Florian Weimer dixit: >for shipping corresponding source code that was actually compiled, and >not just upstream tarballs plus downstream patches.
upstream tarballs plus downstream patches is preferred form of modification, though >It's not unreasonable to do this for link-time optimization purposes. but precisely that would not match the requirements, I’d rather embed a tarball Cem Karan dixit: >That said, I for one would find it *highly* amusing if gcc/clang added >a switch to embed the complete project but gcc/clang can’t know the complete corresponding source, which adds build systems… and IMHO also e.g. the debian/ subdirectory of packaging as separate(!) entity goodnight, //mirabilos -- [...] if maybe ext3fs wasn't a better pick, or jfs, or maybe reiserfs, oh but what about xfs, and if only i had waited until reiser4 was ready... in the be- ginning, there was ffs, and in the middle, there was ffs, and at the end, there was still ffs, and the sys admins knew it was good. :) -- Ted Unangst über *fs _______________________________________________ License-discuss mailing list License-discuss@lists.opensource.org http://lists.opensource.org/mailman/listinfo/license-discuss_lists.opensource.org
