Despite all the worries, the demo went off without a hitch. The multi-touch user interface seemed almost magical as Steve showed off the little nips and tucks that made it truly engaging. Scrolling through lists had an almost liquid smoothness. A double tap on a website column would make it fill the screen. The Google Maps application that came built in was already far more useful and versatile than most dedicated GPS devices, which had only recently shrunk down to pocket size. It was a delightful presentation of a delightful device. There was just one problem, and it was obvious to everyone except Steve.

PETER LEWIS, FORTUNE magazine’s technology critic, had arranged for one of the handful of short, private interviews that Steve granted after the keynote that day, so I tagged along. It was the first time I’d seen him in the flesh since shortly after I had started a sabbatical from the magazine to start work on a book project eighteen months earlier. This was the longest stretch of time that we hadn’t spoken in the entire time I knew him, so I was looking forward to the visit. Steve was visibly relieved that the demo had gone so well, but he bristled a little when Pete and I kept trying to steer him back to one particular subject: Why wasn’t Apple allowing software developers to build applications for the iPhone? After all, it was as powerful a computing device as an early Mac or PC, wasn’t it? I mentioned that Google Maps and the YouTube video-viewing app both demonstrated that it was perfectly possible to “open up” the iPhone to third-party software developers. “We had to help them build those apps, you know,” Steve said. “So we know what went into them.” Then he said he was concerned about how third-party apps could be vetted and policed, to make sure there would be no chance of software viruses infecting the phones. “We want to understand better how apps affect the network, too, before we throw things wide open,” he added. “We don’t want to create a monster.” He also suggested that if developers really wanted to create custom applications for the device, they could always design special websites that would perform the computing tasks on Web servers, with the phone acting simply as a terminal.

Steve had already heard from a slew of people, inside and outside the company, that he had whiffed by not opening up the iPhone to outsiders’ applications from the get-go. John Doerr, the managing partner of the most prominent venture capital firm of them all—Kleiner Perkins Caufield & Byers—had become neighborhood friends with Steve after their daughters had met at Palo Alto’s Castilleja School and started having sleepovers. Doerr had never had direct business dealings with Apple, but he knew all the main players there and was tapped into everything in Silicon Valley. Steve had first showed him an iPhone several months before they shipped. Doerr immediately asked Steve the very same question: Why wasn’t he allowing third-party applications? “At the end of that conversation, I said, ‘Look, I disagree with you,’ ” Doerr recalls. “ ‘And if you ever do decide you want to put applications on it, I’d like to form a fund to encourage people to build them. I think there’s a big opportunity there.’ He said, ‘Okay, I’ll call you back if we change our mind.’ ”

When the iPhone finally shipped on June 29, 2007, the biggest problem customers encountered wasn’t the lack of applications—it was the fact that AT&T’s network coverage was so spotty. To cite just one high-profile case, Mike Slade couldn’t get any reception at his house in Seattle on either of the two phones Steve had sent him. When Slade teased Steve about this in an email, Steve called the CEO of AT&T. The next day, a service rep visited Slade’s home. But there was no solution, and Slade wasn’t able to try out the phones until he traveled out of Seattle.

Worse yet, AT&T’s network was weaker than Verizon’s in the San Francisco Bay Area, so the early-adopter techies who’d bought their units on day one found their calls being dropped regularly as they commuted up and down I-280, which connects San Francisco and San Jose. In those areas where AT&T had sporadic voice coverage, Internet connectivity was even more of a hassle.

Apple and AT&T sold about 1.5 million units in the first quarter the iPhone was on sale, but they probably could have sold many more. Between its cellular woes and the absence of more applications like the ones supplied by Apple and Google, the iPhone proved to be a tougher sell than many would have imagined. People had expected something that would support video games and reference books and fancy calculators and word processors and financial spreadsheets right out of the box. The phone they got couldn’t yet do that. Jean-Louis Gassée, Steve’s former Apple nemesis who had segued into venture capital, puts it bluntly: “The iPhone was crippled when it first came out.”

This time Steve turned around even faster than he had when his team had convinced him to go for iTunes rather than pursue iMovie any further. He didn’t do so gracefully—“Oh, hell, just go for it and leave me alone,” is how Eddy Cue recalls his edict—but he did so quickly. In the fall of 2007, Doerr got a phone call. “From out of the blue, Steve said, ‘I think we should talk. Come on down to Cupertino and tell me about this fund idea that you have.’ So I went to work, and we hastily pulled some materials together and proposed something we called the iFund. I told him we’d commit fifty million dollars to it. Scott Forstall, the Apple guy then in charge of the iPhone operating system, was in the meeting. He said, ‘Come on, John, fifty million dollars? Surely, you could do a hundred.’ So we bumped it up to one hundred million.”

In November, just over four months after shipping its first iPhone, Apple revealed that it would make available a software development kit for anyone who wanted to develop apps. “That’s when we knew Steve had finally come to see the light,” Gassée says. “Suddenly, that was all anyone was talking about in the Valley and in the VC community. Hundreds of little guys signed up, and the race was on. Then they announced the App Store. And then they released the iPhone 3G [the second version, which shipped in July 2008, and had better wireless and a faster microprocessor]. It was only then that the iPhone was truly finished, that it had all its basics, all its organs. It needed to grow, to muscle up, but it was complete as a child is complete.”

IN THE EIGHT years since that January 2007 MacWorld, Apple has sold more than a half billion iPhones. It is the most successful, most profitable consumer electronics product ever, by just about any measure—units sold, dollars of profit generated, number of global carriers that sell it, the number of apps written for it. When you think of it, who sells a half billion of anything costing hundreds of dollars? Sure, Procter & Gamble sells billions of tubes of toothpaste and Gillette sells billions of razor blades. But those don’t come with two-year service contracts that can effectively drive the price of ownership to nearly $1,000 over the life of the product.

When it first appeared in the summer of 2007, there were other devices on the market that described themselves as smartphones. Palm had been selling its Treo for several years, and a Canadian company, Research In Motion, had done well with its BlackBerry. All of these models had pint-sized keyboards and squarish screens. They were all adequate for checking email, looking at your calendar, or finding contacts in your address book. And their businesses were basically doomed, although BlackBerry would hang on for years. The iPhone changed the category forever. Google understood this, and within eighteen months developed Android, a free knockoff of the iPhone’s operating system software that powered phones made by the likes of Samsung, LG, HTC, and later an upstart Chinese handset maker named Xiaomi. A new race was on, and Apple had the lead. Android handsets would eventually outsell iPhones, but this has not been a redux of the Macintosh experience. At least not yet.