[Date Prev] [Date Next] [Thread Prev] [Thread Next] Indexes: Main | Date | Thread | Author

Re: [ba-ohs-talk] new version of graph model paper posted

At 04:12 PM 4/15/02 -0700, Eric Armstrong wrote:
>The perception represents a change in thinking that is always
>astounding the first time it is encountered. I had a similar insight
>driving back from one of the colloqium sessions one evening.
>I shared that insight in a subsequent message to the group --
>that a "document" was really a collection of nodes (although
>"2-D projection of a multi-dimensional network" is more
>accurate).    (01)

Not even a 2-D projection.  Written text is really a 1-Dimensional 
construction.  As I've written before, http://qwer.org/tglt.html , the 
experience of reading a document would not be much compromised if one had 
read the same information on a ticker tape.  (And a ticker tape is 1D)    (02)

Here is another nice analogy that I've thought of recently to describe the 
process of reading:    (03)

Imagine an experiment with two subjects, both experienced artists, in 
separate rooms.  Subject1 is given a drawing and a camera, and asked to 
reproduce the drawing by moving a non writing utensil over a blank piece of 
paper.  The motions of subject1 are recorded, and the tape given to 
subject2.  Subject2 then has the task of watching the tape, and answering 
questions about the picture.  Also, both subject are aware that this is the 
task before them, and aim to increase the success of the communication.    (04)

The recorded motions on the tape are an analogy to writing.  When reading, 
one slowly builds up in ones mind a picture of what the writer is trying to 
communicate, while holding in memory the content of the whole document that 
came before.  Good writing is done in such a way that the reader Is able to 
hold the whole document in mind, and does not have to go back and reread 
what came before.  To remind the reader what one is talking about, one can 
repeat the content of a previous paragraph.  Similarly, in our imaginary 
experiment, subject1 might repeat some of the strokes to remind subject2 of 
what has already been drawn.    (05)

The point of all this is that a picture is a much more effective way to 
communicate then movements on tape, and that similarly text should be made 
more visual by making the associations between sentences explicit.  And ... 
one way of doing this is to assign nodes to paragraphs, and edges to 
references.    (06)

--Alex    (07)