Paul Lawton. A personal look at the early days of Apple Computer

Memories of Paul Lawton - the developer of the first operating system for the computer Apple II - Apple DOS 3.1. The photo Paul in his home office in 1980. Behind him is the Apple II, which was delivered to him personally by Steve Jobs to work on the operating system, which will be discussed further. Next to the monitor are two Disk Disk Disk II with serial numbers 3 and 4.

In 1977, I fired up the then-new microcomputers. By that time, I had been working at IBM for eight years as a system programmer on large mainframes. The IBM operating system has grown to enormous size (many megabytes!). She became so great that no man could understand her thoroughly. The limited address space (64k) and the possibility of owning my own personal computer inspired me.
One day I went to the Byte Shop (the first personal computer store) and found a brochure about a new Apple II personal computer and an order form . The Apple II was amazing. I wanted such a computer. It so happened that almost simultaneously with this, I came across a small advertisement in the newspaper for Shepardson Microsystems. They were looking for a programmer. Early in the morning I sent them my resume. A few days later I came to the interview. It turned out that Bob Shepardson had just signed a contract with Apple Computer to develop an interpreter for the Basic language. He offered to write an interpreter to me.
Two weeks later, I left my solid and reliable position at IBM to start working in a company of three (now four) people. A few days later I had my own Apple II workstation. Computer brought me to Steve Jobs. I learned that Basic will be included in the next generation of Apple computers, codenamed Apple Annie. Woz (Steve Wozniak - approx. Trans.) Was very enthusiastic about the new car. She had to have ROM cartridges with programs and a lot of custom-made microcircuits.
One of the problems we had to solve was the download of files with object code 6502 from our development system in the Apple II. The development system consisted of a COPS microprocessor with a 6502 compiler developed by Bob Shepardson. At the compiler's input, there was a deck of punched cards, and at the exit - a paper tape. How to read a paper tape in Apple II?
Woz found a solution. He designed a card for the Apple II card reader. Once, when Woz and I were setting up this device, I noticed that Woz looks depressed. I asked him what was the matter. He said that he developed a floppy disk controller for the Apple II. He was proud of it, but Apple’s leadership set a completely unrealistic schedule for him to develop both the hardware and the disk operating system. I said that I can take on a disk operating system (DOS). Woz was delighted, and after a short meeting with Bob Shepardson and Steve Jobs, we began work.
When Woz showed me his projects of the disk controller and the software driver for it, I was absolutely shocked. At that time, all disk controllers were healthy circuit boards with dozens of large and small chips. Woz designed the controller, consisting of only seven small integrated circuits. More strikingly, the Woz controller had significantly better performance (performance, data density, reliability, cost) than the existing controllers.
When Woz started this project, he did not look at how others had done before. He thought about how it should have been done. And so he created something absolutely wonderful. This gave me an example of what was later called "thinking from the outside of the box." Later, when I was in charge of a group of engineers, I often told the story of the creation of this controller by Cart to encourage them to think outside the box. Unfortunately, most engineers never reach the creative level of Steve Wozniak.
Bob Shepardson and Steve Jobs spent two weeks to agree on all the details (cost, work schedule, etc.), after which they signed a contract . If you look at this contract, you will be amazed. The price Apple paid was very small compared to the value of the product it received. The deadlines for this amount of work were extremely tight. Product specifications, terms of delivery, quality criteria, penalties, premiums and other legal abracadabra, which became familiar to me in subsequent years, were completely absent. Such were the good old days!
Since I now stopped practicing BASIC, and since we had another job, Bob had to hire another programmer. I knew a great candidate — my life friend Kathleen O'Brien was a very good programmer, and we worked very well in tandem. A few weeks later, Kathleen became employee number 5 at Shepardson Microsystems.
While I was working on DOS, there were big changes at Apple. They moved from a small office behind the Good Earth restaurant in Cupertino to their new “World Wide Headquarters” at Bendley Drive. We loved to poke fun at the "universality". Apple was a tiny startup that was just starting to deliver the first samples of its products to customers. Shepardson Microsystems has already done many projects for such small start-up companies, and all of them have sunk into oblivion. Why won't Apple do the same?
One day, Steve Jobs offered to buy Shepardson Microsystems to form the core of the Apple software development team. Bob refused. The fact that Steve could pay for Shepardson Microsystems only shares of Apple. It is possible that Bob would have shown more interest if Steve had offered a significant stake, but Steve would not agree to give up more than 10% of Apple.
One of these big changes happened when Apple recruited Jeff Raskin as a team leader for technical writers. Jeff was left with the task of writing user guides for DOS and the new Basic. The task was extremely complicated by the fact that none of these products had any specifications. Jeff had an almost complete version of DOS, but Basic was worse - by then only the parser was ready. Jeff decided to write a manual that would also serve as specifications. All this would be good, but Jeff had a lot of his own ideas. As the manual was written, the specifications for both products went far beyond the original agreements. As for DOS, it was necessary to conclude at the last minute additional contracts, including those made by Jeff. Fortunately, the ability to make changes to DOS was held back by the timing of product delivery.
But with Basic these constraints were not, and our little Basic grew, grew, and grew. He stopped to fit in 48k of RAM. We had to develop code segmentation and overlays. Jeff recognized the size of the resulting monster, calling it NOTZO BASIC (word play with not so basic). We called it NutSo Basic (apparently, the "kooky" Basic - approx. Transl.).
The final chapter of our collaboration with Apple began shortly after the joint meeting in October 1978. I scanned the meeting minutes that Randy Wigginton hosted. The meeting.gif file contains an introduction, and the meeting.txt file contains the full meeting protocol. As you can see, the first part of the meeting was devoted to correcting some errors in the DOS 3.0 we submit. The second part of the meeting turned into a discussion of bloated NOTZO Basic. I fixed the errors in DOS on the spot by presenting an Apple-marked listing . The problem with BASIC was resolved very soon after Apple closed the Apple Annie project and canceled the Basic contract. Shepardson and I were not particularly saddened. We were waiting for Atari and the contract to write Basic for the new Atari 800 computer. But that's another story ...
People often ask why the first release of Apple DOS was called Apple DOS 3.1. The fact is that every time I recompiled the code, I increased the version counter. The counter started with Rev 0.1. When I got to (n) .9, I went to (n + 1) .0. The first listing I gave to Apple had the Rev 2.8 number. Someone at Apple (I forgot who exactly) decided that they could not call the operating system DOS 2.8, so the name was changed to DOS 3.0. It was with this version (2.8 renamed to 3.0) that Apple conducted the beta tests, and when they started shipping this DOS, they increased the number by 3.1 to show that the code changed from the beta version. When I last transferred the source code to Apple in October 1978, the version number reached Rev 6.3.