Windows 9: Microsoft faces four daunting challenges

The rumor mill says a public preview of the next big Windows release will appear this fall. But don't get fixated by features. This release isn't a "big bang" but is actually just the starting gun for the next stage in a very long race.
Written by Ed Bott, Senior Contributing Editor

According to the rumor mill, Microsoft will unveil the next big release of Windows, code-named “Threshold,” at the end of September, with a preview version available to the public shortly after.

We already know a little about what will be in what will probably end up being called Windows 9. Microsoft has officially announced the return of the Start menu, with a new, modern design, as well as the ability of mobile (Metro-style) apps to run in windows on the Windows desktop. More recent rumors suggest that virtual desktops will be added and Windows 8’s signature charms menu will vanish. Cortana might even make an appearance.

Enterprise customers, who pay dearly for Windows licenses, want as little change as possible. 

I’ve read a lot of discussion about whether this is Microsoft’s last “big bang” Windows release. As far as I’m concerned, that question was answered nearly two years ago. That honor belongs to Windows 8, which was perhaps the biggest big-bang release ever, introducing a completely new app model and blowing up a lot of the user interface conventions that Windows users had previously taken for granted.

In a world of big-bang releases, the Windows 8 feature set would have been frozen when it shipped. The many new features that have been added to Windows 8 in a series of updates over the past 18 months would have been saved for “Threshold,” which in turn would have been frozen when it ships next year.

A better metaphor for this release might be the crack of a starter’s pistol, marking the beginning of a new stage in a very long race. What gets delivered this fall as a preview will be updated in a few months as a final release, which will then be revised and expanded again and again over the next few years as Microsoft’s new, faster update cadence continues.

While it’s interesting to look at specific features that will be in this next release, that’s ultimately a myopic perspective. It’s more important to look at how that release will evolve (and, one hopes, improve) over the next two or three years.

Tech Pro Research

    So, rather than focus on features, I’ve decided to zero in on the big problem areas that those new and changed features should be designed to resolve. In a somewhat chaotic and ever-changing world dominated by mobile devices and online services, this is what I’m hoping we’ll see as Windows 9 evolves.

    Sharpen the line between business and consumer Windows.

    Remember back in 2001, when Microsoft unified the business and consumer versions of Windows in Windows XP?

    That was a good idea at the time. In that pre-tablet, pre-smartphone era, there was effectively no difference between the hardware in a business PC and a consumer PC.

    Today there are profound differences between business and consumer devices, and the tension between those two markets explains much of the turmoil that began with the release of Windows 8 nearly two years ago.

    You can’t ignore the business market, but you can’t expect much innovation in this legacy business either. Many businesses are buying desktop PCs to serve as single-purpose devices (in call centers or on factory floors, for example). Most conventional business laptops are running Office and a browser and little else. When was the last time you saw a brand-new desktop program or a new class of peripherals for conventional PCs? All the innovation is happening on mobile devices, with software delivered as apps and web services.

    And there’s the problem: Enterprise customers, who pay dearly for Windows licenses, want as little change as possible. Consumers, for whom Windows is an increasingly smaller part of the cost of a mobile device, want the newest features and apps without the headaches of managing a PC’s complexity.

    I suspect that sometime in the next few years Microsoft is going to have to let these two branches of the Windows line drift apart again. That might be the only way to keep conservative IT pros happy while not slowing down the pace of innovation on consumer-focused mobile devices that happen to run Windows.

    Improve the desktop experience.

    More than a year before the final release of Windows 8, then-Windows boss Steven Sinofsky noted that “for the foreseeable future, the desktop is going to continue to play a key role in many people’s lives.” And so, he promised, “we are going to improve it.”

    Nearly two years later, despite the addition of some desktop-friendly features, there’s still more room to improve.

    A new Start menu and apps that run in a window on the desktop will be part of Windows 9.

    That doesn’t mean a large investment in new features or utilities for the desktop. Nor does it require ripping out the genuine improvements that debuted in Windows 8 and slapping a Windows 7 interface pack on a Windows 8.1 kernel. Instead, it means more refinements in management tools and continued usability improvements in the transitions between classic desktop elements and the new modern pieces of the user experience.

    The goal? To dramatically increase the percentage of the Windows installed base that are willing to hop on the current version and stay current. To make that happen, Microsoft has to remove objections from PC users who might otherwise decide to stay with the earlier Windows 7 even as it approaches its end-of-support date in five years.

    And then there are two very big challenges: the headaches of Internet Explorer and Google's insistence on playing hardball, which I discuss on the next page.

    [Continued from previous page]

    Solve the Internet Explorer problem.

    Internet Explorer is the Rodney Dangerfield of browsers, getting no respect at all from Web developers and experienced Windows users.

    That reputation dates back years, to the middle of the last decade, when Microsoft seemed to give up on its flagship browser and on Windows Media Player after antitrust authorities on two continents won giant settlements over the tying of those products. For a brief time in 2009, Microsoft planned to ship Windows 7 without including Internet Explorer at all, although the company eventually backtracked on that issue.

    In a recent Reddit Ask Me Anything chat, IE developers joked that they had considered changing the browser’s name as a result of its less-than-stellar reputation.

    But the problem isn’t the name. The problem is the web itself. Microsoft continues to support IE 8 and 9, which require an unacceptable amount of hacking to render properly. Internet Explorer 11 is much better, but some developers, through benign neglect or aggressive “screw Microsoft” coding, create pages that don’t work properly in the latest version of Internet Explorer. In many cases, the problems would vanish if developers would simply render the page in a browser-neutral way.

    It doesn’t help when Microsoft’s own site for registered partners throws up this error message more than a year after Internet Explorer 11 was released:

    I saw this error when I visited Microsoft's partner site using Internet Explorer on Windows 8.1.

    Visitors have to manually add Microsoft.com to a Compatibility View list? If Redmond's marketers can't even fix their own website for business partners, why should they expect third-party web developers to make the effort?

    As I can attest from personal experience, Internet Explorer 11 on Windows 8.1 delivers, by and large, a very good experience. But there are and will continue to be enough rough edges and incompatibilities to drive most users into adopting third-party alternatives instead.

    Microsoft can chip away at those incompatibilities over time and make Internet Explorer a better browser. Whether it can ever overcome hardened negative perceptions from developers and power users is still an open question. And if it can’t, what’s Plan B?

    Neutralize Google.

    I saved the biggest challenge for last.

    Google is the most successful publisher of mobile apps in the world, as Dan Frommer noted on Quartz recently. Google has 5 of the 10 most popular mobile apps, as judged by Comscore data, including YouTube, Google Search, Google Maps, and Gmail.

    Of those apps, only Google Search is available for the Windows platform, either on Windows Phone or in the Windows Store for PCs running Windows 8.1 (and, soon, Windows 9).

    That’s a huge problem for Windows 8.1, which is continually dinged by reviewers for its lack of native apps for Gmail, YouTube, and other popular Google services. On mobile devices—tablets and laptops powered by Windows, including Microsoft’s own Surface devices—anyone who is invested in Google services is likely to be frustrated.

    Eventually, that frustration will lead them to install Google Chrome on their Windows PC. Which is, of course, the very reason for Google's intransigence.

    Google has neither the incentive nor the desire to cooperate in any way with Microsoft. In fact, Google deliberately removed features from the free version of Gmail shortly after Windows 8 shipped, breaking Gmail support in Microsoft’s Windows Mail client and in Outlook 2013.


    Google publishes APIs for interacting with its services, but prohibits developers from using them to build an app "for full-fledged email client access." And Microsoft is specifically prohibited from reverse-engineering Google’s services to build its own first-party apps, as it discovered when it tried to build a YouTube app for Windows Phone last year.

    I’m not sure what the answer is. A Windows-specific version of Chromium, minus Google’s tracking? A virtual machine running the Android Open Source Project prepopulated with sideloaded Google apps?

    Ironically, with Google having an effective monopoly over search (and search advertising) and Microsoft in the underdog’s role in mobile platforms, perhaps antitrust action is the answer?

    Nah, that’ll never happen.

    That's a pretty daunting list of challenges for Windows 9. I'm looking forward to seeing how Microsoft begins to tackle them.

    Editorial standards