On 2/4/2010 5:58 PM, Nick Kew wrote: > > On 4 Feb 2010, at 21:03, Tollef Fog Heen wrote: > >> ]] Nick Kew >> >> | I don't know if it comes under any of the FSF's exceptions for the >> | core toolchain (as in, compiling with gcc and linking glibc doesn't >> | bring you under GPL). >> >> It's a shell script. It's hardly linked into expat or apr-util and >> there's no way it can make the generated binaries fall under the GPL. > > Yes, I know it's a shell script. > > The point is, we *are* distributing it!
We aren't disagreeing; that is the point of the RAT tool, to catch things like this which we weren't paying attention to, once they had been checked out of subversion *and then* packaged for release. It would be great for that tool to be more widely used by existing projects, not simply the incubating ones :) And yes, it makes a great argument against autocrap at the ASF; it's a shell script, not compiled 'into' anything else, it did not benefit from its GPL 'protection' and yet causes nausea for any non-copyleftist who is still diligent about following all licenses, whether they agree with them or not. In this particular case, it was actually checked in a long, long time ago; http://svn.apache.org/viewvc?view=revision&revision=58140 and should have been averted then, but unfortunately was not. That said, neither the ASF nor any downstream packager can reasonably be shown to violate the license of that specific file. If you can come up with such a scenario, feel free to share it with legal-internal@ for review, and in the meantime please don't use public forums to raise existing licensing concerns, but only to discuss the future concerns or provide the list with the final conclusion of the legal committee's determination. I fail to see that this is a firedrill, and as was already pointed out, this entire component will no longer be distributed with the package as of the next minor version of releases. Lesson learned. Bill
