Windows 10: Fixing the Windows 8 Mistake

on October 2, 2014
Reading Time: 6 minutes

Back in February, I wrote a post called “The Windows 8 Mistake“, which outlined my perception of the reasons why Microsoft did what it did with Windows 8, why it was wrong to do so, and why Windows 8 was so poorly received. I also gave some thought to how Microsoft might go about fixing the problems it caused with Windows 8. With the announcement this week that work is underway on Windows 10, I thought it would be worth revisiting that piece and evaluating whether the new version is likely to solve the problems created by Windows 8 or perpetuate them. I think Windows 10 can be seen in many ways as Microsoft’s attempt to fix the Windows 8 mistake, but I’m not yet convinced it’s the right approach.

A brief recap of the mistake and its consequences

To recap briefly what I said in that previous piece:

  • Microsoft’s strategy for Windows 8 was driven by the desire to create the tablet in the image of the PC, not the smartphone. As such, it created a touch-centric OS and completely rewrote the UI in a way which made some sense for touch devices but felt very alien to the vast majority of existing Windows users more accustomed to a keyboard and mouse.
  • Microsoft also put dividing lines between devices in the wrong places, creating both PC-like and smartphone-like operating systems for tablets (full Windows 8 and Windows RT) and sowed further confusion about Windows 8 tablets.
  • This strategy failed both in the objective of driving tablets down a more PC-like path and in terms of creating a compelling operating system. Windows 8 has become another dud in a series of misfired versions of Windows over the years, more Vista than XP.

My to-do list for fixing Windows 8

The to-do list from my previous piece was as follows:

  • Merge Windows Phone and Windows RT, mirroring the existing iOS and Android structures, and rename Windows RT as Windows Tablet.
  • Make both flavors of the merged mobile OS free for users and OEMs, eliminating licensing fees
  • Do much more to promote consumer services, notably Microsoft’s own Music, Video and Gaming stores and offerings, across its consumer devices (smartphones, tablets and Xbox)
  • Continue with Windows 8 as a separate operating system, making Metro an optional overlay UI for touch screens, but allowing users to choose the old fashioned desktop UI as their primary or only UI if they so choose.

Microsoft has now either done, or announced plans to do, almost all of these things. Windows RT is going away and, in fact, it’s going further in unifying the different Windows flavors by creating a single version called Windows 10, with no more Windows RT or Windows Phone. It announced at Build that versions of Windows running on smaller devices would be free – this doesn’t go quite as far as making all tablet versions of Windows free, but it goes some way towards doing what I suggested. It has also significantly tweaked Windows 8 and in Windows 10 is restoring the traditional start menu and desktop as default features. The third bullet on my list is a longer term goal and one where Microsoft has made some progress, and where it is placing new emphasis with things like OneDrive, but there’s a long way still do go.

Windows 10 fixes many of the problems with Windows 8

Windows 10 goes a long way towards fixing what was wrong with Windows 8. It restores the desktop and Start Menu as familiar features – arguably, keystone features – of Windows, it removes the weird schizophrenia between the Modern UI and desktop worlds, and above all it recognizes that touch is not going to be the dominant mode of interaction for devices running this OS. That’s obviously a good thing. If Microsoft wants to increase (or even maintain) Windows adoption and if it wants to drive upgrade cycles, fixing the fundamental flaws in Windows 8 was critical.

But an update years in the making can’t just fix problems with the last one

However, fixing mistakes from the previous version of your operating system can’t be the sole focus, especially when you’re taking years between releases while everyone else is moving to annual or faster updates. As Apple gets ready to launch OS X Yosemite just a year after the major overhaul of Mavericks, Microsoft is talking about an update which won’t be available for about a year. And the whole focus of what we’ve heard so far has been about putting right what went wrong in Windows 8. Now, to be fair, there is bound to be more coming later on, and as we’ve already said fixing the Windows 8 mistakes was critical, but it’s a bad sign the entire focus so far is on fixing mistakes rather than creating true delighters. That will have to change by the time Windows 10 launches.

Microsoft risks repeating the Windows 8 mistake

There is one other major feature of Windows 10. Unifying the various threads of Windows into a single platform, with a single store and a single approach for developers. But here I believe Microsoft risks repeating one of the mistakes it made with Windows 8. Doing something because it appears to make good strategic sense for Microsoft rather than because it’s best for consumers.

Remember the key mistake with Windows 8 was attempting to parlay Microsoft’s desktop hegemony into mobile dominance. As a result, making decisions bad for customers, especially on mouse-and-keyboard devices (i.e. the vast majority of Microsoft’s installed base). With Windows 10, Microsoft risks repeating that mistake by attempting to again parlay its dominance on the desktop into dominance in mobile, but this time from a developer perspective. Microsoft hopes by unifying the flavors of Windows into a single platform, it can boost development for its mobile devices, because the large base of Windows developers will instantly become Windows Phone developers too.

This is flawed for two major reasons:

  • First, it puts developers rather than users first: unifying these operating systems has very little benefit to end users. Microsoft risks once again forcing a change on them to meet its own strategic objectives. In the process, it risks making decisions which hurt end users rather than helping them.
  • Second, it’s likely to fail even on its own merits, just as Microsoft’s attempt to reshape tablets in the PC’s image has largely failed. The simple reason? The vast majority of app development is now mobile first and the vast majority of the apps people use on their mobiles don’t exist as desktop apps at all, but as web apps. Think of Facebook, Pinterest, Twitter, Instagram, Snapchat and so on: none of them exist as fully fledged Windows apps today, because on desktops people use them in the browser. Whereas it mattered greatly in the past that Windows had masses of developers and Microsoft might have been able to parlay this strength into a strength in mobile, that ship has sailed.

As such, Microsoft’s strategy here risks both alienating users, who don’t want a single operating system across smartphones, tablets and PCs, and failing in its very objective of stimulating cross-device Windows development and boosting interest in developing for Windows smartphones.

Diversity: an asset and a liability

One interesting theme in Microsoft’s launch event was its recognition of the diversity of the Windows user base. This diversity was very well captured by the speakers, who talked about the diversity in the levels of expertise of users of Windows, the diversity in the devices in the market, the diversity of the use cases for those devices, and so on. This diversity is one of Microsoft’s great strengths: the 1.5 billion devices running Windows which Terry Myerson cited would be impossible without supporting a very diverse set of users and use cases. But it’s also Windows’ biggest liabilities, because it means each new version of Windows has to support all these users and use cases.

That diversity, and the frequent mentions of it, seem at odds with the One Windows vision at first, but it appears Microsoft intends to solve it with a multitude of different UIs built on top of Windows. In that context, then, one wonders to what extent developers truly will be able to develop apps once to run anywhere, when the UI contexts in which they’ll run might be very different (all full-screen apps on smartphones, for example, versus mouse and keyboard on many PCs). This is always the problem with any attempt at cross-device and cross-platform app development: it often leads to a lowest common denominator approach where nothing works excellently anywhere, and only the most painstaking developers end up crafting apps that truly work well on every device and in every UI context.

Windows 10 needs to build the foundation for faster iteration too

For all that we did hear about Windows 10 at the event, there were two key things we heard nothing about:

  • Pricing – it’s been widely rumored that Windows 10 may be a free update, which would make good strategic sense both as a sop to people disappointed with Windows 8 and as a way to drive a massive upgrade cycle and therefore get much of the base on the single-OS system.
  • Delivery – there’s also been talk about Windows as a service, and a new streaming model for Windows along the lines of what Microsoft has done with Office 365. This would also make a lot of sense and would further Microsoft’s shift from device purchase models to software-as-a-service models.

Both of these mysteries lead to a third: the cadence for future Windows releases. With Apple doing annual OS X releases and both Apple and Google doing at least annual mobile OS releases, Microsoft needs to get itself into a position where it can iterate more rapidly. Both lower pricing and new delivery models might feed into this, but it’s particularly critical that Windows 10 be built in such a way that it can be rapidly updated, especially in the variant that runs on smartphones. Windows Phone has been a fairly slow-moving OS, with only one major update since launch several years ago. That needs to change, and that’s something else I’m hoping we’ll hear more about as the release gets closer.