• silicon.com
• Management
• CIO Insights

By Brunel University, 19 September 2003 09:30

COMMENT Brunel University's Professor Robert Macredie and Dr Mark Lycett ask how organisations can create living, breathing systems. What are the right approaches? Everyone who’s worked in IT will, at some time, feel that their particular project or system is jinxed. The IT system takes on human attributes. It becomes stubborn, stupid or malign. While this may be just a passing feeling, it’s not so far wrong to equate IT systems with the living. That’s what they are – living systems that evolve according to users’ needs. Previous articles in this series have touched on this issue of flexibility but it’s a topic that deserves a deeper look. Businesses have traditionally adopted static IT models that can take considerable time to change. Legacy systems have been dealt with by creating more legacy systems, mistakes are ‘patched’ and layer is built upon layer, creating an over-complex base for a system that will be extremely difficult to adapt in the future. Looking at the IT system in the long term, it’s obvious that this is a recipe for disaster: half the IT department will be engaged in costly and time consuming maintenance of an over-complicated system. The other half will be working on how to add to that system. Given this, it’s not surprising that companies are finding it harder than ever to adapt with speed to an ever-changing set of market rules. This method of IT development does not always solve the problems it was devised to remedy. Even during the initial process of development, a company’s understanding of its situation and requirements may change dramatically. As large projects are often split up into manageable chunks, each part of the project risks being diverted – leaving the overall strategy far behind. This is where projects stand a very real risk of becoming costly failures. A possible solution to both of these problems – and it’s one that was mentioned in the very first article of this series - is to ‘componentise’, building small IT systems quickly and in a modular system so that these systems can be integrated into a larger, cohesive whole. Implementing IT on a small, modular scale will help ensure that objectives can be tightly monitored and that the system produced truly addresses particular pain points within the business. The obvious key to the success of this approach is integration. Ensuring that these small, flexible systems integrate with each other will mean that the system can still work within the framework of an overall strategy. This ideal would produce a fluid system that grows and evolves easily, rather than a static, monolithic system created in one big bang. So far, so good, but writing the theory is a lot easier than putting it into practice. Changing to this model of IT development will be a major departure for many organisations – and one that will involve a change of mindset from the board downwards. It’s particularly tough because it requires a new way of thinking - businesses will need to begin compartmentalising internal processes into specific services. Only by mapping each business capability clearly like this will the IT department or external consultant be able to translate this into a technical specification – and so design a technical solution that is understood and agreed by both the business and the IT management. Previous articles have discussed communication, and how often problems arise when either users cannot articulate what they want, or they don’t have an understanding of what it is they want in the first place. Breaking a business down by services offered rather than by activity and tasks may help to avoid this common stumbling block – it’s certainly a different approach. An important point to consider here is that, with this high level of user input, the ‘living system’ that is beginning to take shape should be your system – it should not belong to your suppliers. Any IT director responsible for developing a modular based solution must ensure that their business owns the underlying IT architecture for each module in case there is a need to change suppliers mid way through any particular project. It sounds obvious but disputes over ownership of the developed source code can lead to projects being re-started from scratch – with the business incurring enormous costs. This linking together of small specialisations within a business and its IT systems will mean that the IT systems have a better chance of functioning like a business ‘brain’ – working as a whole, while retaining the capacity to function independently if one section requires change. It should also be much easier to add to it as new requirements emerge – and less painful to jettison parts that have become unnecessary. With a different way of thinking and the right level of input, the living IT system that you create should last longer and support, rather than get in the way of, your business. Professor Robert Macredie and Dr Mark Lycett are leaders of ‘Fluid Business’, a groundbreaking research project within the Department of Information Systems and Computing at Brunel University. They can be contacted by emailing FluidBusinessTeam@fusepr.com. Or email us with your opinions at editorial@silicon.com.