Re-using copyrighted code

[Benjamin Kaplan]

On Sun, Jun 9, 2013 at 1:32 PM, Mark Janssen <dreamingforward at gmail.com> wrote:
> On Sun, Jun 9, 2013 at 12:50 PM, Michael Torrie <torriem at gmail.com> wrote:
>> On 06/09/2013 11:18 AM, Mark Janssen wrote:
>>> You actually do not.  Attaching a legal document is purely a secondary
>>> protection from those who would take away right already granted by US
>>> copyright.
>>
>> You are correct, except that the OP has already stated he wishes to have
>> his code distributed. Without granting a license, the code cannot be
>> distributed beyond the people he personally gives the code too.  PyPi
>> cannot legally allow others to download it without a license.
>
> That's not entirely correct.  If he *publishes* his code (I'm using
> this term "publish" technically to mean "put forth in a way where
> anyone of the general public can or is encouraged to view"), then he
> is *tacitly* giving up protections that secrecy (or *not* disclosing
> it) would *automatically* grant.  The only preserved right is
> authorship after that.   So it can be re-distributed freely, if
> authorship is preserved.  The only issue after that is "fair use" and
> that includes running the program (not merely copying the source).
>

No, the original author retains all rights except those explicitly
granted. The same way that obtaining the "source" to a song does not
give you the right to redistribute the song all you want.

> Re-selling for money violates fair-use, as does redistribution without
> preserving credit assignment (unless they've naively waived those
> rights away).  I will have to take a look at  PyPi.  But if you are
> *publishing*, there's no court which can protect your IP afterwards
> from redistribution, unless you explicitly *restrict* it.  In which
> case, if you restrict terms of re-use, you're putting the court in
> jeopardy because you making two actions opposed to one another.  The
> only thing the court can easily uphold is your authorship and
> non-exploitation from a violation of fair-use (note again the key word
> is "use", nor merely copying the code).  But then if you waive *that*
> right away, you put the court in jeopardy again.
>

Fair use has nothing to do with money. It depends on how the work is
used and how you've changed it. Weird Al's song parodies are fair use,
even though he sells them. You distributing copies of a commercial
software to everyone is not fair use, even though you aren't making
money.

>> Here's how the GPL puts it, and of course this applies to any and all
>> licenses, even proprietary ones:
>>
>> "However, nothing else [besides the License] grants you permission to
>> modify or distribute the Program or its derivative works. These actions
>> are prohibited by law if you do not accept this License. Therefore, by
>> modifying or distributing the Program (or any work based on the
>> Program), you indicate your acceptance of this License to do so, and all
>> its terms and conditions for copying..."
>
> Well this is where one must make a distinction with fair-use -- if I
> re-publish my modifications then the code is still subject to the
> terms by the original author.  If I make a copy for myself and run the
> problem for personal, non-commercial use, then I am in the domain of
> fair use and have no other obligations.
>

Again, no. The GPL does not restrict your rights when running on
machines you control, but that's just because of the terms of the
license. Most commercial licenses include terms like "no reverse
engineering the software" that have nothing to do with distribution.