David Kastrup wrote:
All the library parts accomplish their intended purpose in the binary, and compiling and linking their source transforms them into the binary.
Here is a quote from the Compendium II, which the U.S. Copyright Office uses as a manual of procedure: <http://ipmall.info/hosted_resources/CopyrightCompendium/chapter_0300.asp> 323 Derivative computer programs. A derivative computer program is one that is based on or incorporates material from a previously published or registered or public domain program that has been revised, augmented, abridged, or otherwise modified so that the modifications, as a whole, represent an original work of authorship. The automated transformations involved in going from the source code of a library to a binary executable are not "an original work of authorship".
In short: I read and understand your words and explanations, but they don't seem to apply at all.
That is because you have an erroneous idea of what it means for a work to be derivative in the eyes of copyright law. Perhaps it is because you do not understand that the law sometimes uses English words in a technical sense such that their conventional dictionary definitions do not apply. In the everyday sense, one might say that a program which incorporates a library is derivative of that library, but in the technical sense used by copyright law, it is not. _______________________________________________ gnu-misc-discuss mailing list gnu-misc-discuss@gnu.org http://lists.gnu.org/mailman/listinfo/gnu-misc-discuss