A bottom up approach to design

Looking at figures 6.4 and 6.7, it would seem to be illustrating a complete Web presence solution that has been broken up into different sections to create the components. This process is known as decomposition and is associated with top down approaches to object oriented design.

The idea of a top down design approach is that you carefully plan out what you want to achieve and the way you are going to go about getting the desired results. Costs and time scales are entered into his plan to produce a model that is then used to monitor and control the progress of the construction. By splitting the project up into different components – object oriented design (OOD) – different parts can be given to different specialist teams so that they can work simultaneously on the same large project without getting in each others way or conflicting with each other's designs.

This top down approach to object oriented design is used by most large companies and major contractors because it facilitates efficient organisation and control as well as meeting the requirements of most traditional funding or investment bodies.

Most people's idea of object oriented design doesn't extend beyond this top down approach: with its decomposition of a master plan into components. Unfortunately, a top down approach is totally unsuited to the volatile environment of the Internet environment because rapidly evolving technology and fast competitive reactions often cause master plans to be out of date even before they are finalised and approved.

The only realistic way to work in such a volatile environment as e-business is to use object oriented design techniques with a bottom up approach. With this approach there is no master plan; solutions are grown, one component at a time. Each component is integrated sequentially into an expanding whole: added separately to a dynamic design that is constantly adapting and evolving.

To the business mind set of the Industrial Age, such an approach is unthinkable because without a master plan such a system would seem to be disorganised and liable to spin out of control. Yet, in the volatile and unpredictable environment of e-business, the reverse is true. It is the master plan that is liable to rapidly become unstable and the gradually evolving system that is most easily controllable.

This state of affairs throws many conventional business into total disarray. This is typified by the following post that was sent in by one of the readers in the virtual cafe after reading the last chapter:

I am head of marketing for a fast growing young company that trains business people in the management, marketing and strategy issues of the InternetÉ

I've just finished reading chapter 5 of Web Presence. It's fascinating for me, as it is moving so close to problems and issues I come across every day. Previously, my attempts to interest my work colleagues in the ideas contained in The Entrepreneurial Web - which I was very excited by - were hampered by the perception that you were engaging with problems at a very conceptual level, and the company prides itself on providing help and advice of a very practical 'real-work' kind.

My own limited experience of managing an e-business project - when I was put in charge of project managing a piece of development on our own Web site - has shown me how completely on-the-money your views are about the efficacy of managed teams and forward planning. The team kept asking me for my vision of what the finished site overhaul would look like, and I kept refusing to provide it.

I told them that any vision we had now would be completely out of date in 3 week time, and that instead we were going to evolve the site from where it was at that point towards something we would all feel was a site we wanted to use.

When I started talking about green frogs they looked at me as if I were mad. However, despite serious doubts being cast on my sanity, I succeeded through this method in opening our membership area and getting the site - which contains some 1000 pages - more or less completely up to date for the first time in a year.

My successor in the hot seat has adopted a more traditional approach. He currently faces familiar problems of overruns, loss of key members of his team and the consequent non appearance of deliverables within the original timeframe.

This post illustrates a dilemma common to many people at the cutting edge of e-business solutions. Their professional position forces them to hold conventional views, but, their personal experiences tell them that these ways aren't working. It's a catch-22 situation: if you propose or advise a strategy that is unacceptable you lose credibility – yet, if you propose or advise on a strategy that is acceptable you know it won't work.

The only way around this impasse is to go back to basics: go though the steps of conventional decision making processes to see where they might be going wrong. Isolating e-business problems in this way. Not only might find solutions but also lead to discovering unexpected e-business opportunities.