News Stay informed about the latest enterprise technology news and product updates.

One-on-one with Richard Stallman, part 1GNU founder opines on SCO-IBM suit

The SCO-IBM dispute raises many questions. If SCO Group prevails in court, could the GNU/Linux operating system be stripped of important code? Should users worry about future challenges from other groups? And will the flap slow enterprise adoption of free software? Richard Stallman's answers: No, no, and who cares?

In other words, Stallman -- founder of the free software movement, the GNU Project and the Free Software Foundation -- isn't too worked up about the SCO-IBM battle, which began March 6, when SCO alleged in a Utah court that IBM had improperly donated SCO's proprietary Unix code to GNU/Linux.

The GNU founder's name actually comes up in SCO's complaint, in a historical section littered with terminology of the sort Stallman can't abide, like the phrase "Linux operating system." That phrase is incorrect, Stallman says, because "Linux" is the name of only one part of the system: its kernel. The complete OS contains much more, including, for one thing, an awful lot of GNU Project software.

The distinction is important to Stallman, which makes some sense, given that Stallman's vision for and work toward a free operating system predate the Linux kernel by a number of years. (That's "free" as in "free speech," he's likely to tell you, not as in "free beer"; that is, Stallman has no objection to the selling of software. But he believes users should be free to modify its source code.) The GNU Project's OS was near complete, he says, when it was combined with Linus Torvalds' kernel. In its complaint, SCO doesn't take note of these distinctions -- to its detriment, Stallman says.

The following interview shows excerpts of our talk with him about these issues and more.

What do you think of the SCO lawsuit?
The first thing to realize is that SCO was completely confused by the misnomer I've been talking to you about, because if you read their complaint, it says 'Linux' all the time. And there is no single thing which fits even all their background statements that they make about 'Linux.' Some of them fit the kernel. Some of them fit the whole system. But you can't make all of them fit either the kernel or the whole system. So it's not even a coherent position. You've probably read the complaint.
I haven't read all of it. I've seen some parts of it, enough to see how completely confused they are. The complaint, in many instances, expresses skepticism that free-software developers could have, without help, created a system as powerful as the present-day GNU/Linux. For example, the complaint describes Unix as a 'luxury car' and says GNU/Linux was a 'bicycle' until IBM came along, which I guess would have been about 2000.
It's [bull]. People found GNU/Linux to be a fine replacement for Unix six years ago, even somewhat longer. It was reliable. It was solid. It compiled better code. Is that going to hurt SCO, do you think, down the road?
I don't know. I would be glad if it does, but I really have no idea what the courts are going to do.

Another point to notice is that the GNU/Linux system was going quite strong before IBM started making any substantial contribution. So we're talking about marginal parts of the system. Whatever the outcome of the court case may be, it won't affect the system overall. I know the GNU Project has for many years had guidelines in place stating that developers shouldn't look at the Unix source code. Do you feel confident that those guidelines have been followed?
Mostly. You can never be absolutely certain -- and this is true whether you're writing free software or non-free software. If you've hired people, how do you know whether your staff copied something they weren't supposed to copy? You can't just look at it and tell. So you tell them not to do it. And you established this policy when?
In the '80s, based on talking with a lawyer about what we should do. Now, I don't think it would be wrong to copy source code without authorization, but we wanted to make a system that people could use without fear of lawsuits. So, therefore, we adopted the policy [that] we're going to write it all ourselves -- we being, of course, whoever joins the project. Given what you said a moment ago, that it's impossible to know whether the rules were always followed, should users of GNU/Linux be concerned?
People using any software should be concerned, I suppose, because it's a possibility with any software. You can never rule this out with absolute certainty. But, you know, absolute certainty is more than you can ever have in the physical universe. What if IBM made some contribution that enhanced a key part of the system?
To enhance a key part of the system doesn't mean that the enhancement is necessary.

In any case, I see no reason to doubt that IBM wrote these things itself. Because IBM is perfectly capable of developing powerful software. This is an absurdity in this complaint, the assumption that no one in the world could possibly make something that is the equal of what was in Unix. I mean, I don't know about those areas [that SCO is concerned with], but I've done things that were better than what was in Unix. You know, GCC is tremendously better than the Unix C compiler and C++ compiler. And I did that myself, to the point where it was better. So, they're asking you to assume that it is inconceivable that anyone could possibly have done such a job independently of them, which is ridiculous. Given that SCO hired someone like David Boies, who can't be a cheap attorney--
Who knows? That doesn't prove that they've got a case. Do you think this lawsuit is likely to spur other challenges to GNU/Linux?
No. First of all, it's a mistake to think of this as a challenge to GNU/Linux. Because it probably concerns specific code in specific programs, and we could live without all of them. So it doesn't really affect whether we can go on using the system. But SCO must have something that he thought was credible.
They may think they'll get IBM to pay them some money. Remember, what is 'credible'? All that needs to be credible [is] not that they're right, but that they have a chance of getting money.

I'm just pointing out that you don't need to assume that they even expect to win. Do you think this dispute could affect enterprise adoption of free software?
I don't know, but I tend to think that you can talk yourself into worrying about it too much. But aren't businesses sometimes prone to that?
Who knows? I'm not very interested in the subject. Keep in mind that we didn't develop GNU for the sake of having it be used by businesses. We welcome businesses to use it, and everybody, every user of computers should be free to study and change and redistribute software, all the software they use, and that includes businesses, if they're using computers. But we don't give any particular priority to businesses. Our main goal is to see that individuals can live in freedom, and it's nice if businesses can live in freedom to, but individuals are more important than businesses.

Click here for part 2 of's interview with Richard Stallman


Best Web Links on newsmakers, leaders news exclusive: "SCO suit could set nasty precedents for Linux" news exclusive: " Linux analysts, users offer varied predictions on SCO suit"

Dig Deeper on Linux servers

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.