“If we are paying for this research, aren’t we entitled to scrutinise the results?”
That’s the call-to-arms I’m hearing at the moment, and I thought (as a research physicist and taxpayer) it might be helpful to point out a few seemingly overlooked aspects to add to the discussion.
The latest episode in the ongoing saga is a deal brokered by the Sponsoring Consortium for Open Access Publishing in Particle Physics (SCOAP³) to make all particle physics papers open access.
The deal means journal subscription fees paid by university libraries would be directed towards 12 selected journals in return for public access to those journals. SCOAP³ claims the deal will result in more than 90% of particle physics papers being available for everyone.
An immediate query I have is whether this “90%” conclusion takes into account any push to publish in these (and only these) open-access journals.
Pushed to publish?
Currently, university libraries pay subscription fees to allow their researchers and students access to useful journals. Neglecting for a moment the obvious possibility that libraries simply cut their funding of journals and receive the free service that everyone else will, if universities are paying a fee towards these journals, they may “gently encourage” their researchers to target these journals, interfering with the established metric used in grant applications (impact factors).
I currently do my best to publish my articles in journals where I consider the relevant “conversation” to be occurring: usually where a substantial body of work already exists on my topic. While I would like think this is universally true, I am well aware many authors publish in the highest-ranking journals they can, regardless of the conversations already in place.
A paper in Physical Review Letters is highly prized and well regarded, but the journal did not make the cut for open access because it charged too high a price per article.
Something for nothing?
It’s worth highlighting that the SCOAP³ deal is not a case of “free and open access” - the university libraries are still paying for access, but under this deal that access would be for everyone.
I should also point out that public libraries in the US already benefit from free access to many journals. In the case of the American Physical Society (APS), which publishes some of the most widely read articles in particle physics – Physical Review Letters, Physical Review, and Reviews of Modern Physics – public library journal access has been free since 2010, with the hope of extending this model to other countries in the future.
The APS already allows researchers to publish their papers as barrier-free open-access under a Creative Commons Attribution 3.0 License (CC-BY), but for a substantial article-processing charge. It is this charge that the SCOAP³ will essentially be paying on the behalf of particle physicists, but is this even necessary?
The APS Copyright policy seems to answer this point quite clearly for me:
The author is permitted to provide, for research purposes and as long as a fee is not charged, a PDF copy of his/her article using either the APS-prepared version or the author prepared version.
This shouldn’t be much of a surprise - the content of a published paper is available to the author (be it formatted in the published style, pre-print style, or in plain text) and the APS could never prevent researchers from distributing this. If you are a researcher, and you don’t have (but need) access to one of my published papers … just ask me.
While SCOAP³ is pushing for “free” open-access to particle physics journals, we physicists already have something along these lines available. The online pre-print server arXiv, which is hosted by Cornell University library, hosts papers not just from particle physics, but from a range of physics sub-branches. The site is a well-known resource among physicists, because it provides an immediate, free source of papers.
The vast majority of high-energy physicists submit a copy of their manuscript to the arXiv prior to publication. It is here that I consider the real “peer-review” to occur.
The arXiv provides a mailing service that sends out daily updates of new additions in a chosen field. Some of my work is in nuclear theory physics, so I subscribe to a mailing list that alerts me to the title, authors, and abstract of every new addition to the “nucl-th” section of archived pre-prints. Each day I can catch up on what is in the process of publication in that field. Importantly, I can do this away from the university - at home, or abroad - and so can the public.
Similarly, someone interested in my work will see my papers appear as I submit them. There is a also an online listing, and a search facility.
The papers submitted and distributed at arXiv are essentially the draft versions of what will hopefully be published papers. Other researchers in the field can inspect the paper, consider the method and findings presented, and even contact the author if they spot something possibly missing or erroneous. Where it differs from conventional “peer-review” is that there is no control over whether an author makes any revisions based on feedback recieved.
A good article should, but does not always, describe the problem you sought to solve, your method, and your findings. Ideally, your method will include all the steps not immediately obvious to an outsider such that they could logically and reproducibly arrive at your findings independently.
Your findings will ideally include any caveats and assumptions you’ve made in reaching them, and comments on their applicability to your problem. I presume that this reproducibility is the driving force behind the “openness” issue.
Many of the papers on arXiv will not progress further for a variety of reasons: perhaps the findings were found to have a logical error, a typo in the program, or a later-realised poor working assumption. Perhaps the author never got around to submitting for final publication with a journal, or perhaps the paper was rejected by the journal.
This means the content hosted on arXiv doesn’t necessary reflect the articles published in the peer-reviewed literature (to be fair, sometimes even the published literature doesn’t reflect reality). What it does do, though, is provide a listing of researchers involved in a current or past field, and hints towards their contributions.
Right idea, wrong field?
So, at least in many areas of physics, open access to published findings is still an available option. If you want to check the validity of a statement, given a well-written paper hosted on arXiv, it should be possible to do that.
There’s a lot to be said for open-access, and I think it’s a great thing to have research available to anyone who wants it. It would be great if other fields would adopt an arXiv-like model of pre-publication.
For the general public interested in research, I find it interesting there would be a push towards a further opening up of physics research, and not more obvious fields such as medical research.
I would love to have media sources cite the research articles they cover, and have those articles available to scrutinise. This would of course reduce the impact of headlines reading “Y cures Z” if, with just a click, you could see that, “Y mildly correlates with Z (r=0.35), in mice (n=6), under lab-conditions”.
While the motivation for open-access in the case of SCOAP³ and particle physics is well-meaning, it may be the wrong approach, and the wrong field.
As with all articles on The Conversation, this article is published under a Creative Commons license, meaning you can share or even republish it, in its entirety, as long as you follow a few simple rules.