less_retarded_wiki/copyleft.md
2021-12-11 16:08:32 -06:00

4.7 KiB

Copyleft

Copyleft (also share-alike) is a concept of sharing something on the condition that others will share it under the same terms; this is practically always used by a subset of free (as in freedom) software to legally ensure this software and its modifications will always remain free. This kind of hacks copyright to de-facto remove copyright by its own power.

Copyleft has been by its mechanisms likened to a virus because once it is applied to certain software, it "infects" it and will force its conditions on any descendants of that software, i.e. it will spread itself (in this case the word virus does not bear a negative connotation, at least to some, they see it as a good virus).

For free/open-source software the alternative to copyleft is so called permissive licensing which (same as with copyleft) grants all the necessary freedom rights, but does NOT require modified versions to grant these rights as well. This allows free software being forked and developed into proprietary software and is what copyleft proponents criticize.

In the FOSS world there is a huge battle between the copyleft camp and permissive camp (LRS advocates permissive licenses with a preference for 100% public domain).

Issues With Copyleft

In the great debate of copyleft vs permissive free-licenses we, as technological anarchists, stand on the permissive side. Here are some reasons for why we reject copyleft:

  • Copyleft can force harmful conditions as long as they don't violate software freedom. For example just as there may be an obligation of crediting the author, there may as well be an obligation to keep included some text that promotes some unethical political ideology and so everyone who wants to execute the basic freedoms has to promote this politics with which he may not necessarily agree.
  • By adopting copyleft one is embracing and supporting the copyright laws ("marrying the lawyers") because copyleft relies on and uses copyright laws. Copyleft chooses to play along with the capitalist bullshit intellectual property game and threatens to use force and bullying in order to enforce correct usage of information.
  • In a way it is bloat. Copyleft introduces legal complexity, friction and takes programmers' head space, especially when copyleft is probably mostly ineffective as detecting its violation and actual legal enforcement is difficult, expensive and without a guaranteed positive outcome (FSF encourages programmers to hand over their copyright to them so they can defend their programs which just confirms existence and relevance of this issue). Sure, corporations can probably "abuse" permissive (non-copyleft) software easier, but we argue that this is a problem whose roots lie in the broken basic principles of our society (capitalism) and so the issue should be addressed by improving our socioeconomic system rather than by bullshit legal techniques that just imperfectly cure the symptoms.
  • The scope of copyleft is highly debatable (which is why we have different kind of copyleft such as strong, weak, network etc.). I.e. it can't be objectively said what exactly should classify as violation of copyleft AND increasing copyleft scope leads to copylefted software being practically unusable. Consider this example: Linux is copylefted which means we can't create a proprietary version of Linux, nevertheless we can create a proprietary operating system of which Linux is part (e.g. Android in which its proprietary app store makes it de-facto owned by Google), and so Linux is effectively used as a part of proprietary software -- the copyleft is bypassed. One might try to increase the copyleft scope here by saying "everything Linux ever touches has to be free software" which would however render Linux unusable on practically any computer as most computers contain at least some small proprietary software and hardware. The restriction would be too great.
  • In practice, copyleft licenses have to be complex and ugly because they have to strictly describe the copyleft scope and include lots of legal boilerplate in order to make them well defendable in court -- and as we know, complexity comes with bugs, vulnerabilities and other burden. Indeed, we see this in practice: the only practically used copyleft licenses are the various versions of GPL of which all are ugly and have historically shown many faults (which is again evident from e.g. looking at GPL v1 vs v2 vs v3). Permissive licenses on the other hand are simple, clear and well understandable.