Who is telling the truth, McBride or Torvalds?

Who is telling the truth here? Logic tells us that both of them aren’t. On one hand, we have Darl McBride, the man who apparently confuses threats to litigate with titillate, saying that a small number of files (less than 10, unless you count each multiple times the way SCO does) in Linux were copied from Unix. On the other, we have Linus Torvalds, the father of the Linux revolution, saying he knows firsthand that’s not so.

The proof of McBride’s assertions seems to be nothing more telling than the fact that there are files by the same name in both Linux and SCO Unix.
Why make preposterous claims with so little evidence to offer? That’s a good question, and it deserves a good answer.

Here it is. The first reason is money. SCO’s case against IBM ran into a brick wall when the judge told them to put up or shut up. It may very well be tossed out of court in January. The DMCA claims are new, and they are separate from the IBM case. If SCO’s case against IBM gets thrown out, the DMCA claims offer them the opportunity to pursue actions against foes who are smaller and weaker than Big Blue.

The second reason is also money. McBride stands to benefit financially whether the claim is true or not, if only he can get enough people to believe him.

McBride doesn’t even have to win in court in order to profit from those claims, bogus or valid. One way he can profit from them is by frightening other firms into buying SCO licenses to use Linux. Another way is by convincing investors his claims have merit long enough that he can cash in his stock options.

On the other hand, we have Linus Torvalds, the father of the Linux revolution. Torvalds has taken the time to recount his firsthand experience in writing several of the files Darl claims were copied from Unix. Torvalds told the New York Times (free registration required) yesterday, “In short, for the files where I personally checked the history, I can definitely say that those files were trivially written by me personally, with no copying from any Unix code, ever.”

Unfortunately for Darl McBride, it is the transparency of the open source process itself (a process he attacks regularly) that reveals the truth. Torvalds’ statements about the code in question can be proven true by simply looking at the code.

The source code doesn’t lie. It speaks in the unemotional language of computers, and it offers binary evidence in the same way that it does binary executables. The code says, “Yes: it is as Torvalds describes.” It also says, “No: it is not as McBride describes.”

McBride is quoted in the same New York Times story that carried Torvald’s account of personally writing some of the files as saying that said he stood by the company’s assertions. He says an expert some day will prove him right in court.

Of course, if he wants his story to remain consistent, what else could he say?