My Beginnings in Deep Tech

The new job was the start of what would be more than 35 years reporting on semiconductors in what some called the deep tech trades. 

Like ACM, Electronic World News carried a grand mission on the backs of a skeleton staff of writers and salespeople. In EWN’s case, most of the crew had day jobs at EE Times or EBN. In those days, working for EE Times was a prestigious, but demanding gig. The weekly newspaper had recently risen to the top of the heap in its niche as the Bible for the folks who made chips or anything that used them.

The staffing synergies CMP hoped EWN would find eventually became points of contention, cracks that started to show. And like other colonists, the newspaper’s publishers failed to see their so-called global market was really a discrete collection of local markets that preferred local publications in their local languages.

At the time, my concerns were more basic. I wasn’t really sure what semiconductors were or who made them with the exception of Intel and its processors. Rob convinced me that was a fine start since Intel, thanks to its 386 CPUs, was on its way to becoming the king of the hill.

With more learning and luck, I soon made a name among my New York editors by mining nuggets of Intel corporate secrets from a handful of Hong Kong’s computer engineers.

One of first sources was Johnny, an affable and talkative engineer at a small computer design outfit near my office in Wan Chai, the dingy district in the shadow of Central’s towers where, back in the day, sailors spent their money on booze and women.

I heard a rumor Intel was canceling its 486SL, an integrated chip for the next generation of laptops. It was a big deal since laptops were seen as the vanguard of a next big thing -- mobile computing -- and the 486SL was positioned as their primary engine. So, I called Johnny. He said he didn’t know, but would check with his Intel rep and get back to me. An hour later he called back with all the glorious details.

It was the first time I made the lead slot on page 1 of EE Times, something I discovered amid a pile of faxes in my office the following Monday morning.

The Tai Yau building in Wan Chai where 

I worked for EWN in a four-person office.

Johnny taught me a great career lesson: Engineers tell the truth. They don’t even think about putting a spin on it. Truth is a natural byproduct of the engineering mind which is built to be curious and appreciate the little ironies of how things actually work or don’t. I realized that if you talk to the engineers, you’ll get the real story, the story the marketing and PR folks typically don’t want you to know because it dims the halo they are constantly polishing around their corporate image and brand.

For thirty years of my career, I lived this credo: Engineers are the key to tech news, and EWN and EE Times were all about engineers.

Ironically, I came to believe a corollary that wasn’t always so useful: PR people are the enemy, don’t let them know what you are doing because they will alert others and put up road blocks, delays and spin to get you off the real story. What’s more, I started to think they were just plain dumb.

Turns out that was only sometimes true. Years later, in my first corporate job, I worked with plenty of PR people smart enough to be making twice my salary doing work that required more common sense, emotional intelligence and calm under pressure than I possessed. I saw how they gave out scoops to reporters who played the game and bought into the mainly legitimate stories they were offering.

So, starting in my Hong Kong days, I cultivated sources like the director of engineering at Wongs Electronics, Hong Kong’s largest contract manufacturer, but a minnow in the broader computer industry. When I asked him for an interview, he was amazed anyone wanted to listen to him talk about his job and its frustrations. I suspect even his wife didn’t want to hear it. In those days, companies like Wongs didn’t have a PR department or at least not one that knew the company had an engineering department that needed to be secured against cold calls from young and hungry tech reporters.

I don’t recall any names now, but one Wongs engineer hinted that Intel was about to embed interconnects in a future CPU that would make it easy for engineers to pack four processors in one computer, all sharing one pool of memory. It was a big step in Intel’s strategy to go after the market for servers, big back-end office systems. As various forms of memory sharing evolved, Intel’s interconnect strategy stretched my understanding of how computers worked.

Tracking interconnects became a pastime that would lead to many stories as networked PCs were ganging up to take on minicomputers and mainframes. To this day, interconnects remain a secret sauce that continues to give computer makers a competitive edge. The on-chip links in today’s processors let engineers build sophisticated silicon complexes, Lego-like, out of many small chiplets. Intel used its interconnects (among other things) to slowly kill off more expensive minicomputers with PC servers that were nearly as powerful and much cheaper, laying the foundation for cloud computing. But circa 1990, these were just forward-thinking concepts in the heads of a few smart computer engineers.

A Small Cast of Characters

In the time before anyone heard of a browser, I was having a field day as the only semiconductor reporter in Hong Kong, but it was a small pond. One of the biggest fish in it was Tam Chung-Ding aka C.D. Tam (below), a short, gregarious and sophisticated Motorola exec who looked and moved like a Chinese Jean-Paul Sartre if the French philosopher had consumed a gallon of coffee.

Tam aspired to get Motorola to build Hong Kong’s first big fab for making semiconductors. It wasn’t to be, but the company’s much less glamorous chip assembly plant almost singlehandedly established the colony’s first tech park. It sat next to Motorola’s regional office on the shore of Tai Po Wan, a bay that zigzagged through an unspoiled northern nook of Hong Kong, a good 90-minute subway and train ride from Central and just a few train stations short of China’s border.

Tam’s position and outgoing ways made him a media darling of the local press. One of his pet projects was the Motorola Dragonball processor, an integrated chip designed in Hong Kong that powered the Sharp Zaurus (below), a sort of calculator on steroids. It supported handwriting recognition, such as it was in those days, and was one of the many cool but somewhat clunky and limited portable gadgets that were harbingers of the smartphone.

Hong Kong loved gadgets of all sorts. Office messengers prided themselves by carrying pagers on their belts. The most elite messengers had two-way pagers. For a while, Motorola made the MicroTAC, an analog handset with a single-line, text-only display. It was a precursor of the flip phone and looked like the handsets used in Star Trek. I only recall seeing Microtac ads. I think they were too expensive for messengers and too limited for execs. But they sure looked cool.

Despite his enthusiasm and his way with the press, Tam never managed to convince his corporate bosses to build a fab in Hong Kong, but he did spark a project to make chips for cars in Tianjin, a distant north China industrial town with aspirations to be the Detroit of the East. Unfortunately, neither the Tianjin plant nor Dragonball, or even Motorola for that matter, had long to thrive in the cutthroat chip industry.

A colleague in Taiwan later reported the shell of the Tianjin plant, never outfitted, was sold eventually to China’s first successful chip company, Semiconductor Manufacturing International Co. or SMIC.

Looking around the Tai Po district for other stories, I met Alan Wong (below), a handsome entrepreneur behind Vtech, another local success story. Wong came up with an idea for making cool-looking calculators for kids, cross-pollenating the emerging PC craze with the skills of the local toy manufacturing sector. The products foreshadowed the consumer PCs that were still several years in the future. Vtech’s products appealed to upwardly mobile parents and became a proud focus for Hong Kong’s toy business that otherwise focused on trendy no-tech playthings.

Our communications technology was also at a primitive stage in those pre-internet days. I filed stories using a complex script on my Macintosh Plus to dial a private phone number and issue commands that invoked EasyLink, an expensive service AT&T hosted. Sending a typical 750-word story took several minutes and might require a re-try or two if there was a burp in the phone line.

Getting feedback on my stories took even longer. We didn’t have email or any other messaging service. All other communications traveled by fax machines, crude networked printers that could send and receive black-and-white image files (slowly) over the phone network.

On Friday afternoons, the New York office would send me a fax of the newspaper’s front page and its table of contents, so I could see where my stories landed. Sometimes I came into the office on Saturday mornings to pick up the fax, too eager to wait for Monday morning. I read the smudgy pages like an Egyptologist deciphering what my bosses 7,000 miles away were trying to tell me about what they liked.

In those days, fax machines were considered cool. Whenever I asked for an interview with an executive, the Hong Kong PR person or admin would routinely say, “send me a fax.” Even if I had just explained the request in detail, they still wanted a fax, perhaps, I thought, just to savor the cool factor.

My biggest story of those days didn’t require any fancy communications technology. It arrived over lunch with Valdis Dunis (below). A young and entrepreneurial Australian drawn to Hong Kong’s frothy energy, Valdis sported a trim mustache that looked like it would jump off his face with excitement as he shared stories, his hands stabbing the air over his head for emphasis. We met at some cocktail reception when he was a newly minted Asia marketing manager for Texas Instruments with a budget to travel the region spreading the good news about semiconductors.

At our first lunch, he gave me a full mind dump about TI’s plans to attack Intel with a family of 386-like CPUs. I kept thinking this was an amazing story that he shouldn’t be telling me, but I wasn’t about to stop him. Instead, I excused myself to the rest room every so often so I could jot down as many details as I could remember.

When I got back to the office the story wrote itself. I dashed it off to my editors, bragging that the Thai lunch was yummy, too. Again, I hit the lede slot on EE Times’s front page, giving my ego and reputation a boost.

Later, Valdis called letting me know his mistake. The “confidential” tag had come off the Powerpoint slides he and his boss were about to present to computer makers in the region, so he assumed the information had become public knowledge. It wasn’t of course, and Leon Adams who ran the Rivers chip program at TI, was livid about my scoop.

Luckily, Valdis kept his job and his irrepressibly bubbly nature, and we maintained our friendship. He even participated in a silly scavenger hunt my wife organized for one of my birthdays. And a decade after I left Hong Kong, he gave me a tour of Macao, the Portuguese colony that became, in China’s hands, a gambling destination bigger than Vegas.

Next: An Expat’s Life