“Metaphor, far from being just a figure of speech, is actually central to everyday communication and learning.”
The evolution of computing technology is characterised by the continual desire to move away from concrete machinations towards human-centred services. Business, application, infrastructure and data architectures have all moved towards models that abstract the physical manifestations upon which they are built and provide, instead, higher level functions that in many cases were barely conceivable a few years ago.
Like any scientific pursuit, the path of IT discovery never ends. It is a journey characterised by a continual need to release ourselves from the shackles of presumption and past experience. Our imaginations are limited and we appear to be capable of taking only small technological steps forward at any one time. To cope with these cognitive limitations we routinely use metaphor to link new concepts with familiar ones. We talk of filing cabinets and desktops limiting ourselves to the world of concrete artefacts.
Yet as computer scientist Edsgar Dijkstra, creator of the shortest-path algorithm for programming, claims: computing has presented society with such a radical novelty and sharp discontinuity that our attempts to use metaphor and analogy to model and explain computing concepts is doomed to failure!
In cyberspace the only real limitations are those we create ourselves. Nevertheless, unshackled by the limitations of worldly form and substance, we still return to old and familiar concrete objects in an attempt to describe new concepts and information structures. Why is this so?
Well, as contemporary research shows, metaphor, far from being just a figure of speech, is actually central to everyday communication and learning. It has even been argued that without recourse to metaphor it is impossible for humans to carry out abstract thinking. The key then is not to throw it away but to select and apply metaphors with care and consideration—we need to steer away from metaphors that are based on restrictive, concrete sources as these limit the potential usefulness of computing to those features available within archaic physical structures such as filing cabinets, clock faces and so on. Instead, we should select metaphors that are based on flexible and expandable concepts.
We cannot escape the use of the metaphor, yet we still have much to learn to harness the power of metaphor in new domains such as computing. Metaphors are needed because we can only advance incrementally. We can only discover a new layer of technological ability once we have mastered the last one. However, in order to keep making progress we must keep an open mind to the possibilities that still exist. We are babes in the woods when it comes to exploiting the full potential of computing and it is with this attitude that we must investigate emerging technologies and concepts and evaluate their implications for business.
Service-oriented architecture, or SOA, falls into this category and is the subject of one of our articles in this issue. A highly loaded convolution of metaphors, the term is likely to generate strong feelings even before the substance of the concept is investigated. One thing is certain: SOA is generating a lot of interest. The concept of service is one that is close to the hearts of many business people. In fact, a consultant working for a leading vendor recently told me: “We don’t have to generate SOA work. The interest in the market is already there. Unlike object orientation, business people understand services.”
Perhaps SOA represents our best hope yet for aligning our IT activities to the needs and interests of the business. Isn’t that where we’ve been trying to drive IT for decades?
Gerald Khoury lectures in IT at UTS and is managing director of Mackerel Sky, a company that provides independent consulting in enterprise systems management and architecture. He can be contacted at gerald@mackerelsky.com.au
Write to the Editor at Technology and Business
* All fields are mandatory.