by Polly Bennett. Published: 30 March 2012

Standing like sentinels at the doors of CERN’s library are rows and rows of red filing cabinets. Inside each fat, red drawer lie dozens of examples of the original concept of Open Access publishing. Preprints, as yet unpublished drafts of research papers, were a way of communicating research to colleagues at collaborating labs and universities, while publishers took over a year to peer-review and finally publish the same articles. Research developed quickly in accelerator labs of the 1960s and a researcher could not wait such a length of time before reading results from other High Energy Physics (HEP) experiments. This new age of scholarly communication sped up the research process and saw the original role of academic journals, to communicate research, become de facto superseded by this alternative communication culture and ante-litteram Open Access.

CERN itself has long had a policy of open research. The original terms of the 1953 CERN Convention state that “...results...should be published or otherwise generally available,” and the impressive archive of preprints held from labs and universities worldwide illustrates one of the earliest examples. Over the last few decades, Open Access in HEP has picked up steam with the first ever online preprint platform, arXiv, which replaced the worldwide paper-based preprint system as early as 1991. This culture of innovation expands now to published journal articles and comes in the form of an international consortium, SCOAP3 (Sponsoring Consortium for Open Access Publishing in Particle Physics), aiming to convert the peer-reviewed literature of the field to Open Access.

Polly Bennett

CERN’s archive of preprints outside the library

Open Access publishing breaks free of the traditional subscription-based access to research and instead champions free online content for journal readers. Research published under Open Access terms is free from copyright and usually licensed under Creative Commons terms, providing readers with the right to share and re-use the research provided the original creator is acknowledged. As a precursor to SCOAP3, CERN and publishers of HEP journals have negotiated win-win solutions to assure that research articles from the LHC experiments are published Open Access. This adheres with the CERN Convention and, most importantly, has major advantages for the general public, who can access scientific results financed by taxpayer money, and in particular for the HEP community, who can freely re-use information.

Salvatore Mele (CERN Spokesperson for SCOAP3), Anne Gentil-Becot (CERN Library) and Travis Brooks (SLAC) have been able to measure the advantages of Open Access by studying citations to arXiv preprints and journals in the field. HEP researchers put papers on arXiv’s central server as they wait for an external publisher to organise peer-review and officially publish their work. One particular graph in their study illustrates the main benefit of this. Pointing at the screen Salvatore explains, “This graph [see below] shows you the average number of citations an article gets in the HEP field. These are articles which were also submitted to arXiv before publishing. You can see that 4 – 8 months before a paper is published it starts to be cited by other papers, which are in turn submitted to arXiv. The peak in citations is at month 1 [of its official publication]. After this it goes down as the article gets less hot or timely. So there is 1 year where research happens, based on results that are still in the process of being published. This is only possible because information is Open Access via arXiv and everybody can read it and use it. Open Access has shaved one year off the research life cycle compared to waiting for something to be published!”

Salvatore Mele, Anne Gentil-Becot, Travis Brooks

Average number of citations per article per month as a function of the time of the citation relative to the time of publication. Citations at negative times occurred while articles were in their preprint form. Citations at positive times occurred after the publication of the articles. Data is from 26741 articles from the Journal of High Energy Physics and Physical Review D over the period from 1998 to 2007.

The advantage of freely accessing and re-using research is only possible because Open Access allows authors to retain copyright. In the traditional system, publishers often require authors to transfer copyright to the publisher. As such the authors lose many rights to re-use or redistribute their own work. This stifles not only publication of core research but also wider forms of academic communication. Salvatore explains, “if a retired ALICE member wanted to write a book he might have to pay large fees for the use of a graph ‘owned’ by a publisher.”

The issue of copyright gives a particular headache to the ALICE Editorial Board, which is responsible for ensuring the quality of ALICE papers and submitting them for external publication. Due to the unique situation CERN has with many HEP publishers the issue of copyright transfer is often confusing for ALICE members submitting their final papers. Barbara Erazmus, Editorial Board co-chair says, “I asked all ALICE members not to sign copyright transfer. I receive a lot of emails from people asking what they should do, because I think it’s not very well known. The general rule is not to sign copyright so that we retain it and everybody can re-use the results. If ALICE members have any doubts they should contact me.”

The philosophy of Open Access begs the question of why the academic community needs journals at all, especially when journal subscription prices are becoming so high. The answer is clear: journals today organise the peer-review of research articles, a cornerstone of the way science and scientists are evaluated and funded today. However this has a cost. The organisation of peer-review, and other editorial, services requires publishers to charge subscription fees in order that they can recoup costs, and in some cases to make a profit. However, the fact that many researchers read HEP articles on arXiv, means that journals are increasingly de-facto selling a service to the community (peer-review), but this is being indirectly paid by the sale of content (articles) to readers.

SCOAP3 proposes to transform scientific publishing (in the field of HEP) by paying for the quality assurance service directly; Rather than having the subscriptions paid by research institutes and libraries only indirectly supporting editorial services. SCOAP3 is asking them to redirect this money into a central fund which would pay directly for the quality assurance service and wide re-use rights, while making the articles freely available.“Journals are selling information that was born free on arXiv! The basic concept of SCOAP3 is that this doesn’t make sense. Why can’t we just pay for the quality assurance service directly, with a XXIst century service-industry approach, without having this convoluted loop through a medieval content-industry concept,” Salvatore explains. “…and as a result you have as wide a use as possible of rights. The more information is available and re-usable, the more you can build on it and the more you can generate and create.”

SCOAP3 has spent the last four years talking with funding agencies and libraries worldwide; building partnerships to re-direct the funds normally used for journal subscriptions to this central fund. “The potential audience is in the order of a thousand libraries in 40 countries, but luckily in many cases a single national point does exist to aggregate a country contribution” Salvatore says. So far a partnership of over 300 institutes in 27 countries has pledged 10 million CHF, 80% of the SCOAP3 budget envelope. The other task for the SCOAP3 partnership is engaging with publishers to assess their willingness to change their business models to accommodate this (r)evolution in HEP; mostly they are.

The challenge of SCOAP3 now is to merge these two conversations into actual contractual agreements; something never tried before in scientific publishing. “In CERN style, we are building something that is the prototype of itself, and there are unknowns to be dealt with,” Salvatore says, but the SCOAP3 partners hope the system can be in operation by 2013. In his concluding remarks Salvatore adds, “With the leveraging power and infrastructure of CERN, SCOAP3 can solve a problem [for the whole HEP community] that we were already solving for ALICE and the other LHC experiments. We would like for everybody, wherever they are, to enjoy the same opportunities as CERN has been able to create for the LHC experiments.”

If any ALICE member has a question about publishing please contact the ALICE Editorial Board: Helmut Oeschler (; Barbara Erazmus (