[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: What exactly is Derivative ?



-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

> Remove gcc and try to compile again.  It won't work.  Does that mean the
> binary is a derivative of gcc?

that's actually an interesting question, though its answer has fairly
obviously been answered long ago. gcc *does* do some fairly unique
things to a bit of source to turn it into a binary; its optimisations
and special dialectic features are a distinct part of the program, and
most other compilers don't do the same thing -- in fact i'd willing to
bet that any given compiler would produce different output (object code)
from any given set of inputs (source files). even *extremely* simple
things like 'int main(){}' produce different executable code under gcc
and sun's cc (when called with no extra options on an ultra-1 running
solaris 2.5; interestingly the gcc code weighs in at a full kilobyte
less than the sun cc code). so clearly the object code depends in a very
significant way on the compiler used; doesn't this mean that the output
of gcc is a derived work and must be released under gpl?

again, the answer is easily empirically proven to be no, because there
are lots of non-free -- even proprietary -- programs compiled with gcc,
but could someone explain the rationale behind the decision to me? were
rights to the output of gcc specially exempted from the gpl, or is it
determined that it really isn't derivative for some reason? if the
latter, what reason? maybe it just hasn't been thought of and we can go
force hundreds of programs to be re-released as free (or more likely
withdrawn or recompiled with something else)? ;P

- --phouchg
"Reasoning is partly insane" --Rush, "Anagram (for Mongo)"
PGP 5.0 key (0xE024447449) at http://cif.rochester.edu/~phouchg/pgpkey.txt

-----BEGIN PGP SIGNATURE-----
Version: PGPfreeware 5.0i for non-commercial use
Comment: Processed by Mailcrypt 3.5.1, an Emacs/PGP interface
Charset: noconv

iQA/AwUBNvbvbp49M+7gJHRJEQLvbACgy/CKiCRAYDvHziGVBP9m1rrhQZAAn3+y
D/cLG7zW+UxFUD/uXFIkIOFC
=QHlC
-----END PGP SIGNATURE-----


Reply to: