Copyright Law and Linking

Why do I spend so much time dealing with issues of software linking? Does this topic really matter to anyone but open source zealots?

Consider the metaphor of the World Wide Web, a huge collection of individually written web pages that anyone can access and display just by linking. Those pages are individually copyrighted works, made available to all by their authors, generally for free (i.e., at zero price). Under the copyright law, you do not create a derivative work of someone's web page by linking to it, nor is it a derivative work of your web page if it links to you.

At most, such linkages create collective works. A web page, for example, that contains links to articles about open source may present those links in an original, copyrightable way. That list of links is a copyrightable original work of authorship , and the links operate to create a collective work . But the original articles remain the copyrightable works of their own authors.

Not that we can't envision using the Internet to create derivative works of web pages. You can find a web page you like and make changes to it, using the modified version as your own. You can translate a web page from one language to another. You can provide editorial revisions, annotate the web pages, or elaborate upon them. You can then link to your new versions. In doing so, you create derivative works . But it is not the linking that made the difference.

I do not want to discourage the creation of collective works . To do so would be inconsistent with the goals of free and open source software, just as it would be inconsistent with the goals of a free and open World Wide Web. Are the GPL's Programs so different from other copyrightable works that they deserve a narrower range of freedom?

One could, of course, ask the authors of the GPL how to interpret their license provisions, and they have indeed spoken out about this topic on their website, www.fsf.org , and in other public venues . But it is legally unnecessary to know what the drafter of a license ”usually just an attorney with no stake in the matter ” meant to say. That is why I can legally ignore the advisory notice that is published with the GPL after its terms and conditions have ended:

This General Public License does not permit incorporating your program into proprietary programs. (See "How to Apply These Terms to Your New Programs.")

Under the law, only the common understanding of a licensor and his licensees matters, as reflected in the written terms and conditions of the license agreement between them. It is Linus Torvalds, and the thousands of other licensors under the GPL, who have standing under the law to assert their interpretations of the GPL, not the Free Software Foundation (except for that software for which they own the copyrights). And it is a judge who would ultimately decide such an issue if it reaches that level of conflict.

One final warning: If there is an ambiguity or uncertainty of interpretation in a license, the license will generally be interpreted against the licensor regardless of what the license drafter meant to say. It is up to the authors of the GPL to make their license clear, not up to licensees to seek outside guidance to interpret it. I explore that issue further in Chapter 12.

I won't give legal advice of a general nature to the readers of this book. So you can take with a grain of salt my belief that these interrelated sections of the GPL quoted earlier will ultimately be read by the courts to mean that derivative works are subject to the GPL's reciprocity provision, but collective works are not. And as I shall argue again more fully in the discussion of derivative works litigation in Chapter 12, the legal analysis of what constitutes a derivative work simply doesn't depend upon the style or mechanism of inter-program linking.

This, by the way, is also the only interpretation that is consistent with item 5 of the Open Source Principles listed in Chapter 1, that allows licensees freely to combine open source and other software.