ºÃÉ«ÏÈÉúTV

So far, this column has considered the situation in the UK, but I think it is worthwhile considering some ideas from outside the British building industry. For example, a friend who has been working in the Far East on IT development for the design and procurement processes, gave me the following recipe.

Take a good-quality specification and carefully cut it into small chunks. Separate the headings and set aside. Store each chunk in a container and label it with the same name as its original heading. Take the headings and arrange in an attractive and easily digestible form. Place the containers and heading ensemble in a microprocessor and set the controls to create one or more of the following:

• A list of components that will be found in the finished building

• Individual specifications for each component or group of components

• Individual specifications for each trade

• Individual specifications for performance requirement type – fire, acoustics and so on

• The original specification.

The list of components is highly recommended. Not only are specifications attached to each, but also the performance requirements that lead to their selection, potential suppliers,programming, cost and so on.

We discussed the recipe over a drink. I said that it seemed like a good idea to break up information so it could be joined with other information in different combinations, but it all seemed very futuristic.

•  A computer can turn a building spec into an interactive information system
•  The data is virtually 3D, making the transitions between the brief, design, procurement and assembly easier

"Oh yes? Go to your a supermarket and have a look at what's happening there," he replied.

He explained what he meant. With one pass over the bar code at the check-out, the store's inventory is adjusted, market research data is recorded and, in some cases, manufacturers are told what is next for the production line. He suggested that the building industry would do well to acknowledge the role of automated information in the success of modern businesses.

I told him about some of the developments here and asked him what he thought. Surprisingly, we agreed almost immediately that the idea was a significant advance, but should the industry really be satisfied with just providing high-tech filing cabinets? Surely we should be looking for the next step.

We had a long discussion. We talked about how information was most often seen as a separate entity, when in fact the information was the work itself. Work, after all, is really the management or manipulation of information. He told me that designing and building a web application is not so different from designing and building a building. There is a whole set of objects that you can select, adapt and assemble. It started to sound familiar.

"The main thing is to get relationships right so that the result is both functional and attractive," he said. It was very refreshing to discuss this with someone from outside the UK building industry, and although I cannot say I fully understand all the ideas and how they might be implemented, I have set out a synopsis below:

For information to interact it has to be machine processible. There are many schemes for classification of information, but there is one that is universal – the words we use to communicate.

Using the words that represent the physical aspects of a building, all the relationships, constraints, subjective judgements, performance requirements and component details can be collected in a single project file in English. Fortunately, the only information retrieved is that which is pertinent to the task. The data is virtually 3D to begin with, making the transitions between the brief, design, procurement and assembly processes easier to manage.

For a new system to take hold, it needs first to establish itself in an area not well served by systems that exist. This is the information that is generated during the period between the initial brief and the production of technical drawings. Whether design solutions arise from sketching, modelling, thinking, dreaming or whatever, the importance of trapping the criteria on which they are based cannot be overstated. How else can a designer know the real status of the design at any one time? How else can the employer know his requirements will make it through to the end result? The answer lies in distinguishing creative achievement from measurable effort; what only humans can do from what machines could do.