Re: [libreplanet-discuss] GFDL with Invariant Sections or other unmodifiable parts. Was: Final Thesis: H-node

[Michał 'rysiek' Woźniak]

Dnia niedziela, 19 maja 2013 o 15:39:57 Michael Dorrington napisał(a):
> On 19/05/13 14:20, Thomas Harding wrote:
> > Le 19/05/2013 14:52, Michael Dorrington a écrit :
> [...]
> 
> >> How can we get the FSF to recognise this and so change the licence it
> >> uses for its manuals to be a free one?
> > 
> > This is one historical point of disagreement between
> > Debian(FSGuidelines)/FSF,
> > and has been highly discussed. Unfortunately...
> 
> I don't see it is a disagreement between Debian and FSF.  More of a
> disagreement between the FSF and itself, as its philosophy and practise
> disagree on manuals.  The FSF philosophy says that manuals should be
> free but the FSF practise is to distribute non-free manuals.  That's
> non-free by the FSF's own measure.
> 
> > Sometimes opinions are not flexible: you should wait for a good
> > circumstance.
> 
> Be sure to point out when the good circumstance occurs. :)
> 
> > This thread and issued document by Michal could emphasize the problem,
> > but you're pleased to find a solution ;)
> 
> I wish that Michał would post his text into the thread so we can discuss
> it.

Ask and ye shall receive:
http://rys.io/en/101.txt

WHY I FIND -ND UNNECESSARY AND HARMFUL
======================================================================

  *'UPDATE: highlighted the harmfulness of incompatibility of "no
derivatives" licenses with libre licenses (including other
CC-licenses); heartfelt thanks to Carlos Solís[1] for the Spanish
translation[2]. ¡Gracias! *

  There are two basic arguments for licensing some works under a "no
derivatives" license (e.g. any -ND Creative Commons license[3], or the
GNU Verbatim Copying License[4]):
 - some authors do not wish for their works to be modified, twisted,
used in ways they do not approve of;
 - some works (for example, expressing somebody's opinion) are
fundamentally different from other kinds of works and should remain
invariant.

  I believe both are specious. And I feel "no derivatives" licenses
are both ineffective and counter-productive. Here's why.

"I DON'T WANT MY WORK TWISTED!"
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 
   So you're an author and you do not wish your work to be twisted or
modified to say something you didn't want to say. There are two
possibilities of such modification:
 - somebody takes your work, twists it and publishes it under your
name, suggesting you wrote this;
 - somebody modifies your work and publishes it under their own name,
as a derivative work.

  /The first possibility would be illegal regardless of the license!
/Nobody has the right to claim your authorship over something you did
not create; nobody has the right to modify your work and claim it is
still your work. -ND licenses are unnecessary for that purpose, it's
already in the copyright law.

  As far as derivative works that build upon the original but change
the meaning are concerned (/without misrepresentation of authorship/),
I do not feel we need licensing restrictions for that. That feels too
close to censorship for my liking -- /"thou shall not use my own words
against me"/; /"I don't like what you're trying to say so I will use
copyright law to stiffle your speech"/.

  Besides, creating parodies is fair use and no amount of "no
derivatives" licensing clauses willl change that. Same goes for
quotes. Your words /will/ be used in works that say something you do
not wish to say, whether you like it or not!

  In that sense, "no derivatives" licenses are ineffective.

"SOME WORKS SHOULD BE INVARIANT!"
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 
   This argument hinges upon an assumption that some works (memoirs,
documentation, opinion pieces) are fundamentally different than others
and hence should be preserved as they are.

  First of all, all of what I wrote above applies here. Such works
cannot be "modified" anyway, "modification" is in fact creation of a
derivative work, nobody can (legally) misrepresent authorship or the
derived work as the original; such works, also, can be quoted and
parodied, regardless of the license. "No derivatives" is ineffective.

  /"No derivatives" licensing stops people from doing things most of
us would say are desirable/. Like /improving/ upon some work, creating
better arguments or updated versions. Like /translating/ into another
language to disseminate knowledge and argumentation. These things are
genuinely /good/, but people that would like to do them /will/ look at
the license, and will then get the message they cannot proceed...

  More importantly, however, this argument assumes that there is only
one context in which a given work can be used. E.g. "an essay on free
software" as an article to read and get argumentation from. Or a
"memoir" as a historical document describing one's views and furtune.

  Thing is, any work can be used in any context, and often is.

  Think for a moment about a teacher in an IT class using an essay on
free software as educational material, modifying it just a bit so that
the class can better understand it or relate to it, or using it as a
basis for an in-class discussion. "No derivatives" would not allow for
that.

  Think about how artists used different kinds of "materials" for
their works of art -- for example Duchamp's "Fountain"[5]. A "memoir"
or an "opinion piece" could easily be used in an "art" context, for
example as a basis of vocabulary for some free-software scripted (as
in: done by scripts in interpreted programming languages) poem writing
contest. An example of something similar is HaikuLeaks[6].

  I bet we could find similar haikus in GNU documentation, FSF policy
papers, Linux documentation. "No derivatives" would not allow anybody
to use these in such a context -- and I would say that's a genuine
loss.

  And in that sense "no derivatives" licenses are counter-productive.

MUDDYING THE WATERS
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 
   "No derivatives" licensing is also truly harmful.

  They make it harder to explain what libre licensing is. Many people
believe that any CC or GNU license is "free as in freedom", while in
fact CC-*-ND and GNU Verbatim licenses cannot be considered as such.
This distinction is both crucial and hard to convey.

  They also cause segmentation within the CC/GNU-licensed group of
works: some such works (often co-existing on a single OS or in a
single[7] repository[8]) are licensed in a way that makes them
incompatible with others. Some cannot be modified or used (namely, "no
derivatives"-licensed) in new works, while others can.

  This muddies the waters, makes libre licensing this much harder to
explain and this much harder to take advantage of.

TL;DR
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 

 - "No derivatives" licensing does not protect against things we want
it to protect against (either because these are explicitly allowed for
by the copyright law, or because the copyright law already disallows
them);
 - and at the same time stops people from doing things we might
consider interesting or beneficial;
 - while making it harder to promote libre licensing and
libre-licensed works.


Links:
------
[1] http://azkware.ticoblogger.com
[2] http://rys.io/es/101
[3] http://creativecommons.org/licenses/by-nd/3.0/legalcode
[4] http://www.gnu.org/licenses/license-list.html#GNUVerbatim
[5] http://en.wikipedia.org/wiki/Fountain_%28Duchamp%29
[6] http://haikuleaks.tetalab.org/
[7] http://www.flickr.com/search/advanced/
[8] http://www.jamendo.com/en/search

-- 
Pozdrawiam
Michał "rysiek" Woźniak

Fundacja Wolnego i Otwartego Oprogramowania

signature.asc
Description: This is a digitally signed message part.