The Little Engine that Did.
Microwave Bypass & the Start of Fixed Wireless Access

I was a struggling 25-year-old microwave reseller in 1986, working from my kitchen table, not knowing that as I cried in my cereal, my business was about to explode. A year before that, I was selling telco services at MCI. I knew nothing about microwave, except that it could save some of my clients a fortune. I'm talking tens of thousands a month. I learned that from a guest speaker who talked our sales group about the future trend of "microwave bypassing". I found it so convincing that I left at lunch, walked across Boston Common, and registered "Microwave Bypass Systems" as a business name.

Months later, I convinced a radio vendor, M/A Com, to set me up as a reseller, fronting me $700 a week for six-months so I could quit my job and sell their radios around New England. I reasoned that a single link order, based on a 35% margin, would easily get them their money back, and worst case, they would have invested $17,500 in marketing, including gaining access to my former accounts, several of which were hot prospects. 

​

My boss at MCI thought it was ridiculous to leave a job where I was doing well, so I could sell a highly technical product that I had no understanding of. The president of MCI didn't know the first thing about microwave either—and "microwave" was what the "M" stood for—but that was beside the point. 

The way I saw it, I had a "one-in-a-lifetime" opportunity to save businesses thousands a month, work for myself and get rich in the process. If I could sell leased lines, I could sell microwave. Whatever I didn't know, M/A Com could help me with.

I sent out mailers advertising a turnkey installation, performance guaranteed. Payback was so good it seemed illegal. Still, I was on ten months and still not a single sale. The VP at M/A Com extended my $700 weekly draw, because it seemed I was on the verge of an order, but I was on a short leash and things were getting tense. 

When I started out, someone told me that a CFO was practically obligated to invest in anything that brought better than a two year payback. My proposals were coming in at under six months. Something didn't add up.

At last, I learned that something was bigger than savings. Buyers were afraid that going with microwave would upset carriers they depended on for all their other needs. Meanwhile, carrier account managers who caught wind of any potential microwave plans, were well versed in rumors about health hazards and employee lawsuits from radiation exposure.

My situation was clear. If I had to keep selling T1 radios, I would die. 

Ethernet users were a different breed, mainly universities, tech firms, research centers and hospitals. Their priority was anything that facilitated network access to tap education and research opportunities. They didn't worry about upsetting carriers. They tweaked them for fun. 

​

My interest in Ethernet was sparked by my first customer order, a job I landed from a marketing piece I mailed to area businesses. Talk about the "right place at the right time”, the order was from Symbolics, the first dot com address on the internet.

Symbolics was an AI firm and they had me cutting my teeth on a twelve mile, two-hop licensed microwave network that uniquely combined voice (T2), broadband video and a proprietary Token Ring interface. It was the first installation of its kind and interesting enough that DataPro Research published this report. 

I thought the Token Ring part of the job was super interesting, but my roommate worked at Digital Equipment Corp. (DEC) and they were behind a competing standard, Ethernet. Between the two, I became convinced that Ethernet would emerge the dominant one.

​

Armed with some insights, I lobbied radio vendors to develop an Ethernet solution, but they were smarter about market timing and turned me away. Their rejection only spurred me on and so I hired my first RF tech, Frank Miani, who I met through M/A Com, my radio vendor for the Symbolics job.

Frank liked the path I was pursuing and left M/A Com after twenty years to help me as a 1099 consultant. He explained that the solution I wanted was doable, assuming we could find someone to design a LAN/microwave interface, and for his part, he thought that he could handle the radio modifications.

​

Frank thought that if I could convince the LAN engineer who designed the Token Ring interfaces for Symbolics, to give us an Ethernet design, we might have a shot. Weeks later, I gave that engineer—whose identity has never been revealed—a $25,000 down payment, which came from earnings on the Symbolics account, which amounted to just over $130,000. If I was careful, I’d have enough money for two prototypes and twenty-five finished units.   

​

The three of us were a team, but seldom together. Frank was older than my dad and a creature of corporate habit. He showed up to work at 6 a.m. and was gone by noon. The LAN engineer, who came to be called “Henry Coolidge”, could only work after 7 p.m. and did his best work while I slept. And I shuttled back and forth, encouraging, pushing, prodding, begging and occasionally being useful.

​

In the darkest hours, Henry might be up against a design obstacle that he feared could sink us. He was a star in his day job, but I had everything to lose if we failed and so I’d press for an explanation and out of that came a process, aided by some translation—analogies where bits of data become water and buckets or buses and trains—where a person like me might advance a worthy proposition.  

Henry was patient when I urged him forward and pressed him to understand anything he might have been hung up on. We’d bat around thoughts until something sparked a connection in his mind and *that* moment brought a profound sense of satisfaction.

I never questioned the value of our mission, though business on a shoe-string means a constant scramble. Things like, sneaking in back doors to use lab gear we couldn’t afford and testing after hours on Boston University’s network, because we had no way of generating random traffic. I managed the tests, taking copious notes and communicating them to Henry, who could never be seen in public with me or his employer might lay claim to the tech he was developing for me.

Remarkably, nine months into development, we had our first Etherwave Transceiver prototypes (what we called our LAN/microwave interface). And not a moment too soon for a beta test we arranged at Boston’s famous Mass General Hospital, where I'd invited local university IT managers and Laura DiDio, a reporter from Network World. Laura's been a friend ever since, and we laugh when I remind her what she told me on the day of our test. She said that if we failed, it would still be a good story. The radios for our test came from International Microwave, which was still burning them in on the heels of our demo. Afraid for them to be shipped, I drove them back to Boston and carried them like eggs.

Our test connected a terminal to a host machine across a data center floor, no antennas, just waveguide, and between our Etherwave and the customer’s network was a DEC bridge, the first such device to hit the market [see product sheet]. I had no clue what was happening as our customer, Dave Murphy, went back and forth pinging machines. Frank kept ducking out to smoke and Henry of course, couldn’t be present. The shame of it was that he and my customer would have hit it off famously, which would have made me proud, only that never could happen.

​

Our test succeeded in demonstrating that packets of various sizes, burst rates and protocols traveled seamlessly across the microwave connection—spanning a few yards of air—and that our Etherwave was compatible with the hospital’s bridge and network platform. [Check out this testimonial from Boston University's network manager]. It was game on from there and weeks later, we had our first purchase orders; two 10Mbps links for Mass General and one to connect Boston University to Harvard.

Our early clients were all network gurus, including at MIT and UC Berkeley. They knew more than we ever would and there was no hiding the fact. Yet, we had the solution everyone needed. Regional networks were springing up everywhere, institutions were eager to be connected and fiber, if it could be had, often cost millions. By contrast, our solution met the full 10Mbps Ethernet rate, had a 90-day turnaround (pending FCC licensing) and cost $35,000, installed.

​

The company was Microwave Bypass Systems and in our time (1986-2000), we developed the first fixed wireless tech platform—network interfaces, bridges and radios—type accepted the first LAN radios, collaborated with Cisco, partnered with Motorola and empowered applications from distance learning to telemedicine.  We brought the first internet access to hundreds of world-leading universities, hospitals, tech firms and research centers, giving them their first experience with microwave and connecting millions across 46-states and half a dozen countries.