Talking With Michael About Software

"No Michael," I said emphatically, "the cost isn't the big thing."

I was kicking back in Michael's office finishing off a gin and tonic and avoiding smoking the clove cigarettes scattered about the room. Michael was one of the company's two directors and had the largest office in the building. Before taking over as director of our company, he was a professional shepherd in northern England. Artifacts of his ovisiary past littered the room: wool seat covers, wool macrame art, small woolen tapestries. It was the type of room a sensitive male from 1978 would be proud of, but this was 1990.

"The cost is the only thing," Michael shot back, eyes perched atop a nose slanted just far back to make it clear i was being looked down upon, "you lived through the home computer wars; you saw what Commodore did to Texas Instruments."

Whenever Michael wanted to make a point about how small upstarts can take on large incumbents, he always went back to Commodore. This time he was being doubly-unkind. He knew my uncle was a vice president at TI and that I had been a "true believer" in the 9900 architecture a decade earlier.

"Yes. Cost is important," I started, "if Apple had released the Mac at a price even close to discount PCs, it would completely decimate IBM's market."

Michael looked confused; he must not have thought I would capitulate so quickly. I jumped in quickly before he could respond, "But costs always go down. It's the one truth of our industry."

He was about to talk, but I cut him off again, "What is expensive today is average in eighteen months and cheap in three years. Our development cycle is about nine-months, so that gives us four releases to build a core product people want to buy."

"But our burn rate is much too high," Michael shot back. I couldn't argue with that. I had done a back of the envelope calculation and realized there was no way we could continue with our current staffing levels.

"We're done with the Casio and Sharp ports next month," I said, "we can move those resources to OS development then; there's more than enough work to keep them busy."

Michael looked unconvinced. He leaned back in his chair, his clasped hands in front of his mouth, his face dour.

"Tech support moves to DAC after we deliver the ports and we can let go of our hardware business," I continued.

"We can't get rid of the hardware. It's Sam's baby. He'll never let it happen," Michael responded.

"How could he block it?" I asked.

"Sam's a senior shareholder in the venture," Michael answered; I knew our other director, Sam, had sunk some of his own money into the venture, but this was the first time I realized how much.

Michael took this moment to take back the conversation I had dominated. "We have to make something work with commodity PC hardware," he said, "Putera's made it clear we can't build our own PCs."

"Hunh?" i responded, confused.

"Putera's happy to make modems and act as an integrator of small PCs, but the OS can't depend on custom hardware," Michael said.

"Ah," I responded, finally understanding, "Daddy Warbucks will pay for a truck-stop but not the Taj Mahal."

"Yup," Michael replied with his best American accent.

"So there's no chance we'll be making a NeXT-killer," I said, dejectedly.

"Nope."

"What if we added some of the NeXT software development bits to our OS?" I asked.

"What bits?" Michael asked.

"We could rip off Interface Builder," I replied.

"Why bother?" Michael responded, "it's just ResEdit on steroids."

"Have you looked at it?" I asked.

"Sure," Michael answered.

"No. Have you built an app with it?" I asked.

"Should I?" Michael shot back.

"Yup."

Michael sighed and continued, "I worry it won't matter. Microsoft has beat everyone to the punch. Have you seen Windows 3?"

"It still has a lot of bugs," I replied, "and is still pretty confusing. People will never buy it."

"Compaq is supposedly going to bundle it with their new PCs," Michael said.

"So what?" I asked, "So a lot of people see how crappy it is."

"You remember when the PC first launched?" Michael asked, "you could get PC-DOS, CP/M-86 or the UCSD p-system. PC-DOS was effectively free while the other two cost money and had an existing user base."

"I see where you're going," I replied, "CP/M and p-system are effectively dead and Microsoft made millions off PC-DOS."

"People like free things," Michael said.

"Yeah. Damn Them," I replied, "Damn Them."