Summary

Two systems developed and evolved independently will share some superficial semantics. Perhaps they will both be concerned with sales orders. They may have been built on different technologies or purchased from different application software vendors. This situation leads many to believe that the technical barriers may be difficult to overcome, but once done, the resolution of the data will be easy. The truth is almost always the opposite: There are technical difficulties, and sometimes they are tricky, but they rarely create major overruns in integration projects. The semantic issues, which are routinely discounted, assert themselves late in the project or even in the post-conversion phase, where they are very expensive to deal with.

Hundreds of billions of dollars are at stake in doing this well, to say nothing of getting companies in a position where they can respond and adapt more easily. This chapter examined this issue in detail, and outlined two strategies for mitigating this problem with the enterprise: the development of an enterprise message model as a reference point for flexible and economic integration, and the use of a semantic broker so that each application would not have to understand the semantics of every other application. For external integration (B2B), we outlined the key role of cross-industry consortia and industry-specific consortia and outlined strategies for dealing with each. In Chapter 13 we examine another technology that is growing in popularity and that is often used in connection with enterprise integration: Web Services.