Paul Allen saw the January 1975 edition of Popular Electronics one afternoon at ‘Out of Town News’—a newspaper kiosk in the center of Harvard Square. The cover touted the “World’s First Microcomputer Kit.”

Paul knew what those words really meant: The first true personal computer had arrived. He immediately ran to find me and said, “It’s happening without us!” I knew we had no time to waste.

Paul wrote a short letter to Micro Instrumentation and Telemetry Systems (MITS), claiming that we had a version of BASIC for the Intel 8080 chip that the Altair 8800 ran on.

Weeks went by without a response. Eventually, we followed up on our letter with a call. Ed Roberts, the president of MITS, answered. He said the first group who could produce a working version would get the deal. We were determined it would be us.

Paul and I worked day and night for the next two months to come up with the software product we had already said existed.

Paul figured out how to make Harvard’s PDP-10 simulate the Intel 8080, and I designed and wrote the BASIC interpreter using assembly language.

After a lot of hard work—and not nearly enough time—Paul flew to MITS’ headquarters in Albuquerque, New Mexico to show them our demo. He even had to finish a small bit of code we had forgotten to write as the plane was landing.

At MITS, Paul loaded the Altair—the machine on the cover of Popular Electronics—with our BASIC interpreter on paper tape. It was a tense moment. We never had the chance to test anything on the Altair. In fact, neither of us had ever seen one in person until Paul got to the meeting. 

It was clear at that moment that everything was about to change. Paul’s demo resulted in a pivotal deal with MITS to distribute the interpreter on their machines. I once called it the “coolest code I ever wrote.” I think that’s probably still true today.