April 25, 2016
While more than half of open access papers are published under the terms of a liberal Creative Commons Attribution Licence, the majority of authors of open access works seem not to accept the terms of either this or any other Creative Commons license.
Despite the fact that the majority of journals indexed in the Directory of Open Access Journals use liberal Creative Commons Attribution (CC-BY) licence as a default, and that probably more than half of all articles published in open access serials are published under the terms of this licence, academic authors seem not to support liberal licensing. How is it possible? Are authors of more than 600 thousand CC-BY licensed works invisible in surveys? Or do they publish under the terms of this license against their will?
Let’s have a closer look at this problem using the data from the Key Challenges of Research Communication De Gruyter Open Author’s Survey. According to our research only 6.7% of academic authors claim that its OK for them to have their work “republished by a commercial company without their approval”. This is of course allowed under the terms of the CC-BY licence, which allows everybody to republish the work for any purpose. This suggests that 94.3% of academic authors do not accept the terms of this licence. Of course, CC-BY requires attribution of the original work, which was not highlighted in our survey question. This might be seen as a reason why we obtained smaller share of positive answers than Taylor and Francis in 2014. In a survey by T&F, 19% of authors accepted their works to be “used for commercial gain”. However, in both cases the vast majority of authors did not accept commercial use without their authorisation.
Where are the authors of more than 600K open access papers (licensed under CC-BY licenses) in these two researches? Well, we still do not know. Only 7.5% of authors of gold open access papers, books or chapters taking part in our survey accept commercial re-use. This is still only a small minority of gold open access authors, and it is hard to accept that this minority publish the majority of all open access works.
There is also no significant disciplinary difference here; number of publications, career level and type of contract do not seem to influence the likelihood of acceptance of commercial reuse without authorization. It is hard to say anything for sure for such a small fractions but authors for peripheral countries are a little bit more friendly toward the concept of commercial re-use.
Another important issue that we have examined in the survey is acceptance for the reader’s right to translate the work without the author’s approval. Again, the majority of authors do not like this idea. Only 28.3% of respondents agreed to grant such a permission to their readers. Of course, this permission is granted when the work is published under the terms of CC-BY licence. Experience with gold open access has no visible impact on this attitude.
Researchers living in the peripheral countries more often agree to give this permission to all their readers.
Early Career Researchers are also more friendly towards the concept of translation without approval than others (35.8 % accept that), especially than established researchers (only 25.6% respondents from this group would like to grant this right to their readers). Students (graduate and undergraduate) are in the middle (30.4% accepting).
STEM researchers are more likely to grant their readers permission to translate their works without their approval than their HSS colleagues. This is quite surprising, since humanists write in regional languages much more often, and therefore can benefit more from translations than STEM researchers.
Authors who published more papers than their average disciplinary colleagues in the last 3 years are more eager to protect their text from unauthorised translations (23% of them agree to grant the rights to translate to all readers) than others (31.1%).
Including a text in an anthology
Republishing academic work (a book chapter or an article) as a part of an anthology of academic texts is allowed by any Creative Commons licence (CC-BY-NC exclude re-publishing by a commercial company or for commercial purposes only). And since CC licenses are the standard of open access academic publishing, it is quite surprising that only 29.1% of authors accept republication of their works. In the last research by Taylor and Francis, 39% of authors expressed a positive attitude toward this concept.
Our research has also shown that only 30.9% of authors who published gold open access articles themselves accept possible republication in an academic anthology. This is in contradiction with the fact that probably a vast majority of them had published their works under the terms of a Creative Commons license and their open access works actually might be re-published.
Authors living in the core countries are less likely to accept republication without their approval (19.1% of authors accept it in the core countries versus 37.7% in peripheral countries). Disciplinary differences show similar patterns than in case of translations. Other factors do not seem to influence this issue.
Text and data mining
The last issue concerning licensing that was analysed in our research is so called ‘text and data mining’. Text and data mining is allowed by the CC-BY licence, and now it is treated as a copyright exception in the UK and USA, so in these countries it is always legal to mine text and data. However this research technique is still not popular and as you will see, not accepted.
We asked our respondents “Do you agree to grant your readers the right to extract data from the text of your work by automatic software without your approval”. Only 30.9% of them agreed. Taylor and Francis get 53% of positive answers regarding this issue, however there were several differences in the formulation of the question. First of all, in T&F, survey respondents were asked to answer on a 1 to 5 scale, which in my opinion is less appropriate in this case. When a researcher is asked to sign a copyright agreement with a publisher (s)he accept sthe contract or not, there is no middle ground here. And in the T&F survey only 22% of researchers answered the question about text and data mining with a ‘5’ which meant “Strongly agree”, while 31% chose the answer ‘4’. What is more, T&F simply asked if researchers “agree to use his/her work in text or data mining”, which, in my opinion, might be less clear for authors who are not familiar with this term.
Anyway, this is a surprisingly low outcome in both cases since it is hard to understand what researchers may lose when their works are being mined by other scholars (or robots). Researchers who published their own work in open access seem to accept text and data mining more often than those who don’t (32.7% versus 25.8%). Authors from peripheral countries are also more liberal here. Other factors seem not to influence attitudes toward text and data mining.
Again, it turns out that the main factor that influences attitudes of academic authors is geographical location. It seems possible to me that some authors from outside of the leading countries see liberal licenses as a chance for wider dissemination of their work. This is important to them, probably because they are looking for recognition in core countries.
It remains a mystery how more than half of open access papers are published under the terms of a CC-BY licence, while at the same time, the majority of authors of open access works seem not to accept the terms of either this, nor any other Creative Commons licenses.
The possible solution of this conundrum is that academic authors probably do not care about licensing too much. Finding a good place to publish their work is complicated enough. They have to balance Impact Factor and prestige of the venue with the chances of getting a particular paper published, and they simply do not want to make the whole decision process harder for them. However, it should be made clear that, when somebody convinces them to pay more attention to licensing the results might be quite surprising, with the majority of authors choosing very restrictive solutions.