Re: apache non-free?
On Fri, Dec 07, 2001 at 10:51:57AM -0500, Ben Collins wrote:
> As I said before, I am arguing my opinion, not the law.
You should have said this before. The way you wrote it you made it a
> If Apache truly
> enforces the definition (and it is up to their definition, not the law)
> of "derivative" as any modification, no matter how slight, then surely
> Apache is non-free, as far as the DFSG is concerned.
We do not interpret licenses in favor of the copyright holder. There
is sufficient precendence to this (biggest example being KDE) where we knew
the copyright holder would never enforce a problematic clause, but did not
consider it to be free anyway. So, if you are right about the DFSG
interpretation, then Apache is indeed non-free. The only point one can
argue about is if Debian wants such programs which require modified versions
(or derived works, whatever) to be renamed to be included. We can discuss
this point seperately though.
And it is not up to their interpretation, because if they don't offer a
written interpretation in the license (if they don't define the term), the
broadest meaning of the term as applicable by law is assumed by the court.
In copyright law, the term derivation is pretty good understood by the law
without a definition. Only if you want to restrict that term (for example,
if you want to declare that a "patched" version is not a derivative), you
have to include a definition in their license.
I have snipped the last part of you reply already, but let me say that if
you want maximum protection under law, you are well advised to *not* define
what a derived work is in your license. I would definitely advise against
defining terms in licenses which have a common meaning in copyright law.
Because this way, you are in most cases restricting the meaning of the
terms, and loose protection, and open loopholes or backdoors.
`Rhubarb is no Egyptian god.' Debian http://www.debian.org email@example.com
Marcus Brinkmann GNU http://www.gnu.org firstname.lastname@example.org