Version 1.1 of the definition has been released. Please help updating it, contribute translations, and help us with the design of logos and buttons to identify free cultural works and licenses!

Elements of journal freedom: Difference between revisions

From Definition of Free Cultural Works
Jump to navigation Jump to search
(→‎Only BY and SA are free and open: Additionally, because the non-commercial provision obfuscates intent, using such a license increases implicit legal costs for everyone, except perhaps for giant for-profit companies which thrive on copyright confusion.)
(→‎References: +Farge)
Line 62: Line 62:

* [ Diamond OA]
* [ Diamond OA]
* Marie Farge and Jean Gasnault, ''[ Open Science, Open Doctrine, How to Share Knowledge?]'', 2019. doi:[ 10.3233/FAIA190013]


Revision as of 13:49, 14 September 2021

What does an academic journal need to consider if researchers want to keep their freedoms? This essay attempts to dispel some myths about copyright and corporate power in open access and open science.

In short:

  • in our society, there is no way to eliminate (the right to) commercial usage and profit (or otherwise cancel the power of money): they are granted automatically and cannot be destroyed,[1] so you can only fragment and distribute them as much as possible or needed;
  • there are also few legal and technical ways to ensure your work is "owned" forever by its scientific community (i.e. that it stays in the commons), while there are many ways it can become private property of someone to the exclusion of others.

False promises of non-commercial clauses

Some adopt NC restrictions hoping they harm the big evil publishers. The opposite is true. To make the rich publishers powerless, you have to ensure that their money cannot buy anything; failing that, ensure that if they buy something they cannot have a financial return on it (turn the market and their shareholders against them).

"Non-commercial" is a misleading name for a public licenses module. A descriptive name would be "commercial monopoly" (of the rightsholder): the author rights for commercial usage are not given away but reserved to the original owner, who can still sell them to others.

What problems arise if a journal carries a "non-commercial" clause?

  • If you run a journal, you don't know who could at some point use it commercially. Maybe it's nobody, not even you. Maybe it's a small set of authors. Maybe it's some entity out there.
  • As long as no public commercial license exists, somebody can always buy out commercial rights from the authors: maybe some publisher or entity is already doing it secretly or surreptitiously, since so many copyright contracts are secret. Or at some point somebody will make an offer which can't be refused and will end up having more rights on the journal than you have.

Buying out exclusive rights on a number of small titles, previously thought to be worthless, is exactly how the big publishers became the powers they are now and how the serials crisis came to be in the 1970s. Only pre-emptive neutralisation of such rights can ensure that they don't end up in the wrong hands and that they don't become a weapon against the scholarly community.

What are some things to consider when selecting a license for a paper or journal? Some links follow.

  • "Commercial" probably doesn't mean what you hope. Make sure you're not accidentally allowing something you dislike while disallowing something you like. You need to check applicable laws and case law at least in the jurisdictions which are most interesting to you. If you don't have time to do this, don't use a non-commercial license.
  • It makes sense to keep your commercial rights only if you plan to actually use them (e.g. by selling printed copies) and sue whoever infringes them (e.g. by posting them on an ads-ridden website or subscription-based database). You think that if Elsevier uses your NC paper you'll bring them to court with some top lawyers and claim hefty payments for their economical damages to you? Then fine, use NC.
  • If you use a permissive license like CC-BY, you're ensuring that, as far as copyright goes, there will always be perfect competition in distributing your work: nobody can gain monopoly on it and use high market power to slap an high price on it.
  • If you use a copyleft license like CC-BY-SA, any investment by deep-pocketed publishers to use your work will need to be turned back to you and the commons as a publicly shared output under the same license. Elsevier wants to pay someone to wrap your journal in a fancier layout? How generous of them, it means we get at least the resulting PDF for free under the same license.

Only BY and SA are free and open

In the Creative Commons toolset, only the "Attribution" and "Share-Alike" clauses qualify for the freedom definition and open definition.

Unfree licenses such as NC and ND (non-commercial and no-derivatives) alienate anyone who believes in free knowledge or open content: because they cannot know whether at some point the journal will become somebody's exclusive property, it makes no sense for well-meaning hackers and technologists to provide technical support. You wouldn't drop a gold bar at the front door of a billionaire's office hoping that a good person picks it up, would you?

Additionally, because the non-commercial provision obfuscates intent, using such a license increases implicit legal costs for everyone, except perhaps for giant for-profit companies which thrive on copyright confusion.

Trademarks and other

Beyond copyrights, many systems are available to gain monopoly or market power, which can be used against an OA journal. Intellectual property is a controversial name for systems which allow anything immaterial to be bought and excluded from the use of others. This is a poorly researched area of OA journals (business) models.

Some suggestions:

  • Follow the EFF advice on patents: public interest patent pledge.
  • Keep in mind that most journals have a generic name which cannot be trademarked [citation needed].
    • If your name is not a generic name yet, you might manage to make it one, by doing the opposite of what every trademark lawyer suggests to those who want to keep their trademark. However, there is no certainty of success.
  • Ensure any logo or name is public property, or better nobody's property (example).

Trademarks are sometimes considered a possible tool for a specific kind of business model for open-source software, although this is controversial as of 2020. It might be possible to devise a way for a trademark to be both valuable and innocuous in the long-term for the scholarly community around an academic journal, but that requires at a minimum:

  1. a significant initial investment in legal services to make sure that the trademark does what you want (probably tens or hundreds of thousands of dollars);
  2. the existence and perennial upkeep of some kind of entity to which the trademark can be assigned and which can be trusted to forever be better than any other possible entity or person out there which might want to use the trademark.

We have learnt from history that any incorporated entity is at risk of being captured by forces hostile to the academic community, and when it comes to scholarly communication that's a near certainty. Once again, therefore, the only solution might be a creative way to fragment the property of the trademark, possibly by splitting it among thousands of individuals, for instance scholars in advanced age whose property will then be further fragmented by inheritance after their death. However, trademarks are subject to a number of requirements, including actual use, and no example of such public property of a trademark exists yet.


The content of a journal or project may be free and open, but if it relies on proprietary software and otherwise exclusive technology someone may still be able to exercise unwanted influence on it. Always ensure:


  1. As Pam Chestek put it This Is Why You Don't Call It "Intellectual Property", hoping that no new exclusive rights will not be created is like hoping that the sun won't rise tomorrow. You can write it in a contract too, but it won't help much.