Steve Blank

This is Part IV of how I came to write “The Secret History of Silicon Valley“.
Read Part III first and it will make a bit more sense.

All You Can Read Without a Library Card

It was 1978. Here I was, a very junior employee of ESL, a company with its hands in the heart of our Cold War strategy. Clueless about the chess game being played in Washington, I was just a minion in a corporate halfway house in between my military career and entrepreneurship.

ESL sent me overseas to a secret site run by one of the company’s “customers.”  It was so secret the entire site was could have qualified as one of Dick Cheney’s “undisclosed locations.” As a going away gift my roommates got me a joke disguise kit with a fake nose, glasses and mustache.

The ESL equipment were racks of the latest semiconductors designed into a system so complicated that the mean-time-between-failure was measured in days. Before leaving California, the engineers gave me a course in this specialized receiver design. Since I had spent the last four years working on advanced Air Force electronic intelligence receivers, I thought there wouldn’t be anything new.  The reality was pretty humbling. Here was a real-world example of the Cold War “offset strategy.” Taking concepts that had been only abstract Ph.D theses, ESL had built receivers so sensitive they seemed like science fiction.  For the first time we were able to process analog signals (think radio waves) and manipulate them in the digital domain. We were combining Stanford Engineering theory with ESL design engineers and implementing it with chips so new we were debugging the silicon as we were debugging the entire system.  And we were using thousands of chips in a configuration no rational commercial customer could imagine or afford.  The concepts were so radically different that I spent weeks dreaming about the system theory and waking up with headaches. Nothing I would work on in the next 30 years was as bleeding edge.

Now half a world away on the customer site, my very small role was to keep our equipment running and train the “customer.” As complex as it was, our subsystem was only maybe one-twentieth of what was contained in that entire site. Since this was a location that worked 24/7, I was on the night shift (my favorite time of the day.) Because I could get through what I needed to do quickly, there wasn’t much else to do except to read.  As the sun came up, I’d step out of the chilled buildings and go for an early morning run outside the perimeter fence to beat the desert heat.  As I ran, if I looked at the base behind the fence I was staring at the most advanced technology of the 20th century.  Yet if turned my head the other way, I’d stare out at a landscape that was untouched by humans.  I was in between the two thinking of this movie scene.  (At the end of a run I used to lay out and relax on the rocks to rest – at least I did, until the guards asked if I knew that there were more poisonous things per square foot here than anywhere in the world.)

Before long I realized that down the hall sat all the manuals for all the equipment at the entire site. Twenty times more technical reading than just my equipment. Although all the manuals were in safes, the whole site was so secure that anybody who had access to that site had access to everything – including other compartmentalized systems that had nothing to do with me – and that I wasn’t cleared for.  Back home at ESL control of compartmentalized documents were incredibly strict. As a contractor handling the “customer’s” information, ESL went by the book with librarians inside the vaults and had strict document access and control procedures.  In contrast, this site belonged to the “customer.”  They set their own rules about how documents were handled, and the safes were open to everyone.

I was now inside the firewall with access to everything.  It never dawned on me that this might not be a good idea.

Starting on the safe on the left side, moving to the safe on the right side, I planned to read my way through every technical manual of every customer system.  We’re talking about a row of 20 or so safes each with five drawers, and each drawer full of manuals. Because I kept finding interesting connections and new facts, I kept notes, and since the whole place was classified, I thought, “Oh, I’ll keep the notes in one of these safes.” So I started a notebook, dutifully putting the classification on the top and bottom of each page.  As I ran into more systems I added the additional code words that on the classification headers.  Soon each page of my notes had a header and footer that read something like this:  Top Secret / codeword/ codeword / codeword / codeword / codeword / codeword / codeword.

I was in one of the most isolated places on earth yet here I was wired into everywhere on earth.  Coming to work I would walk down the very long, silent, empty corridors, open a non-descript door and enter the operations floor (which looked like a miniature NASA Mission Control), plug a headset into the networked audio that connected all the console operators — and hear the Rolling Stones “Sympathy for the Devil.”  (With no apparent irony.) But when the targets lit up, the music and chatter would stop, and the communications would get very professional.

Nine months into my year tour, and seven months into my reading program, I was learning something interesting every day. (We could do what!?  From where??)  Then one day I got a call from the head of security to say, “Hey, Steve can you stop into my office when you get a chance?”

Are These Yours?

Now this was a small site, about 100-200 people, and here was the head of security was asking me over for coffee.  Why how nice, I thought, he just wants to get to know me better. (Duh.)  When I got to his office, we made some small talk and then he opened up a small envelope, tapped it on a white sheet of paper, and low and behold, three or four long black curly hairs fall out.  “Are these yours?” he asked me.

This the one of the very few times I’ve been, really, really impressed.  I said, “Why yes they are, where did you get them?”  He replied, ‘They were found in the ‘name of system I should have absolutely no knowledge or access to’ manuals. Were you reading those?”  I said, “Absolutely.”  When he asked me, “Were you reading anything else?”  I explained, “Well I started on the safe on the left, and have been reading my way through and I’m about three quarters of the way done.”

Now it was his turn to be surprised. He just stared at me for awhile.  “Why on earth are you doing that?” he said in a real quiet voice. I blurted out, “Oh, it’s really interesting, I never knew all this stuff and I’ve been making all these notes, and …”  I never quite understood the word “startled” before this moment.  He did a double-take out of the movies and interrupted me, “You’ve been making notes?”  I said, “Yeah, it’s like a puzzle,” I explained.  “I found out all this great stuff and kept notes and stored in the safe on the bottom right under all the…”  And he literally ran out of the office to the safes and got my notebook and started reading it in front of me.

And the joke (now) was that even though this was the secret, secret, secret, secret site, the document I had created was more secret than the site.

While the manuals described technical equipment, I was reading about all the equipment and making connections and seeing patterns across 20 systems. And when I wasn’t reading, I was also teaching operations which gave me a pretty good understanding of what we were looking for on the other side.  At times we got the end product reports from the “customer” back at the site, and these allowed me to understand how our system was cued by other sensors collecting other parts of the electromagnetic spectrum, and to start looking for them, then figuring out what their capabilities were.

Pattern Recognition

As I acquired a new piece of data, it would light up a new set of my neurons, and I would correlate it, write it down and go back through reams of manuals remembering that there was a mention elsewhere of something connected.  By the time the security chief and I were having our ‘curly hairs in the envelope’ conversation, not only did I know what every single part of our site did, but what scared the security guy is that I had also put together a pretty good guesstimate of what other systems we had in place worldwide.

For one small moment in time, I may have assembled a picture of the sum of the state of U.S. signals intelligence in 1978 − the breadth and depth of the integrated system of technical assets we had in space, air, land, and other places all focused on collection. (If you’re a techie, you’d be blown away even 40 years later.) And the document that the head of security had in his hand and was reading, as he told me later, he wasn’t cleared to read – and I wasn’t cleared to write or see.  I’m sure I knew just a very small fraction of what was going on, but still it was much more than I was cleared for.

At the time this seemed quite funny to me probably because I was completely clueless about what I had done, and thought that no one could believe there was another intent.  But in hindsight, rather than the career I did have, I could now just be getting out of federal prison. It still sends shivers up my back.  After what I assume were a few phone calls back to Washington, the rules said they couldn’t destroy my notebook, but they couldn’t keep it at the site either.  Instead my notebook was couriered to Washington – back to the “customer.”  (I picture it still sitting in some secure warehouse.)  The head of security and I agreed my library hours were over and I would take up another hobby until I went home.

Thank you to the security people who could tell the difference between an idiot and a spy.

When I got back to Sunnyvale, my biggest surprise was that I didn’t get into trouble. Instead someone realized that the knowledge I had accumulated could provide the big picture to brief new guys “read in” to this compartmentalized program.  Of course I had to work with the customer to scrub the information to get its classification back down to our compartmental clearance. (My officemate who would replace me on the site, Richard Farley, would go on to a more tragic career.)  I continued to give these briefings as a consultant to ESL even after I had joined my first chip startup; Zilog.

Two Roads Diverged in a Wood and I took the Road Less Traveled By,  And That Has Made All the Difference

Extraordinary times bring extraordinary people to the front. Bill Perry the founder of ESL (and later Secretary of Defense)  is now acknowledged as one of the founders of the entire field of National Reconnaissance, working with the NSA, CIA and the NRO to develop systems to intercept and evaluate Soviet missile telemetry and communications intelligence.

ESL had no marketing people.  It had no PR agency.  It shunned publicity.  It was the model for almost every military startup that followed, and its alumni who lived through its engineering and customer-centric culture had a profound effect on the rest of the valley, the intelligence community and the country. And during the Cold War it sat side by side with commercial firms in Silicon Valley, with its nondescript sign on the front lawn. It had Hidden in Plain Sight.

As for me, after a few years I decided that into was time to turn swords into plowshares. I left ESL and the black world for a career in startups; semiconductors, supercomputers, consumer electronics, video games and enterprise software.

I never looked back.

It would be decades before I understood what an extraordinary company I had worked for.

Thank you Bill Perry for one heck of a start in Silicon Valley.

I was 24.

My first class of students at ESL: Guardrail V Training Class (note the long black curly hairs)

Part IVb of the Secret History of Silicon Valley continues here.

Filed under: ESL, Secret History of Silicon Valley, Technology | Tagged: Bill Perry, Cold War, ESL, Signals Intelligence, Steve Blank | 10 Comments »

This is Part III of how I came to write “The Secret History of Silicon Valley“.

1978. Two years out of the Air Force, serendipity (which would be my lifelong form of career planning) found me in Silicon Valley working for my first company: ESL. If you’re an entrepreneur, ESL is the most important company you’ve never heard of. If you are a practitioner of Customer Development, ESL was doing it before most us were born. If you think the Cold War turned out the right side up (i.e. Communism being a bad science experiment) ESL’s founder Bill Perry was moving the chess pieces. And no one who really knew could tell you.

Bill Perry’s public life as Secretary of Defense and his subsequent work in preventing nuclear proliferation and nuclear terrorism is public knowledge. But part of his life that that doesn’t even merit a Wikipedia entry is that Bill Perry used Silicon Valley to help end the cold war.

Fred Terman Sent Us

In 1953 the U.S. Army needed to build missile and proximity fuse jammers and Quick Reaction Capability (QRC) systems (translation: the other side just came up with something that’s killing us in a shooting war, get us a fix quick.) The Army offered Fred Terman, the Dean of Engineering at Stanford, a $5M contract to build an electronics countermeasures lab. When Terman said no, Sylvania, a tube company which built proximity fuse tubes in WWII, won the contract and set up its Electronic Defense Lab (EDL) in Mountain View California in the middle of an orchard. Terman became a consultant to the company.

In ten years Sylvania EDL grew to be one the largest companies in the valley − 1300 people were working on electronic countermeasures and electronic intelligence. By 1961 its customers now included our intelligence agencies. (BTW, when the customers were “three-letter” intelligence agencies, contractors used an oblique way of talking about who they were working for: they were all referred to as simply the “customer.”)

In 1964, Bill Perry, the head of the lab, frustrated with GTE’s management, quit (GTE, a phone company had bought Sylvania in 1959.) And in the tradition of great startups, on the way out Perry took 6 of his best managers with him.

At ESL Military Intelligence Was No Longer an Oxymoron

Perry not only took his best managers, but he also took his customers, and his desire to build a company culture that was the antithesis of working for a phone company. In building ESL Perry made a conscious choice to emulate Hewlett Packard (then considered the “gold standard” of a great technology company.) HP had an ethical culture, entrepreneurial spirit, and deep Stanford engineering department connections. One key difference: unlike HP, which had restricted stock ownership to the founders and top management, Perry made sure everyone at ESL had stock. There were no venture investors. The “customers’” contracts funded the company. Seven years later in 1971 ESL went public.

Not surprising with a CEO with a PhD in Math, at ESL the engineers ran the company, pursuing bleeding-edge designs in antennas, receivers and microwaves – at times hand in hand with Stanford’s engineering department. (Some of this stuff was so advanced that the rumors were that we got it from the alien spacecraft hidden at Wright-Patterson Air Force base.)

ESL was unique among the “we do microwaves” that the Valley specialized in before it was Silicon Valley. ESL was a systems company that used computers, and in the mid-1960’s using computers for electronic intelligence was considered revolutionary. ESL specialized in embedding minicomputers in electronic intelligence systems, turning a tedious manual process into one that looked like magic. The “customers” in Washington had never seen anything like it.

While those computer-based systems paid the bills, Perry’s even more profound insight would change the outcome of the Cold War.  Up until ESL, radio and radar signals had always been received by analog receivers.  ESL realized that by turning these radio waves into computer bits, ones and zeros, they could be processed in ways that had been considered theoretically impossible.  ESL’s systems allowed signal extraction and correlation against targets the Soviet Union thought were undetectable and impenetrable. But this digital world required new theories, and new devices – two items provided by Silicon Valley in the form of Stanford’s engineering department and the emerging/booming semiconductor business.

ESL and “the Customer” – No Such Agency

ESL kept getting business and growing mostly through unsolicited bids. Because they were extremely good at what they did, most of the contracts they won were “sole source.” However, it didn’t hurt that Perry several allies at the “customer.” One of them, Bud Wheelon, had been a classmate of Perry’s at Stanford and they both had worked on the electronic intelligence collection problem, Perry at Sylvania EDL and Wheelon at the Space Technology Lab at Ramo Woolridge. In 1962 Wheelon left for a new job as the first director of the CIA’s Directorate of Science and Technology where he was responsible for development of OXCART, the A-12 Spyplane, and three major satellite reconnaissance systems.  These would be the heart of ESL’s business.

A-12 OXCART CIA Spyplane

ESL found other ways to stay very close to its customers. Forty years before Customer and Agile Development methodologies became popular, ESL had analysts from its “customer” sitting side-by-side with ESL engineers designing new equipment together. And in the 1960s ESL’s customers asked the company to analyze and interpret telemetry data even though this was a traditional function of the “customer.” In five years, ESL went from a plucky startup to the market leader in Sigint and telemetry intercepts. While it was a for-profit company, Perry believed ESL’s goal was to serve the national interest instead of just the stockholders. He identified with their customers, not shareholders. If there was a conflict between profits and doing the right thing, at ESL the goal was to “think of the country first.” Yet ESL was just act one for Bill Perry.

Yes We Can – Dumping Detente – Bill Perry and “the Revolution in Military Affairs”

After 20 years of an escalating arms race, the Nixon administration decided to take a new approach to dealing with the Soviet Union: Détente. Kissenger’s thinking was: history may be tilting to the Communists and we may not be able to win the struggle with the Soviet Union so let’s settle for parity. Yet while the U.S. had been engaged in the Vietnam War, and had agreed to parity in nuclear weapons, Soviet forces in Europe had built a 3 to 1 advantage in tanks, artillery, armored personnel carriers, and soldiers, all under Détente.

In response the U.S. dumped Détente and embraced a new strategy to counter the Warsaw Pact by not matching them tank for tank or solider to solider. The new insight was that we could change the game completely and take advantage of a lead we had that was getting longer every day – by using our computer and chip technology to aggressively build a new generation of weapons that the Soviet Union could not.

At the heart of this idea was something called “precision strike,” what we would today call smart bombs or precision guided munitions. But this new strategy was more than making the bombs smarter. It involved building stealth aircraft to deliver these precision weapons unseen by any enemy radar, and designing intelligence and reconnaissance systems that would target for them. Smart weapons, smart sensors, and stealth.  And the heart of all of this were microwaves, silicon chips, electronics and computers that only the U.S. could design and produce, and a good part of it was coming from Silicon Valley.

The Arms Factories that Won the Cold War Were Semiconductor Factories

Who was the government official pushing all of this? It was none other than Bill Perry, who had become the head of Research and Engineering for the Defense Department. From 1977 to 1981 Perry cranked up spending for research and development on a massive scale. The budget for the Defense Advanced Research Projects Agency (DARPA) doubled (with the Assault Breaker to turn U.S. asymmetric technology advances into weapons), and huge “smart weapons” defense programs like the F-117 stealth ground attack plane and the B-2 stealth bomber; precision guided munitions; JSTARS, a surveillance system; and the satellite Global Positioning System (GPS); MX missile; Trident submarine; and Tomahawk cruise missiles.

F-117 Nighthawk – 1st Stealth Ground Attack Plane

These changes in American defense policy spooked the Soviets. The Chief of Staff of the Red Army said that this “Offset Strategy” was revolutionizing contemporary warfare and posed a military threat that the Red Army could not match. “We cannot equal the quality of US arms for a generation or two. . . . We will never be able to catch up with you in modern arms until we have an economic revolution. And the question is whether we can have an economic revolution without a political revolution.”

The U.S. Cold War strategy had gone from a “let’s be friends” to a “yes we can win” strategy. By the mid 1980s Ronald Reagan was cranking U.S. defense spending even higher. Gorbachev, now the Soviet Premier, had to grapple with the spiraling cost of military systems that weren’t amortized by consumer purchases. Arms control with the U.S. and massive cuts in weapons and the military seemed like the only way out. And the rest is history.

Bill Perry was my first boss in Silicon Valley.
He put us on the path to use Silicon Valley as a weapon in the cold war.

My small part as a foot solider in this adventure is in the next post.

Part IV of the Secret History of Silicon Valley continues here.

Filed under: ESL, Secret History of Silicon Valley, Technology | Tagged: Bill Perry, Cold War, ESL, Signals Intelligence, Steve Blank | 15 Comments »