I have been thinking about Open Access (and Open Data) since I ran into misunderstandings about what is an Open Access journal. As a result I have asked for clarification on one of the most prominent OA mailing lists and have received considerable information and help both publicly and privately. As a result I'll try and share my thoughts.
I'm not going to cover this comprehensively - for example Bill Hooker has done an excellent job in one of his blogs. But I'll try to clarify some of the things I have found difficult. And I'll try to be objective.
A word of warning. Although the basic idea of OA is simple, its practice is more complicated. There are also confusing terms: OAI (Open Archive Initiative) is nothing to do with Open Access (and OAIS has nothing to do with either). The various declarations (Bethesda, Budapest and Berlin) have very similar acronyms - BOAI - (but at least are very similar and sometimes referred to as BBB).
On the positive side OA is happening and is unstoppable. So anything critical I say should be viewed in the greater light. The movement has now enough momentum that it can tolerate robust discussion. But it is clear that although there is a generally shared vision there are many reasons why people want it and this sometimes causes tensions.
Firstly to remind people of the motivation:
- Access. I belong to the eMinerals project funded by the Natural Environment Research Council. NERC is "...tackling the 21st century's major environmental issues such as climate change, biodiversity and natural hazards."). The project (run by Martin Dove) comprises several UK universities and research establishments. Towards the end of its first phase we thought it was a good idea to publish different aspects of the work in a single issue of a journal, and here it is: the complete journal issue. I hope you liked reading the paper that I was a co-author of (CML tools and information flow in atomic scale simulations). Well if you did you were luckier than me and the rest of the project because none of us could! What? The University of Cambridge? One of the UK deposit libraries couldn't read a journal? Well, yes - I could probably have gone to our splendid Giles Gilbert Scott building and tried to find a physical copy (but it could have been somewhere else and I simply wanted to skim through the issue, not an adventure). So, yes, the University of Cambridge does not have an online subscription to Molecular Simulation. Can the authors put a copy of the final manuscript on the project page? Yes! "You are able to post, after a 12-month embargo (STM) or 18-month embargo (SSH), your revised text version of the final article after editing and peer review on your home page". (see Copyright Transfer FAQs - this is packed with interesting information which I may revisit in later posts). So, I've waited 18 months and now I can read about the projects.
- Permission. I have already blogged about how publishers' permissions restrict modern usage to published papers. Despite the BOAI, publishers actively frustrate the text-mining (e.g cut off by publisher) and robotic indexing of papers (see earlier post where I argue that licenses restrict legitimate re-use).
So the two issues I care about are access and permissions. I'll come back to these later and try to portray some of the views that OA proponents have.
I'll finish with the slightly provocative statement that Open Access is still a Movement rather than a Code of practice. The declarations urge general motivation rather than precise actions that should be carried out. A declaration of rights might say that all humans should be treated equally, but may not spell out how this is to be done or whether certain actions conform. Similarly some body can say it is an "Open Access" funder/publisher/repositoryProvider, etc. but is it really? Without agreement on the practice of OA we can't tell. And, as far as I can tell, there isn't currently any organization that decides.
Does this matter? Yes - it does now. Many funders are requiring publication as "Open Access". Do authors and publishers conform? I don't know. And I'm specifically worried about the permissions aspect of OA - I believe that mandates must require that re-use is supported as strongly as access - that is why I have campaigned for Open Data.
For me, if my robots cannot read the articles then as a human I have no interest at all in reading the "fulltext". And if I cannot guarantee that they can do this without publishers shutting off the supply of journals then OA will not have delivered.