Foxes Propose New Guidelines for Henhouse Design: Comments on NISO’s Proposed Open Access Metadata Standards

This commentary is in response to: NISO RP-22-201x, Open Access Metadata and Indicators (draft for comment), which is available at: http://www.niso.org/apps/group_public/document.php?document_id=12047

NISO is the National Information Standards Organization, a non-profit industry organization whose mission statement reads: "NISO fosters the development and maintenance of standards that facilitate the creation, persistent management, and effective interchange of information so that it can be trusted for use in research and learning." Their recently-issued proposed guidelines for new metadata fields to be attached to scholarly works purport to address and clarify issues of access and re-licensing surrounding the electronic distribution of journal articles.Briefly, they propose to add two fields to the standard metadata: one called "free to read" to identify documents that may be accessed without restriction or registration, and a second called "license_ref " to point users to a uniform resource identifier (URI) that will purportedly explain the re-use terms that apply to the document in question.
The draft proposal is not long-18 pages-and is thankfully plain-language and not overly technical.NISO is to be complimented for that.
The first problem with the proposal, though, and in many ways the root of the subsequent ones, is the composition of the 16-member working or advisory group.It is overwhelmingly made up of publishers, publishing consultants, and publisher associations.There is a single representative of an American library (Indiana University), two from British libraries (Wellcome Library and University of Birmingham Library), and one from a British library association (JISC).The representative of the American library organization (SPARC) is actually the owner of a publishing consulting company (ScholarNext).There is no one to represent the interests of the potential users, the faculty authors, or the institutional disseminators of the content.
In fact, the group contains 75% representatives of publishers and publishing services, many of whom have opposed, misrepresented, and sought to limit legitimate fair use of published materials.

Author's Note:
This commentary is in response to: NISO RP-22-201x, Open Access Metadata and Indicators (draft for comment), which is available at: http://www.niso.org/apps/group_public/document.php?document_id=12047 JL SC Commons, CrossRef, Ex Libris, Inc., EDItEUR, and Kennisland.They have been offered a seat at the table while the writers and researchers who create and use scholarship have not.This has the appearance of a selfinterested cabal setting up standards that further entrench their control over content on which they have managed to secure a near monopoly.
Propagation of the standards lends an unearned aura of credibility, legitimacy, and authority to untested publisher claims of ownership or proprietorship.It also threatens to marginalize non-commercial and nontraditional publishers or disseminators of scholarly content.The guidelines are targeted at a relatively narrow (but profitable) band of communication, where content purportedly owned or licensed by large publishers is furnished to libraries and faculty, conceived only as passive recipients.Adoption of the proposed practices would only further normalize the institutionalized theft of intellectual property from the creators and originators of knowledge, who remain subject to the predatory and monopolistic practices of the majority of these guideline writers.

"free to read"
While this designation makes some sense in the narrow universe of subscription, hybrid, and gold OA journals publication, it does not address the myriad of forms in which scholarly content exists online."Free to read" is defined as "accessible to anyone with an internet connection and without registration."So public domain works in Google Books would not qualify (sign-in is required).Nor would pdf files from the National Academies Press website (which requires login for pdf 's), though their low-res non-log-in-requiring page-by-page .giffiles theoretically would.What about works that can be accessed only as page images and only a page at a time, such as those in the Library of Congress's American Memory project?These would seem to qualify sensu stricto, but how would the attribute be attached or applied.Works in Hathi Trust, or the Internet Archive might or might not.A 400 Mb file from the Internet Archive might be theoretically accessible to someone with dial-up access to the internet, but in practice-not so much.

"license_ref"
More disturbing is the "license_ref" attribute, purporting to clear up the murky waters of re-use rights.Publishers frequently misunderstand, misstate, or misrepresent the rights they hold over academic materials.Springer, for example, labels articles authored by US government employees as "Copyright US Government," which is an impossibility and an absurdity.Elsevier and others frequently publish without a copyright statement other than "Published by Elsevier"; this is purportedly done when some or all coauthors have declined or failed to sign the copyright transfer agreement.Many publishers commonly place a copyright ownership statement on materials that are public domain (see Figures 1-8, Appendix).In practice, copyright statements are routinely placed by copy-editors who have little or no understanding of copyright law or knowledge of the specific transfer agreements.It is also common practice for copyright to be settled on or remain with the author(s), while all publication, dissemination, and re-use rights are held by the publisher.Copyright transfer agreements are frequently 10 or more pages of single-spaced fine print.The "license_ref " attribute will do little to sort out this mess, and it will, in fact, lend greater credence and assumed authority to mistaken, misguided, or fraudulent claims of ownership.
A further complication to the "license_ref " attribute is that the re-use rights granted by various publishers differ so widely in what is granted to various classes of users and for various versions of material.With some publishers, a work may or may not be disseminated by the institution based on whether that institution has a policy mandating accessible dissemination.For ACS, the institution must have a mandate; for Elsevier, it must not-and both these publishers are represented in the working group.Moreover, while most commercial publishers have declared policies regarding accessible re-use of materials they have published, many scholarly societies, university presses, and smaller publishers have not.Certainly, this is an inconvenience to repository managers and authors seeking to provide access to their materials, but the adoption of the proposed NISO standard will do little to address this issue.
JL SC Yet another complication is the frequent discrepancy between the stated policies of publishers regarding reuse and the actual rights granted or transferred by contract.Several publishers have issued re-use and author posting policies that flatly contradict the terms of their standard author agreements, in all cases claiming controls over reuse licenses that have no legal or contractual basis.
Further, the idea that an attribute referring to a licensing URI will persist over the long term or point to the same content over time is either innocent or absurd.To believe that the era of publisher acquisition and consolidation is ended is naïve; and online resources, even uniquely identified ones, often do not persist beyond the next website redesign or revision or software update.Today's URI is next year's Error 404.
Frankly, the publishers need to put their house in order before presuming to prescribe new metadata standards that will perpetuate their uneven and self-serving administration of the rights they have wrested from the academic laboring class.Enshrining their sometimes questionable claims in dedicated metadata fields would be good business practice for them, but not for those who want to see scholarly communication conducted on more equitable and fairly competitive terms.
A further issue is that these proposed metadata standards raise the bar for small-scale, start-up, and occasional publishing efforts-such as those from university libraries, departments, centers, or student organizations.Not all publishers have the expertise or the infrastructure to comply with the standards and practices recommended by NISO.This appears to be yet another occasion where the promotion of "open access" has become the sheep's clothing under whose guise non-standard, noncommercial, and innovative publishing models are pushed beyond the pale of acceptability or feasibility.The standards proposed here are another example of an existing monopolistic profit-based cabal of large commercial publishers (and their sometimes unwitting allies) asserting control over the language and practice of academic discourse.
I urge the NISO organization to re-think its objectives, means, and priorities here, and to open up the discussion to those stakeholders whose interests were ignored or suppressed.
Paul Royster serves on the JLSC editorial board.The opinions expressed here are those of the author, and do not necessarily reflect JLSC's position.

Coordinator of Scholarly Communication
University of Nebraska-Lincoln Libraries Lincoln, NE 68588-4100 proyster@unl.edu

APPENDIX Figure 1. Journal of the American Water Resources Association
Note that the American Water Research Association both claims to own the copyright to this article and simultaneously admits that it is in the public domain.

Figure 2. Wetlands
Springer declares copyright in the name of the "U.S. Government," but the U.S. government cannot hold a copyright by virtue of authorship or creation.There is no copyright in this document; it cannot belong to anyone because it does not exist.

Figure 3. Environmental Monitoring and Assessment
Here Springer claims to own copyright of a U.S. government work outside the United States, but a work not subject to copyright is not subject to international copyright treaties or conventions, so this claim is erroneous, if not fraudulent.

Figure 4. Aquatic Toxicology
Elsevier's claim here is unclear.Copyright is asserted, but Elsevier is identified only as the publisher, not the copyright holder.In any event, the work is not copyrightable, despite the somewhat dodgy assertion.

Figure 5. Science of the Total Environment
No shyness here.Elsevier confidently claims ownership of a public domain U.S. government work, probably based on an author transfer agreement in a transaction tantamount to purchasing the deed to the Brooklyn Bridge.

Figure 6. Meteoritics & Planetary Science
The Meteoritical Society might be excused for its misstatement of ownership; they are, after all, concerned about events on a higher plane.But this is still an inaccurate and misleading claim.