The Mythical 80%

This post is really an open letter to Paul Thurrott and Mary Jo Foley regarding how Microsoft introduced the now-renamed Metro interface in Windows 8. Paul runs the, while Mary Jo is at Clearly, they are popular technology analysts of notable repute. They also appear with Leo Laporte on Windows Weekly, where Paul has repeatedly contended Microsoft had to release the “Metro” tiled interface as a mandatory part of Windows 8 in order to assure it some level of adoption. Well Paul, just how well has Windows 8 been adopted by Microsoft’s established Windows userbase? I actually do have a more substantive response to your position, but that question was too relevant to let slip.

Your assertion (reviewing here for everybody else) was that Microsoft really didn’t have many other options, and that the inertia of its huge—roughly a billion—base of established users of its x86-based software offerings gave it a guaranteed income which it could rely on while introducing its touch interface. You referred to this userbase as an enormous “runway.” Taking your aviation analogy a step further, I submit that userbase effectively makes Microsoft the equivalent of an An-125, the world’s largest aircraft. You bet it’s on a long runway, the thing can’t exactly hop into the air because of its incredible inertia! You’re quite right to say no new major applications have appeared for Windows in years, but that only confirms to me that the market Windows dominates is stabile. Paul, I believe you’ve previously said you drive an older car, which tells me you don’t subscribe to the notion of buying the latest and greatest every two years so why are our expectations different for core software? Of course, the answer is that our economic model stresses unbridled growth in practice if not in theory, and companies are driven to maximize profit even if that means selling customers on things of dubious value. Otherwise, why replace software that we are satisfied with? The average consumer isn’t savvy about his or her digital needs so deception isn’t difficult.

Touch technology is essentially meant to make computing desirable to a broader population, and that’s fine as far as it goes. Even I have a Nexus 7. However, touch technology in its current incarnation is basically a conduit for consuming content. Relatively few people can thumb-type paragraphs into their device, and I’m certainly not composing this post on my Nexus. As writers, you and Mary Jo have often said that you need honest-to-God keyboards on which to do your work. Windows (and OS X) has historically been agnostic about whether it’s used to consume or create content, meaning that its capability extends beyond what touch hardware presently makes easy for someone to utilize. Therefore, I would say that Microsoft’s decision to require all Windows 8 users to contend with the Metro interface immediately upon logon was a bonehead move that once again undermined user confidence in the company’s ability to make wise strategic choices. It is as if they don’t recognize why Windows continues to be purchased in mass quantities. Sure, part of that is momentum, but the momentum comes from all those wonderfully critical true x86-compatible applications that keep companies and so many people—like me—in operation today. Anything that impedes users from getting to that compatibility blows a large crater in that runway you referred to. Although an An-125 is a large airplane with tolerance for some runway irregularities, there are limits.

Coincidentally, I encountered this article by Karl Mattson Monday morning that does a very good job of presenting the product design problem that I’m trying to illuminate. Karl focuses on the global markets, but his points are equally valid in the domestic arena. You said it yourselves, Microsoft basically copied the approach used in recent years by the likes of Apple and Google, and didn’t give Windows 8 customers much choice initially. Yet, the very reason why Apple succeeded with the iPad, iPhone, and yes, even OS X, is precisely because their respective markets were largely untapped. The situation with OS X would’ve been closest to Windows in that existing Mac users were compelled to replace all their software within a year. The crucial difference, though, was that Mac only had roughly 5% of the personal computer market at that time so Apple literally had little to lose. Microsoft potentially has the opposite scenario with Windows. Yes, they must innovate and incorporate touch technologies, but that needed to be done in a manner that co-existed peacefully with the established user interface. This is why Karl’s article is so poignant as it highlights steps a company should take to avoid an one-size-fits-all mindset in creating a product. Would that Microsoft had been more assiduous in applying those processes to development of 8. Instead, they made the same sort of mistakes they committed with Vista, a technical success but a marketing titanic.

For the record, I think Microsoft needed to introduce Metro in exactly the manner you say they could not, gradually. Not as a separate operating system, but as an optional subsystem available under Windows 8 on machines capable of properly employing it. No doubt initial adoption would have been abysmal, but that brings us back to the question I posed at the top of this lengthy post. Given the 900-million-dollar write-off Microsoft took last year on the Windows RT variant of Surface for Windows 8, a true non-starter in my opinion, would that money not have been better invested in evolving Metro to work better within Windows 8 on the x86 Surface Pro? We’ll never know, but that certainly would’ve given them time to develop Metro versions of their heavy-hitters like Office. Again, industry analysts tend to under-value that big base of existing Windows applications, and I grant that it may not seem cool, interesting, or maybe even relevant. Microsoft itself now regards Windows as a product line, but the userbase begs to differ. To us, that musty old pile of applications is the core of Windows’ relevancy. That’s why johnny-come-latelies like Windows Phone and Windows RT have so much difficulty gaining traction, at least domestically. “Windows” has specific meaning to us, and, unlike the situation Apple had with OS X, this userbase is large enough to inflict significant damage to the bottom line if bullied.

That probably explains much in the reports you’re hearing about “Threshold,” AKA Windows 9, initially scheduled for April, 2015. Only offering Metro in certain skews, what most people think of as “editions,” would be an improvement. However, I fear Microsoft still hasn’t received the correct message if the reports that they don’t intend to make Metro available to all skews are true. I’m not aware of anyone who objects to the availability of Metro. Rather, the issue has been Microsoft’s decision to make it the initial default on all system configurations. That choice created numerous problems. Yes, Metro needs to be part of every skews, including those sold to enterprises. I spent many years as an application developer for a multi-national energy company, and I can easily envision scenarios where a combination of touch and traditional keyboard would be quite productive. We know there are already laptops that can open flat. Touch-enabled versions would be invaluable in plants and factories as control stations mounted on walls and machines. Use touch for broad input, and the keyboard when precision is required. Somebody needs to shout this message too loudly for Microsoft to miss: Leave choices like the initial user interface up to user preferences and hardware capabilities.

It is at this point that I mentally heave a sigh of relief. I am a physically disabled person, and the primary focus of this blog is on how I encounter in everyday life affect, or are affected by, my disabilities. You’ll understand momentarily how news regarding Microsoft, Metro, touch technology, and Windows 8/9 all pertains to my blog, but I felt it essential to start with your take on the Metro kerfuffle and my intellectual response as someone with a Computing Science BS (a rather appropriate abbreviation) degree and a good chunk of enterprise experience (yes Mary Jo, “I went there”). Frankly, hell yeah, it was a boatload of material to cover, particularly when I have to type with one knuckle. Still, it seemed necessary to lead with the Microsoft stuff to garner your attention.

The true thrust of this post is somewhat broader, being another variation of Karl’s article that deals with the impact of this “consensual bias” on users with disabilities. What Microsoft did with Metro in 8 was an example of the problem. Although Metro is officially usable with any pointing device, we know it’s a bear without a touchscreen. Paul and Mary Jo, I don’t genuinely expect substantial replies from either of you really; what I sincerely want is to further the conversation on how Silicon Valley design decisions are increasingly being pushed out with fewer and fewer choices left open to the user. There’s the argument that the approach has served Apple well. True, but there’s no question in my mind that it also created much of the demand for Android. Several people without any technology interest have told me they found IOS to be too restrictive. I think Microsoft was equally overbearing in their original position regarding Windows 8 and Metro.

I have never told anyone this, but I decided in 1979 to formally enter the computer-relating field because I saw computers as a “great equalizer.” Like dogs, computers couldn’t care less about someone’s disability. As long as the person can effectively communicate to them what to do, the results of the cooperative effort will be just as good as everyone else’s. That equalization effect is key to enabling people with moderate to severe disabilities to function as effective members of our society. I am in a power wheelchair, and I cannot speak. Yet, just like anyone else, I graduated college, had a career, and continue to do pretty much do normal everyday activities, largely thanks to the Digital Age. In fact, it isn’t truly accurate to say I can’t talk. I can, because of an application I began developing back in 1997. I needed the ability to communicate verbally as part of my job, and my employer was willing to purchase any of the existing assistive communication systems. It turned out, however, that none provided the required speech quality so I rolled my own, using enterprise-level technology and based on a former DoD project. Even today, my application compares favorably to more modern equivalents, and I continue to occasionally “pop the hood” and modify something to better handle a scenario I come across, such as use in Amateur Radio.

Perhaps sharing this sort of information gives you a glimmer of why I continue to have a personal stake in that tiresome old x86 pile of software. J I remain somewhat active, and I have a couple of high performance laptops (Dell/Alienware, actually) that share the extremely hazardous duty of always being on my wheelchair. I literally refer to that as “active duty,” and it’s why I have to keep the service contracts current on those machines. This in turn requires me to replace one of them every two years (it’s a four-year lifecycle) so I have to care about whatever Microsoft is doing with Windows.

What prompted to write this epistle is a disturbing trend that seems to be undoing much of the equalization effect intrinsic to computers, specifically the devices and methods companies are inventing to provide a user interface. Understandably, this is being propelled by the imperative to minimize cost while making computers accessible to an ever-increasing population of potential customers in an ever-widening range of scenarios. I have no fundamental issue with that. It’s laudable, in fact. My objection is that companies no longer care—if they ever did—that not everybody can use every input device or method. Suppose you went blind tomorrow—Heaven forefend, imagine using anything with a touchscreen. There are workarounds in a few cases and I’m no expert on blind accessibility, but this example would be very vexing at best. Accessibility seems to be passé, and I don’t know that things will improve again until we have brain-reading implants. (“Brain Bluetooth,” anyone?)

In any case, disability access is only one dimension to what appears to be decreasing efforts to include any appreciable flexibility into new “solutions.” Ironically, Microsoft resolved the general issue back in the mid-90s in their own guidelines for applications written for Windows. One of the dictas could be paraphrased as thou shall provide the user with at least three ways of doing anything. Yes, exactly! That’s truly all it takes to give anybody a reasonable chance of doing or using something. Naturally, implementing that could be expensive in practice, especially if you’re developing a physical device. But again, Karl’s article spells out the basic steps. Or, Microsoft, read your own old documentation.

Yes, I am a demographic outlier, but think about that. By definition, anyone who appreciates technology for its own sake is a demographic outlier, because most people couldn’t care less about it. They use it either because they are required to (by work, for instance), or for a specific task. And yet, Silicone Valley is basing its latest innovations on what it believes those users will want. Excuse me? That largest group doesn’t really “want” anything until it’s “sold” to them so why are we relying on them to drive innovation? That’s the basis for the title of this post; there really isn’t a monolithic majority in any demographic. I would counter that it is the outliers who define the use cases, since we will be the first to try a new whatchamacallit and give feedback useful for refining it. So why do we seem to be forgotten?

And finally, the catalyst that triggered this whole long post (the general topic has been rattling around in my head for a decade) was a game, namely a MMO called World-of-Tanks. (Yes, I know, Mary Jo’s eyes are rolling, but hey, former enterprise geeks still need some recreation.) In that game, you can map any control in the game to any key on your keyboard, but that’s it. WoT is essentially all about driving tanks, and I’m dubious about the efficacy of driving a tank with a keyboard. Another similar MMO, War Thunder, takes a broader approach and allows you to map any game control to anything that Windows sees as an input event. The differences between those two MMOs is a fairly good microcosm of what happens when you do and don’t have flexibility.