Move Slowly and Fix Things (Premium)

He hasn’t, but if Panos Panay were to ask me my advice on running Windows, I’d tell him to do what he’s been doing with Surface. Yes, I’m sometimes critical of the slow pace at Surface, but this is exactly what Windows needs right now.

As you may know, the phrase “move fast and break things” originated at Facebook, and it was the internal motto of that firm until 2014. “Unless you are breaking stuff, you are not moving fast enough,” Facebook founder Mark Zuckerberg once said, referring to this motto. But let’s not give Zuckerberg or Facebook credit for this creation: This motto is really just a rephrasing of the “competing on Internet time” mantra of Netscape in the 1990s, and it was used for the same reason, to separate the new and fast from the old and slow. It was seen as a differentiator.

Microsoft has certainly applied this philosophy to various degrees of success across the firm’s business units. Office, for example, has evolved from a monolithic suite of desktop applications to a broad set of cross-platform apps and services that literally received dozens of new features every single month. And somehow, it’s worked.

But with Windows, we’ve seen the opposite end of the spectrum to disastrous results: Windows as a Service (WaaS), Microsoft’s strategy for pretending that its legacy desktop platform is an online service and updating it accordingly, has failed in dramatic fashion. Windows 10 software updates are known now for only two things: The speed at which they appear and the reliability problems that they inevitably rain down on its users.

Microsoft puts on a brave face when it comes to Windows updates, and it often touts the success of its telemetry-based rollouts, despite the obvious disconnect with reality. But truth is always found in actions, not words: In both 2019 and 2020, Microsoft has halved the speed at which it delivers feature updates, which are really full version upgrades, from twice per year to once per year, matching the speed at which other major platforms, like Android, iOS, and macOS, are upgraded. Microsoft can talk up WaaS all it wants. But it has slowed down specifically because WaaS has failed.

This is the mess that Panos Panay inherited when he was handed the leadership role at Windows after he attempted to flee the software giant for a similar role at Apple. One might see this as a power play, but whatever, it worked: Now Panay runs both Surface and Windows, and it’s easy to imagine that he will use his new role to drive innovation forward in both in lock-step.

Less clear is how or whether his experience at Surface will influence how things are done in Windows. Or whether he will be influenced by how things are done in Windows and switch gears accordingly in Surface.

I’m hoping for the former.

And there is evidence to suggest this will be the direction Panay takes. If you think about the criticisms I’ve had of Surface over the years, the most obvious being its inability to add Thunderbolt 3 capabilities to its premium PCs despite the fact that the rest of the industry embraced this technology several years ago, you can kind of boil it down to the fact that Panay, and thus Microsoft, has evolved its hardware very slowly and rarely makes big leaps forward, especially with existing designs.

But what company does this remind you of? Right, Apple. Apple does the same thing: It’s been milking the MacBook Air design for what, 300 years now? And it stupidly pushed back at the reliability complaints for its terrible butterfly keyboard design for several years before finally giving in to the inevitable and moving—yes, slowly—back to a more traditional scissor-switch design.

This slow pace can be maddening to enthusiasts, but it makes sense for hardware, which is expensive to design, create, and manufacture. And reusing designs over several years isn’t just cheaper, it’s good for customers, especially business customers, since all your existing peripherals will continue working when you upgrade the PC. That’s why Microsoft has stuck with the Surface Connector.

But if you think about what I was really asking for in Redmond, Start Your Photocopiers. Please (Premium), it was for Microsoft to stop adding major new features to its legacy desktop system and focus, instead, on consistency and quality. In other words, take the same evolutionary approach that both it and Apple already take to their hardware and apply that thinking to software. Stop adding crap to Windows. Move more slowly. And fix or remove things that have been broken or inconsistent, in some cases, for decades.

It’s the right way forward for Windows. And while it’s not clear yet which direction Panay will choose, I’m hoping that his history and the obvious example provided by Apple will show him overwhelmingly that there is only one right way to do this.