By the fall of 1983, we had committed to announcing and shipping the Macintosh at Apple's next annual shareholders' meeting, to be held on January 24th, 1984. The failure of the Twiggy disk drive almost caused us to be late (see Quick, Hide In This Closet!), but the new Sony 3.5-inch drive seemed to solve all our problems, and the rest of the hardware was ready. The Macintosh ROM was frozen in early September and sent for fabrication. All that remained was finishing the System Disk and our two applications, MacWrite and MacPaint.
The software team worked hard over the 1983 Christmas break. The Finder still wasn't finished, and performance problems persisted—especially when copying files between disks, which took interminably long. Integration testing was extensive: cutting and pasting between applications, and interactions between apps and desk accessories. As the New Year arrived, it was clear we were running out of time.
By the first week of January, the software team worked around the clock, testing and fixing issues. Every employee became a tester, and Apple provided catered dinners for those staying late (see 90 Hours A Week And Loving It!).
With less than a week until the deadline, it was obvious too many bugs remained to ship. Late on a Friday evening, we concluded we needed an extra week or two. Steve Jobs was on the East Coast with Bob Belleville and Mike Murray for press events, so we arranged a Sunday morning conference call to inform him.
Jerome Coonen, our software manager, spoke for the team as we gathered around the speakerphone. We were exhausted, progress was slow, and unresolved bugs remained. Jerome proposed shipping "demo" software to dealers for the launch, then delivering final software to customers a few weeks later. We thought he was persuasive, holding our breath as we waited for Steve’s response.
"No way. We're not slipping!" Steve replied. The room gasped. "You've worked on this for months—another few weeks won't make a difference. Get it done. Make it as good as you can. Now get back to work!"
We managed to gain a few extra days by working through the weekend and shifting the deadline to 6 a.m. Monday, when the factory opened, instead of Friday afternoon. We agreed to rest and return Monday for the final push.
The final week was the most intense I'd ever experienced. Steve wanted Bill Atkinson and me to fly to New York to present a Mac to Mick Jagger (see A Mac For Mick), but I stayed in Cupertino to help fix bugs. Some team members paused work for magazine photos—Newsweek, Rolling Stone—which upset others. Tensions ran high.
Friday arrived, and bugs still plagued the Finder and MacWrite. Randy Wigginton brought in a huge bag of chocolate-covered espresso beans, which, along with medicinal doses of caffeine, kept us awake for the final days. We began releasing builds every few hours, re-releasing with each significant fix.
Each new release triggered a new round of testing. Around 2 a.m. Sunday, I found a clipboard bug. I thought I knew the cause but was too tired to act. I tried to ignore it, but Steve Capps noticed my expression, grilled me, and then helped craft a fix.
Around 4 a.m., a release went sideways—even MacPaint crashed, despite its usual stability. But our 5:30 a.m. build was much better; the worst issues seemed resolved, and we had a potential release candidate.
We tested it intensely until 6 a.m., when Jerome had to leave for the factory. It looked good, but someone found a potential showstopper: the system hung when a blank disk was inserted in MacWrite—it didn’t start formatting. I realized it was waiting for an event, so I tapped the spacebar, and formatting began. Jerome considered the bug severe enough to delay, but he left for the factory anyway, knowing duplication needed to start, even if it was just a demo.
By sunrise, the team began to scatter, exhausted. After working so intensely for so long, it felt surreal to have nothing to do. Donn Denman and I sat dazed on a lobby couch, watching accounting and marketing staff arrive around 7:30 a.m. We were a sight—everyone knew we’d been up all night (I hadn’t been home or showered in three days).
Around 8:30, Steve arrived and immediately asked if we’d made it. I explained the formatting bug; he deemed it not a showstopper. We were done. I drove home to Palo Alto around 9 a.m., collapsed into bed, and expected to sleep for days.