It Was the Best of Times, and Then It Was the Worst of Times (Premium)

That the same leadership and teams made both Windows 7 and Windows 8 is an incredible fact of history that’s still hard to fathom. But Windows 7 and Windows 8 are both misunderstood, too, all these years later. For all its accolades, Windows 7 was simply a service pack for Windows Vista, an easily-achieved goal. And Windows 8? Yeah, it was a freaking disaster from a UX perspective, but the underpinnings and desktop advancements were far more impressive than what Microsoft had delivered in its predecessor.

What a world.

I covered the history of these two Windows versions, and a lot more, in my epic Programming Windows series, which I’m now working to turn into an eBook akin to the Windows 11 Field Guide. But the story is important enough for a high-level overview: after biting off more than it could chew with Longhorn, Microsoft was forced to continue updating Windows XP while it plotted a much more modest new release that came to market in 2006 as Windows Vista. Vista … was not loved, to put it bluntly. But it was also widely misunderstood, and its problems were quickly solved via two service packs and other updates. And, it should be noted, by the release of an Intel integrated graphics chipset that could finally render its attractive Aero graphics. Time heals all wounds.

But Windows Vista was also Microsoft’s first major stumble with its still-dominant PC platform: driven by years of non-stop success, the ego-driven software giant didn’t grasp that public perception would sink this system until it was too late, and then it responded slowly. Windows Vista is now understood to be the point in time when Microsoft’s market power peaked and then began its inevitable decline. And by the time Microsoft shipped Windows Vista to consumers, Apple had already announced the iPhone, changing everything.

As bad, the transparency of the Vista regime, led by Jim Allchin, was replaced with a Politburo-style veil of secrecy under new Windows chief Steven Sinofsky, a divider who saw only loyal faithful and enemies of the state inside of his team, elsewhere at Microsoft, and out in the world. You were either with him or you were against him. It was a recipe for disaster.

But at the beginning of his tenure, it was easy—perhaps too easy—to be with him. Sinofsky’s first job was clear, and those inside and outside of Microsoft rallied behind the mission to right the wrongs of Windows Vista. He would use—then rely on, and then overly rely on—telemetry data, automated feedback that came out of the product, to guide the direction Vista’s successor took. He was a fixer, a man with small ideas who was somehow paradoxically also convinced he could be Microsoft’s Steve Jobs, an industry seer with a finger on what would come next. It’s worth pointing out that Jobs never relied on telemetry data to figure out where to go next.

But fixing Vista was easy, and telemetry data did, for this one release, tell the tale: Vista was bloated and slow. It was incompatible with too much hardware. And its vaunted Aero graphics mode worked on too few PCs.

And so Sinofsky continued the componentization work that had begun under his far more technical predecessor and debloated the successor to Vista while time fixed the other issues. The result could have been called Windows Vista Service Pack 3. But it had inexplicably taken three years to come to market, thanks to Sinofsky’s plodding ways. And it just as inexplicably was widely celebrated as the single best version of Windows yet released. Indeed, it also went on to become the best-selling version of Windows ever, and the most widely used, to that point.

Continuing this success would be difficult for any leader or team, but Sinofsky and his loyal lieutenants somehow managed to do the impossible and deliver a turd of historic proportions. Windows 8, as the next release was imaginatively titled, would likewise take 3 years to come to market. But this time, three years wasn’t long enough because Sinofsky was blinded by two obsessions: his desire to out-Jobs Steve Jobs and defeat Apple’s growing multitouch kingdom—the iPhone was changing personal computing forever and was soon joined by the iPad—and his hatred of .NET, the Microsoft technology stack that his predecessor had used for Windows Vista.

Windows 8, in Sinofsky’s small-minded view of the world, would dump the decades of expectations that his customer base had built up and would target a coming generation of Windows PCs that were more tablet that PC, despite the fact that none of these devices existed yet. It would likewise dump the technically excellent .NET and take major steps backward and deliver a mobile apps platform, inside of Windows, that had nothing to do with the apps of the past and relied on dated COM (component object model) technologies. And he would deliver this trainwreck, this Frankenstein’s monster, in just three years.

It was, Sinofsky later said, “too much too soon.” But that’s one of those cute descriptions that sounds right but in no way explains what really happened. It wasn’t just too soon: the nascent mobile apps platform that Microsoft delivered in Windows 8 wouldn’t be ready for prime time three years after that when it delivered Windows 8’s successor, called Windows 10. It was also just wrong: while the personal computing market was moving ahead with the iPhone and Google’s Android platform, desktop Windows was in no position to ride that wave too. Worse, a team inside Microsoft was building a Windows Phone product that Sinofsky refused to collaborate with. And so the Windows 8 app model was different from that used on Windows Phone. Can I get a slowly building golf clap for Microsoft’s industry seer? He really saw where the industry was going.

If Sinofsky’s biggest problem was that he lacked imagination, Windows would have been fine, and the solid desktop advances that the team made in Windows 8 would have served nicely as an obvious follow-up to Windows 7. But Sinofsky was that rare combination of undeserved ego, lack of imagination, and no sense at all of where the industry was heading. He was a follower who was tasked to lead. And so he led by following. Apple most obviously, but more generally, the mobile world that had already surpassed Windows. And he failed. Badly: Windows 8 was the wrong product at the wrong time. It addressed a market that did not exist then and would never materialize. It fought bogeymen competitors like the iPad that, yes, would be successful, but still don’t threaten the PC today, 10 years later.

Microsoft saw the Windows 8 trainwreck coming, but it was too late to stop it from happening because the software giant doesn’t release new Windows versions in a vacuum: instead, an industry of hardware and software makers had spent the past three years building products for its release and, unlike Microsoft, they couldn’t afford any delays. And so Microsoft did what it could: it fired Steven Sinofsky, who was at that time waging an overt campaign to unseat Steve Ballmer as CEO, and instituted a plan to upgrade Windows 8 rapidly to actually meet the needs of its customers, all while methodically removing Sinofsky’s key lieutenants and every last vestige of that regime, save one: Panos Panay, a hardware peripheral designer who would one day run Windows himself somehow. Oh, how history repeats itself.

With Sinofsky gone, the transparency returned, and Microsoft quickly delivered Windows 8.1, Windows 8.1.1, and Windows 8.1.2, reversing the worst UI mistakes of Windows 8. But the new mobile apps platform remained, and when Windows 10 finally arrived in 2015, that platform was what it should have been years later minus the .NET bits: something that worked consistently across PCs, tablets, and phones (and other devices). But it was too late, and Microsoft killed Windows Phone in 2015 too. Thanks, Steven.

Sinofsky is the real reason Windows Phone failed. But it is impossible to calculate the damage that Sinofsky did to big Windows as a platform. It was a near death blow that is as remarkable for the ineptitude of his work as it is for its timing. Windows 10 would go on to outperform Windows 7, but that was somewhat artificial because it was also kept in-market for several years, far longer than the three-year timeframes of its predecessors. And Windows 10 had its own controversies, with new Windows chief Terry Myerson—formerly head of the Windows Phone team, go figure—inflating the upgrade count to boost his pay package. He was finally outed after a few years, too, and lasted even less time than his disgraced predecessor.

It doesn’t matter. Sinofsky presided over and at least partially caused the downfall of Windows, and that will always be his legacy. Today, it is perhaps fitting that both of his creations, Windows 7 and Windows 8 are being put out to pasture together, with both products reaching the end of their support lifecycles. These things should not be happening at the same time, of course. But Windows 7 was so popular that Microsoft was forced to support it for longer than it wanted to. And Windows 8 was so unpopular, that few users remain to mourn its passing.

In 1998, I started the SuperSite for Windows—originally called the Windows NT 5.0 SuperSite—with the specific aim of looking at the future of this platform. And it is in that spirit that I continue covering Windows today, looking to the future. Windows 7 and Windows 8 are the past, and one is undeservedly well-respected while the other is deservedly inglorious. What an incredible mess.