It's hard to imagine that Linus Torvalds could have launched Linux without directly using earlier operating system work, according to a report that has become controversial even before its scheduled publication Thursday.
The 92-page report, from a 14-person Washington, D.C., think tank called the , suggests more Linux credit should go to Minix. A Unix clone, Minix that was designed by Andrew Tanenbaum at Vrije University in Amsterdam for the study of operating systems and software, which Torvalds used before he embarked on Linux development in 1991.
In an e-mail interview, Torvalds strongly disputed the study's conclusions.
According to the study, it's safe to argue that Tanenbaum, who had years of OS experience and who had seen the Unix source code, could create Minix in three years. "However, it is highly questionable that Linus, still just a student, with virtually no operating systems development experience, could do the same, especially in one-sixth of the time.
"Why are the most brilliant business minds in the history of PC technology, with hundreds of millions of dollars in capital, licensing Unix source code, if it is as simple as writing it from scratch with little help or experience? Is it possible that building a Unix operating system really only takes a few months—and, oh by the way, you don't even need the source code to do it?" the study questioned.
Gordon Haff, an engineer and analyst for analyst firm Illuminata, took a more measured view. "I think we can all stipulate that Linux is not a 'clean room' creation. Whether that makes it a derivative work is a question for the lawyers and the philosophers," he said. As for suspicions about Torvalds' rapid early progress, it should be noted "that the original product was quite primitive," he added.
The study comes not long after several attacks on Linux—many of them , whose Windows OS competes with Linux. More significantly, it arrives in the midst of a by the SCO Group, which .
Although the new study raises more questions than it answers, in an interview author Ken Brown was bolder in his claims.
"It's clear to me, at least from quotes from Tanenbaum, that Linus started from Minix...He just sat down with Minix and wrote this product. By definition, that is not an invention," Brown said. "If you sit down with the Ford blueprints and build a Chrysler and don't give Ford any credit, that's not invention."
In an interview conducted for the study, Tanenbaum said Minix "was the base that Linus used to create Linux. He also took many ideas from Minix, including the file system, source tree and much more."
If Linux is a derivative work of Minix, that makes Linux vulnerable to charges of intellectual property infringement by Prentice Hall, which published books and the Minix source code but restricted its use until 2000, the study said. "Arguably, Prentice Hall has lost out on tens of millions of dollars" because of lost book sales, the study said.
But Torvalds argued that he and other Linux developers have given proper credit.
"Linux never used Minix code...We never credited anybody else's code, because we never used anybody else's code," Torvalds said. But Unix, he said, did provide ideas: "Linux has always credited Unix. There has never been any question about the fact that Linux was very open about taking a lot of good ideas from Unix."
Minix, he said, was simply a platform on top of which Torvalds did his programming work.
The study suggested that Torvalds might have gradually replaced Minix code with Linux, but Torvalds denied it.
"I didn't 'write the Minix code out of Linux," Torvalds said. "I was using Minix when I wrote Linux, but that's in the same sense that you are using Windows when you write your columns. Do your articles contain Windows source code because you use Windows to write them?"
Torvalds isn't the only one to dispute the study: Tanenbaum himself sided against Brown.
"Linus didn't sit down in a vacuum and suddenly type in the Linux source code. He had my book, was running Minix and undoubtedly knew the history (since it is in my book). But the code was his," Tanenbaum said in a about his interview.
"By the time Linus started, five people had independently implemented Unix or something approximating it...All of this was perfectly legal and nobody stole anything. Given this history, it is pretty hard to make a case that one person can't implement a system of the complexity of Linux."
Fueling the flames
When the institute announced the pending publication of the report earlier this week—saying it "directly challenges Linus Torvalds' claim to be the inventor of Linux"—it immediately drew from open-source advocates who suggested Linux foe Microsoft was behind the report.
Microsoft indeed has provided funding to the Alexis de Tocqueville Institution for five years, a Microsoft representative said, without disclosing how much has been granted. Microsoft funds several public policy institutes, including the American Enterprise Institute, the Center for Strategic and International Studies, the Heritage Foundation, and the Cato Institute, the representative said.
Brown declined to discuss his organization's funding sources, but said there are several and that its research is independent. "I publish what I think and that's it. I don't work for anybody's PR machine," he said.
One area where Brown and Torvalds agree is that Torvalds shouldn't bear the title of Linux "inventor."
"I'd agree that 'inventor' is not necessarily the right word" to describe his role in Linux, Torvalds said.
The study also raises the issue that Torvalds saw Unix source code. This was available in annotated source code that John Lions, a professor at the University of New South Wales in Australia, made available to his classes. The notes were widely distributed illegally afterward, and "many suspect that Linus also had the Lions notes," the report said.
Not true, Torvalds said. "I've never seen the Lions book, although I've obviously heard of it. And no, no Unix source code either."
Brown and his colleagues interviewed more than two dozen people for the book, but Torvalds "didn't get back to us" with requests for comment. Torvalds said he never received any e-mail from the institution.
The Linux issue underlies Brown's concern that open-source software makes it easier for other countries to benefit from U.S. technological prowess. "How are you going to have an intellectual property economy if you can just rip off stuff?" he asked.
Such political and business issues likely will get more attention in a book Brown plans to publish in coming months that will expand on the study.
The study will be sold by an outside e-book seller, Brown said. Although his organization usually makes studies available on its own, outsiders have crashed its Web site twice in recent days after the institute published the Linux study press release, Brown said.
The study is at times controversial, but in the end, it isn't revolutionary, Illuminata analyst Haff said: "It doesn't ultimately tell me anything surprising that would cause me to rethink the role of open source."