Bottom Up object oriented design

The idea of object oriented design, splitting a project up into communicating modules, is reasonably easy to understand if it is used with an overall, formulated plan. However, as we saw in the last chapter, with the extreme conditions of uncertainty and competition in the Information Age, using a formulated plan will produce severe "Christmas tree light" problems.

The only way round this is not to use a plan at all, but, to use a bottom up strategy where a design is grown upwards without any planned way for it to grow. This involves starting with a single module then creating new modules, one at a time to add on to the growing structure. Each new module is designed, tested and completed in a working environment before going on to start the next. In terms of Christmas tree lights, this is equivalent to constructing a string of lights one light at a time. At each addition, a check is made to ensure that they are all working properly together. Only when this partially built string is seen to be working properly is another light added.

The full set of modules is not decided beforehand. There is no overall plan. However, designing without a plan is not a concept that is readily accepted by the psyche of the Industrial Age: it is the antithesis of all corporate thinking. This was apparent in my dealings with the large telecom company and their academic advisors from the university. The idea of working without a plan seemed to them to be completely crazy.

It does seem crazy, until you realise that a plan, however sensible, cannot take into account any unknowns that crop up, such as emergent effects that appear only when a structure is in use. No plan can plan for these, so, you have to have a strategy that allows for these unknowns to appear singly, leaving you room to deal with them before too many get together to make the problem of sorting them out too complex.

Not everyone though is confused and disoriented by bottom up strategies where you create a design without a plan. Truly creative people grab hold of the concept immediately because to them it is just plain common sense. An artist, working in oils may start out with just a hazy idea of what he or she wants to paint and let the painting evolve into a work of art as inspirations take hold during the painting process. Often, the results are as much of a surprise and delight to the artist as they are to the viewing audiences.

A writer might start with a single character: placing the character into a situation, then, by giving the character random personality traits visualise how the character might react. The interaction will prompt other randomised characters to be introduced and their interactions trigger ideas for a story line. The writer might well begin with only a starting place for a plot and then let the story develop by itself. Many writers have expressed the mystical nature of this process, explaining that the characters seem to acquire a life of their own and in some mysterious way the story tells itself.

All this is anathema to the corporate minds of the Industrial Age. They cannot imagine a world where you just set the workers free to go off and do their own thing. But, there again, it would seem ridiculous to anyone: until they click onto the bottom up paradigm.

Chris Heape, commenting on the above in the virtual cafe, put his finger on the problem when he wrote:

If one just takes the artist for a moment, at his canvas. To a certain degree you're right, Peter, but you also have a very romantic idea of what it is to stand and paint that wretched canvas. It looks at you every morning, in it's virgin canvas colour, grinning at you until you make a move. Things move along, and mark after mark, brush stroke after brush stroke is applied and one is reasonably satisfied that one is "attempting to reach that goal that one set in the beginning".

But as the whole canvas gets more complicated with each new brushstroke, one is in an *intuitive dance* with the canvas and oneself, balancing each new stroke with THE WHOLE CANVAS of MARKS. Each stroke affects the rest of the picture. No stroke is something in itself. This is a scenario I have obviously been involved in many times and I know is what many other artists have battled with. It is the balancing of many entities into a very subjective harmonic whole.

Chris Heape

The way Chris sees the painting of a picture it does seem to be a frustrating and perplexing process, but, this is the very way many people view the creation of an e-business or e-commerce solution. There are too many alternatives to choose from. The possibilities are endless, hardware and software is evolving even as you try to work out the solution. Every new item added increases the complexity. There seems to be no way to logically approach the problem of design.

Yet, the way out of this dilemma is simple, once you click upon the right paradigm. It needs a trick way of thinking that allows you to approach the problem from a different perspective. In this case, the trick is to see the artist as three people: 1) a creator of initiatives 2) an expert craft worker; 3) a decision maker. It is only when these three different roles are separated out that the problem seems possible to solve.

Seeing the painting of a picture as a team effort where creative initiators and craftsmen are trying to satisfy a decision maker can break the problem up into three neat categories that can be dealt with separately. The enigmatic problem then comes down to determining what is the basis of the decision maker's decisions.

Once you get into this area you then have to start asking questions about the purpose of the artwork. Is it to please the decision maker, or, is it to please the people who will see the finished work: the viewers? If it is being designed to please the viewers this immediately puts a restraint on the decision maker's vision. The vision then becomes the decision maker's interpretation of what the viewers would like to see as a vision.

This shifts the role of visionary over to the viewers and the role of the decision maker then becomes one of interpreting their requirements. The decision makers role thus changes from a visionary to a communication specialist who communicates the requirements of the viewers to the craftsmen and initiators.

This begs the question: "Who is now the designer? Is it the initiators and craft workers? Is it the decision maker" Or, is it the viewers? It is how this question is answered that lies at the heart of e-business and e-commerce solution design.