|
Burroughs' Future in Electronics
MRS. MARY HAWES* - NMAA** 1959 Proceedings - Data Processing (*Mrs. Mary Hawes is Senior Product Planning Analyst with the Burroughs Corporation -Electro Data Division.) |
|
DATA PROCESSING Page 20 Burroughs' future is electronics. In fact we don't have to wait for the future, Burroughs' present is electronics. We have so many developments in progress that I could easily spend the next hour, my allotted time, describing some of the very interesting phases of our work. I could even limit myself to those involved with data processing equipment without ever touching on some of the problems with which I believe you are most vitally concerned. What I should like to do before discussing my favorite subject of data processing and the use of electronic computers, is to take you with me for the next five or ten minutes during which time we will forget our present problems. You can even forget. you are in this room if you wish. You" see, I had the privilege of having a math professor who used to take us on such jaunts in which he would describe how various businesses, industries and government agencies were using mathematical tools about which we were studying or which might be under development. It was great fun and furthermore, it broadened our horizon and helped us to realize that our problems, though large in themselves, were part of a much larger group of problems. Slightly more than ten years ago today, the guiding minds of Burroughs recognized that Burroughs' future lay in electronics. They started then setting up the research laboratories and obtaining the facilities which they knew would be required if we were to develop the electronic tools which business, industry and our national defense would need. I am well aware that most of you are in business or industry. Before you set up a mental block against my including "national defense" so much in my introductory remarks, I would like to emphasize the very important role it plays in the overall scheme. The most important requirement to our national defense is that we have equipment that will do a specified job in no more than a given elapsed time, with the utmost in reliability; and that such equipment be ready for this job no later than a given date. In other words the costs involved with not being able to do a job or not being able to do a job in the specified time, is of much greater significance than the cost in dollars and cents of doing the job. Business and industry come into the picture when we have the equipment that can do the specified job for no more than "x" dollars and cents. Perhaps now you can understand why it is that various developments are first made available to our national defense. There may be a considerable time lag before they are offered to you. A great amount of additional research, including the development of different techniques, usually must be spent to make some of these tools practical for your consideration.
|
|
DATA PROCESSING Page 20 All of you know something about the very large data processing national defense projects known as the Sage and Atlas programs. The Sage project is the Air Force radar and data processing network developed to protect us against air attack. Burroughs has been principally concerned with the problem of developing data processing and transmitting equipment for this system. Again with the large scale data processing. equipment for the guidance of the Atlas missile, first at Cape Canaveral and now around the nation, Burroughs is playing a leading role in developing computers which can control the path of a missile even after it has been fired. Furthermore, since the control comes from a program, improved guidance mechanics can be incorporated into the system. To give you some idea of the size of these programs, Burroughs contracts on these two program alone amount to something in the neighborhood of $220,000,000. From the very large electronic data processing systems we can go to the very small and compact computing systems for submarines and aircraft, and to electric timing devices which could be used in the warhead of a missile. We are doing a great amount of work on miniaturization, sub-miniaturization and micro-miniaturization. Miniaturization has been explained as the process by which a lot of little pieces spread over a lot of square inches are all crammed together in the smallest possible sealed container. For example, a one cubic inch plastic block may contain as many parts and pieces as the average portable radio. Yes, I know you are already asking yourself, how do you get inside when something goes wrong? The answer is, you don't. To begin with, it isn't necessary to "get inside" very often; and if you should, you throw the offending unit away and replace it with another, Life tests run on some of these units indicate you should expect to replace an offending unit every ten years or more.We are carrying on some very interesting work in the area, of memory devices. Even though the results of some of our research using some of these memories on a small scale seem fantastic even to us, we realize it will probably be considerable time before we can make them available to you in the size you will require. To come a little closer to home, I am sure most of you have. heard of the 'work we are doing in the area of electrostatic printing. A prototype Whippet Printer was built for the Signal Corps which prints 3,000 words per minute. However, the electrostatic technique makes possible the printing of alpha-numeric information at speeds up to 30,000 characters per second. A great amount of research has gone into developing techniques by which printed information, such as that recorded on bank checks with magnetic ink, can be automatically read by electronic devices. The ability to automatically read this information together with the ability to control the rapid movement of pieces of paper varying in size, permits us to sort bank checks at speeds of 1,500 checks per minute. This development is a major key in making practical the use of electronic data processing systems for banks, including the area of commercial bookkeeping. The value of research has always been recognized by Burroughs. Many |
|
BURROUGHS' FUTURE IN ELECTRONICS Page 21
of the projects initiated in the laboratory today will not reach you for two, five or even ten years. Last year, almost 60% of our sales were for products not known ten years ago. It is expected that an even larger percentage of our income ten years from today will be from products and developments now in the laboratory or not yet on the drawing board. But enough of the future and near future. I have not begun to cover the many projects in which Burroughs is currently participating, nor have I even mentioned any of our electronic computer systems with which you may have worked in the past or with which you are currently working. In fact, from this point on, any reference to equipment will merely be to help put across a point. I wish to discuss with you what we believe to be some of the problems you have encountered or will encounter in your data processing using electronic computer systems. It was in 1886 that William Seward Burroughs developed the first practical adding machine. Burroughs still makes adding machines but I dare say there is as little resemblance between the appearance and anatomy of that adding machine and the one our salesman offers you today, as there is between the personnel who used the equipment in 1886 as in contrast to today's user. However, the reason for the development of such a tool is not so different. Business and industry recognized that more accurate control was necessary if their businesses were to grow and prosper. In the early days of electronic data processing computers, a great amount of effort went into proving that we could use them effectively. Having put data processing applications on these electronic computer systems and also having gone into companies and seen the same jobs being performed principally by young ladies, I have never ceased to be amazed. I am certain that it would take me a long time to be as dexterous as some of these young ladies are as they sort pieces of paper or insert and withdraw cards from a file. I have even seen row after row of persons operating desk calculators with one hand and recording results with the other. However, it is also very alarming to realize that the greatest speed and accuracy is obtained when their actions become automatic and all but bypass the thinking mechanism of the human being. I well recall one of our early studies in which we were asked if we could process 50,000 transactions per hour against a master file, creating an up-to-date master file together with a comprehensive tabulation of the results. We were finally permitted to actually put the problem on the computer and demonstrated we could process 117,000 such transactions per hour more than twice as many as required. We were immediately asked if we could introduce a large rate table, which was in the same sequence as the file, and compute various required values as we processed the transactions. The procedure was altered, with the result that our rate of processing transactions now dropped to approximately 105,000 per hour. Although we had been given a week to incorporate this change, we were able to do so and check out the altered procedure in 8 clock hours, including one hour of computer time. I shall never forget the meeting which followed, for it was then I was told, "Mary, it is far more important and significant that
|
|
DATA PROCESSING Page 22 you were able to incorporate such a major change in procedure in so short a time, than it was that you could process the transactions at the rate of 117,000 per hour rather than 50,000 per hour." Normally about 30 persons would have been involved with this phase of their data processing because of the large volumes of data involved. The length of the training period required in effecting a major change in procedure, including the correction of the errors made en route, could easily have taken a major part of a year. I might add one of the major headaches in effecting new procedures where a number of persons are involved, comes with the way exceptions are handled. And as you know so well, data processing problems sometimes seem to be one exception followed by another exception. Of course, it isn't quite this bad. But it must be remembered that most of our data processing is concerned with "SERVICE" of one type or another and "SERVICE" is very closely allied to human beings. Where human beings are involved, you will always have exceptions. Thus, the most important feature of electronic computer systems is the ability to change. Let us carry this idea a bit farther and say that one of the prerequisites of data processing is the ability to change. Business is never stagnant, it never remains the same for long. A rather striking example of this that comes to mind is the company that was making foam rubber sofa pillows, . and mattresses. They had a very successful business and decided to automate their production line to speed up their production and reduce their expenses. In a relatively short period of time, their sales of sofa pillows zoomed to new heights, while the sales of mattresses decreased slightly. Production was adjusted; the same amount of raw foam material was used but a much larger percentage went into sofa pillows. The only difficulty came near the end of the fixed production line where pillows were coming at such a rate that it was impossible for the girls to handle them fast enough in the space available. A sea of pillows soon developed. You can imagine the results of having built a production line without full realization of the need to change. In this case they brought in more women and set up an inefficient but workable system to stem the tide until they were able to cope with their "changing" world. Somewhere, in a large percentage of the' articles you read today on electronic data processing, you will find the term "systems approach." This is not just a catch phrase; it has a very real and deep significance. I am tempted to say that the need for a "systems approach" is no greater today than it was ten years ago. This would not be true. The world in which we live today is not the same as it was ten years ago. The requirements in business and industry have altered in proportion to our personal lives. Once it was sufficient to produce a better mousetrap.. Today you must also merchandise it more cleverly and be prepared to service it more effectively than does your competitor. However I do believe, looking backward, that the need for seeing your requirements as a whole, has always been present. But when you go into electronic data processing, you are face to face with making many decisions in advance. It sometimes appears that because of this requirement to plan "What will we do if ---?", has forced you to examine your
|
|
BURROUGHS' FUTURE IN ELECTRONICS Page 23
problems more closely than you have for a long time. So many of these "little" questions: are confounded with company policy. When you realize that every time a certain exception presents itself, it will be processed in the same prescribed manner, you find yourself giving it more thought than you might have on a "one-of-a-kind, every-so-often theory." This scrutiny at a detail level can be the eye opener. Once you have started the analysis, it is not too long before you realize that the large savings in paper work management come at the systems level. You also realize that now you have a tool which not only lends itself to a systems approach but which can produce a far more effective piece of work if a systems approach is used. I have not heard of a single complaint from a company that has spent a large sum of money looking into its data processing procedures and requirements as a part of an electronic data processing system evaluation, since they have found that money was well spent whether they decided in favor of or against the acquisition of such a system. Not so long ago, I read of a company who had decided in favor of an electronic data processing system. They also decided not to wait until the electronic equipment was installed to incorporate their overall systems approach into their methods and procedures. They estimated the savings resulting from changed procedures would not only pay for the study but also for the first year's rental on the new system. It is my contention, and I have said it in front of many an audience, that the majority of users of electronic data processing know more about their problems, at both the overall and the detail levels, at the end of their first year of operation than they had ever thought of. In the majority of cases, the computer has been used for procedures not possible with manual methods or with tabulating equipment. The results have been extremely interesting. It has been estimated that between 65% and 85% of the time spent in getting a data processing problem ready for a computer is spent in the definition of the problem. I have experienced running a "realistic" demonstration analysis on a "system," having this system undergo six major revisions during preparation for actual installation (meaning we now incorporated the various exceptions and idiosyncrasies which had not been defined as part of the original problem), and one more major revision after installation. I should add that this system naturally enough included a file maintenance job and that the creation of the master file from a composite of several card files, turned out to be almost as difficult a job as the data processing system once the master file existed. In attempting to explain why the problem definition seems to be so involved, it is my opinion that the difficulty lies in the fact that so many different people know only a small fraction of the job. In some instances, pieces are missing; in other instances, functions overlap and may even be handled slightly differently depending upon who does it. Data processing is never simple. The only job that is really simple is the one you haven't yet tackled. I well recall the first file maintenance system we developed as a demonstration problem for an electronic data processor. It was a beautiful system; and it worked fine on paper. It was one of those "straight forward" life insurance jobs where
|
|
DATA PROCESSING Page 24 we took care of everything since we allowed every field to vary. It took us several man years to do the job and naturally having "received the problem," we didn't have to work too closely with the people who knew the problem more intimately. Even today I almost blush at our innocence and ignorance. The processing procedures were all right as far as they went, but they didn't go far enough. We had attempted to be very sophisticated and achieve the utmost as far as computer usage was concerned. However, the problems we created for the operator were fantastic. We had not faced up to the control and general flow of information or paper work within the company whatsoever. In other words, we had taken a very real application and put it on an electronic data processor without looking at the entire system or how we were to control it in a day-to-day environment. We had a group of procedures each of which was fine in itself; but we had no system and very little control. Time is so very important today. There never seems to be enough of it. As a result of too little time, of attempting to make use of a new tool before the ground has been prepared, a number of electronic data processing systems have been installed prematurely. This may have resulted in an expensive lesson but it may also have helped us to learn our lesson earlier and better than we might otherwise have done. A greater number of persons have found their problems to be larger and more complicated than to be smaller and simpler. Also they have found that as the complications rise, the cost also rises. On the other side of the ledger, they have discovered the potential of their hew-found tool for the solution of problems they could not have tackled otherwise. But perhaps most important of all, they have found that by using an electronic data processing system, they obtain control to a degree not possible heretofore. We talk about management reports, about management by exception. We talk of maximizing profits and minimizing costs. We dream about the time it will be possible to have the month's financial report available on the first, or at most the second of the following month; or furthermore, for the Manager to be able to dial into, a computer from his desk, and ask for certain detail figures on a particular part of that report. He might even ask for an analysis of the data related to this particular part of the system over say the past 3 or 6 months. He might ask that the system analyze what the results would have been if "such and such" had happened 6 months ago, 3 months ago, or predict what would be the result of incorporating a particular change assuming everything else will follow its current trend. There is no question but that being able to obtain this type of information when you want it, is worth a very great deal. How many dollars and cents? No one has been willing to estimate it for me. I was talking to an industrialist less than a year ago about what he thought the computer climate might be four years in the future. He forecasted that his company would rate priorities to computer applications as: A. Design Applications--20 %
|
|
BURROUGHS' FUTURE IN ELECTRONICS Page 25
He went on to say that the reason for the priorities was that the design problems just had to get on the computer. The savings accruing from computer usage for these type problems, in time, and dollars and cents meant life blood to his company. In the area of optimization problems, he told of a relatively recent experience in which they had put a given problem on one of the faster computers with a large memory. Never before had it been possible to analyze such a large amount of related information; what was more, the results from this particular problem were available to management before they made their decision relative to the problem area. The story went on. Management took time to evaluate what their decision would have been had Fhis information not been available, in contrast to what their decision actually was. The saving to the company was a low order six digit dollar figure. They place data processing applications last only because the others must be placed first. The number of computing systems being sold for data processing applications is steadily increasing. The amount of paper work is still increasing in volume while the number of available competent clerks is decreasing. But still more important than the requirement to process paper work is the requirement for more accurate, up-to-date, and relevant information to form a better basis for management decisions. I have talked in very general terms about certain overall data processing requirements including:1. Approach your problem as a whole. To me, ,the two greatest problem areas for data processing applications lie with input-output and with the development of automatic programming techniques that will permit us to do the job at hand effectively and efficiently. You are hard headed businessmen who want all that I have mentioned and ,then some. You want answers to all those problems which have been bothering you for the past urn-teen months but for which you see no solution; in some cases you haven't been able to formulate the problem but you want the answer nevertheless. What is more. you want all of this, faster and more accurately than you have been getting your work done at a price less than you are currently paying. The only trouble with this is that too many of you are sitting back with a "show me" or a "watch and wait" attitude. For many years there were very few who felt that data processing type problems would lend themselves to automatic coding techniques. This was due primarily to the fact that there are seldom any two problems which are identical. Not only are the problems different, but the sub-sections are different. However, the approach to many problems is similar and the structure of the sub-sections is similar. The solution lies in the use of what we call
|
|
DATA PROCESSING Page 26 generators rather than the fixed-subroutines which are common in the mathematical type problems. In using a generator, you start with a structure of skeletal coding and build a particular subroutine to fit a particular set of specifications or parameters. You might say you make use of generators to produce "custom coding."It is interesting to note the difference in the reason for the development of automatic coding techniques for mathematical type problems as opposed to the data processing type problems. Automatic techniques for the mathematical type problem were of direct benefit to the person doing the programming as a labor saving device. The programmer found he could use the same section of coding, with only a few address modifications, in another problem or many problems. He found it more interesting to let the computer incorporate these sections or subroutines not only because the computer could do it faster and more accurately than he, but also the programmer was now free to tackle new problems. From here, the next step was to devise methods by which the programmer could state his problem in general mathematical terms to eliminate still more coding. The form in which the problem was stated incorporated, to a certain extent, the idiosyncrasies of the particular computer for which the compiler was written and also of the individuals who developed the compiler. This necessitated restating the same problem in a slightly different form for different computers. Especially was this true when the computers were developed by different manufacturers. All of you know of the efforts which have been expended recently toward defining a common algebraic language so that it will be possible to state a mathematical problem so that it can be understood not only by many individuals, but also by many different computers.For the data processing type problems, first came some generalized routines such as sorts, merges, and report writers. These were separate routines in which the parameters were specified so that various addresses and switches could be modified by the computer to make the resultant routine work on data consistent with the specified parameters. These are generators and were the forerunners of the generators which are now used to develop subroutines or larger segments of routines. It was recognized at a relatively early date that the major difficulty encountered with data processing problems was with handling of information or the flow of information in and out of a data processing system. Furthermore, the greatest percentage of inefficiency lay in this area. From the operational standpoint, the greatest amount of difficulty lay in the same area since the input-output equipment, being electro-mechanical, was more subject to down time than were the electronic circuits. Coupled with all of this was the necessity to incorporate restart points not only because of the length of time a particular problem might take, but also for recovery in the event of malfunction on the part of the data processing system. Towering above all these considerations was the fact that the same routines will be run by different operators and that changes must be incorporated from time to time within these routines, most often by someone other than the original programmer. Furthermore, the work must be accomplished on time. Paychecks must be delivered on time.
|
|
DATA PROCESSING Page 27 Vendors must be paid within prescribed periods of time. Orders must be filled on time. Therefore, if changes are to be applied, they must be applied correctly. 'Not only must the corrections be applied correctly within one routine, but in all routines where the change is reflected. Out of this comes the realization that in data processing applications, the emphasis must be based on the entire system and' upon its maintenance. Much time, analysis and experimentation have been spent in an effort to determine the best way to represent a data processing system taking into account the difficulties associated with its: 1. Definition We at Burroughs are now on our way toward developing automatic coding tools based upon the use of multi-level flow charts with problem definition in English words, algebraic expressions and mathematical equations. Operational and procedural standards that contribute toward good systems organization are further incorporated by use of a library of subroutines, generators and data descriptions. The entire automatic approach is biased to favor revision and maintenance rather than initial solution. Let me take a few minutes to attempt to indicate some of the problems we faced, and which contributed to our approach. In overall, an automatic data processing system must recognize that human beings initiate the action, human beings control the action, and human beings also make use of the end product. Furthermore, the abilities and characteristics of the various types of individuals who are part of the system must neither be overestimated nor underestimated if they are to be able to turn in a "best" performance. Procedures must be kept simple to use; if complicated or sophisticated techniques are necessary, they must be internal. Emphasis must also be placed on economical operation as well as on most effective results. It has been estimated that about 65 % to 85 % of the cost of getting ready a data processing problem for a computer lies in the definition of the problem It is also one area that does not lend itself to the compression of time. If you do ,not do sufficient planning and flow charting initially, you add time at an exponential rate later. One of the reasons for the complexities involved in defining the problem lies with the fact that a properly designed data processing problem normally cuts across department lines. In different groups, practices differ, language differs and occasionally similar overlapping functions are processed somewhat differently. Since many persons collaborate in defining the problem, the language used must be understood by all. Not only must' all parts of the problem work, but they must work together as a unit The easiest place to locate a logical error or weakness is at this time when the problem is laid out in flow chart form. First is the general flow chart which includes how it ties into the company organization; second is the process flow chart which presents the systems organization of the
|
|
DATA PROCESSING Page 28 problem itself, in terms of its various parts; third are the detail flow charts for each part. This detail flow chart is organized according to functions which are to be performed and more oriented toward the problem than toward the computer. The "language" used to define the problem must be understood by the person who knows the problem. We find all too often when procedures are presented verbally, one is so busy following the speaker that he is not free to explore alternate paths which may occur to him. On the other hand, if he does explore, he loses some information which is being presented and that is not good. Broad written paragraphs usually leave material out because of the voluminous writing which is required. The detail flow chart organized by function, not so detailed as instruction level, together with the larger view flow charts seem to answer the problem with regard to: 1. Ability to obtain overall logic as well as detail logic Now if we add a fifth and sixth advantage: 5. Easy to translate to machine code we really may have an excellent start toward reducing some of our major headaches. It has been said many times, the greatest advantage of an electronic computer is not its ability to do a job faster but rather to do it better and also retain the ability to change and improve as we develop better procedures and techniques. We must not lose sight of the desirability to process our problems on more than one computer. In the early days we recognized this by seeing that it was possible to write a tape file on one tape unit and read it from any other connected to the same data processor. We soon recognized the necessity to read or write tapes on different computers of the same model as a requirement for back-up. This was enlarged to include different data processing systems by the same manufacturers as the installation grew in size. Now we recognize the necessity to be able to easily take the same data processing problem and put both it and its associated data on different systems. It is possible in a centralized operation that the central computing system is fed by several different smaller computing systems each of which is doing "practically" the Same job. -It would be desirable both from the information and from the cost standpoint to have the problem defined once and only once with the variations added.Again, let me emphasize the maintenance and revision problems. It has been estimated by a number of users of large scale equipment that as much as 25 % of programming talent may be used in taking care of production runs. This consists of supervising the operation, updating the system and also making changes that would improve the system. This need to change easily and correctly exists long before an operation is in production. The chances
|
|
BURROUGHS' FUTURE IN ELECTRONICS Page 29
are excellent that at least one major and several minor changes will be made in a system after it is "frozen." These changes are not unwarranted so, therefore, we need to recognize this as a pattern and be able to cope with it. During the last few minutes, I have attempted to point out just a few of the difficulties you have encountered or will soon encounter. They may be "old hat" to you because the same type of problem cropped up in the last system you were responsible for. With electronic speeds, they just crop up over a shorter period of time.Burroughs is not only doing research which allows them to produce equipment which will better meet your requirements of capability, reliability and cost but also programming tools so you can use that equipment more effectively and efficiently. We recognize that you in the field have the problems which you alone can solve. We are attempting to give you the ' best tools possible so that together we can contribute to better business, Industry and government management and thus continue as a leader of the world of tomorrow.
*Mrs. Mary Hawes is Senior Product Planning Analyst with the Burroughs Corporation -Electro Data Division. |
|
**Editors Note: Who Was the NMAA? In the following paragraphs there is a brief overview of the NMAA, which became DPMA (my era!) and then evolved into AITP --Ed Sharpe Archivist for SMECC History of AITP The year was 1951. Harry S. Truman was President of the United States. A 3 bedroom home cost $9,000.00. A new Ford listed for $1,480.00; postage was $.03; and a loaf of bread cost $.16. Joe DiMaggio retired from baseball; I Love Lucy premiered; and peace talks began in Korea. In Chicago, a group of machine accountants got together and decided that the future was only beginning for the TAB machines they were operating. They were members of a local group called the Machine Accountants Association (MAA). The technology was new; something few people understood and managing this new technology was a skill that even fewer people possessed. The machine accountants recognized the need to form a professional support group, a national association, to address the growing issues of this new technology. Thus on December 26, 1951, after a constitutional convention was held in Chicago, the State of Illinois granted a charter and the National Machine Accountants Association (NMAA) was founded. Groups from Houston, Columbus, Wabash Valley, the Twin Cities, Penn-Del, and 22 others were the first to join NMAA. Robert L. Jenal, systems manager for Toni Company, was elected the first International President at the 1952 First Annual Convention in Minneapolis. In 1960, the association sponsored a meeting of educators and businessmen with the purpose of establishing the Certificate in Data Processing (CDP) professional examination program. The first CDP exam was held in 1962 in New York. 1962 was also the year that the association leaders recognized the changing nature of information processing techniques brought about by the introduction of the computer. Thus, the members decided in 1962 to adopt a more progressive name, the Data Processing Management Association (DPMA), to reflect the changing industry. Always striving to promote the continued education of the members, the leadership of DPMA created the Registered Business Programmer (RBP) examination in 1970. Both the CDP and the RBP exams were given annually under the rules established by the Certification Council, at test centers in colleges and universities across North America. Eventually, DPMA decided to help establish the Institute for the Certification of Computer Professionals (ICCP) to stimulate more widespread interest and industry acceptance of the examinations. ICCP began administering the CDP program in early 1974. The association has always acknowledged the contributions of prominent professionals within the Information Technology field. Beginning in 1969 with the creation of the annual Computer Sciences Man-of-the-Year Award for outstanding contributions to the information processing industry, DPMA has established a long-standing tradition of honoring IT professionals from every aspect of the industry. This prestigious award was renamed the Distinguished Information Sciences Award in 1980 and is awarded every year at the Annual Meeting of the Members. As the industry has evolved, so has the association. Starting as the NMAA, evolving into the DPMA, and then into our current evolution in 1996 of the Association of INFORMATION TECHNOLOGY PROFESSIONALS (AITP), the association has kept pace with the changing needs and interests of our members. AITP members span every level of the IT industry from mainframe systems, to micro systems, to PC based LAN and WAN systems, to virtual systems and the internet. AITP has special niches created that cater to the special interests of our members. Our members are found in every facet of society as well. They're in colleges and universities; banking; industry; retail; the armed forces; local, state and federal governments; hospitals; etc. Copyright © 1998 Association of INFORMATION TECHNOLOGY PROFESSIONALS To further information, or to become a member of AITP go to:
|
| We are actively
seeking information related to any of the topics presented on this
page... please email info@smecc.org if
you can be of assistance!
Thanks, Ed Sharpe archivist for SMECC
|
