Vista at the tipping point

By Jeremy Allison
Posted on ZDNet News: Jan 18, 2007 2:02:00 PM

[The opinions expressed here are mine alone, and not those of Google, Inc. my current employer.]

Commentary--What if they released an operating system and nobody cared?

Five years after the release of Windows XP and costing a staggering $9 billion to create, Windows Vista has been released to manufacturing by Microsoft. The mighty marketing machine has swung into action. "Microsoft's most important product launch ever" blare the headlines in the trade press. Yet the silence from businesses and customers is deafening. No one cares. Contrast this with what most people would consider Microsoft's most successful Windows launch ever: Windows 95. People actually queued outside stores to be the first to buy this exciting new product, the launch itself was covered as news; real news, actually covered by the mainstream press as a real media event; not just in the computer trade press.

No matter how much spin is put on this launch, it's a disaster. There's simply no excitement about it. Most quotes from businesses are about how much of a chore it will be to upgrade, with warnings about how much old software will be incompatible and how people will have to buy new machines just to run it. No one actually wants this new system, except Microsoft and some of the hardware vendors who are desperately hoping Vista will revitalize moribund computer sales.

I think the day of the big-bang operating system release will die with Vista. This kind of upgrade has become obsolete. It might have made sense in the age of disconnected computers, where an upgrade involved a PC technician going to each desktop with a CD-ROM, but with the advent of Internet-connected PCs it's crazy. People want to simply keep patching their existing systems remotely and securely until eventually all of the original code has been replaced and you're running a new operating system. This at least is something we in the Open Source/Free Software community have become very good at, as it mirrors the very environment we need to create our software in the first place.

Where are the new features?
Why is Vista such a catastrophe and how does this affect the Open Source/Free Software community? Part of the problem, I think, is that Vista essentially does nothing new, and has no new features that are of interest to the general computer using public. The veteran IT journalist Nicholas Petreley (now editor of Linux Journal) created his first law of computer journalism, which is "No technology exists until Microsoft invents it". This held true while Microsoft systems were so primitive that every new release was a vast improvement on the previous one. The public "oohed" and "ahhed" over such exciting new features as multi-tasking, and overlapping windows; even as people in the industry tried to point out that every new feature was merely copied from other, more sophisticated systems. The problem for Microsoft is that most of this copying has already been done. Windows XP actually has most of the features of Linux and the Mac, though I'd complain they're implemented poorly in Windows. Even if Vista has improved on the implementation, what kind of a marketing message is "we now do things properly"?

The second problem is the Vista hardware requirements. Windows has always been notorious for being created to run well only the next generation of hardware, and Vista is certainly no exception. This might have worked back in the day when PC hardware was relatively young, and had not reached saturation point, at least in the Western world. But now every office already has at least one PC for every employee, and businesses resent having to spend more money on new hardware simply to satisfy a basic operating system upgrade. Most businesses have at least learned to live with Windows XP and see no need to completely replace existing working systems. Linux vendors have learned this to their cost; it's really hard to get businesses to throw out existing Windows desktops and replace them with Linux desktops even though the Windows desktops are painfully insecure and vulnerable to viruses and other malware. The pain of change is just too great.

The problem for Microsoft is that they need to make people upgrade in order to get the license fees that fuel the massive profit machine up in Redmond. You can't spend $9 billion on a new system only to tell your customers "we made a mistake, we'll just keep adding patches to Windows XP" even if most of their customers would rather they do just that. Microsoft needs to obsolete Windows XP and force people to upgrade to Vista, and as fast as possible. Microsoft's real competitor isn't Linux or Apple, it's the hundreds of millions of old Windows systems already out there that they have to replace -- or watch their income stagnate. This is going to get ugly.

It's a delicate balancing act. They have to replace these old systems and migrate customers onto a dependence on new features whilst still allowing interoperability with their old systems out there. But they can't allow too much interoperability, or people will chose to stay with their existing systems. And they can't allow too little, or people will look at alternatives such as Linux or the Mac if they feel they're going to have to replace everything anyway. This is the tipping point, and it's a time of great opportunity for Linux and Apple.

Get a refund
Over the next year or two, businesses will be looking long and hard at Vista. Consumers will use Vista, they'll have no choice. Microsoft's partnership with OEMs will ensure that. When you buy a new PC for home you'll get Vista – like it or not. My only advice to the home user is to do what my good friend Dave Mitchell in Sheffield, U.K. did and get yourself a refund for Windows from the vendor; Dell in Dave's case. It's hard work. Getting a refund is so rare that he's still coping with his 15 minutes of fame as media such as the BBC call him up for interviews.

Businesses have more options than consumers. The bigger ones do side agreements with Microsoft to allow them to have "obsolete" systems like Windows XP pre-installed on their new computers even though they're licensed for Vista and Microsoft hates allowing them to do that. Smaller businesses don't get that choice. But if they don't consider Linux or Mac systems, then I can guarantee that they'll end up wincing at the cost of implementing Vista. As I usually say in my talks "if you're not at least running a pilot Linux implementation you're paying too much for your Microsoft software."

What should the Open Source/Free Software community do about Vista? In my opinion, we need to just keep doing what we're doing. Various pundits have predicted doom and gloom if we don't compromise on our principles and adopt proprietary drivers, or license proprietary media codecs to allow Apple iPods to work with Linux. I don't agree. You don't change a winning strategy just as you're starting to succeed. Linux has become completely dominant in some areas such as high performance computing and outshines Windows in areas like Web serving. It's just in the workgroup server space, which is tied into the Windows desktop, that the Microsoft monopoly still holds. We need to keep improving our products and, most importantly, keep offering the one feature that only we have, the one that Microsoft can't copy without completely changing their businesses model. We need to keep offering our users software Freedom. If Microsoft wants to copy that, then I'd welcome it and welcome them to our community. But I think they have to go through the tipping point with Windows Vista first.

---

Jeremy Allison is one of the lead developers on the Samba Team, a group of programmers developing an Open Source Windows compatible file and print server product for UNIX systems. Developed over the Internet in a distributed manner similar to the Linux system, Samba is used by all Linux distributions as well as many thousands of corporations worldwide. Jeremy handles the co-ordination of Samba development efforts and acts as a corporate liason to companies using the Samba code commercially.

He works for Google, Inc. who fund him to work full-time on improving Samba and solving the problems of Windows and Linux interoperability.