This is a true story of how a meeting with a 14-year-old kid, contributes to the kid becoming a multi-billionaire. It is the story of how an off-hand gift contributes to the formation of an empire worth billions of dollars. It focuses on how the gift-giver lives in a forty-three year-old fantasy about payback – his silver bullet.
The Story
Ever wonder what happened to "old what's-his-name"? My brush with what's-his-name began on Monday morning March 4, 1969, when I excitedly told my boss at Boeing, that I was leaving for a real job with Digital Equipment Corporation, or DEC.
My hiring was justified by the selling a second large-scale computer system to Computer Center Corporation (CCC). CCC was a Seattle-based, start up timesharing service bureau. My job as a Software Support Specialist in Digital's local office came out of that sale. The CCC types of companies were the progenitors of today's online services such as America On Line (AOL).
I had never used nor heard of DEC’s large-scale PDP-10 computer, in spite of my extensive systems and applications programming experience with large-scale IBM computers,. The PDP-10 computer system was the industry's first commercially viable timesharing system, could support 32 simultaneous users. In late March 1969, I went to the Company's headquarters in Maynard, Massachusetts, a Boston suburb for five months of training to be a Software Support Specialist.
Returning to Seattle in September, I went to CCC and spent a half-day meeting its chief technical people -- Dick Gruen, and Bill Weir. Dick had previously worked at Digital Equipment Corporation's (DEC's) headquarters where he programmed portions of the computer’s operating system.
In Dick's new role, he helped CCC write a very stringent purchase contract for its second large-scale computer system. The key terms of this computer purchase contract were based on what is called the "acceptance criteria." Acceptance is the point in time when both parties agree that the computer has performed sufficiently well for payments to commence. This tough contract stated that for a seven-day period, there could be "no more that four faults that caused the system to stop."
DEC must have been desperate for this sale, because that type of acceptance criteria was barely within the 1969 times-sharing system state-of-the-art. Being a contract participant, Dick seemed to be the only person involved with the contract, who fully understood that fact.
Bill Weir, with five years of previous systems experience with DEC's large-scale computers, had the ability to process machine instructions at the speed of thought. While reading a listing of machine code (assembly language) instructions he could point out errors as he read.
With only five months of training, I realized that I was facing a formidable technical challenge. It seemed that the only way to "win" was to focus on getting the new system accepted rather than engaging in technical masturbation with CCC's technical heavyweights.
During my first week at CCC, I was handed a list of seven system stopping problems (bugs). With help from my boss, Larry Wade, The Western Regional Software Manager, I managed to get the list down to three items, so we could start acceptance testing.
DEC PDP-10 Console
Four days into acceptance testing, Bill and Dick took me aside, and acting like hyperactive kids, gleefully asked me to watch the execution of a program they had written. Bill hit a few keys on the computer console to start the program. A message started to be typed on the computer console -- "By the time this message completes, all memory will be zeerroooeess!!! 1….2……3…….. In that brief period, all memory was indeed overwritten with zeroes. They had exercised a bug that allowed unlimited input/output (I/O) of any chosen data to any place in computer memory, including the area where the operating system resided.
"Does running this program four times count as four separate faults?" Bill politely asked while affecting a puzzled face.
"No." I said. "One bug caused the problem and executing it more than once is still only one bug." As I made an embarrassed retreat to my office, I said, "I'll have to check with headquarters to see where we go from here,"
Back at my office, I held a strategy meeting with the sales manager. We concluded that CCC wanted to hold off the completion of acceptance until it loaded the new computer with paying customers. It wanted to make money without making payments to DEC.
We wrote an angry letter to CCC's management informing them that the machine "could not be used under any circumstances for its paying customers until acceptance testing was completed." The letter achieved its purpose.
I called headquarters and the decision was made to send an operating system designer to Seattle to solve the I/O bug and some other outstanding bugs. After a week of intensive effort, we restarted acceptance testing with only one serious bug on the list. This stop-and-start pattern continued for several weeks.
About the first week in November, CCC shifted its strategy to focusing on the non-operating system software such as high-level computer languages (Basic, Fortran, etc.). Dick and Bill introduced me to a group of six kids from a private school in the suburbs. CCC agreed to give them almost unlimited, free computer time if they would try to find bugs in the non-operating system software. The kids were thrilled to be involved in the bug finding game. I was particularly impressed with one kid, who seemed to be the leader of this little band of software terrorists. He was very thin with bushy blonde hair and large glasses that seemed to cover his entire face. In spite of our adversarial situation, I soon recognized that these kids were unusually bright, friendly and respectful.
With mutual agreement that these non-operating system bugs were not part of the acceptance testing, I kept the pressure on CCC to allow the acceptance testing to continue. The game came to a screeching halt when, on November 15th, the phone company shut off CCC's phone system. At that point, we all had to acknowledge that this was the end, and that Computer Center Corporation had died an agonizing death.
About a month later the receptionist rang my phone and said, "There are some gentlemen here to see you."
My mouth nearly dropped to my knees when I walked out into the lobby. Three of the software kamikazes from Lakeside School were standing there beaming from ear to ear. Their leader spoke first.
"Remember us?"
"I most certainly do. You're the kids from CCC."
"Yes, we wondered if we could ask a favor?"
"No problem, what can I do for you?" I said
"We have a computer club and we would like to buy your latest PDP-10 software manuals."
"Come into my office and I'll see what we can find."
The four of us squeezed into my tiny office and we started going through the stack of manuals on my bookshelf.
"All of these are extras." I said, "You guys can have them at no charge."
"Gee, thanks, we really appreciate that."
They marched out of the office with an armload of manuals and that was the last time I saw them. Six months later I was transferred to DEC's Regional Office in California. My promotion to the position of Western Regional Software Support Manager meant that the CCC situation had not damaged my career.
About three years ago I read a description of the CCC debacle in a biography about one of the world's richest men. It is still hard to believe that I was involved, if only tangentially, in the development of his success.
What's-his-name, turned out to be Bill Gates.
Bill Gates
For thirty-four years now, I have been walking around unknowingly with a silver bullet. Over those years, I have watched the phenomenal growth of Microsoft Corporation. One day I will crank up the chutzpah to contact Bill Gates, Microsoft's Founder and Chairman, and tell him that I made a mistake about the PDP-10 manuals and that $100,000 is now due to cover their cost. I doubt that I will get a reply, but it is a very comforting thought.
PS – I will take a check.
