Delirium wrote in part:
-- Does requiring attribution (like old-style BSD) make the license non-free? Richard Stallman seems to think so, from what I can gather from gnu.org.
I hope that this isn't what Richard Stallman thinks!, because his own licences (GNU FDL, also GNU GPL and GNU LGPL) also require attribution. Actually, according to <licenses/license-list.html#OriginalBSD>, the problem with the old BSD licence is "the ``obnoxious BSD advertising clause''", and even then, RMS says that the licence is (while flawed) still free.
The ``obnoxious BSD advertising clause'' is more than attribution; it requires attribution on all ''advertising'' for the software, and ''specifically'' for the software's origin at UC Berkeley. This is obnoxious because it's more than just a Credits file; even so, in RMS's opinion, it's still free. Source: http://www.gnu.org/philosophy/bsd.html.
-- Does permitting only verbatim redistribution make the license non-free? Many people (including myself) would argue so.
I would argue so too, and I believe that even Stallman would agree. RMS often permits only verbatim redistribution of his essays, but I don't believe that he would characterise those essays as free. He doesn't want them to be free, since they're ''his'' opinion -- in contrast both to free software and to free documents like Wikipedia, which belong to the world.
Neither RMS, nor Debian, nor the Open Source Initiative, give clear definitions of "freedom" except for programming code, and it's not clear that the same rules should apply to everything else. But all of them that require free software must be modifiable.
-- Toby