I’m tired, tired, tired of reading self-proclaimed experts who insist that Windows has more malware because Windows has more market share, with the corollary that low market share is the main thing that keeps Macs safe.

This assumption is so wrong that I hardly know where to begin. As it happens, I recently found myself commenting on a particularly inane blog article, and that helped me to focus my thoughts.

The discussion about the reason why Macs don’t get viruses inevitably invokes the old canard about a direct correlation between market share and malware share. Many claim it as fact. I wish they would realize that their assertion, though self-apparent to them, is a theoretical explanation for an observed phenomenon based on mere correlation.

Observed phenomenon: OS X has not, to date, been seriously attacked by malware, while malware is rampant on the Windows platform.

Correlation: Macs have less market share than Windows.

Faulty logic: The correlation therefore is the cause of the phenomenon.

Well, correlations do not prove causation. Global temperature increase is not owing to the declining number of pirates, although such a correlation can be shown (just ask the Flying Spaghetti Monster).

Assertions of fact require evidence, not opinion. We’ve known that for, oh, about four hundred years now. Aristotle thought that the sun revolved around the earth and that heavy objects fall faster than light objects. These notions sound logical enough. But they are wrong, as Galileo demonstrably proved. Thinking something does not make it so.

For those who purport to “know” that malware authors avoid OS X because of market share, rather than for some other reason, I say, you might be as smart as Aristotle. And as wrong as Aristotle.

I’m going to assert a fact here, and let you decide if it is accurate. Ready? Here’s the fact: you, friendly reader, do not know why I use a Mac. Let’s go on: you do not know why I am typing this in MarsEdit. You do not know why I host this site on a server in a colocation facility. You do not know why I spend so much time developing my abilities as a Cocoa programmer.

OK—the assertions are true? You agree that I am the only person who knows those things because I am the only person who knows what’s going on inside my own head? (For the purposes of this thought experiment let’s avoid for now psychological complications such as denial.)

Now let’s turn our attention to an author of malware—or rather, to all of them, because some people like that blogger I mentioned insist that they know what is going on inside all those authors’ heads. I don’t think I’m going out on a limb here when I say that the only people who know why malware gets written on a particular platform are the people who write the malware. You don’t know, and I don’t know. Only they know.

Real evidence of those authors’ intentions is awfully hard to come by. If all the malware writers swore an awesome oath that they avoid OS X because of market share, I suppose we would have at least some evidence. It wouldn’t be dispositive, like dropping two cannonballs off the Leaning Tower of Pisa, but it would be something.

Of course we don’t have any such evidence, dispositive or not. At least I’ve never seen any cited in the long history of this debate. The closest thing to objective fact I’ve seen is the observation that even in areas where Microsoft does not hold the majority of market share (such as web servers and database servers), it still has the most malware, but even that comprises a mere correlation.

It stands to reason that an electron can’t go through two holes at the same time. But it does. Sometimes what stands to reason is not what stands up to evidence. It might very well be true that market share plays a part in the malware author’s decision how to spend his time. But that’s an opinion. It’s not a fact.