The blurring of organizational boundaries between departments, corporations, and even countries in the post-industrial world has rapidly increased the need for information sharing and communication. Business managers, aware of costs associated with inadequate interoperability and related inefficiencies and lost opportunities, are calling information to flow smoothly within and between their organizations across time zones and geographic boundaries. This results in increased pressure on CIOs and IT departments to make it happen.
How expensive can lack of interoperability be? The real costs are often hidden and not easy to quantify. A study of the US automobile manufacturing industry conducted by Research Triangle Institute (RTI) for the National Institute of Standards and Technology attributed $1 billion annual cost directly to poor information interoperability. Another RTI study, of the US construction industry, found that the difficulty of converting files from one computer format to another, and the errors resulting from improper conversion, were major contributors to the staggering $15.1 billion the industry wasted each year because of inadequate interoperability.
Today there is no argument that interoperability is a critical element for optimizing business performance and maintaining competitive advantage. The problem is, how to achieve it. IT buyers have experienced excitement over the latest technology fad, only to find themselves disillusioned as the latest big idea faded.
Let’s have a closer look at what interoperability means. At first we used to think about one computer being able to connect to another. Then came the Internet, with its magical ability to interconnect computers anywhere. But this was like installing a telephone cable between San Francisco and Shanghai, and finding that there was still a need to translate between English and Chinese. So we started talking about the problem of information interoperability. The World Wide Web and XML, which help computers to understand and translate the structure of information, go some way towards solving this problem. However, they don’t help with understanding or translating meaning – this is the issue of semantic interoperability. There is of course a danger here with this name: “Semantics” is a philosophical term, and we shouldn’t spend years arguing about the meaning of meaning. What we want is an engineering approach, backed by standards, and that approach is through the use of a robust IT architecture built upon open standards.
1