Air University Review, May-June 1980

When is a Computer Like a Pickup Truck?

The Title question has picked up some answers. For example, a computer is like a pickup truck

• When it can be found at every level of the Air Force organization from flight line to chapel.
• When it can be used by persons with a minimum of training.
• When it costs less than $5000.
• When it does generalized jobs like hauling and sorting and specific jobs like working on power and telephone systems,
• When maintenance can be done by a centralized pool, by local contract, or, on a limited basis, by the user,
• And when units from many different manufacturers are found around a base, each type chosen because of its availability or features,

Earlier in this journal I wrote that the micros are coming and would be with us soon.* I predicted that these micros (stand-alone microcomputers built around a single microprocessor chip) would be a common sight in Air Force offices within the next five years. At that, we would be lagging far behind private industry because of the vagaries of the military procurement process.

*Captain Frank J. Derfler, Jr., USAF, "The Micros Are Coming," Air University Review, September-October 1977, pp. 30-35.

Like many who dare to guess the future, I erred on the side of conservatism. We are seeing a proliferation of microprocessor-controlled devices entering the inventory right now. I will leave discussion of the use and implications of microprocessors in weapon systems to those doing the work in that field.¹  

What I will focus on is the neglected area, the computer with the attributes of a pickup truck that is proliferating into flight-line shops and headquarters offices.² Usually coming without much fanfare, sometimes almost in secrecy, and even sometimes paid for out of private pockets, this growing group of invisible computers often is unregulated by the typical automatic data processing (ADP), accounting, or communications experts in the Air Force.³ Should it be?

According to authoritative sources, the Tandy Corporation (Radio Shack) sold over 100,000 of its TRS-80 consumer style microcomputers during 1978.⁴ If sales from all other companies equal this number, then there were about 200,000 microcomputers owned by private individuals in early 1979, and the number is continuing to grow by at least 10,000 a month. We can postulate that these individuals have some rather obvious traits. They are probably well educated, technically inclined, and relatively secure financially, and they may have some relationship with computers in their jobs.

Certainly, many present or future individual computer owners can be found in the military. These people know that while their small computer has a number of the attributes of an "adult toy" (average price about $1000), it can also do many useful things around the house, such as keeping inventories, accounting, cross-referencing, indexing, and otherwise "handling" information (words) and data (numbers).

One day, when faced with perhaps a new and possibly dreary task at work ("Harry, the Old Man wants to know how much the Ops guys spent on TDY trips to Peoria over the last two years."), our uniformed computerist might turn to his machine at home for help. As Harry and others around him begin to see the value of applying some computer power to their primary or additional duties, more home machines may come to the office. As the personally owned machines prove their value, unit funds may be expended to "buy one like Harry's."

Through this and other similar scenarios, the consumer-style microcomputer will find a home in the USAF office or shop next to the hand calculators, typewriters, and automatic coffee makers there already. They will (and did) enter without review by the software compatibility, hardware reliability, and systems analysis boards and committees that claim responsibility for such things. These computers are arriving in response to a perceived need that can be met expeditiously and at a low cost.

What is wrong with that? Well, let's see.

Ten years ago, consultants who were pricing large-scale computer installations used this rule of thumb: 50 percent of the total price for hardware and 50 percent for software. That ratio has changed recently but not because the programming of big time-shared user systems has gotten cheaper, The ratio has changed because hardware, judged by capability, has been reduced to a small fraction of its total price while programming costs continue upward.⁵ The hardware is just jacks for openers in the computer game. Software now makes up about 80 percent of system development costs.⁶ Several Air Force agencies, therefore, pay a great deal of attention to the software used on big systems in a valiant effort to control this part of the price equation.⁷ Indeed, the search for a common Air Force-wide programming language seems close to its goal. But large multi-user programming has long been a black art shrouded in its own dogma, doctrine, and dialect. This art has been complex because it tries to be all things to many users and often falls miserably short in quality and almost always short in timing.⁸

The antithesis of big system programming is found in mini- and, particularly, in microcomputer systems. They try to be only one thing at a time to only one user at a time, so they usually succeed. Microcomputer programming is at the other end of the scale in cost and marketing, too. There are many, many systems, so programs are mass produced and distributed, Most sorting, merging, inventory, word processing, and bookkeeping programs can be obtained for retail prices between $7 and $150 each.

"Software maintenance," a euphemism that means either "we didn't get it right the first time," or "the user changed his mind," is a heavy programming cost in large systems.⁹ The small systems are so unfettered by complicated program interrelationships that program maintenance (updates) can be done by almost any experienced user. Most of these small computers are programmed in a language called BASIC. BASIC was originally developed at Dartmouth College by Professors John G. Kemeny and Thomas Kurtz, who conceived of BASIC as a computer language simple enough to be used by beginners yet powerful enough to carry out sophisticated computation. This year, about 25,000 U.S. students are learning BASIC in secondary schools.¹⁰ Eight-year-olds have written some very credible programs for microcomputers, The number of people who are able to tailor a prepackaged, low-cost, mass-produced computer program to their own needs is growing daily. Certainly, all of these programs will not be compatible among different brands or configurations of equipment, but neither are typewriter ribbons or truck tires.

A spin-off of this growing familiarity with programming will be an exorcising of the demons that now haunt large computer programming. The work is precise and tedious, but it should not be and will no longer be mystical.¹¹

Initially, the invisible computers coming to work for the Air Force should not result in any costly programming effort. In fact, some relief may be felt on the Worldwide Military Command and Control System (WWMCCS) and base level systems as programs and customers move from hard-to-use, fussy, remote devices to desk-top personal computing service. Later, after the population has been established, consideration should be given to providing some standard programs unique to the Air Force, but this void may well fill itself, too. Just as certain in-house publications now list unique military programs for the popular programmable hand calculators (ask any field artillery type about "double-checking" trajectories with an SR-52), so will they soon list programs and even sources of prerecorded program cassettes for the military computer user.

The subject of repairs introduces some good questions, but if we stick with our pickup truck analogy, these answers become less frightening. Service is usually available from local dealers through a purchase order. Many kinds of office machines are already serviced in this manner. Because of the lack of redundant equipment, some backup or manual mode of operation should be maintained. Important data should not be stored away on a tape or disk where it cannot be retrieved if the system is down. This point argues for compatible systems so that the computer down the hall could be used when yours is not available. Compatibility, it should be noted, does not have to mean sole source or (shudder) military specifications. Compatibility simply means that anyone of several similar devices will carry the load, just like pickup trucks. Compatibility simplifies maintenance without mandating strict uniformity.

The Broad View

Earlier, I asked some pointed questions about the need for regulation in the purchasing and introduction of microcomputer devices. These questions seem simplistic until you understand the scope of the problem. The computer like a pickup truck is only the tip of the iceberg. While many of us are still trying to convince our typists that "word processing centers" (read "typing pools") are good for them because such consolidation is the only way we can afford $20,000 word processors, the price of microcomputer-based word processing devices has suddenly fallen to less than $8000.¹² Typewriters with microprocessor augmentation are now available from at least two major manufacturers. Are these just typewriters or are they computers? Obviously, they are both and neither one. The General Services Administration has recognized this situation and has attempted to deal with it by classifying equipment according to how many lines of text it can display or how it uses a printer. These are operational definitions that are reasonably easy to apply, but they beg the question, "Why bother?" Some of these systems have communications options that enable them to transmit pages of text any distance over existing telephone lines. Are they then telecommunications devices? Yes, but not exactly. We have communications computers and we have computers that communicate. We have typewriters that look like TV sets and computers in our coffee pots. Different people manage resources that are beginning to look more and more alike.¹³

I am tempted to advise allowing this raucous cacophony of digital midgets to continue until it reaches some equilibrium of its own accord. But, in the interest of good management practices, we have got to try to give some guidance to the computer identity crisis. The Proceedings of the U.S. Naval Institute has a section called "nobody asked me, but. . ." Borrowing an oar from our naval friends, I will stick it into the old bit stream. I would propose that the Air Force step out in front of the microprocessor revolution by issuing one simple directive: Any computer devices, aside from weapon systems and other than test equipment, that can talk to other computer devices must have one common Air Force-wide standard for transmission. This is not technically difficult. Some very good standards exist.¹⁴ We just have to choose one and stick with it. If we do this, we will have developed, de facto, a computer communications system with great redundancy and flexibility that would otherwise take millions to program and procure.

Many will argue that the invisible computer (both stand-alone mircocomputers and microprocessors imbedded in other devices) must be rigidly controlled. I would like to take a stand against this position, but I don't think I would be successful. The bureaucracy involved is seriously threatened by the flow of information that such systems provide. Their opposition will be couched in terms of security and conservation of funds--and there is just enough truth there to show good color--but actually there is a strong perceived threat to the existing channels of communications and ways of doing things. Such resistance to change must always be expected, no matter how seemingly clear the value of the change is. So regulation has been mandated, and the mandate will be carried out.¹⁵ The only question is: By whom? Digitally speaking, who will control the number crunching, data handling devices? We have proved that they are mutations of old ideas with new capabilities, so where should limits lie?

The placement of the data processing function in commercial organizations seems to run on a wheel of life with a yin and yang of its own. Management of data processing began in the accounting or comptroller structure because they were usually the first corporate users of data systems in the form of electronic accounting machines. (Remember the phrase EAM cards?) The management function then moved out into the manufacturing and processing sections, which made use of automation in many areas. Control of computers was often decentralized. Now, driven by the high cost of software, control of data processing is seen wandering back into the financial fold again.

I suggest that this is the wrong place for it for two reasons: First, data processing is a service that is needed at every organizational level. The financial folks are certainly pervasive in scope, but they are not a service organization and do not think like one. They do not think in terms of users or "customers." Rather, they are more of a watchdog or "auditor" mentality. This mental set is valuable and needed, but it is not compatible with a service function. Second, the upper levels of management in the controller/comptroller structure usually have a financial rather than a technical background. The computer field is breaking new technical ground faster than any other area. These new technologies are often reaching the "hardware" stage within 12 months of development. This represents a deluge of high technology products that should be managed by persons with some" technical background.¹⁶

It could be reasoned that the administrative types in the military control the paper that is frequently the product of a computing device, they know about things with keyboards, and they are a service agency, so perhaps they should be the managers of things that nearly think? They have an input, certainly, but no more than the security police who control vehicle traffic have an input into what kind of vehicles are procured. My position is that the standardization and review of "smart" typewriters, microcomputers, and all sorts of other devices with communications augmentation should be the job of the communicators on any base. They would not validate needs, but they would regulate compatibility in the same way that they ensure that intrabase radios programmed under the appropriate table of allowances can talk together. As Robert Angliss, executive vice-president of RCA Global Communications wrote recently, communicators are becoming "movers of information" instead of providers of circuits. Computers provide the means of moving this information. Their management is now an integral part of the job of communications.¹⁷  

Communicators are, by mission, training, and experience, providers of high technology service. It is logical and natural to combine the management of communications with the management of other computerized services. An example is provided by the Commander in Chief Pacific Command (CINCPAC), which calls its J6 "Director, Communications and Data Processing." ADCOM has a similar integrated function. The Air Staff once combined communications and computers under the KR office symbol, but this connection was broken in a 1978 reorganization. The inclusion of several important computer programming and acquisition agencies under the Air Force Communications Command is wise recognition of the inevitability of the love affair between computers and communications. This was a match made not in heaven but in the high technology Silicon Valley in California. It is better that the offspring be born with the benefit of formal marriage.

The use of small computers is expanding rapidly. They are a low-cost high-return investment useful in a variety of ways. Their growth will affect many existing and planned systems and capabilities. They are relatively simple in use and concept and do not need the management scrutiny given to the expensive hybrid systems they resemble. They are made of humble and sturdy stuff--like a pickup truck.

1. See the "Semiconduct Focus" feature in any issue of Electronic Warfare Defense Electronics magazine.

2. Portial Isaacson, "Personal Computing," Datamation, July 1978, p. 219.

3. "Personal Computers Undercut the Control of DP Departments," Administrative Management, May 1978, p. 71.

4. Wayne Green, "Publisher’s Remarks, Kilobaud, November 1978, p. 6.

5. Wayne Green, "Publisher’s Remarks, Kilobaud, October 1978, p. 7.

6. M. I. Bernstein, "Hardware Is Easy: It’s Software That’s Hard," Datamation, November 1978, p. 32.

7. See the "Data Systems Authorization Directory," AP 300-16.

8. Dennis Streveler, "Designing by Committee Works—Sometimes," Datamation, March 1978, p. 117.

9. Sarah Rolph, "The Micro Momentum," Datamation, December 1978, p. 75.

10. Bernard Banet, "Computers and Early Learning," Creative Computing, September/October 1978, p. 90.

11. John Lehman, "How Software Projects Are Really Managed," Datamation, January 1979, p. 119; Howard Bromberg, "The Consequences of Minicomputers," Datamation, November 1978, p. 102.

12. For instance, the Memorite System I retails for $7900.

13. Amy D. Wohl, "Communicating Word Processors," Datamation, March 1978, p. 102.

14. "First Steps and Caveats to Network Development," Administrative Management, December 1978, p. 33.

15. Air Force attention was focused on the management control of microprocessors as early as 1 August 1977. A Hq USAF FRAX letter of that date clearly spells out the applicability of the ‘300 series Air Force regulations to microprocessors and microcomputers.

16. Jerome Kanter, The Computer and the Executive (Prentice Hall Books, 1967). Chapters 4 and 6 make an avid case for the separation of the data processing and comptrollers functions. More reasoned discussions are contained in Computerized Management Information Systems by Joseph F. Kelly (Macmillan Company, 1970), chapter 3, and in the article "DP’s Role Is Changing" by C.W. Getz in Datamation, February 1978, p. 117.

17. Robert Angliss, "Movers of Information," Communications News, January 1979, p. 44.

Author’s note

Anyone interested in further reading on this subject should also see the following excellent articles.

Major Frank J. Derfler, Jr. (M.Ed., University of South Carolina) is a student at Air Command and Staff College, Air University (ATC). He recently served as Chief of the Tactical Communication Operations Division at Hq PACAF. Major Derfler is a frequent contributor to technical, hobby, and professional military publications and a writer on technological matters with military, social, and economic implications. He is a previous contributor to the Review.

The conclusions and opinions expressed in this document are those of the author cultivated in the freedom of expression, academic environment of Air University. They do not reflect the official position of the U.S. Government, Department of Defense, the United States Air Force or the Air University.

Air & Space Power Home Page | Feedback? Email the Editor