Against the Grain

Volume 23 | Issue 2 Article 9

April 2011 Open Access Self-Archiving of Refereed Research: A Post-Gutenberg Compromise Stevan Harnad University of Southhampton, [email protected]

Follow this and additional works at: https://docs.lib.purdue.edu/atg Part of the Library and Information Science Commons

Recommended Citation Harnad, Stevan (2011) "Open Access Self-Archiving of Refereed Research: A Post-Gutenberg Compromise," Against the Grain: Vol. 23: Iss. 2, Article 9. DOI: https://doi.org/10.7771/2380-176X.5786

This document has been made available through Purdue e-Pubs, a service of the Purdue University Libraries. Please contact [email protected] for additional information. However, many publishers either do not make our collective understanding of the type of de- Journal Article Versioning is … any changes to the version of record or display mands from scholarly readers and practitioners from page 20 non-standard indicators when such changes oc- for article versioning standards. cur. SAGE is prepared to contribute to shared practices to release material of the highest pos- efforts toward clear and acceptable practices sible quality, published within known patterns for iterations beyond the version of record. Endnotes and bearing standard mechanisms of cataloging SAGE is prepared to launch another wave of 1. The summer 2010 survey was conducted and archiving, such as ISSNs and DOIs. Jour- production and platform enhancements to our by NISO across representatives of key nal article version metadata are now a facet of journals publishing program that allow clear groups — repository managers, librarians, publishers’ responsibilities in disseminating indications of changes to an article’s version and journal publishers and editors — to learn scholarly material online. And, while SAGE of record. We look forward to partnering with more about their interest and engagement cannot satisfy every researcher and every in journal article version terms and related other members of the scholarly community to practices. The objective was assessment member of its community, SAGE is taking a examine the conceptual and logistical impli- of the scholarly community’s investment significant step toward an industry-wide solu- cations of this change within to all aspects of in new routines to steward online article tion for standard versioning practices. our industry. versions at every stage of public distribu- The next major hurdle in the evolution The 2010 NISO study supports this need tion. Specifically, the focus was on the of journal article versioning is industry ac- for more discussion and awareness on these uptake of terminology recommended by the NISO/ALPSP Journal Article Versions ceptance of post-publication corrections and topics that will bring us closer to versioning (JAV) Technical Working Group, www.niso. enhanced versions of record. Many recom- standardization. Today, there exists a troubling org/publications/rp/RP-8-2008.pdf, in 2008. mended standards, such as NISO’s JAV terms, disconnect between the needs of scholarly This recommended practice is managed by incorporate support for any iterations following researchers and the obstacles encountered by the NISO Content and Collection Manage- what was known in the print-only world as those in publishing and dissemination roles. ment (CCM) Topic Committee. the “final” issue version or version of record. Further research of this sort is needed to expand

Open Access Self-Archiving of Refereed Research: A Post-Gutenberg Compromise by Stevan Harnad (Canada Research Chair in Cognitive Sciences, Université du Québec à Montréal & School of Electronics and Computer Science, University of Southampton) http://openaccess.eprints.org

f asked what would be ideal for their re- research published annually; it also means these days is exceedingly light1) has corrected search access needs, most scientific and that the authors of all those annual articles anything substantive, the author can update the Ischolarly researchers would say that they only have access to a fraction of their potential final draft to incorporate that too. would wish to have access to every piece of users worldwide. Access, usage, impact, and Author self-archiving is called “Green research relevant to their own work, rigorously research progress are being lost, annually, be- Open Access” (Green OA). The majority of peer-reviewed, conscientiously copyedited, cause access falls short of being universal. journals today (and almost all the top journals) and elegantly formatted, online and on paper, A solution has existed ever since the onset have already given their official green light to as soon as it is ready for publication. (In some of the Post-Gutenberg (online) era (Oker- immediate author self-archiving of their final fields — e.g., high-energy physics — research- son and O’Donnell 1995). The solution is drafts. For the minority of articles published ers also want access to research before it is known, and it is (belatedly) beginning to be in the journals that do not yet endorse Green peer-reviewed, but so far this is the exception implemented: authors can make their OA, the final draft can and should rather than the rule.) Moreover, because in peer-reviewed research accessible be deposited in the author’s insti- most fields the research users and the research free for all online by self-archiving tutional repository immediately authors are the same population, wearing dif- their peer-reviewed final drafts in upon acceptance for publica- ferent hats, what is ideal for the user is also their institutional repository imme- tion in any case. If the author ideal for the author: researchers conduct and diately upon acceptance for pub- wishes to observe a journal’s publish research so it can be accessed, used, lication, and their institutions embargo on OA, access to applied, and built upon by other researchers and funders can mandate such the deposit can be set as in further ongoing research. The progress self-archiving (Harnad et “Closed Access” rather than and funding of their scholarly work — not to al. 2003). The author’s self- “Open Access” during the mention their careers and salaries — depend archived final draft is not the embargo. The bibliographic on the uptake and impact of their research publisher’s version of record metadata (author, title, jour- findings. Hence the broader and earlier the — it is peer-reviewed, but it nal, abstract, etc.) of Closed access to their findings, the better for authors is not copyedited nor in the Access deposits are immedi- (Gargouri et al. 2010). publisher’s final format. So ately visible to all, webwide, So much for ideals. Now, what is the real- the solution is a compromise; and the institutional reposito- ity? There are about 25,000 peer-reviewed but it is a compromise that ries can implement an “eprint scholarly and scientific journals, across all is incomparably better than request” button that allows disciplines, nations, and languages, publish- the status quo. It means that would-be users to request and authors ing about 2.5 million articles per year. No refereed research findings are to provide a single copy for research purposes university or research institution in the world immediately available to all potential users, not (Sale et al. 2010). This too is a compromise: can afford to subscribe to all, most, or even just to the fraction that are at subscribing insti- it is not OA; it is Almost-OA. many of those 25,000 journals; most can only tutions. The published version’s formatting is But universal Green OA self-archiving afford to subscribe to a small fraction of them. of no importance to the many would-be users mandates, adopted by universities, research That means that most researchers worldwide who would otherwise have no access at all; institutions, and research funders worldwide only have access to a small fraction of the and if the copyediting (which for most journals continued on page 24

22 Against the Grain / April 2011 Unlike with OA’s primary target, journal Archivangelism (http://openaccess.eprints. Open Access Self-Archiving ... articles, the deposit of the full-texts of books org/index.php?/archives/467-Open-Access- from page 22 in Open Access Repositories (http://roar. Book-Impact-and-Demotic-Metrics.html). eprints.org/) cannot be mandated (http:// Friday, October 10, 2008. will not only remedy the research access www.eprints.org/openaccess/policysignup/), Harnad, S. (2009) The PostGutenberg problem immediately, but it may eventually only encouraged. However, the deposit of Open Access Journal (http://eprints.ecs.soton. lead to an even better solution, and the natural book metadata + plus + reference-lists can ac.uk/15617/). In: Cope, B., and Phillips, A. one for the online era: once the final drafts of and should be mandated by universities and (Eds.) The Future of the Academic Journal. all refereed research articles are being self-ar- funders. That will create the metric that the Chandos. chived and hence freely accessible to all users book-based disciplines need most: a book online, institutions may well decide that they citation index. Thompson-Reuters Web of Harnad, S. (2010) No-Fault Peer Review no longer need to subscribe to the journals in Science only covers citations of books by Charges: The Price of Selectivity Need Not which they are published. Cancelation pressure (indexed) journal articles, but book-based Be Access Denied or Delayed (http://eprints. will induce journals to cut costs by elminating disciplines’ biggest need is book-to-book ecs.soton.ac.uk/21348/). D-Lib Magazine 16 obsolete products and services, beginning with citations. Citebase (http://citebase.eprints. (7/8). the print edition, and then the online edition. org/) could provide that, once the book ref- Harnad, S. (2011) Gold Open Access All production, access-provision, and archiving erence metadata are being deposited in their Publishing Must Not Be Allowed to Retard the will be offloaded onto the network of insti- authors’ institutional repositories too, rather Progress of Green Open Access Self-Archiving tutional repositories. The author’s refereed, than just journal articles. (Google Books (http://eprints.ecs.soton.ac.uk/21818/). Logos revised, accepted final draft, self-archived in and Google Scholar are already providing a (in press) his institutional repository, will become the first approximation to a book citation count.) Harnad, S., Brody, T., Vallieres, F., Carr, version of record, and the only service still Analogues of “download” metrics for books L., Hitchcock, S., Gingras, Y., Oppenheim, provided by the journal publisher will be peer are also potentially obtainable from book C., Stamerjohanns, H., and Hilf, E. (2004) review (and possibly some copyediting). vendors, beginning with Amazon Sales Rank The green and the gold roads to Open Access The true cost of peer review alone, per (http://www.rampant-books.com/mgt_ama- (http://www.nature.com/nature/focus/access- article, is only a fraction of what is being paid zon_sales_rank.htm). In the humanities it debate/21.html). Nature Web Focus. per article by institutional subscriptions today. also matters for credit and impact how much Harnad, S., Carr, L., Brody, T., and Op- Institutions will easily be able to cover the the nonacademic (hence nonciting) public penheim, C. (2003) Mandated online RAE peer-review costs for their annual outgoing is reading their books (“Demotic Metrics”). CVs Linked to University Eprint Archives articles out of just a fraction of their annual Institutional repositories can not only (1) add (http://www.ariadne.ac.uk/issue35/harnad/). windfall savings from the cancelation of their book-metadata/reference deposit to their OA Ariadne 35. incoming journal subscriptions. That cost will Deposit Mandates, but they can (2) harvest be even lower if charged per individual round Amazon book-sales metrics for their book Houghton, J.W., Rasmussen, B., Shee- of refereeing as no-fault refereeing fees rather metadata deposits, to add to their IR stats han, P.J., Oppenheim, C., Morris, A., Creas- than as acceptance/publication fees (which (http://trac.eprints.org/projects/irstats). Re- er, C., Greenwood, H., Summers, M., and require factoring in all the costs of the rejected positories can also already harvest Google Gourlay, A. (2009). Economic Implications articles into the fee for the accepted articles) Books, http://books.google.com/books?q= of Alternative Scholarly Publishing Models: (Harnad 2010). %22decline+and+fall+of+the+roman+emp Exploring the Costs and Benefits (http://www. Covering publication costs through per- ire%22+gibbon&btnG=Search+Books, (and jisc.ac.uk/publications/reports/2009/econom- article publication fees instead of through Google Scholar, http://scholar.google.com/ icpublishingmodelsfinalreport.aspx), London per-journal subscription fees is called “Gold scholar?q=%22decline and fall of the roman and Bristol: The Joint Information Systems OA publishing” (Harnad et al. 2004). It is the empire%22 gibbon&sa=N&tab=ps) book- Committee (JISC). natural, stable solution for refereed research citation counts today, as a first step toward Okerson, A., and O’Donnell, J. (Eds.) publishing in the Post-Gutenberg era (Harnad constructing a distributed, universal OA book- (1995) Scholarly Journals at the Crossroads; 2009), but it is only possible if Green OA self- citation index. The Dublin Humanities Met- A Subversive Proposal for Electronic Publish- archiving is universally mandated first, so that rics Conference (http://www.coimbra-group. ing (http://www.arl.org/scomm/subversive/toc. (1) the access-provision and archiving costs eu/DOCUMENTS/coimbra-groups-semimars/ html). Washington, DC, Association of Re- can be offloaded onto institutional repositories, metrics workshop programme2.pdf) was also search Libraries, June 1995. concerned about other kinds of online works, (2) the journals can downsize to peer-review Sale, A., Couture, M., Rodrigues, E., and how to measure and credit their impact: service provision alone, and (3) institutional Carr, L., and Harnad, S. (2010) Open Ac- metrics don’t stop with citation counts and subscription cancelations can release the funds cess Mandates and the “Fair Dealing” Button download counts. Among the many “De- to pay for the peer-review fees. Universal (http://eprints.ecs.soton.ac.uk/18511/). In: motic metrics” that can also be counted are Green OA mandates followed by downsizing Dynamic Fair Dealing: Creating Canadian link-counts, tag-counts, blog-mentions, and to Gold OA saves a good deal of money overall Culture Online (Rosemary J. Coombe and Web-mentions. This applies to books/au- (Houghton et al. 2009), whereas trying to do it Darren Wershler, Eds.) the other way round costs more money and fails thors, as well as to data, to courseware, and to generate universal OA (Harnad 2011). to other identifiable online resources. We should hasten the progress of book metrics, Does this solution generalize to schol- Endnotes arly monographs? The economics of book and that will in turn accelerate the growth in OA’s primary target content, journal articles, 1. Copyediting is the lightest in STM publishing and journal publishing are not the journals; it may still be somewhat more same. Nor is it true of all authors of scholarly as well as increasing support for institutional and funder OA Deposit Mandates. substantive in humanities and arts journals, monographs, as it is true of all authors of peer- as well as in books. This would need to be reviewed journal articles, that they write solely References examined systematically, but it seems almost for uptake and impact, not for royalty revenue. certain that the practice and the demand for But research is research, and book authors Gargouri, Y., Hajjem, C., Lariviere, V., copyediting are declining in the online era, too benefit, both in their research and in their Gingras, Y., Brody, T., Carr, L., and Harnad, and it may make more sense to offer it for S. (2010) Self-Selected or Mandated, Open a fee as an optional extra service to authors careers and funding, from the impact of their and their institutions. findings. So perhaps once a book citation index Access Increases Citation Impact for Higher is created and shows the impact to be gained Quality Research (http://eprints.ecs.soton. from making monographs OA, monographs ac.uk/18493/). PLOS ONE 5 (10). e13636 too will take the Green and eventually the Gold Harnad, S. (2008) Open Access Book- road to OA (Harnad 2008). Impact and “Demotic” Metrics. Open Access

24 Against the Grain / April 2011