14.5 Research Pays

There is a mentality that has developed over the years in the computer hardware industry that hardware is the most important aspect of a computer system. Once the hardware is designed, it is tossed over the wall for the software developers to do something with it. This was clearly the mentality of the developers of the IBM PC. IBM thought that building the box was the most important aspect of the project. So pervasive was this mentality that IBM decided to give the operating system component of the system to the "kid in Seattle." That single decision made IBM an also-ran of the computer industry. Microsoft could easily acquire IBM.

The nature of the computer industry has changed. Software is now king. The bulk of the revenues have shifted to the software side of the house. Unfortunately, this news does not seem to have penetrated very far into the upper reaches of the bastions of the hardware companies. Most of these companies invest heavily in basic hardware research. These same companies have no formal software research program. The upper-level management of these organizations came from the hardware side of the house. They clearly think that people are buying hardware. The customer's focus, however, is clearly on functionality. This functionality is provided by software, not hardware.

It is clear that there is substantial return on investment for monies spent in hardware research. This research creates intellectual property. The resulting intellectual property can be developed into a product within the company or it can return substantial revenue under licensing agreements. A case does not have to be made to justify the benefits derived from both basic research and product-oriented research. The case that does have to be made is that investment in software research is equally important.

The state of the software discipline in regard to research funding in the commercial and governmental sectors is abysmal. The need for serious funding in this area is great. Software now drives the control systems for safety-critical applications. The liabilities are far too great to discount the importance of software reliability. The opportunities for revenue enhancement through software are also very great. Someday in the near future, computer companies will proudly show off their new multibillion-dollar software research facilities. Software quality and security will soon be the driving force behind the successful survivors in the competition for the consumer dollar. Those companies that are best able to respond to these needs will be those that have a future. It is clear that we have reached the technological limits of handcrafted software systems. The time has come to make a serious investment in software research to provide a solid foundation for the development of engineered software systems.