Parannoyed Delusions

Today I am proposing a new diagramming tool to model conversations between people. I will be using a deriviation of UML's Activity and Sequence Diagram and will be referencing Gordon Pask's Conversation Theory. For those who are not familiar with software engineering or cybernetics, this exercise will not be fun and I recommend you not to read on. But if you are interested or think you might be, please do...

This exercise was first thought of after having quite an interesting conversation at the Mamak's yesterday. I noticed (actually I have been noticing this for quite some time already) how conversations tend to wildly "evolve" from one topic to another. So I wondered, how? And why? And decided to play around with this wonderment with a little exercise. Because of my extreme level of boredom at that time, an idea came naturally. And because of my natural geek tendencies to 'nerdify' things, I had a little fight with my common sense...

"Let's try and model these flows of conversation using a diagram," I thought.
What the-?

"I can even use a subset of UML like the Activity Diagram or Collaboration Diagram!"
Why the f**k you want to do that?!

"Ooh, this is going to be cool!"
Sigh.

And so I began my research with highly charged nerd 'semangat' and to my surprise and joy, I stumbled upon a lot of interesting things. Like Gordon Pask's Conversation Theory. This theory pratically answered everything I wanted to know about conversations. It abstracts the conversation into a neat little pragmatic package that can be used to successfully implement a diagramming model. Here's an excerpt from a paper:

A "P-individual" (= participant) originates (by internal conversation) a conceptual procedure which when "applied" (i.e. executed) produces a representation or an action. Another P-individual tries to do the same, if the representations or actions that they produce and display in shared space, are regarded by each other to be about the same, then an agreement has been reached, and the agreed concept can be given one label which both participants can confidently use in further conversation. If however, the productions differ, then the participants need to exhibit the ways in which they are executing their concepts in order to establish a distinction between the two. At this point they become able to assign two different agreed labels. In which case each participant gains a new concept. There is however in CT, a strange recursion underlying such conversational learning, because the participants themselves are taken to be no more than emergents from previous conversational agreements, all executing in some informationally coupled M-individuals (= biological-bodies &/or machines). (See also: Glanville, "Pask a Slight Primer", (1993).

And so based on that theory, I modelled up simple notations for my so-called Conversation Flow Diagram. Here's an example of the CFD that depicts (partially) the aforementioned the conversation I had:

         (1)                           (2)
    Participant 1                 Participant 2

 .-----------------.       .--------------------------.
 | Eating disorder |   X   | Craving fat is a natural |
 '-----------------'   '   | (evolved) human instinct |
                       '   '--------------------------'
                       '
          .- - - - - - '- - - - - - - - .
          '                             '
 .--------'--------.       .------------'-------------.
 | Evolution is    |   X   | Evolution is partially   |
 | total bullshit  |   '   | true                     |
 '-----------------'   '   '--------------------------'
                       '
          .- - - - - - '- - - - - - - - .
          '                             '
 .--------'--------.       .------------'-------------.
 | We came from    |   Y   | Therefore Nabi Adam is   |
 | Nabi Adam       |   '   | an ape                   |
 '-----------------'   '   '--------------------------'
                       '
                       '
                  .----'-----.
                  | Hmmmm... |
                  '----------'

So obviously, there are two people talking. The boxes represent conversation topics. When there's a disagreement between the topics, an X is marked between them and a follow up topic is presented next. When there is agrement, a Y is marked, and the concluding topic is presented.

Quite a remarkable attempt in bullshiting actually.

References:

1. Reflections on the Conversation Theory of Gordon Pask
2. The Conversation of Theories & A Theory of Conversations
3. Conversation Theory: Reasoning about significance and mutuality
4. Unified Modeling Language (UML), version 1.5