Re: Future of openLilyLib

[Carl Sorensen]

On Tue, Sep 22, 2020 at 3:08 AM Karsten Reincke <k.reincke@fodina.de> wrote:

Dear Carl;

here is my explanation using the method of showing an analogy:

(1.A) MIT/GNU Scheme:
(a) is an interpreter of the Scheme programming language
(b) is licensed under the GPL
(c) takes Scheme Code & delivers output

(https://www.gnu.org/software/mit-scheme/)

=> Any Scheme code being taken as input for the MIT/GNU Scheme interpreter may be licensed under any other license. The GPL copy left effect only affects the development of the MIT/GNU Scheme interpreter itself.


(1.B) guile-aspell
(a) is a Guile Scheme library for comparing a string against a dictionary
(b) is licensed under GPL 3+
(c) is included into an overarching program functionally using this library
(d) is interpreted by MIT/GNU Scheme in combination with the overarching program

https://www.gnu.org/software/guile/libraries/

=> Due to §5.c of GPL-v3 (https://www.gnu.de/documents/gpl-3.0.en.html) the overarching program using the library guile-aspell must be released under the terms of GPL-v3 too because it depends of the functionally of the guile-aspell library

But here is where you lost me.  The program using guile-aspell must be released under the terms of GPL-v3.  But the output of the program need not be released under the terms of GPL-v3.  The output can be released under any terms the user desires to use. 

(2.A) Lilypond:
(a) is an interpreter of the Lilypond notation language
(b) is licensed under the GPL
(c) takes Lilypond code & delivers output

(https://lilypond.org/index.html)

=>  Any Lilypond code being taken as input for the Lilypond program may be licensed under any other license. The GPL copy left effect only affects the development of the Lilypond program itself


(2.B) OpenLilyLib
(a) is a Lilypond library for the GNU LilyPond music notation software
(b) is licensed under GPL 3+
(c) is included into an overarching lilypond music code functionally using this library
(d) is interpreted by the LilyPond interpeter in combination with the overarching lilypond music code

https://openlilylib.org/

=> Due to §5.c of GPL-v3 (https://www.gnu.de/documents/gpl-3.0.en.html) the overarching lilypond code using the library OpenLilyLib must be released under the terms of GPL-v3 too because it depends of the functionally of the OpenLilyLib library

Yes, if you choose to release the lilypond code, it must be released under GPL-v3.  But if you release only the output of the code, no such restriction applies.

 

Summary:

If I wrote a piece of music using LilyPond Code (for being interpreted by the Lilypond interpreter) and if I included OpenLilyLib into my code, you as the Copyright owners of OpenLilyLib could me enforce to distribute my work (music'score) under the terms of the GPL simply by using this analogy. By using my explanation, you would win every trial in every legal area which accepted the GPL as an effective license. That's the risk I would have to take, if I used OpenLilyLib to ease my work.

This is false.  The music score is output, not a program.  It is not a derivative work.

 

Supplement:

Andrew Bernard asked in another mail, whether it would be "my intent to shut down lilypond altogether" by pointing out this license issue. He fears the "deleterious effect of my post", that "the users and readers of this list" could be "deterred from using it". He does not want to " see OpenLilyLib because of a complicated legal misrepresentation". Finally, he declared my hint "as an attempt to damage openlilylib in people's minds" and demanded "Really, enough now."

Once for all:

1) It was my intention to be enabled to use OpenLilyLib.

2) It is not a misinterpretation, it is the established viewpoint. Another 3rd. example: Due to the same reason, MariaDB has re-licensed its connectors under the LGPL for preventing its users from the trap of the GPL licensed mysql connectors.

But this doesn't require GPL license on reports generated from the database. 

3) Mr. Bernard did not consider the difference between the interpreting program and the interpreted code. That's his fault and leads to his misunderstandings.

4) Mr. Bernard wants to protect his work (a well understandable objective) and tries to protect it be declaring his disputant (me) as an enemy with a dishonest hidden agenda. He can do so. But he - and therefore: Urs, who has asked for help - has lost a potential contributor.

5) There would be a simple solution for all, for him, for me, and for the readers of this mailing list: he simply could add an 'including exception' to his licensing statement, which clearly and explicitly says, that including OpenLilyLib does not cause the copyleft effect to the including lilypond code. He would not lose anything: each improvement to OpenLilyLib had to be released under the terms of the GPL v3, too. And - as he might signal by his words in this discussion - the using music code remains private.

Since the music code is not included in the pdf output, the pdf output is not subject to GPL.

The fact, that Mr. Bernard apparently does not want to realize point 5., puzzles me.

I remain respectfully yours

Karsten Reincke

The inability to recognize the difference between a program and the output of a program puzzles me.

Sincerely,

Carl