The GPL and license infection [LWN.net]

This disappointing Financial Times article has been more than adequately refuted by commenters on LWN and many other places. As FUD attacks go, this one was one of the more laughable in recent times. However, there is one point this article raises which is still occasionally trotted out by those trying to make people afraid of the GPL. It has been a while since we have looked at this claim, so it is worth a quick review pass.

Here's what "distinguished professor" Richard Epstein has to say:

First, as a straight interpretive matter, [GPL section 2b] only states what the obligation of each programmer is with his own private improvements. It does not in so many words specify the appropriate remedy when some portion of the open source code is incorporated into an otherwise proprietary program. The apparent intention of the provision is to "infect" that new program so that all of its content becomes open source software subject to the GPL. In principle, the entire Microsoft operating system could count as "the work" that becomes open source because a few lines of open source code have been incorporated into it by inadvertence.

Mr. Epstein does not, of course, tell his readers just where he obtains his information about the "apparent intention" of the GPL. Certainly it does not come from the vast amounts of text written by the creators and supporters of the GPL, who have never made this claim. Only the SCO group believes it has a license with this sort of power, and they seem to be having a hard time convincing others of this fact.

The relevant section of the GPL is this:

b) You must cause any work that you distribute or publish, that in whole or in part contains or is derived from the Program or any part thereof, to be licensed as a whole at no charge to all third parties under the terms of this License.

What that means is that, if, say, Windows were to be combined with GPL-licensed code in such a way so to create a derived product, the only way of distributing Windows which would comply with the license would be to put the whole thing under the terms of the GPL. Note that the GPL does not address the use of a combined program at all - only its distribution. Distribution under non-compliant terms would indeed be a violation of the license.

What happens then? Unlicensed distribution of copyrighted material is a straightforward legal matter. The person or company doing this sort of distribution can be sued for copyright infringement. Fines can be imposed, and distribution of the offending product can be halted with an injunction. Failure to comply with the license can also cause the infringer to lose the right to use the software in the first place.

These can be heavy penalties. In particular, a company which has worked hard to get a product to market can be devastated by a court-ordered halt to that product's distribution. Such are the risks of working with other peoples' copyrighted code; there is nothing unique to the GPL here. Mr. Epstein is right to say that no court would force proprietary code into the open as a result of a GPL violation. But it is only people like Mr. Epstein who raise that issue in the first place. It remains true that straw men are the easiest to knock down. What the community needs to do is to help ensure that such straw men are recognized for what they are.

(Log in to post comments)

The GPL and license infection

Posted Oct 28, 2004 18:38 UTC (Thu) by jre (guest, #2807) [Link]

Alert readers may have paid special note to the following:

The movement, whose principles have been expertly analysed by James DeLong of the Progress and Freedom Foundation ...

Ah, yes -- the Progress and Freedom Foundation. Now, where do you suppose they get their funding?

You suppose correctly.

Prof. Epstein appears to belong to the closed circle of GPL-attackers funded covertly or openly by Microsoft.
Fortunately for the rest of us, they tend to quote each other so extensively as to blow their respective covers.

See Tim Lambert's superb investigative post for a list of the usual suspects.

The less calloused among us may also be irritated by the distinguished professor's misuse of the phrase begs the question.
But perhaps we should cut him some slack. He may have gone to the same classes as Ken "not the sharpest pencil" Brown.

The GPL and license infection

Posted Oct 29, 2004 21:06 UTC (Fri) by giraffedata (subscriber, #1954) [Link]

I have to agree with Epstein more than LWN about the intent of GPL.

It is more than apparent to me that the FSF's goal is to get publishers to liberate their code, not to sue publishers for damages and collect royalties.

Epstein does not suggest that a court would order source code released, or that those who license code under GPL intend for a court to do so.

If the GPL resulted only in authors of free software winning lawsuits and collecting royalties, I'm sure RMS would consider the movement a failure.

So yes, the apparent intent of the GPL viral clause is to induce a copyright owner of a new program to open source his stuff too. Not force him -- just tempt him with the offer of royalty-free use of some code.