Suppose I want to err on the safe side, let's consider for a while that what I asked before is not allowed.
Does it seem less questionable (or more clearly allowed) for me to prepare a zip file containing GPL'ed script files and put it as a separate download on my site (telling the users to unzip this file in a certain directory of my app) ? The GPL'ed script files will be unmodified (containing the license and all the original code) - but the file and directory structure may not be as in the original distribution. Thanks, Andrei Stefaan A Eeckels wrote: > On Thu, 14 Dec 2006 18:03:37 +0100 (CET) > "Alfred M. Szmidt" <[EMAIL PROTECTED]> wrote: > > > > My application consists of a collection of scripts (full source > > > distributed) licensed under a license that I'll call A. I want > > > to distribute with my application, the source version of a script > > > licensed under GPL. This script (source) will be imported and > > > used at runtime by my application. > > > > > > Is this allowed when license A is not compatible with GPL ? > > > > Yes. Your scripts are not based on or derived from that script, so > > it cannot have an influence on the copyright status of your > > original work. It's what is called "mere aggregation" in the GPL. > > > > This is completely wrong. The source code is not merly aggregated, it > > is actually imported by the program. And thus constitutes a deriviate > > work. It is exactly the same situation with linking a binary. > > > > It would be a different situation if it was a program that would > > execute the GPL script. > > In which case the result of this process on the computer where it is > running would be a derivative (not deriviate - you must be thinking of > "opiate" :) work of both scripts. > > The scripts of the OP are written by the OP, and contain no code of the > GPLed script. They cannot be a derivative work of the GPLed script. > They use the functionality of the script, just as they use the > functionality of the script interpreter that is used to run them. A > bash script is not a derivative work of bash. A script that sources or > calls another script is not a derivative work of that script. > > Where I can have some (but not much) sympathy for the claim that a > compiled program that is dynamically linked to a library should be > considered a derivative work (a functionally identical, but statically > linked program would contain material from the library), the idea that > "include" type statements create derivative works is asinine. > > It is especially asinine because it makes every source program a > derivative work of the OS or at least the libraries it uses. If this > type of reference is recognised as creating a derivative work, it > becomes impossible to write a program, because the copyright statutes > forbid the preparation of derivative works without the consent of the > copyright holder. > > If writing ". /foo/bar" in a shell script makes the _source_ code of the > shell script a derivative work of /foo/bar, programming essentially > becomes impossible, because then you cannot write those 10 characters > without the permission of the author of /foo/bar. > > -- > Stefaan A Eeckels > -- > When the need is strong, there are those who will believe anything. > -- Arnold Lobel _______________________________________________ gnu-misc-discuss mailing list [email protected] http://lists.gnu.org/mailman/listinfo/gnu-misc-discuss
