seven", because it gives me an opportunity to mention my paper
identified as follows:
John Warfield (1988), "The Magical Number Three--Plus or Minus Zero",
Cybernetics and Systems 19, 339-358.
The basic idea is this: if you take three concepts, say A, B, and C, and
then you imagine them in all possible combinations, i.e., (A,B), (A,C),
(B,C), and (A,B,C); lo and behold, you have seven things: the three
originals and their interactions in combinations: whereupon, starting
with three things, you have used up all of your liberty (as set forth by
Miller) in dealing with those three plus their four interactions.
If you move on to four things, you find that the total of 4 plus 11
varieties of interaction give you 15; well beyond Miller's magical number.
If you then say well, in case of interactions, we'd better stick to three
things instead of seven, you may encounter what has been called "Peirce's
remarkable theorem" that all complex relationships can be constructed by
using only monads (which you start with), dyads, or triads.
Let's then go theological and say that humans have been limited to the
ability to deal with (at any one time) at most three things and their
interactions while, the good news is that with sufficient
"Peirce-a-verance", you can gradually construct the most complex
relationships with that elementary capacity.
-- JOHN WARFIELD Jwarfiel@osf1.gmu.edu