Why Windows 8 matters for real work, and so will Windows 9

The bottom line is this: to do real work with real computers you need a real operating system. You need Windows. Nothing else, really, will do.
Written by David Gewirtz, Senior Contributing Editor

Listen, forget about smartphones and tablets for five minutes, will ya? I want to talk to you about real computers, computers that do actual work, not just play Angry Birds and help you check in when you're at Starbucks again.

In fact, I want to talk to you about Windows computers.

Windows. You remember Microsoft Windows, right? That's the operating system, that as of January 2012, is used on 92.03% of all non-phone and non-tablet computers.

Windows is important. Very, very important. I'm writing this article because my colleague here on ZDNet, Steven J. Vaughan-Nichols, wrote Five reasons why Windows 8 is DOA. And that, my Dear Readers, is an assertion that deserves a response.

In this dissenting opinion, I'm going to approach this discussion from two vectors: the importance of "real" computers and the Microsoft Windows release cycle. I'm not going to go after SJVN's individual points, because -- while some of them are valid -- it's the assertion that "Windows 8 is DOA" that needs to be explored.

Also, I want to say that I really respect Steven. In fact, he's one of the technology journalists I respect most in the industry. That said, here's my considered rebuttal.

The importance of "real" computers

The market for processor-based technology is huge. It ranges from appliances like slow cookers, that use small processors to replace arrays of logic chips, to the giant server farms that power Facebook and Google.

When it comes to the very small processor-based devices and the very largest servers, there are teams of engineers that make operating system choices based on engineering necessity. And while there are versions of Windows that will work at both extremes, let's leave those extremes out of the picture.

Instead, let's talk about the computing devices we've all come to know and love these last 35 years or so: PCs. We've used PCs for a wide variety of work, a whole pile of office productivity applications, creative endeavors, vertical market specialty work, entertainment, and -- with a huge surge in the last five years or so -- a high level of social interaction.

We mix our use, so that one PC might be used for programming at 3pm and for playing Star Wars: The Old Republic at 9pm. On Tuesday, we might be doing bookkeeping, and on Thursday, we might be in Photoshop, editing an image for a Web site.

With PCs, anything has been possible. But that broad range of possibility has come at a price: complexity. Not everyone needs a PC that can do everything. My Mom, for example, only wants to do four things: email, writing, Web browsing, and checkbook management. That's it. She doesn't care about or need to do anything else.

My across-the-street neighbor has a pile of kids. All they want to do is tweet and text and log into Facebook to see if they've been mentioned by the cool kids. For them, an iPhone or Android phone is all they need. Of course, when they have to do their homework, they borrow their Dad's PC (and download some tunes on the sly while they're at it).

My point is that not everyone needs the power of a PC. For many people, a tablet with a keyboard is enough (although, in a future article, I'll show you why it's still not really possible to live with just an iPad as your sole home computing device). For other people, just a smartphone is enough.

Next: A predictable, flexible OS »

« Previous: A dissenting opinion

So, where the PC (and I include the Mac in this, as well as Linux desktops) had complete dominance of the user computing experience from about 1975 to about 2005, the emergence of smartphones and tablets has peeled off a large portion of the appliance-user market.


Even so, many of us with real work to do will need real computers. We need these computers because the locked-down experience provided by iOS doesn't provide enough freedom to get the job done. And yes, I acknowledge that Android isn't as locked down, but only the very fringe of Android users are tweakers.

For a lot of real work (and you know what I mean, from full-tilt video editing, to software development, to serious office work, and on and on and on), we need fully powered machines with lots of available hardware configurations and options. We need machines where we can choose the amount and speed of RAM, machines with huge and fast local storage capabilities, machines with the ability to string all manner of wacky peripherals, machines that can communicate on a wide variety of transports and protocols.

Put simply, for most real work, you need a real PC. Today, that's a PC running Windows, OS X, or Linux. And it's this need that showcases why Windows is so important to so many people.

Apple appears to want to move away from flexible, general-purpose operating systems like OS X, in favor of the locked-down, sandboxed environment of iOS. There are benefits to this, the reduction of malware being a big one. But this strategy also eliminates the flexibility to use whatever software is needed to get the job done.

Take, for example, WireTap Anywhere, a Mac OS X program that's used to re-route sound between various programs. It's quite astonishing and something I wanted to use in the Skype Studio project.

But WireTap Anywhere won't work in OS X Lion. According to the company:

Changes in the latest Mac OS will prevent you from capturing application specific audio...We are investigating potential solutions, though it's not simply a compatibility issue, and it's not as simple as it might seem. We do have a solution that will work, but 1) Apple wouldn't allow it in the MacAppStore, 2) It won't work with applications that are "sandboxed".

Because Apple is locking down its operating system, utilities like WireTap Anywhere, tools real people need to do real work, are being shut out of the ecosystem. We don't know if we'll eventually see Mac OS X go away, but it's clear that the flexibility we had in previous versions is being systematically routed out of the system.

And that means that -- over time -- Mac OS X won't be able to do much of the real work we need it to do.

So that leaves Windows and Linux as our workhorse operating systems. Both are capable operating systems, but they come from very different places -- and have very different temperaments. Windows is entirely commercial, supported by Microsoft. Linux is free-wheeling, with some combination of commercial support and some combination of loyal, if nearly insane, user base.

Because Linux has many different distributions, all of which don't support the same thing, it can be virtually impossible to get anything done in a reliable fashion.

I wrote about some of the challenges in Why I’ve finally had it with my Linux server and I’m moving back to Windows and Mea culpa: coming clean about my n00b Linux mistakes and while -- as many of the commenters will assert -- it is possible to be productive with Linux, there's a whole lot more folklore than some of us busy folk are comfortable with. By the way, if you want to see some bile, read those comments. Whew!

Anyway, the point is that Windows is predictable, documented in the extreme, supported completely, and consistent, and it's backed by one major corporation that does its best to keep the system running. This is in stark contrast to Linux, which is backed by a lot of fringe companies and professionals who traffic in guild-level secrets and passed-along folklore to keep their systems running.

This all brings us back to Windows. Windows is mission critical in today's computing world. It's the only real solution that's completely flexible, is completely supported, and is expected to remain so going forward into the future.

Since we're always going to need to do real work (even though the appliance users like my Mom and my neighbors may well migrate to tablets or locked-down systems), Windows will continue to be necessary and relevant.

All of that brings us to the next vector in this rebuttal, the argument that Windows 8 will be DOA.

Next: The Windows release cycle »

« Previous: A predictable, flexible OS

The Microsoft Windows release cycle

Microsoft has always had a strange pattern with its Windows releases. One release is reviled by techies, and the next is celebrated. Windows 3.0 was mediocre, while Windows 3.1 was a home run. Windows 95 was a big change but had its rough edges, while Windows 98 hit it out of the park. Windows Me was an embarrassment, while Windows XP was an epic success.

Windows Vista was a disappointment, while Windows 7 was -- by virtually every measure -- an exceptional operating system.

Microsoft is a machine, a Terminator, the Borg. They learn from their failures, adapt, and refine. It's almost like one operating system release is a multi-year trial balloon beta release, packaged in a profit center, and the next incorporates everything learned, everything loved, fixes everything hated, and simply dominates the computing landscape.

By this metric, Windows 8 is a trial balloon, experimenting with ideas and themes, those themes to be adopted, tweaked, or dropped in Windows 9. The Metro interface is one of those trial balloons. Microsoft has tried others.

Remember Bob? It failed miserably, but the help system technology fiddled with in Bob eventually found its way into Office and lived there for years. Or what about the Windows Sidebar in Windows Vista? It was celebrated as a new UI feature, ignored by most users, and (although its still usable) pretty much vanished into history. And yet, some of what we see in the enormously popular Task bar of Windows 7 came from the sidebar (and was similarly inspired by the OS X dock).

The point is, Microsoft isn't afraid to experiment with UI advances, take the slings and arrows of criticism from pundits and purchasers alike, and adapt, creating solid releases that build from questionable or experimental ideas.

All that would seem to imply that I agree with Steven's assertion that Windows 8 will be DOA. After all, if Windows 8 is one of the trial-balloon releases, doesn't that mean it's predestined to be a failure?

Oh, if only I could have a failure like Windows 8 will be. Seriously. Almost no one contends that Windows Vista was a raucous success, and yet, by just October of 2007 (about 11 months into commercial sales), Windows Vista had sold 88 million copies.

88 million copies. Even if you factor in OEM sales and the various editions, you're still talking about 5 to 10 billion dollars in sales. In 11 months.

If Windows 8 is to be "DOA" like Windows Vista was a "failure," well, then I wish I had such DOA and failure experiences. I could give up this gig, build me a ratrod, make the most over-the-top trebuchet punkin' chucker ever, and never, ever have to read another email message again.

Back to our rebuttal

Steven says no one needs Windows 8 -- and that's probably true. But we always like improvement in our computing environments, and while we may not think we need more capabilities, we usually find ways to use them.

Here's a short story about that. Back in the early 1980s, my buddy Jim and I decided we were going to go into business supercharging IBM PC ATs. It turned out that you could swap out a crystal and boost performance by something like 50%. So we packaged up the crystals with some instructions, planted ourselves at the San Mateo Fair Grounds for an early computer fair, and tried to sell these early overclocking kits for something like 20 bucks.

We actually broke even, which was good for a first venture. But we also were shocked, because almost everyone we met couldn't figure out why we'd ever want the PC AT to go faster. After all, it was so very, very fast. It ran at 6 Mhz (not Ghz, Mhz, or a thousand times-ish slower than what we have today).

Someone saying we don't need Windows 8 is like that. Our failure of imagination doesn't mean we won't later see an opportunity for the product or a key capability. Of course, most of us won't rush out to get Windows 8. I probably won't run it in production for a year or more after it's released.

But that's not the point. Windows evolves into adoption. It slowly oozes into production. As one machine breaks down, new machines replace it, and those machines often use the new OS. No doubt there'll be a feature or two worth loving, and those will encourage further adoption.

Windows 8 will eventually make its way onto our desktops. It may not sell 88 million copies in the first 11 months, just because people now have tablets and smartphones. On the other hand, PCs are still selling and more and more previously marginalized populations are now using computer technology, so there is likely to still be substantial international growth in the desktop operating system market.

Many of us will install Windows 8 on a couple of machines, this year (if it comes out on time) and certainly in 2013. In a few years, we'll start hearing rumors of Windows 9, and we'll then move the rest of our then-aging fleet of Windows 7 machines up to Windows 9, and the cycle will begin all over again.

Will we also have Windows tablets and Windows phones as mainstream, major players in the market? Will we also have Arm-based machines running Windows? Probably, to some degree.

But the bottom line is this: to do real work with real computers you need a real operating system. You need Windows. Nothing else, really, will do.

Editorial standards