When a developer distributes a modified version of open source code as his own without attribution to the original author, is that a copyright violation? Earlier, a district court said no. Yesterday, an appeals court strongly disagreed.

A typical copyright violation is the variety where someone makes money from the sale of a product or service whose idea or whose content belonged to someone else. The usual reason someone would license an idea or content to anyone else is in order to share in the proceeds. In the open source community, the motivation is different: The author seeks only credit, some measure of validation, and for others not to claim his work as their own.

So when a license has no monetary value, from the perspective of the law, is it really an infringement of copyright -- rather than just an ordinary breach of agreement -- when the terms of that license aren't adhered to? That's been a very difficult question for the law to answer, but yesterday, the Federal Circuit Court of Appeals gave a very affirmative response.

The case involved two producers of open source software with more than one thing in common: Both Robert Jacobsen and Kamind Associates produce software to let model train enthusiasts program their train controller chips. And the software from both was, in large part, the DecoderPro package created by Jacobsen's company, JMRI.

Kamind's Matthew Katzer used code that Jacobsen made freely available through SourceForge, and no one disputed that. Jacobsen's attorneys notified Katzer of that discovery in March 2005.

As an amicus brief filed on Jacobsen's behalf by a team led by the Creative Commons Association (PDF available here) explained, "Katzer and KAMIND admitted that they copied the Decoder Definition Files from the DecoderPro software. They also admitted that they stripped out the 'JMRI credit information' and 'comment fields.' The information removed from each copied file includes references to the copyright owner, author, copyright date and information about the Artistic License. Therefore, Katzer and KAMIND violated the Artistic License, by not including standard versions of the executable files, not explaining where to get the Standard Version, and not indicating how the decoder definition files were changed."

After Katzer apparently acknowledged that he had sold about 7,000 copies of his software, even though it gave no attribution to Jacobsen or follow-on to his public license, Jacobsen's attorneys decided that Katzer really did owe him a patent license fee per copy sold. In August 2005, Katzer was sent an invoice for $203,000.

Katzer then retaliated by registering the DecoderPro.com domain name so Jacobsen couldn't do so, and the fireworks began.

As Katzer's attorneys argued in their legal brief (PDF available here), "Katzer concedes, for the purposes of this appeal, that Jacobsen is the owner or assignee of the copyright, and that Katzer downloaded, modified and distributed the Decoder Definition files. However, Jacobsen has failed to show that Katzer has infringed one of Jacobsen's exclusive copyright rights. As a threshold matter, Jacobsen's attempt to use federal copyright law to enforce that terms of the Artistic License that require that the JMRI group receive credit or attribution in derivative works, is misplaced. Copyright law, as it is presently written, does not recognize a cause of action for non-economic rights."

The District Court reasoned that Katzer did what JMRI's license explicitly enabled him to do -- modify the code and redistribute it, even commercially -- and that just because he failed to give Jacobsen credit doesn't translate to an infringement of Jacobsen's copyright.

"The JMRI Project license provides that a user may copy the files verbatim or may otherwise modify the material in any way, including as part of a larger, possibly commercial software distribution," reads the District Court ruling from California District Court Judge Jeffrey White. "The license explicitly gives the users of the material, any member of the public, the right to use and distribute the [material] in a more-or-less customary fashion, plus the right to make reasonable accommodations. The scope of the nonexclusive license is, therefore, intentionally broad. The condition that the user insert a prominent notice of attribution does not limit the scope of the license. Rather, Defendants' alleged violation of the conditions of the license may have constituted a breach of the nonexclusive license, but does not create liability for copyright infringement where it would not otherwise exist."

But here is where the three-judge panel of the Appeals Court found Judge White in error: by presuming that just because Jacobsen and JMRI didn't charge a license fee, he didn't have any economic interest in the product.

As the final CAFC ruling reads (PDF available here), "Traditionally, copyright owners sold their copyrighted material in exchange for money. The lack of money changing hands in open source licensing should not be presumed to mean that there is no economic consideration, however. There are substantial benefits, including economic benefits, to the creation and distribution of copyrighted works under public licenses that range far beyond traditional license royalties. For example, program creators may generate market share for their programs by providing certain components free of charge. Similarly, a programmer or company may increase its national or international reputation by incubating open source projects. Improvement to a product can come rapidly and free of charge from an expert not even known to the copyright holder. The Eleventh Circuit has recognized the economic motives inherent in public licenses, even where profit is not immediate."

The win for Jacobsen drew a statement of support from Stanford University Professor Lawrence Lessig, a principal advocate of open source rights: "In non-technical terms, the Court has held that free licenses such as the [Creative Commons] licenses set conditions (rather than covenants) on the use of copyrighted work," Prof. Lessig wrote. "When you violate the condition, the license disappears, meaning you're simply a copyright infringer. This is the theory of the [General Public License] and all CC licenses. Put precisely, whether or not they are also contracts, they are copyright licenses which expire if you fail to abide by the terms of the license."

In a statement on SourceForge late yesterday, Jacobsen wrote, "Although there will be some further proceedings, this is a huge step forward for us. The appellate judges have ruled that when Katzer took JMRI files, modified them, and distributed them as if there were his own, he was violating copyright law and we can hold him accountable."