SOS license

[Brian Behlendorf]

On Tue, 9 Nov 1999, Alex Nicolaou wrote:
> Brian Behlendorf wrote:
> 
> > > 3. permissive in allowing patch distribution under other license terms
> > 
> > An author can *always* distribute patches under a license of their
> > own choosing, at least by my reading of fair use.  Also, it looks like
> > even if people modify the software for personal use, they have to publicly
> > post their patches.
> 
> Hmm, if this is true then maybe some of the GPL is not enforcible. The
> GPL reads: "2c)  If the modified program normally reads commands
> interactively when run, you must cause it, when started running for such
> interactive use in the most ordinary way, to print or display an
> announcement including an appropriate copyright notice and a notice that
> there is no warranty (or else, saying that you provide a warranty) and
> that users may redistribute the program under these conditions". So, if
> I start with gdb, which prints such a notice, and publish a patch under
> terms that indicate that every end user must pay for the patched
> version, I have generated a useless patch. 

No.  You can charge for the patch, you can charge for the act of giving
the patch to someone, but you can't charge for the patched version.

> Either the user isn't allowed
> to modify the program with my patch under the terms of the license,

The user is always allowed to modify the program for personal use.  Again,
basic copyright law.  I can rip pages out of a book I bought and glue in
pages I've written myself, for my own use, without worrying about what
license those books have.  It's if I go to resell that combined work that
I run into problems.

> because my patch changes the start-up banner of gdb to indicate that it
> is non-free, or the user isn't allowed to apply my patch since they
> cannot do so and still follow my license that says that all users must
> pay me for the patch. 

I see 2c) as applying to a redistributor, not to a user.  I.e., you can't
redistribute a combined work that removes that line, but as a user, do
whatever you want.

> It is true that my intent is that even if the software is modified for
> personal or internal use the patches must either be published or
> returned to the copyright holder. I don't see a problem with that, but
> I'd be interested in having it explained to me if there is an issue.

There are some practical problems with it (are you prepared to handle lots
of bad patches, and does that have value for you anyways), but the bigger
issue is philosophical.  I contend that what people do, privately, with
modifications to software should not be an issue that concerns the
original author, unless I give the combined work to someone else.  It's a
matter of privacy at that point.  I may not want to deal with the hassle
of publishing them and having to answer questions about it.  I also
contest your use of the word "returned", as if those patches were taken
from you in the first place.

> > > 4. explicit about how the "official" version of the software is
> > > distributed
> > 
> > That's not a copyright issue, it's a trademark issue, though the
> > copyright license can tie the use of that trademark to rights granted by
> > the license.
> 
> I just want the license to be clear about it ... 

>From the Apache license:

 * 4. The names "Apache Server" and "Apache Software Foundation" must
 *    not be used to endorse or promote products derived from this
 *    software without prior written permission. For written 
 *    permission, please contact apache at apache.org.
 *
 * 5. Products derived from this software may not be called "Apache",
 *    nor may "Apache" appear in their name, without prior written
 *    permission of the Apache Software Foundation.

Is that not clear?  Substitute in whatever other trademarks you want to
protect.

> I'm tired of the debate over whether free software really has owners or
> not. 

Anyone who argues against that doesn't understand copyright law.  Yes,
Virginia, all software has owners.  Seems your issue is over who owns the
trademarks - still, that's not a hard to understand issue. The only
debatable point is whether someone who contributes a patch to Apache, or
to Linus for the kernel, or to any other project, is implicitly granting
either actual copyright ownership, or an unlimited right to distribute and
relicense, or neither of those two.

> The fact is that the free software development community has norms
> about how things are done, and I wanted to incorporate the norm into
> the license so that it isn't up for debate.

When someone "returns" a patch to you, are they obligated to grant all
rights to it to you?  What if there is a patent issue that neither you nor
the contributor knew about?  What if there was a patent issue that the
contributor *was* aware of but didn't tell you about?  There will always
be debate about licenses, particularly those that go beyond telling you
want you can do when you want to redistribute.

> At the same time, I didn't want to stop others from
> ignoring the norm, but simply restrict how they can do so in order that
> people can see that they are using a forked version of the software.

I, Bruce, and others apparently feel that the current licenses satisfy
that requirement.  

	Brian