1994-02-14 CAD/CAM

I knew I could not expect to keep blatantly end-running the CEO.

When I get an idea, I start at once building it up in my imagination. I change the construction, make improvements and operate the device in my mind.—Nikola Tesla

"You've been able to do things with our product that no one else has. We think you understand it well enough to help us manage it." I almost dropped the phone. There it was, an attractive job offer from the CEO of a software start-up. It actually made sense once I thought about it.

I like to draw. The tactile feedback of rolling a pencil between fingers and thumb to lay a perfectly uniform thickness of graphite on the drafting paper is a visceral and intellectual pleasure that has not dimmed with age. The clarity of line, the precise rules of projection and representation, are a pleasing abstraction and explication of an unclear and unruly reality. Despite the advance of so many imaging technologies in the intervening decades, paper and pencil remain useful and satisfying for me.

I originally learned drafting, technical drawing, in a high school industrial arts program in the late 1970s. This was just before the beginnings of Computer Aided Design, CAD, so the analog technologies we used were highly evolved. (This is often the case just before a technology becomes obsolete.) I enjoyed drafting, and I was good at it. My technical drawings for some of my earliest inventions won an award or two.

Detailed, accurate drawings enabled me to spot problems in my other projects before I had wasted or spoiled my limited resources. This was my first real lesson in the utility of design. Prior to this, I had simply carried ideas in my head, fashioned them by hand, and tested them to see if they worked. This kind of naive empiricism can produce results quickly, but also has a brutal failure rate. Once I started putting initial designs on paper, my failure rate dropped appreciably, and my projects evolved more quickly.

At university I was fortunate to be in the first experimental class in computer engineering graphics. Rather than polish our drafting skills, we learned to program graphics elements using FORTRAN punched into Hollerith cards. Processing and printing out was done from a mainframe computer in overnight batches. In hindsight, the effort appears both laughable and tedious, but it did expose us to the essential ideas. The course was not an unqualified success. Hurling our stacks of Hollerith cards off the top of the computer science lab at the end of the term was an expression of enormous relief.

To my lasting benefit, that course did give me connections with a research group that was doing cutting-edge work in computer graphics. Visualizations and simulations were becoming increasingly feasible. Finite Element Analysis (FEA) in particular grew rapidly into a crucial tool in multiple disciplines. Computer Aided Design (CAD) and Computer Aided Manufacturing (CAM) began to show promise. By the time I began my master’s program in 1988, CAD/CAM was evolving rapidly, and I did my best to keep up on the latest developments as they applied to my materials science studies.

I had just turned thirty when I left aerospace, and the entire industry in Southern California was downsizing. I felt lucky that I was able to parlay my experience in CAD/CAM into work in engineering and information systems support with a consulting company in Long Beach.

I had hoped that working for a smaller, family-based company might be a good antidote for the harsh lessons I'd learned under a huge corporation. It was certainly different. The CFO liked me for my ability to find cheaper solutions to expensive problems. However, it gradually became clear that I was not going to have much leeway for independent research unless I married into the family. That, I was sure, was a set of unwritten rules I would be unable to follow.

I had been there a little over a year when one of the smaller software companies I had been dealing with called me up. I had been writing product reviews and tutorials for magazines on the side, just as a hobby. Based on some of those publications, the CEO offered me the position of PC Product Manager.

It was true that I'd been able to produce some exceptional CAD work with the software. I found it much more intuitive than anything else on the market. I had been able to immerse myself in the creative flow while using it, in contrast to the constant interruptions and work-arounds common with competing products. I thought this program was on the right path for the future.

As it turned out, they wanted me to manage the whole product development. The lead programmer, Gavin, liked me. I knew how to use the program, and I had a number of ideas for improvements and add-ons. Even better, I'd be working in a start-up incubator at University of Texas-Austin, one of the best technology centers in America.

Little did I know that the software company was a snake pit, due primarily to Tim, the CEO, having undiagnosed mental health issues. One aspect of that was that he kept insisting on taking sales calls, saying "I need to qualify this buyer," then talking customers out of buying the product. Another was that he habitually reversed his own decisions the next day.

It was so bad that the staff turnover averaged one hundred percent in the first nine months I worked there. The office admin changed three times. The only people who hung around were the two longtime programmers on the Mac side, and Walter, the CFO. He was Tim's father, a retired CPA, and the main reason the company continued to operate. I suspect he kept the company going as occupational therapy for his son.

The most important part of my job was to keep the CEO away from Gavin, a brilliant programmer (and university lecturer) based in Northern Ireland. Tim's behavior would drive Gavin to quit after each email or phone call. Coaxing the programmer to resume work was time-consuming and stressful. Keeping Tim from firing me for interceding was more of the same. Trying to follow the classic business rule of 'Keep the boss happy' became a bad joke.

This is not to say that it was all bad for me. I was able to keep working at the cutting edge of CAD/CAM software and hardware. I learned more about coding, particularly about writing plug-ins for other people's programs. It all meshed beautifully with my love of design, drawing, and visualization. Every day was a new set of problems to solve.

I remember one long-distance conversation in particular. "Gavin, I'm looking at the latest build you sent me, and everything's reversed right-to-left." I had assumed that this was the result of a glitch in the code or some bizarre compiling error.

"Oh, sorry, Robin. I forgot to flip that switch before compiling."

Gavin's casual admission brought me up straight in my chair. "Wait, what? Your development version is reversed? Why?"

"That's what works best for me. I wanted this program to be like holding the work in your hand, and all the tools to behave like you're in a physical shop. So that's how I'd arrange things."

I thought quickly, trying to reconcile what I already knew about Gavin with the implications of these revelations. "Okay. Um. I have to say, I'm more comfortable with the interface arranged the opposite way. It seems more intuitive. It definitely is like holding the work in your hand. I think you've hit your goal there. But backwards?"

"That's what I hear from pretty much everyone. I guess my brain's a bit different."

No kidding Gavin, I thought to myself. That was the understatement of the year. But I needed to pursue the original point. "So you set up your working version to suit yourself, and you have a single switch that reverses everything?"

"That seemed the simplest solution. Is it causing any problems?" A note of worry had crept into his voice.

I tried to reassure him. "No, Gavin. Everything's fine. That's an elegant piece of coding, but no one ever noticed it?"

"Well, thank you, but no. There are quite a few bits here and there I do for myself. I try not to bring up anything new on my own. You know how it sets off you-know-who."

Gavin's avoidance of the CEO's name had become a habit, as if to ward off the summoning of a malign entity. I could not blame him, and at times had barely stopped myself from speaking the same phrase in the Austin office. That would not have gone over well with the CFO, although I'm certain that the other programmers would have laughed.

"Ah. I understand completely." I thought for a moment. "Tell you what, why don't you send me a build with all the switches set to what you find useful for yourself. I promise I won't share it until we talk it over."

"Are you sure? It's quite different from what you are used to."

"I'm sure, Gavin. I really want to see what you've been using for your own work." I hoped he could hear sincerity in my voice.

"All right, then. Probably be about an hour." He sounded cheerful enough. I think he was happiest when he was working.

Gavin was an amazing programmer. I had his version in half the time he'd estimated, and it was more than I had hoped. It ran faster, had half again the number of tools, and the results were sharper and more precise. It looked like next year's version, after a very busy year of development.

I wheedled Gavin into allowing me to show it to the other programmers and product managers, but not to Tim. As I expected, they were just as impressed and enthusiastic as I was.

I hatched a plan. While Tim was out at his gym (another of his obsessions), I made much of showing off Gavin's version to everyone else in the office, including Walter, and expounding on how well the new features would do in the market. That was my job, after all: PC Product Manager.

Tim came back to find everyone gathered around a set of screens with multiple copies of Gavin's version running, showing off the “new” features to oohs and aahs.

Of course he tried to kill it.

I just kept calmly putting the software through its new paces, over and over again, not saying a thing. By this time, I had learned that any words from me would only complicate the situation.

After a while, Walter called Tim into his office for "some important paperwork" and closed the door. After a longer while, the CFO came out.

"Can you write up all the new features in an email and send it to me? Just what you were telling us earlier."

And that was how Gavin and I got a year's worth of upgrades past the CEO, despite the latter's mental issues. Oh, Tim hated me for that. But the software was so much better! I got an enormous neurochemical rush off that. I had a two-day euphoria and mini-hits for weeks afterward. Simply updating the documentation gave me another buzz. And no headaches!

Gavin was happy, too. He got the programming work he was so proud of out into the world without having to battle with Tim. He trusted me even more after that, and we maintained a very good working relationship.

However, I knew I could not expect to keep blatantly end-running the CEO. Very soon I negotiated a contract to change my status from employee to independent subcontractor, and thereafter worked from home. Working remotely was so much better for me. I was able to talk with Gavin without Tim listening in and interjecting, and I could limit my interactions with the CEO to emails that I was able to craft more carefully. I also CC'd Walter on all communications. I think that had a lot to do with our actually completing the project. It certainly reduced my stress levels. I had fewer migraines from the pressure of masking and spent less time moaning in a darkened room.

Finally, the requirement for software documentation, and the fact that no one else understood the program, meant that I had to learn to write user and reference manuals. It turned out that I had some ability in that area. It amused me that some people returned the software as unusable but wanted to keep the manuals I had written.

We shipped the last version of the software on schedule and with the promised feature set. I packed up, collected my closing check from Walter, and headed back to Ohio. My contract was up, and I wanted to get out of the jurisdiction before Tim pulled any more stunts.

I swore then that I would never again work for a boss with untreated mental health problems. After much thought, I decided to work for myself. Yes, I know, the joke is obvious.