Windows XP SP2 and the Risk of a Linux Backlash

The best method known for getting people extremely angry at you is simply to be right where they’re wrong — especially if you give them any opportunity to read a moral subtext into whatever they’re wrong about. It’s sometimes okay be a tiny bit smarter than the people you work with, but it’s always devastating to working relationships to be proven right if that makes people feel you are somehow morally better than they are.

Have that happen and they’ll soon be calling you smug, self righteous or pious. No matter how humbly you do the self-effacing shuffle, the working relationship will be history. You’ll never work well with them again.

I think the Linux community is at risk of having something like that happen right about now with respect to the people who sell and support Microsoft’s products. Unfortunately, the fact that we’re right and they’re not goes past technology to moral issues because deep down they know they’re misleading their customers.

There are two ways to look at this. The first is through behavioral theory — specifically Leon Festinger’s theory of cognitive dissonance, a theory that gets misstated an awful lot in part because it’s blindingly obvious in retrospect and in part because the theory leads to the comfortingly recursive prediction that most social scientists will hate and deny it. In other words, much of what you read about this theory isn’t outstandingly fair and balanced.

Critical Predictions

In brief, however, the theory makes two critical predictions. First, people will try to strengthen existing beliefs by rejecting contrary information and actively seeking out supportive information. Second, the energy believers put into doing this will increase as the boundaries between believers and others get stronger.

In other words, opposition strengthens belief and the more people believe in something, the harder they’ll fight to keep that belief no matter how obvious the increasing absurdity of their beliefs and actions might be to the uninvolved.

Festinger’s best example of this process at work involves the response to failed religious prophecy: Do “world enders” change their opinions the day after it doesn’t end? You’d think so, but they don’t. On the contrary, they change the due date and double their efforts to convince others that their unique knowledge conveys or reflects moral superiority.

So do Windows people switch en mass to Linux just because it’s better and more secure? You’d think they should, but they don’t. On the contrary, they double down their bets on whatever Microsoft promises will let them catch up to the rest of the world — 64-bit SQL server extensions last week, SP2 this week — and then redouble their efforts to show the rest of us how wrong we are when that too fails to meet expectations.

Confluence of Belief Systems

The most volatile emotional combination comes at the confluence of two belief systems, one culturally unquestionable and the other personally important for financial reasons. Unfortunately, that’s where the Windows to Linux transition is right now: Doing the right thing is culturally unquestionable for most Americans and some Canadians, but Windows is the primary source of income for many of Microsoft’s biggest supporters.

Emotionally and behaviorally, that combination spells crunch time for people like this poster to a PC World forum:

Microsoft is great, who dragged us all out of the dark ages of dos, microsoft did with windows. Today allmost everyone has a pc and internet. People are jealous and its all so common to attack the most succsessfull, They do have a monopoly, sure but who else is offering what they have? I say great job to Bill Gates, keep up the good work, and let me know if I can be of any help. My company recommends microsoft products, builds OEM with Ms installed and generaly I have made my living selling and supporting microsoft producs, so I for one am very happy, I still remember my first pc with dos 3.3 and I think by comparison Windows XP has come along way.

You don’t have to look at this situation through the lens of dissonance theory. You get the same prediction if you simply compare the attitudes of those Americans whose ethical views are situational to those whose ethical views are founded on long-term behavioral standards.

Imagine that you think you’re the victim of a minor medical error. After consulting several lawyers you know that the case probably won’t survive discovery and that the best you can hope for is a few thousand bucks in go-away money. One firm, however, tells you that their star performer is a snake charmer noted for extracting huge settlements from juries. They’ll happily take your case for 65 percent of the proceeds because his name will drive that go-away money into six figures. Do you take their offer?

If you decide yes, the chances are you’re a situational ethicist willing to see shades of grey wherever personal advantage might be at stake. In contrast, if you think blackmailing an insurer for personal gain is a particularly despicable form of theft, you’re probably an absolutist who believes that some things are inherently right or wrong.

That split carries over to attitudes to computing: See everything in nice situational shades of grey and you should hype Windows if that’s where the money is. See some issues as sharply right or wrong, however, and you’ll find yourself sending clients who insist on using Microsoft’s products in e-commerce or other security-sensitive applications to your less scrupulous competitors.

Absolutism and Moral Tension

My own experience suggests that absolutism is a great way of reducing your tax liability, but this unhappy fact doesn’t detract from the fundamental conflict underlying the Windows decision process.

The problem is that most of the Windows people eager and willing to make that sale know perfectly well that they shouldn’t. The result is that every new round of Microsoft failures ratchets up an already increasing moral tension in a debate that ought to be strictly technical and financial.

The result is a further hardening of positions with the Wintel people digging in to resist contrary information more strongly. In the worst-case scenario, the resulting behavioral conflicts will delay the adoption of Linux and BSD desktops in proportion to the degree to which Microsoft influences income among the decision makers — meaning that the United States will be last and slowest to convert to the new desktops coming out of Apple, Sun and open-source development groups like OpenOffice.org.

I think we’ve been starting to see that recently with Linux use in the U.S. falling behind that in the rest of the world, thereby giving earlier adopters in places like China, Europe and India further competitive advantages over Americans.

It wouldn’t take much in the way of publicly “counting coup” by the Linux community to make this much worse. So please, as the horror stories from SP2 start to come in next week, mute the laughter. It’s true that we are smarter and more moral then they are, but this time lets not tell them. Ok?