Re: Org-mode exporters licensing

[Marcin Borkowski <mbork@mbork.pl>]

On 2015-07-27, at 14:17, Daniele Nicolodi <daniele@grinta.net> wrote:

> On 27/07/15 13:52, Marcin Borkowski wrote:
>> I disagree.  Licensing a tutorial with GPL is a stupid thing to do.
>> A tutorial may contain code which people naturally mimic (or even
>> copy).  Such things should definitely be in PD.
>
> As yourself pointed out in one of your emails, in many legal
> ordinations, there is no such concept as public domain: you cannot
> renounce to the copyright on your intellectual production.

That I've already learned.  OTOH, one of the reasons to use PD might be
that I explicitly state that I object the legal system I live in.  (Mind
you: I'm not an anarchist, and I'm very far from that.  But this system
is almost unbearable.)

> Therefore licensing something as public domain is not quite possible. If
> you want to grant the users of your code the most freedom (but do not
> care about this freedom being carried over to others) the 3-Clause BSD
> license http://opensource.org/licenses/BSD-3-Clause, the 2-Clause BSD
> license http://opensource.org/licenses/BSD-2-Clause, or the MIT license
> http://opensource.org/licenses/mit-license.html are good candidate
> licenses formulated in the framework of copyright law as accepted
> internationally.

Thanks for the suggestions!

> However, you cannot derive your work from some other work distributed
> under GPL and license it with a more permissive license (as the ones
> suggested above). What constituted a derived work is however not
> scientifically defined (and you have been rather terse in describing how
> your work build upon code released under the GPLv3). In one place you
> explicitly mention running a query-replace on the source code:
> mechanical transformations of the source code are considered derived
> works, even if the end result does not resemble at all the original.

I agree that I was probably too concise.  In another post I included an
explicit example of what kind of "transformations" (mechanical or
otherwise) I had in mind.  I still personally find hard to believe that
what I have in mind would consitute "derived work".

> I would suggest you to do derive your work from the GPL code and then
> consult with the authors about its licensing. If you are only using the
> GPL code as a skeleton, I think they would not have objections (but you
> could also easily re-implement it from scratch).

This seems wise.  I'm not sure whether I would re-implement it "easily",
especially that I see no point in deliberately not looking at existing
code.  (Besides, I saw it anyway, and I can't unsee it;-).)

> Other than this I would recommend you to refrain from harsh comments on
> a matter on which you hold strong ideas but weak knowledge (as most of
> this thread demonstrates). Especially if your positions seem detrimental
> of the Copyleft model, and you are asking for help in a mailing-list
> devoted to a very successful Copyleft program.

Well, as I mentioned earlier, my knowledge is less and less weak, also
thanks to your explanations.  OTOH, the more I know about these issues,
the more I dislike the status quo, and the more harsh my opinions about
GPL in particular are.  (It is not a secret that I am very critical of
the GPL and of the FSF.  Still, as I said before, I'm very hesitant
about explicitly breaking their rules.)

> Cheers,
> Daniele

Thanks again,

-- 
Marcin Borkowski
http://octd.wmi.amu.edu.pl/en/Marcin_Borkowski
Faculty of Mathematics and Computer Science
Adam Mickiewicz University