Subject: Re: binary-only code is evil
To: NetBSD-current Users's Discussion List <current-users@NetBSD.org>
From: Ben Collver <email@example.com>
Date: 09/30/2006 20:43:16
On Sat, Sep 30, 2006 at 06:50:20PM -0400, Greg A. Woods wrote:
> > My biggest complaint with blobs is lack of portability, [....]
> Lack of portability is, quite simply, just another type of bug.
> I.e. the main real-world problem with binary-only code is inability to
> fix bugs, add and/or remove features, etc.
> Free systems must allow their users to fix any and all bugs, including
> those that affect or prevent portability.
The authors designed and intended their binary-only code to be platform
specific. Since this is true, portability would be a bug.
You probably know better than I that folks have the ability to fix bugs
and add features to binary-only code. For example, the non-commercial
fixing of a bug in a SuperNES ROM, and English translation. Not an
ideal situation, but it was all done without source code.
NetBSD is not in the public domain, and is trademarked. Do you need
prior authorization to call it NetBSD, before selling your locally
patched version? If it is okay for NetBSD to be partially restricted,
can unqualified freedom be the issue?
On another note, can NetBSD include BLOBs without going against the
goals of portability and non-encumbering licensing? The thing I most
appreciate about NetBSD is that, like mutt, it sucks less. Now, a
person can use a non-free program like xv in combination with mutt.
POSIX makes that easy. If we saw standardization for BLOBs, and the
supporting code did not break other parts of the system, then what
are the drawbacks? (Probably, that the "if's" are false.)