The article “After SUBACS” in the October 1983 Submarine Review caused me to reflect back on the experience of over 25 years involvement in the “digital revolution” of the u.s. Navy and to ask the question. “What went wrong?”.
Certainly the promise of great benefits was there. The concepts which we tagged with such impressive names as “Central Computer Complex” and “graceful degradation” were not mere marketing buzz words but were firmly based on what could be achieved through digital technology. The potential benefits they would bring to the Fleet were recognized at the time but somehow got lost between exuberance for the design phase and the realities of the Product.
As a result. we found ourselves going through a series of designs. each one promising more through technology. each falling short in delivering that promise. Is SUBACS another in that series? I think not.
In going back to the early days of NTDS and continuing through the various evolutions of combat systems that occurred first in the surface fleet and later in the submarine fleet. an explanation can be found for what went wrong.
From the perspective of a developer. the evolution of digital combat systems can be divided into three phases. The first phase, beginning with NTDS, was characterized as a learning process or proving ground. The major concern on the part of system developers was how to implement an emerging digital technology into an existing analog combat system. Certainly the potential benefits of digital combat systems over their analog counterparts were recognized at this time. However, the real concern and effort on the part of digital system developers was just to make it work!
The early digital systems did work and much was learned. They also provided certain advantages inherent to digital technology. However, in the final analysis, it is questionable if the benefit to the operating forces was any greater than could have been obtained with an analog approach.
The second phase in the evolution of digital Combat Systems for submarines started about the time of the SSN 688 development. This phase was characterized by a widespread application of digital technology-in the BQQ-5 Sonar, the all digital Attack Center, Ship Control Subsystems, Integrated Radio Room, and others. Virtually every area of the Combat System was converted to digital technology.
Simple control routines gave way to complex operating systems. A host of software development tools were developed–compilers, simulation routines, high order languages. The system design process was formalized through a hierarchy of specifications and design documents.
During this period of progress and maturing, the submarine community developed and deployed a number of sophisticated digital combat systems. Impressive gains in performance and overall system reliability were achieved. In spite of these successes, overall the system fell short of design expectations.
In a shipboard environment they proved to be cumbersome and unfriendly to the user. Maintenance, both afloat and ashore, was time consuming, costly, and frequently required special expertise to resolve. The purported ease of accommodating system growth and incorporating new functions bad not been realized. The result was that in the few short years since the SSN 688 Class and the TRIDENT Class were deployed, the submarine community has been embarked upon the largest effort yet to build a “final” digital combat system. What went wrong?
It would be easy to argue that nothing went wrong; that the events and experience of the past 25 years are a necessary part of progress. Such an argument is probably wrong for it leads to the conclusion that as long as some progress is being made the ways of the past are sufficient for the future.
In the first and second phase of digital Combat System evolution two fundamental mistakes were made by those involved in its development. The first mistake was a failure to recognize that the sole object! ve of a Combat System was to provide submarines with the ability to conduct war. All too often, the means became the end. The challenge of technological inno?ation justified the effort to produce it. Consideration of mission objectives, if considered at all, were typically dismissed on the grounds that technology need not become involved in tactics. The second mistake was that there was no proper assessment of the true state of digital technology which was available at the time, in terms of its ability to fully support the design objectives of the then developing Combat System. The tendency was to assume that what was achievable in principle or had been implemented in certain cases could be applied to the System as a whole.
The net result of these mistakes was a cycle of high expectation followed by limited success. This, in turn, generated an attempt to provide a solution through ad hoc fixes. Ultimately, this led to the realization that a total new design was necessary.
The challenge to SUBACS is to not repeat this cycle. The third phase of the evolution of digital Combat Systems is in progress. Phase three is characterized by a digital technology that has reached maturity in a number of areas.
In the area of hardware, performance has increased while costs have decreased. Reliability is greater, size is less. And this is by several orders of magnitude over the earlier phases.
Software development has progressed from the status of being a black art–practiced by a few specialists–to the point where it is a highly defined, highly automated practice and this practice can be subjected to the same disciplines of management as any other product development.
Significant changes have also occurred in the area of personnel. In the earlier phases, personnel involved with the application of digital technology learned their trade through individual experience and practice as they learned. Now a significant and increasing number of people have been formally educated in the theory and application of digital technology. They are well equipped to address the technical issues of evolving systems.
This maturity of the technological base goes a long way towards alleviating the technical problems experienced in the earlier phases, many of which can be attributed to attempting things which at the time were beyond the state of the art.
This is not to say that SUBACS does not contain elements of technical risk. They do exist. The ADA language, for instance, presents a set of new problems. Distributed Processing and Bug architecture have not been implemented in a real time Combat System to the extent contemplated in SUBACS. Also, the structure and management of the data base necessary to support a distributed environment presents a technical challenge. The point is. that because of the maturity of the technological base. the SUBACS program is in a much better position to address these technical issues and to provide solutions that are general in nature and thus will provide a system design that will accommodate the requirements of the future.
The key then to avoiding mistakes in the application of digital technology is to clearly identify and recognize those areas of risk–areas where there are uncertainties in design. and areas where we are assuming that a technical solution will be available when we need it. These areas must necessarily be addressed and fully resolved before committing the design of the system to development. Only in this way can we avoid the fundamental mistakes that will result in a compromise of the original concept and design of the system
J.A. PETERSON
