Letters from George Snively to Barney Oldfield

In April 1993 Barney Oldfield suddenly surfaced and thus began a lengthy correspondence which ended up with his book King of the Seven Dwarfs, published in 1996 by the IEEE Computer Society Press.

It all began when Barney wrote an article for the Annuals of Computer History to "correct" my article -to, among other things, to refute my statement that he had been fired. He maintains that he resigned for "personal reasons". The book evolved as he got deeper and deeper into the story.

I wrote him some one hundred letters during the course of the project. About 25% of them are reproduced here. Most of them were in response to questions from Barney and reading them may be somewhat like listening to one end of a telephone conversation but I have included only those that more or less stand on their own. I have not included Barney's letters principally as a matter of privacy but also because I could not easily match them with my replies and I had not saved all of them.

I disagreed with both premises of Barney's book:

1. that GE's efforts in the computer business were a failure and

2. that GE's management philosophy was to blame.

My efforts to persuade him in these matters had only minor success. (The final book was toned done considerably from the first draft. His negative interpretations of events was more indicative of one who had been fired - not one who had resigned for personal reasons)

GE’s SPECTACULAR SUCCESS DURING THE LAST HALF OF THE 20^th CENTURY IS SUFFICIENT EVIDENCE OF THE EFFICACY OF GE’s MANAGEMENT PHILOSOPHY

(I hope that those who think otherwise were not "short sellers" of GE stock during this period)

George Trotter called me last night with news of your address. It may surprise you to know that you have been the object of a wide search. WGBH TV, the Boston based PBS station, wanted to interview you concerning ERMA for their series on the computer, called "the machine that changed the world". They had planned an extensive piece on ERMA and the early GE efforts but couldn't find you and both Clair Lasher and Al Zipf were too ill at that time to journey to Boston and appear on camera.

Footage of me (as a last resort) telling the story of the Computer Department was left on the cutting room floor for lack of corroborating evidence and other visual material. (The series was driven by the availability of visuals - i.e. Echert and Mauchlay's home movies of UNIVAC got considerable play, while Von Neumann received minimal voice-over comment with only his photograph on camera.)

There are any number of people who would like to get in touch with you. Nate Norris advises that he has sent you a copy of the Computer Department alumni list and tells me that you have been in contact with Bob Johnson.

You ask so many questions, I hardly know where to start. So I will just start - and ramble.

First, my "tall tales" was not meant to be a spoof - but a memorial to you. My guess is that less then half a dozen people in the audience of approximately one hundred Computer Department Alumni had ever heard of you. I didn't want you and your efforts to be forgotten and didn't know that you were still alive and well and able to speak for yourself. (We couldn't even find a picture of you for WGBH.) My old memory may have been faulty, but it is the story as I remember it. (Some of the episodes were as reported to me by Russ Krapf and Ken McCombs. The later was admittedly not a very good reporter.) As you know, history fades into myth and then myth becomes history.

I stayed until August 1967. At that time there had been roughly 500 people at sub-section level and above. Approximately 80 were still in place. Of the remaining 420, a number resigned, one (I can't remember who,) was promoted, one (me) stayed at the same level throughout the whole period and the balance were either fired or demoted. It seemed time to leave.

Lasher, of course, succeeded you - but as "Acting" and I don't know that he ever was made permanent. Ray Barclay and Ken McCombs stayed until Van Aken fired them both. Ken Gieser (now deceased) was shunted aside and down. He stayed through the Honeywell merger - ending up as the "Major Domo" to entertain visiting dignitaries. Art Newman (still living) was demoted several times - when I left he was in charge of the employee store. I thought that Herb Grosch left on your watch. I distinctly remember you billing IBM for his unpaid expense accounts and for his carpeting. I don't recall how long Bob Johnson stayed. You have Ray Barclay's address. Ken McCombs has disappeared. Ray stopped by to see me several years ago, principally to see if I could enlighten him as to why Van had fired him - I couldn't.

The replacements for Ray Barclay and Ken McCombs were both excellent choices. Cy Statt, whom you may remember from GED in Syracuse, as Manager-Manufacturing and Gil Gillespie from Financial Services as Manager-Finance.

The confusion over the number of ERMA systems is probably due to the "commercial" version being referred to as the GE110. Someone in the GE Alumni group (Vern Schatz I believe) has a listing by serial number and customer. 65 is a believable number. Security Pacific Bank alone ordered 10 and I believe took delivery on all of them.

NCR was not able to market the 304 successfully. It is surprising, but as I understand it, in addition to the machine's costliness NCR did not thoroughly understand the department store application that was supposed to be their strength. It was also likely that IBM was too entrenched in the data processing function. Only a very limited number were sold. There were also extreme corporate culture differences in the manufacturing area between GE and NCR.

Helmet Sassenfeld, one of Von Braun's people, succeeded Herb Grosch at the ASU Computer Center. The Huntsville project was ultimately lost but Helmut built a nationwide network of computer service bureaus. It became a very large and profitable time-sharing network, which GE kept in the merger with Honeywell. As far as I know it is still a large operating network. (Unless the PC's have greatly impacted it).

I'm surprised that Reg Jones stated that all 14 years of the Computer Department were a loss. I was under the impression that he was the only one besides Gerald Phillippi who understood the bookkeeping where the greater the success in a rental business, under GE's bookkeeping policies, the greater the "book loss" (or tax shelter). The principal problem with the Computer Department was that no one understood the financial statements. I will prepare an appendix of explanation.

Lasher was a planner. His first inclination, after he got the "Big Look" approved was to recognize that it was out of date and to start re-planning. It didn't seem to occur to him to implement. (He was also emotionally exhausted). Fortunately, or unfortunately, Lacy Goosetree, as Manager -Marketing, and Bob

Sheeley as National Sales Manager were implementers. Bob Sheeley assembled a national sales organization "Par Excellence". It included the likes of Ken Fisher (later of Prime Computer), Jim Pompa (to become a Honeywell V.P,), Tom O'Rourke (founder of TYMSHARE), Warren Prince, Jim Helm, Vic Casebolt (to become at one point President of Storage Tech. and now V.P. -International Paper), Clint De Gabrielle (to become the Czar of Computing for the State of Washington), Curt Hare (to become President of TERAK and half a dozen other Silicon Valley ventures) and on and on through a whole other host of outstanding capable people like Leo Mott (who formed a Dallas based facilities management company), Vern Schatz, Bill Duster, Paul Shapiro, Jay Kear (President of an Orange Co. software firm), Dick Nosky, Pete Repenning, Gene Ringstad, Len Call, Bill Peake -the list is twice as long. I've only scratched the surface. The legacy of the GE Computer Dept. is the later accomplishments of the people who left. Unfortunately, they out-sold the capacity of the Computer Department to produce, principally in the software and applications area.

In May 1961 Bob Sheeley and Lacy Goosetree, shortly after Lasher received approval for the "Big Look", and planned earlier in anticipation of the OK, held a national sales meeting. This weeklong meeting was held at the Superstition Ho Hotel in Apache Junction. Its purpose was to put the GE brand on the above group, most of who had come from our competitors. This up-beat meeting with everything done very professionally and first class more than accomplished its purpose. Everyone who attended cherishes the experience and it is this group of people who form the nucleus of the Computer Department Alumni.

Lasher was always successful in getting 570 to approve his ambitious plans - but lost his courage in implementing them as the red ink flowed. He would, as also did his successors, abort the plans before they were fully implemented. This created confusion. It was impossible to justify "losses" to 570 when he himself was aghast at them. It seemed "flippant" to month after month explain larger than anticipated losses by suggesting that we had a better month than anticipated. (See the enclosed spreadsheets)

We had nearly 100% of the banking business (with the GE 225), which proves the success of ERMA, a goodly chunk of the internal GE business and were making sales for manufacturing applications based on the strong GE manufacturing experience. However, many of these external sales were "paper tigers" as far

as deliveries, applications and software capabilities were concerned. You were not the last computer salesman to learn that it was easier to sell a computer that had yet to be built then to sell an existing one.

The obscured fact of the computer business at this time was that IBM was essentially selling to themselves. They had trained and placed the Data Processing Managers whose path of progress was for IBM to place them in a better assignment with another company. Their loyalties were to IBM and not the company that employed them. IBM assisted them by having most of the Fortune 500 companies pass Corporate Resolutions forbidding the purchase of computers (for all the old familiar arguments) thus eliminating the need for the DP mangers to bother with cumbersome Appropriation Requests (and raising the cost of entry for competitors). They also convinced upper management that Purchasing was incapable of making procurement decisions in regard to computers. Only the DP Managers were capable of this. As a result, this was a difficult market to crack, particularly when these facts were not understood and marketing strategies and tactics were not developed to counter them.

The consequence was that our customers, to a man, were mavericks. The ones with the courage to jump the IBM traces. They extracted a price for their courage. IBM and their DP Manager customers would enter into a mutual conspiracy of silence if a particular program or piece of hardware was late or failed to perform. On the other hand, it was incumbent for our customers to alert their management at the first sign of potential problems. We built many customer files into what Bob Sheeley called "monuments to negotiations". (Speaking of which, it might interest you to know that the ERMA contract was not finally executed until after the last system was shipped.) IBM also very effectual kept GE's upper management informed of the Computer Department's problems.

I don't recall anything particularly dramatic or traumatic about Clair Lasher's exodus. In my memory, he just quietly slips out and is quietly replaced by ???. (Who was so un-memorable that I can't remember his name). I remember ??? as being from the mold of old line GE General Managers. A mild mannered, pipe smoking, graduate of the "Test" program. He sat calmly and quietly in his office as the computer business swirled around him. In relatively short order he was named the General Manager of the Process Computer Department, which was a natural fit.

Following this lamb, Van Aken came in roaring like a lion. Van became very frustrated and confused when his roaring against the tide of the computer business was to no avail. He also became paranoid when GE's upper management would learn of the Computer Department's problems from IBM before Van knew of them himself. He would become even more paranoid, emotional and irrational when Lou Rader, newly returned from UNIVAC, replaced Harold Strickland and began forcing ex-UNIVAC people on him.

Van is now retired and living in Tempe looking relaxed, tanned, healthy and as dapper as a Gentlemen's Quarterly model. You should contact him for the story of his reign as GM of the Computer Department.

The first to go was Bob Sheeley when Lou Rader "suggested" an ex UNIVAC salesman, Art Ashauer, for that position. Art was likeable, easy going and relatively competent. However, the sales force, which was fiercely loyal to Bob Sheeley and most of whom were more competent and experienced than Art, were astounded by the move. Rader's next move replaced the well liked, competent and effective Tom O'Rourke with Ted Green (also from UNIVAC) as the Western Regional manager. The final straw, which destroyed probably the best sale force ever assembled, was the replacement of Lacy Goosetree (who is also alive, well, and active - you should contact him) with Vern Cooper from GE Supply. Vern was articulate, imposing, vain, intimidating, incompetent, stupid and naive. (I later learned that he was

Herch Cross's son-in-law and the Computer Department was just a temporary assignment on his "fast track".)

Vern Cooper's classic remark concerned the banking business. He didn't think we were doing well in the business since we didn't have any of the important New York banks and "whoever heard of Security Pacific Bank (then #7) or Northwest Bancorp?". The very closely-knit banking industry, which has a very active grapevine, always watched us closely for any signs of wavering in our support of them. Vern could not understand why we had more salesmen at headquarters than in the field and one of his first moves was to disband the headquarters industry support groups and send them to the field. The banking market was lost in one fell swoop when he disbanded the banking support group. I was at Pittsburgh National Bank (a good GE 225 customer) getting the final documentation executed for the lease of two GE 400 systems the day this move was announced. The bank decided to wait to see what was happening and delayed executing the lease. They ultimately went with Boroughs. I was also in the meeting with Martin Marietta(sp?) when we were trying to save the GE 600 installation with a number of non-green dollar concessions such

as delayed starts on net lease rentals and etc. totaling $3 million soft dollars. The meeting ended when Vern, to show Charlie Lighthouser, then Executive V,P, of Martin, that he was tougher than Charlie said, Take your $3 million dollars and give me back my machines!". When I interrupted to point out that the

concessions were in consideration of Martin keeping the machines I was told to "shut up" that he knew what he was doing. Charlie looked at me, shrugged his shoulders and indicated that it was an offer he couldn't refuse - and he didn't!

The follow on to ERMA with the B of A was a contest between two "paper tigers". We were proposing the not yet in production GE 600 system and IBM was proposing a likewise situated upper model of the 360 series. IBM ultimately "bought" the business but you will be delighted to know that part of the price IBM paid was to pay the monthly maintenance bills for ERMA for any months they were late in delivering. For nearly 18 months IBM paid us the monthly bill of $250,000.

Van Aken was followed by John Hanstra from IBM. This was a very un-GE like move (If it was engineered by Lou Rader, he atoned for some of his above sins). John, who had been responsible for the IBM 360 series, was very affable, self confident, easy to talk to, and extremely knowledgeable about the business. He garnered immediate admiration and respect from one and all. Unfortunately, he had hardly settled into the position before he died in a crash of his two engine airplane

John was followed by Lou Wengert. Wengert grew up in the financial function of X-Ray and became General Manager of several GE large motor departments. He came to us from Locomotive. He also came in "roaring", but after Van Aken it seemed like more of a "meow", particularly when afterward he would smile like he didn't mean it. He also became very emotional and irrational as had Van Aken. I left sometime, maybe a year, after Lou arrived. (You will note the lack of dates in this whole narrative - I just

Shortly before I left the Department was split up into half a dozen different Departments. As I recall

Jack Vanderslice took over the Oklahoma operations. I don't remember who the other GM's were. Soon after the split up, Lou Wengert was promoted to V.P. of one of the industrial groups. Stan Smith, replaced Lou Rader about the same time that Wengert came aboard.

I had lunch with Lou Wengert, who is retired and living in Carefree, AZ., several years after he left the Computer Department. He told me that he had been appointed GM of the Computer Department because of his manufacturing expertise and that he was ordered out to phoenix to fix the manufacturing problem. Sure enough there was a big manufacturing problem. After focusing on this problem for nearly eighteen months he suddenly discovered that this problem, as big as it was, paled in comparison to the engineering and marketing problems, but it was too late.

I will end here. I suspect that I have raised more questions then I have answered. I'm sure that I have told you far more than I actually know! I will give you time to digest this missive and will then

PS I think that the only way the story will be clarified is to get half a dozen of the knowledgeable people together so that their individual recollections can trigger each other's recall.

The enclosed spreadsheets will illustrate the problems that the rental business presented in the Computer Department financial statements.

The first Case is probably the most dramatic. When Fred Borsh became President of the Company he let it be known that the results of a study that he had made showed that GE had never been successful in a business where they had less than 30% of the market. This was particularly true in markets, such as computers, where there were few competitors. He let it be known that those Departments who could not demonstrate the ability to capture 30% of their markets might be short lived.

Consequently, Clair planned the "Big Look" to see what it would mean if the Computer Department aggressively pursued a plan to reach 30% of the computer market in 5 years. This was in late 1960 and early 1961 and the 1966 computer market was being estimated at $3 billion. (It actually nearly doubled that amount.)

The Case I. enclosure ignores the real problems of achieving such an objective but shows the impact of the rental business on the financial results should such an eventuality be profitably achieved. As you can see, huge losses are incurred - due to GE's bookkeeping- even if the product’s sales and rentals are quite profitable by themselves. I don't recall the starting point, but the end result was to be an "if sold" volume of $900 million in the 5th year.

Case II. Shows the result of backing the projection down to 12.5% of the 5th year market. This plan had a chance of being achievable and affordable even though the numbers are still large and negative. Defying the industry experience of 80% rental and 20% sales we forecast 30% sales to improve the numbers.

Case III. This case was prepared to show the effect of changing the percentage sold to 50%.

Case IV. This case shows what happens if you stop growing at some point. You need to project out farther then 10 years to end up in a "steady-state" profitable position.

Case V. An extreme case to show that the money comes flowing in if you cease shipments at some point.

I recall at one point calculating that if we were to grow at more then 12% a year that we would theoretically never book a profit. This in a market easily growing at more than 25% a year.

Case II. approximates the "Big Look" that Clair had approved. While it called for a 12.5% market share we had to run like hell to obtain 7% or 8%. However, this smaller percentage was a larger than planned volume as the market grew much faster than forecast.

At the time I was preparing the forecasts for Clair I had to do them manually. Each projection took about 40 hours as I had to project each shipment month by month and layer by layer. Who knows how we may have been able to fine tune them strategically had we had the "what if" capability of current spreadsheets.

As far as I am aware, neither Clair or anyone else had difficulty in getting 570 to understand these projections when they were presented in the "abstract" so to speak in a long range plan. It was when the current month operating results actually reflected these numbers did the lack of understanding suddenly

The problem was that even though it was understood that success would generate huge red numbers, for any one month it was difficult to discern how much of the variance from plan was caused by success, how much from a change in the sales/rental ratio and how much was the consequences of favorable or unfavorable operating problems.

The problem was amplified out of all proportions by the three months rolling forecasts, which I'm sure you remember. The current month's result could vary by a $1 million or more within days of submitting the forecast. A customer for a $1million dollar system which we had forecast as being "sold" would decide, usually the day of shipment (which you know is always the last day of the month) to rent instead. This would cause nearly a $1 million swing. Such large swings in the matter of just several days severely strained the creditability of the Computer Department's management with 570, even though they were completely beyond the control of management.

In answer to your question about 570's attitude toward the Computer Department, I would sum it up by saying that they were always supportive of the business and willing to invest in the big picture. For many of the reasons demonstrated above, plus known operating problems and IBM's input, they were not always

Just a short note on my long view of the people involved in the GE Computer Business and it's profitability. Without diminishing the important contributions of many, many people, there are only five who really made a difference.

It didn't make much difference who was involved after this. All anyone could do was to try to hang onto the tiller as the forces in the market place drove and buffeted the business. (Despite what other's egos might think - the market made most of the decisions.)

Finally, Reg Jones for negotiating the very timely and profitable sale to Honeywell.

I would argue that on an "if sold" basis the business was always reasonably profitable. Gerald Phillippi argued, legitimately, against the use of "if sold". His argument was that any small adjustment to the top line of the operating statement fell right through to the southeast corner where it is magnified. That "ipso facto" we had not sold the rented systems and to have done so would have likely required a reduction in selling price, a decrease in volume, or both, resulting in either case in a reduction from the income calculated on an "if sold" basis.

It was good talking to you today. You have entered into an ambitious project - nothing new for you - and I give you my best wishes and support.

If you haven't already done so, I suggest that you contact Bruce Bruemmer at the Babbage Institute. He has been gathering information, including attending one of the Alumni meetings and interviewing many of the attendees, about the GE computer business. Perhaps the two of you could collaborate in this

In my long view everything that happened before the 1961 Second First Annual Sales Meeting was prelude. The first act curtain raises on this meeting. You will be able to find a goodly number of people who stayed with GE from this point through the sale to Honeywell. (Many of the products and activities mentioned in this meeting were suddenly concocted to meet the needs of the meeting, don't put too much credence on anything presented there.)

I would contend - but neither long nor hard - that with the ERMA deal in hand you would have found some engineer or engineers to execute it. Only you can judge whether Bob Johnson was absolutely essential to this project and should be on my "short list".

Lou Wengert will be able to fill you in on the sale of the Process Computer Business, which he negotiated. ( He may be too embarrassed to mention the proposed sale to Fairchild which garnered a lot of national publicity when it fell through. Your local newspaper morgue or library should have it. Or you can

find it if you are "with it" and have computer access to the various databases).

Manufacturing had to contend with the rapid transitions being made in the industry from wire wrap, to printed circuits, through discrete components, single sided, double sided boards, etc. to IC's. I'm sure that these transitions, and the growth, were much more rapid than 570 had ever seen before and, understandably, there were attendant problems. Cy Statt can ably fill you in on manufacturing.

About midnight on ( You can find the exact date from the Wall Street Journal files) I took a break from running out the "Big Look" numbers required to capture a 30% market share and went down to the cafeteria for some coffee. I picked up a copy of the Wall Street Journal that was lying on the table and there

was Sarnouf (Sp?) claiming that RCA was going to be number two in the computer industry in five years doing $1 billion a year in sales! The same goal as ours, but I had already concluded that GE could not finance such an ambitious goal and that we would have to scale back drastically (even if we could per chance achieve it operationally). I went back upstairs and grabbed a copy of Moody's to look up RCA. here was no way that Sarnouf could back up his claim. RCA was in default on the interest on $50 million of debentures, had nominal earnings, and I don't now recall but presumably wasn't paying any dividends. The stock price was down and it was certainly no time for them to go to market for more money. HE COULDN'T DO IT! This led me to analyze our other competitors. IBM was having problems at this particular juncture ( I believe they were between the 1401 and the 360). Prudential was balking at providing a previously scheduled take-down on a debenture issue and was insisting on a higher interest rate. I don't recall the status of our other competitors but obviously none of them, except IBM, could do it either. I stayed up all night making these analyses and showed them to Clair the first thing the next morning.

Clair incorporated these analyses in his presentation to the Executive Office of the scaled down "Big Look". It was reported back to me, I don't now recall whether directly by Lasher or through McCombs ( and a lot of creditability rests on this question) that Ralph Cordiner called in John Lockton and Gerald

Phillippi to get their views on the competitive analysis and whether GE could finance the venture. In those days John Lockton thought that GE could finance the world if necessary and gave an unqualified yes. As reported, Cordiner then stated, "Up until now I saw no reason for GE to be in the computer business as we had no unique contribution to make. It is essentially an assembly business, not manufacturing, so we can't capitalize on those skills, we had no unique technology as the engineers are freely exchanged, and we would be up against reputedly the greatest marketing organization. However, I now see it as a table stakes game and we have the stakes." And thus the go ahead.

Ultimately the basis for the decision proved to be wrong. It was not the cash required but the earnings impact of the rental business on the widows and orphan that own GE that proved it wrong.

In my financial naiveté of the time, I thought that RCA had no wriggle room. However, Sarnouf took a gamble and called the debentures, which were in default. They converted to equity and I don't know what he would have done if they had opted for the cash. Next, while we were involved in the "mating dance" with Bull, he was busily selling them $50 million worth of computers on a letter of credit. You know what a boost an outright sale like this gives to the financials of a rental business. It seemed to me at the time, and still does, that his approach to Bull was far superior to ours!

To me the Bull episode is an example of what happens when you shunt operating managers aside into staff positions without portfolios. They run around getting into mischief.

PS: An amusing episode about Van Aken concerns his first staff meeting which he called for eight o'clock on a Monday morning. Promptly at eight o'clock he closed and locked the conference room door. This left Lacy Goosetree and Ken Geiser pounding on the door to be let in. (There was no relenting). The word quickly spread to the troops and my poor secretary, who religiously punched in three minutes early,

came in in tears the next morning when she was nearly fifteen minutes late. She had been caught in the biggest traffic jam that the Black Canyon Highway had ever experienced as everyone tried to get to work on time.

As I forthrightly stated, and you have now confirmed, you could never hear the word "no". A rare attribute possessed by only the most superlative of promoters and salesmen.

This meeting, which was about the Appropriation Request to expand the Microwave Laboratory fourfold to build ERMA, was not attended by either me or George Trotter. It was reported to me by Russ Krapf who, in contrast to Ken McCombs, was a good reporter.

I prefer his version as expanding the Microwave Laboratory at that time made no sense in light of the mission of the Laboratories Department and Haller's desire to create a computer laboratory at Penn State. In addition, the expansion didn't happen.

PS: The reason we were so successful with the banks is because they had not opted for the punched card and had no entrenched IBM DP Managers. We had a level playing field.

PSS: The aborted negotiation for the sale of the Process Computer business to Fairchild was between Gene White for Fairchild and Lou Wengert for GE. Subsequently, Gene became President of Amdahl and is now Chairman Emeritus. When Gene was appointed President, he interviewed me to be CFO. In discussing the opportunity with Lou, he was about as uncomplimentary about Gene as one can be about someone. When it later became apparent that Gene was going to be successful, and not fail as miserably as Lou had predicted, Lou then cottoned up to him when they served together on the board of Don Oglesby's Three Phoenix Company. (I was not offered the job and also predicted his failure when he told me that Amdalh's big advantage was that since they were "plug compatible" with IBM at a lower price they didn't need to spend much money on sales & marketing.)

The importance of ERMA as the machine that launched the banks into the use of computers and GE into the computer business, can not be exaggerated. However, don't over emphasize its importance as a product. It required considerable redesign into the GE 210 to make it a commercially feasible product. The GE 210, in turn, was quickly obsoleted by the GE 225 at less than half the cost.

Time-sharing is the principal legacy of the Computer Department. It was made possible, fittingly enough, by another bootlegged product - the DATANET 30 - which married the widely diverse technology of computers and communications. It led to the present net-working of the world. It also put computer terminals on peoples desks and promoted distributed data processing and started a process which has terminated in the personal computer.

Bill Bridge did the hardware design and Dick Smith and Don Knight, working with Dartmouth's John Kemeny, helped develop the BASIC language to make it "user friendly". The first system was installed at Dartmouth. (A GE 225 with the DATANET 30)

PS: You keep requesting me to identify the key problems. This is difficult. There were significant problems in almost every functional area most of the time! The majority of them were "good problems" - those that are caused by success. These problems were caused by the unprecedented growth that stretched and unbalanced all resources, including the human ones, to the breaking point. We went through an ungodly number of good people.

Thanks for the Forbes article. I was always amused when outside sources quoted and interpreted, second hand, GE's computer business numbers. We who were compiling them didn't know what they were or how to interpret or explain them!

Market numbers were equally as illusive. Dollar volumes contained a mix of sales, rentals, maintenance and software. Counting systems, or main frames, did not take into account a great disparity in size and peripheral mix. I'm sure you get the picture. (We, at one point, had a District Manager in Kansas

City -whose name I of course can't remember- who prided himself on knowing where all the computer installations were in his district. He thought he knew them all only to find, when he hired a knowledgeable ex IBM sales manager, that his list included less than half of the IBM installations.)

The Forbes Article states that GE's management did not contribute to, or comment on, the article. Forbes thus was relying exclusively on ex-employees who can not be expected to be particularly complementary. It would take a lengthy essay, with contributions from many more knowledgeable than I, to rectify the impression left by this article.

It is obvious that Herb Grosch got his two cents worth in. The comments about GE's management philosophy and lack of computer industry knowledge is almost a Grosch quote. If the quote from Bob Johnson is accurate, it confirms my recollection of him as more of a "laboratory type", "state of the art" engineer than a product designer.

Lasher was 100% right in dropping the W, X, Y and Z line. It would have been sheer folly to try to take on IBM across a broad spectrum of their line when we were running like hell trying to keep up in our several niche markets.

The GE 600 product line encountered all the usual problems involved in trying to convert a military project into a commercially practicable product.

To put the Forbes article in perspective it must be recognized that we were starting nearly five years behind (May 1956 - May 1961). While we were concentrating on ERMA and the NCR 304 and then suffering through the process computers only period, our competitors in the general purpose computer business were ramping up very rapidly. When we got the green light, we had to ramp up one hell of a lot faster just to stay even (Lasher was appointed Acting-General Manager to succeed you with instructions to complete the ERMA and NCR 304 programs and then to limit the department's scope to process computers. The Forbes article hints at this). I wouldn't read too much into Lasher's leaving. My sense is that he put into motion something neither he, nor anyone else, could control. He was too nice a guy and the job rapidly out grew him - and many others. It was already too big for Van Aken by the time he came aboard. Many jobs outgrew people almost as fast as they accepted them.

My reply to people who complained that a new manager didn't know the business was, "Good! The business is a different business today than it was yesterday and it will be a different business tomorrow. We don't need people who know the business as it was yesterday.".

I'm glad you contacted Chuck Ettinger, though he is somewhat controversial, as he is the only section manager to span the Lasher-Van Aken period. In addition, he was appointed (by Stan Smith I believe) General Manager of the Small Computer Operation which included Olivetti. You should ask Vern Schatz for his poem on the history of the Computer Department.

I do not have a copy of the Metcalf or other reports. It's interesting that we didn't place any historic importance on them at the time. WGBH tried unsuccessfully to find them - but they couldn't find you either. Someone told me that George Metcalf is still alive, he might have a copy.

PS: I need a time line. The only dates that I'm sure of is the November 1956 move to Phoenix, and the May 1961 Sales meeting. Do you have a list of corroborated dates? I thought you left in 1959 not 1958 as Forbes says. This would put the "Big Look" approval much earlier than I recall. What were the dates of the changes in General Managers?

I am 100% with Lacy re Hersh Cross and what he did to Van Aken. I would be as charitable as Lacy towards Van except that Van's attitude as a 100% company man made him shoulder the full responsibility rather than letting on that the firings were not totally his decisions. This posture of his figured into the

I can readily believe that the GE600 schedule was shortened by 12 months over that planned by the engineers and software people. As I have previously indicated, the market forces were irresistible and this accelerated schedule was likely required to compete with other "paper tigers" - or quit the game. (I

recounted in an earlier letter how IBM was 18 months late in delivering the system to the B of A that had beaten out the GE600.)

I would love to see a copy of what Lacy has on the June 1960 "Big Look" presentation. This date puts it about 6 months earlier than my poor recollection. Does George Metcalf have a copy of the Metcalf Report?

I'm sure that Helmut Sassenfeld and Dick Lemon will get back to you. The time sharing service business was not, as you put it, "taken away from the Computer Department". By this time the Computer Department had been broken up into several Departments and a number of operating sections and was now the Information Systems Division. I was gone by the time this move was made, but my guess is that one of the reasons for the move was to change a large user of the Division's products into a large customer. At the time of the sale to Honeywell this time sharing service business was using well in excess of $1 Billion worth of computer products. It may also have been to position it out of the sale to Honeywell.

As indicated above there were two aspects of the time sharing business. (a) the sale of timesharing hardware and (b) the sale of time sharing services. The saga with O'Rourke concerns the former.

As I understand it, Van Aken had two sons neither of whom turned out to be role models. To compensate, Van "adopted" two people as the sons he wished he had had. One was Chuck Thompson, now an Executive V.P. of Motorola and the other was Tom O'Rourke. They socialized together, went on golfing trips together and etc. When Rader returned from UNIVAC he brought Ted Green with him and gave him his choice of marketing regions. Ted chose the Western Region which O'Rourke was managing. Van would rather have cut off his right arm than to replace Tom, but with the attitude as indicated above, he did it and never told Tom that it was other than his (Van's) decision. Once the move was made he wanted nothing more to do with Tom. I was not aware of this posture until later when I inadvertently became directly involved.

After Tom was replaced he decided to resign and to go into the time sharing service business. He came to see me to help him with his business plan. (When the Department was split up into several operating sections each Finance sub-section manager was assigned as a financial counselor to an operating section. I had been assigned to Helmut Sassenfeld and thus had some experience and insight into the business.) We prepared the business plan and five year projection for TYMSHARE on my living room floor in Carefree. I then introduced him to Tom Clausen (later Chairman of B of A) who at that time was managing the National Accounts for the Southern part of the B of A and also had an SBIC under his control. He arranged for his SBIC (which was managed by George Quist and was later spun out of the bank and became the investment banking firm of Hambrech and Quist) to make an investment of $250,000 in Tom's company.

Tom then negotiated an order to rent a GE235 timesharing system. Van was aware of these negotiations but apparently had anticipated turning the order down for credit reasons. When that day came Tom said "but Snively has already approved the credit". Van has no master when it comes to "chewing" someone out and I got the full treatment when I was called to his office. I further compounded my error when, during one of his more histrionic moments, I said " You mean that the B of A is not a good credit?". This led to an unbelievable explosion in the midst of which he declared that it was a stupid business

decision to rent GE235's so near the end of the product life cycle and that they were only to made available for cash sale. "Did Tom have enough money to buy one?" The answer was "no" and the meeting ended on that note.

There were immediate repercussions to this policy change as both Ford Motor Company and Chrysler had GE235's on order to rent. Both had IBM instigated policies prohibiting their purchase of computers and both threatened anti-trust suits on the assumption that the policy was an attempt for Helmut's operation to monopolize the business. It took a lot of negotiating to work out of this problem. (I.L. Stevenson -if he has not suffered too many infirmities- can fill you in on these legal maneuverings.)

Meanwhile, Arnold Spielberg had gone to work as V.P.Engineering for Max Palevsky(sp?) at SDS. Tom approached them to design a time sharing machine for him. They agreed to do it if Tom could find a customer for a second machine. Tom got Jim Pontius of Comdisco in Chicago to place an order with him and SDS proceeded to design a machine. TYMSHARE ultimately purchased a large number of timesharing systems from SDS. The GE235 market never took off.

But it really didn't make much difference. By this time the business was being pulled by the market for the GE600 systems. The banking business was no longer such a major factor that Vern Cooper's thoughtless killing of it made a substantial difference and the time sharing hardware market was soon to be satisfied by the GE600.

I enjoyed your Prelude. The motto of the Alumni group should be "it was one hell of a ride"! (page IV "Frank Walsh" should be Jack Walsh.) Tom Watson was a mover and shaker in the same circles with Ralph Cordiner but IBM was not GE's best customer as assumed by Bob Johnson - to the contrary, GE was IBM's biggest customer. IBM bought some vacuum tubes and fractional horsepower motors but not in the significant volume of other radio and TV and appliance manufacturers. The Federal Government was GE's "biggest" customer. The misinformed notion that IBM was GE's "best" customer seems to be endemic among the engineers.

We did bid on the successor to ERMA (Perhaps not with a "responsive proposal" and I have a vague recollection that the bid specs were written in IBM's favor). I can't place the time frame - but we bid the 600 system. Bob Johnson would have left by this time. Lacy Goosetree may still have been involved -

check with him. Al Zipf had moved up in the Bank and I suspected that IBM had managed to infiltrate the Data Processing operation. They would have had to justify their selection of IBM to the rarefied atmosphere that Al Zipf now occupied as being due to "GE's unresponsiveness".

IBM was still stung by your success in getting ERMA. They could not afford to lose the biggest bank in the country a second time. The goodly sum I mentioned was the $250,000 a month for the maintenance of ERMA that IBM had to pay us each month of the 18 months of their delay in getting the successor up and running.

Have you communicated with Reggie Jones? You must have been surprised to learn that he was behind you all the way! I have no idea where he was during the early days of the computer business. Phillippi was an early staunch supported and ally of the Computer Department.

To my knowledge Jack Walsh was never anywhere near the computer business. His biographer would have had to rely heavily on the same media stuff with which we are familiar. I agree that GE was reluctant to sell businesses but they never hesitated to liquidate them! You picked up on the contradiction about acquisitions. However, GE was always constrained by anti-trust considerations (and a not invented here attitude) from making acquisitions. As a result, GE had limited experience in negotiating or managing them.

I don't recall who requested the Builtmore meeting but Phillippi (sp?) had been instructed to drop everything else and concentrate fully on the measurement problem which was confusing management about the Computer Department's P & L.

The attendees, in addition to Phillippi, were George Smith (and one other whose name I can't recall) from Accounting Research Services, ?? McCarthy from Tax Accounting Service (plus as I recall a tax lawyer from legal), Murray Furgeson the leasing expert and GECC Board member from Credit & Collection Services, John Stanger, President of GECC's Leasing Company and Dewey Mehew, Treasurer of GECC. Ken McCombs, Tom Hage and I represented the Computer Department.

These, with a few exceptions, were all very intelligent and imaginative people. I do not believe that there was any alternative or option that was neglected in the wide-ranging discussions over three days of seclusion without interruptions.

John Stanger, who was at heart a "peddler", salivated over the possibility of GECC handling all the rentals and leases. Dewey Mehew pointed out that CECC was organized under the banking laws of The State of New York and as such would be in violation of banking regulation "Y" if they entered into non-pay-out rentals and leases. McCarthy discussed the possibility of the consolidation of GECC with GE, which would eliminate the assumed arms length advantage in any event.

We were promoting my scheme of "POLO" which involved splitting the Computer Department into a Product Operation that would sell to a Leasing Operation. The issue here became the question of transfer price. McCombs tested everybody's sharpness by making foolish statements. Tommy Hage was a nervous wreck trying to figure out who was winning so that he could be on the winning side. Needless to say the discussions were far ranging and exhaustive. I have previously related Phillippis’s summation.

PS: I suspect that GE's top management was frequently cautioned by the tax lawyers about making public statements explaining "losses" in the computer business as tax shelter.

Very early on (and prior to the Builtmore meeting - which was principally involved with accounting issues) we were imaginatively and creatively structuring "full pay-out net leases" which we could book as "sales" -see note below. These leases were attractive to our banking and utility customers. I also established early relationships with the first personal property leasing companies - Al Zises of Bankers Leasing and Pod Boothe of Boothe Leasing. A little later I established an exclusive relationship with Lease Financing Corporation who administered a number of tax sheltered leasing partnerships set up by Ollie Vanderbilt. Our competitors could not meet the attractive and flexible terms we were able to offer the banks and other customers with this program.

With the advent of the GE 400, working with the principals of Lease Financing Corp., we structured a plan for third party financing of intermediate term non-pay-out rentals which GE could book as "sales'. This was the 4-5-6 Plan, named for it's four, five and six year terms. This was in the era of the 8 year

investment tax credit and 8 year net-net leases. This was an extremely complex plan which required the establishment of an arms length third party -Systems Capital Corporation (SCC). SCC provided part of the risk capital -raised through a public offering underwritten by Blair & Company, Ollie Vanderbilt's investment banking firm. The senior debt for the firm portion of the leases was furnished by AEtna (with the commercial banks warehousing them to a $10million size). Ollie's tax sheltered limited partnerships provided leverage on this debt to generate a low cost of funds and some risk capital. This program was relatively successful by itself, but it was highly successful in getting the customers to realistically assess the useful life of their installations and many were thus persuaded to purchase or lease (not rent).

Obtaining approval of this program within GE was difficult to say the least. To have to admit that there was something they couldn't do between the financial resources of GE and GE Credit was traumatic indeed. The full power of the GE Credit Corp.'s management descended on me a number of times before the program was implemented. But the problem was too big to ignore - and they couldn't provide a better alternative.

The ultimate beneficiary of this program was Nate Norris. Ollie Vanderbuilt called me one day in great distress after one of the banking customers, near the end of the rental term, had called inquiring for his address in order to return three GE 400's. I informed him that Nate Norris had just begun making a

market in used GE 400's and gave him Nate's number. Ollie did not have to clean out his carriage house and Nate got an inexpensive supply of product! (Nate ultimately sold his company for $6,million - $4 million in cash and a note. He later got it back when the buyer defaulted on the note.)

Too bad you and I lost contact. In 1970 I started the Snively Financial Organization to handle short term rentals. I'm sure we could have provided you a better program than Trans Union Leasing Corp. did.

I have only interesting personal recollections of Herb Grosch. I don't recall anything directly relating to the business, except his obtaining the Huntsville contract. In my memory, he wasn't around very long. His "modus operandi" was to operate as a shop steward. He had his programmers convinced that he and only he stood between them and management. That without him they would be subjected to all kinds of indignities, such as coming to work on time, committing to completion dates, budgets and etc. I believe that Hal Norris spent some time in Herb's operation and might have some comments.

Note: The Company-wide task force that both Lacy Goosetree (from the Communications Department) and I served on in late 1956 to early 1958 issued two definitive volumes on leasing. One relating to the Lessor and the other to the Lessee. These extensive reports anticipated the FASB and IRS rulings by nearly 10 years. We were too smart for our own good. While we were smugly anticipating the future FASB, IRS and SEC rulings, others such as Memorex, Storage Technologies and Data Products were furiously playing by the then existing rules.

GE made a "cash-less" entry into the business when progress payments from the B of A for ERMA and NCR for the 304 created positive cash flows for the first two years.

This was followed by years of large "tax shelter" created by GE's bookkeeping for the rental of computers.

Then a profitable sale of the hardware business, structured such that there was no recapture of the tax deferments.

Finally, continuing large profits from the huge time-sharing and networking business that it retained.

I am up to Part IV in my review of the book. I remain impressed with your intuition about conversations that you weren't party to and the captivating flow of the book.

Enclosed are copies of the pages that I have made editing changes or etc. Don't spoil the flow of the story if incorporating any of these suggestions does so - I appreciate the need for a measure of poetic license. Most of the corrections are self explanatory but I expand on some of them below:

Page III. If there was an award for you, don't be too modest to mention it. If not, it's because you "disappeared" and no one knew what had happened to you.

Page 4. They were called Product Scope, not Product Charters. Product Charter nay be a better understood term for non GE people. By the same token, outsiders think the Computer Department is the DP operation - Division is a better understood term. However, substituting it in this saga would lead to no end of confusion. Do you require editorial consistency?

Page 30. At one point, when we were trying to extend the ERMA derivatives to the smaller banks, I lobbied long and hard to market them at x cents per items processed. The banks had a "rule of thumb" that it cost them $.03 per item. (No one was able to tell me where the number came from - but they believed it.) I figured we could get the system in at low volume and grow as the bank grew. Replacements would be at our option as a cost reduction rather than the bank decision based on competitive activities.