Re: Open Test License v1.1 rejection

2004-02-18 Thread Alex Rousskov
On Wed, 18 Feb 2004 [EMAIL PROTECTED] wrote:

> Alex Rousskov scripsit:
>
> > Assuming I did not, let me replace "derived products" with "derived
> > works" since "product" is difficult to define. I will also explicitly
> > include "published test results" in the derived works. A published
> > test result is a derived work, right?
>
> No, at least generally not.  When you compile a program with the
> GPL-licensed GCC, the resulting binary is not a derivative work of
> GCC.

Right. Interesting. So one can require modified test cases (source
files) to be renamed before re-distribution, but one cannot require
results based on modified test cases to be renamed before publication.
So one can rename the test case but leave the old name in results and
still be compliant with the license.

Are there any examples were licenses claim "program results" of some
kind to be derivative work?

Thanks,

Alex.
--
license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3


Re: Open Test License v1.1 rejection

2004-02-18 Thread jcowan
Alex Rousskov scripsit:

> Assuming I did not, let me replace "derived products" with "derived
> works" since "product" is difficult to define. I will also explicitly
> include "published test results" in the derived works. A published
> test result is a derived work, right?

No, at least generally not.  When you compile a program with the GPL-licensed
GCC, the resulting binary is not a derivative work of GCC.  You are still
trying to make a copyright-based license do the work of a trademark, and
it just won't.

-- 
I marvel at the creature: so secret and John Cowan
so sly as he is, to come sporting in the pool   [EMAIL PROTECTED]
before our very window.  Does he think that http://www.reutershealth.com
Men sleep without watch all night?  --Faramir   http://www.ccil.org/~cowan
--
license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3


Re: Open Test License v1.1 rejection

2004-02-18 Thread Alex Rousskov

On Wed, 18 Feb 2004, Ian Lance Taylor wrote:

> As can be seen form the comments, the problem is clause 3:
>
> 3. Publication of results from standardized tests contained within
>this software (, ) must either strictly
>adhere to the execution rules for such tests or be accompanied
>by explicit prior written permission of .



> The distinction between distribution and use is important.  A license
> based on copyright can't restrict the use of the software.  It can
> only restrict copying of the software.

I now better understand the difference and its importance. Thanks for
clarifying it to me!

> I think that you will need to remove any restrictions on the use of
> the software.  That means no restrictions on publishing test
> results.

Understood. Please check whether the following modification of
Apache-v1.1 clauses would pass the "no use restrictions" test.

The original and OSI-certified Apache clauses say:

4. The names "Apache" 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 [EMAIL PROTECTED]

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.

Let me modify that a little. I will replace
"Apache" and "Apache Software Foundation"
with a
standardized tests names (, )
template:

3a) The names of the standardized tests contained within
this software (, ) must not be used
to endorse or promote products derived from this software
without prior written permission.

The names of these standardized tests may not appear in
derived product names without prior written permission. For
written permission, please contact .

The above modification seems identical to the original as far as OSI
certification is concerned. Is it not? Did I introduce any "use
restrictions" at this step compared to the Apache license?

Assuming I did not, let me combine "must not endorse" and "may
not name" sentences into one:

3b) The names of the standardized tests contained within
this software (, ) must not be used
to name, endorse, or promote products derived from this
software without prior written permission. For written
permission, please contact .

Did I introduce any "use restrictions" at this step?

Assuming I did not, let me replace "derived products" with "derived
works" since "product" is difficult to define. I will also explicitly
include "published test results" in the derived works. A published
test result is a derived work, right?

3c) The names of the standardized tests contained within
this software (, ) must not be used
to name, endorse, or promote works derived from this
software (including published test results) without
prior written permission. For written permission,
please contact .

This was probably the most "risky" step. Does "derivative product"
mean something other than "derivative work" from legal point of view?
Did I introduce any "use restrictions" at this step?

Assuming I did not, let me _relax_ the above requirement for written
permission. I will make written permission required only if tests are
not executed using standardized rules:

3d) The names of the standardized tests contained within
this software (, ) must not be used
to name, endorse, or promote works derived from this
software (including published test results) without
either prior written permission or strict adherence to test
execution rules. For written permission or execution rules,
please contact .

Since (3d) does not add any restrictions to (3c), I hope it does not
introduce any OSI certification problems if (3a-c) versions did not.

> I should add that I understand why you want the restrictions.  But
> there are many things which people want which do not fall under the
> actions permitted by open source.

Yes, of course. I hope that by starting with an already OSI-approved
license and applying what seems to be OSI-irrelevant changes, I can
arrive at something OSI-approvable. Did I make any mistakes in the
above derivation? Is (3d) clause compatible with OSI requirements and
open source spirit?

If it is, I will resubmit the license with (3d). If it is not, I would
love to know where I went wrong when reshaping OSI-approved Apache
v1.1 clauses.

Thanks a lot,

Alex.
--
license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3