At the beginning of 1980, three people left Hewlett Packard to form Software Publishing Corporation (SPC). The three were Fred Gibbons, president; Janelle Bedke, chief operating officer; and me—John Page—vice president of engineering. This is the story of how the events that led up to SPC's founding looked through my eyes: "The Engineer's Tale" (with a nod to Chaucer's Canterbury Tales), if you will.
A new opportunity in software: Sometime near the end of 1979
I have been at Hewlett-Packard now for 10 years, and I am deeply involved in writing a database management system for HP's commercial computer systems. These are "minicomputers," so called because they are only the size of a refrigerator but could do the work of room-sized mainframes. The work is intricate, and I am totally immersed in it, to the point of not being very aware of general trends in the outside world.
One day late in 1979, I run into Fred Gibbons in the hallway. He is on the marketing side of the company, more aware of the outside world and very tuned into the industry trends. He asks me if I have seen VisiCalc, the first spreadsheet program, which runs on the new Apple II personal computer. Yes I have, but find it hard to believe that such a small computer could do anything serious.
Fred has worked up a business plan to start a new company, the thrust of which is that software for these new PCs would be sold like music recordings. The computer is a "player" for the "titles" you could buy. People would buy many titles and use them with the player. The idea that the software titles could exist as a consumer market separate from the computer is pretty radical. Fred suggests I try to write something inmy spare time, and an Apple II is placed in the trunk of my car before I can change my mind.
A few days later I am hooking up this strange and tiny computer in my spare bedroom. What is it really? What can it do? How does one develop software for it? It dawns on me pretty quickly—we are trying to cram a quart into a pint pot. Still, the VisiCalc guys did it. Therefore, it is possible, and so I press on.
After a few days I am getting quite nervous. I decide to wander around the local Computer Land store in Los Altos and see what people are buying and what the typical machines look like. It isn't encouraging because the typical systems being sold are so small. Most machines sold have only 32 Kbytes of memory, a cassette tape drive, and use a TV for a monitor. It has color support, but the color is truly awful. The wave of the future is 5.25-inch, 120-Kbyte floppy drives, considered a luxury at this time. Hard drives have not yet been invented for personal computers. The machine is 8-bit and has a clock speed of 1 MHz. It only has uppercase characters.
To me this thing is a toy. Hats off to the VisiCalc authors! It is only their "proof of existence" that stops me from giving up. This has to be possible, so again I press on.
In spite of my misgivings about the machine, it is slowly becoming clear that it represents the beginnings of something big. I shake the doubts from my mind and force myself to think about it in a serious way.
Although the term has not yet been coined, I want to build a set of productivity tools that are powerful yet easy enough to use by the first-time user. Based on that concept, it is becoming quite clear that cassettes are out for storing users' data. To build a database (or Personal Filing System as we wound up calling it), we need random access to the storage. Loading programs by cassette tape is also tedious and erratic—they were designed for music, not digital storage. I also feel strongly that the software has to be very easy to use, which means the internal complexity goes up. The...