Category: E-Prints

Open Access Repository Software Use By Country

Based on data from the OpenDOAR Charts service, here is snapshot of the open access repository software that is in use in the top five countries that offer such repositories.

The countries are abbreviated in the table header column as follows: US = United States, DK = Germany, UK = United Kingdom, AU = Australia, and NL = Netherlands. The number in parentheses is the reported number of repositories in that country.

Read the country percentages downward in each column (they do not total to 100% across the rows).

Excluding "unknown" or "other" systems, the highest in-country percentage is shown in boldface.

Software/Country US (248) DE (109) UK (93) AU (50) NL (44)
Bepress 17% 0% 2% 6% 0%
Cocoon 0% 0% 1% 0% 0%
CONTENTdm 3% 0% 2% 0% 0%
CWIS 1% 0% 0% 0% 0%
DARE 0% 0% 0% 0% 2%
Digitool 0% 0% 1% 0% 0%
DSpace 18% 4% 22% 14% 14%
eDoc 0% 2% 0% 0% 0%
ETD-db 4% 0% 0% 0% 0%
Fedora 0% 0% 0% 2% 0%
Fez 0% 0% 0% 2% 0%
GNU EPrints 19% 8% 46% 22% 0%
HTML 2% 4% 4% 4% 0%
iTor 0% 0% 0% 0% 5%
Milees 0% 2% 0% 0% 0%
MyCoRe 0% 2% 0% 0% 0%
OAICat 0% 0% 0% 2% 0%
Open Repository 0% 0% 3% 0% 2%
OPUS 0% 43% 2% 0% 0%
Other 6% 7% 2% 2% 0%
PORT 0% 0% 0% 0% 2%
Unknown 31% 28% 18% 46% 23%
Wildfire 0% 0% 0% 0% 52%

Snapshot Data from OpenDOAR Charts

OpenDOAR has introduced OpenDOAR Charts, a nifty new service that allows users to create and view charts that summarize data from its database of open access repositories.

Here’s what a selection of the default charts show today. Only double-digit percentage results are discussed.

  • Repositories by continent: Europe is the leader with 49% of repositories. North America places second with 33%.
  • Repositories by country: In light of the above, it is interesting that the US leads the pack with 29% of repositories. Germany (13%) and the UK follow (11%).
  • Repository software: After the 28% of unknown software, EPrints takes the number two slot (21%), followed by DSpace (19%).
  • Repository types: By far, institutional repositories are the leader at 79%. Disciplinary repositories follow (13%).
  • Content types: ETDs lead (53%), followed by unpublished reports/working papers (48%), preprints/postprints (37%), conference/workshop papers (35%), books/chapters/sections (31%), multimedia/av (20%), postprints only (17%), bibliographic references (16%), special items (15%), and learning objects (13%).

This is a great service; however, I’d suggest that University of Nottingham consider licensing it under a Creative Commons license so that snapshot charts could be freely used (at least for noncommercial purposes).

Census of Institutional Repositories in the United States

The Council on Library and Information Resources has published the Census of Institutional Repositories in the United States: MIRACLE Project Research Findings, which was written by members of the University of Michigan School of Information’s MIRACLE (Making Institutional Repositories a Collaborative Learning Environment) Project. The report is freely available in digital form.

Here is an excerpt from the CLIR press release:

In conducting the census, the authors sought to identify the wide range of practices, policies, and operations in effect at institutions where decision makers are contemplating planning, pilot testing, or implementing an IR; they also sought to learn why some institutions have ruled out IRs entirely.

The project team sent surveys to library directors at 2,147 institutions, representing all university main libraries and colleges, except for community colleges, in the United States. About 21% participated in the census. More than half of the responding institutions (53%) have done no IR planning. Twenty percent have begun to plan, 16% are actively planning and pilot testing IRs, and 11% have implemented an operational IR.

While the study confirms a number of previous survey findings on operational IRs—such as the IR’s disproportionate representation at research institutions and the leading role of the library in planning, testing, implementing, and paying for IRs—the census also offers a wealth of new insights. Among them is the striking finding that half of the respondents who had not begun planning an IR intend to do so within 24 months.

Other institutional repository surveys include the ARL Institutional Repositories SPEC Kit and the DSpace community survey.

Economists’ Self-Archiving Behavior

Ted C. Bergstrom and Rosemarie Lavaty have deposited an eprint in eScholarship that studies the self-archiving behavior of economists ("How Often Do Economists Self-Archive?").

They summarize their findings in the paper’s abstract:

To answer the question of the paper’s title, we looked at the tables of contents from two recent issues of 33 economics journals and attempted to find a freely available online version of each article. We found that about 90 percent of articles in the most-cited economics journals and about 50 percent of articles in less-cited journals are available. We conduct a similar exercise for political science and find that only about 30 percent of the articles are freely available. The paper reports a regression analysis of the effects of author and article characteristics on likelihood of posing and it discusses the implications of self-archiving for the pricing of subscription-based academic journals.

Their conclusion suggests that significant changes in journal pricing could result from self-archiving:

As more content becomes available in open access archives, publishers are faced with greater availability of close substitutes for their products and library demand for journals is likely to become more price-elastic. The increased price-responsiveness means that profit-maximizing prices will fall. As a result, it can be hoped that commercial publishers will no longer be able to charge subscription prices greatly in excess of average cost. Thus the benefits of self-archiving to the academic community are twofold. There is the direct effect of making a greater portion of the body of research available to scholars everywhere and the secondary effect of reducing the prices charged by publishers who exploit their monopoly power.

OAIster Hits 10,000,000 Records

Excerpt from the press release:

We live in an information-driven world—one in which access to good information defines success. OAIster’s growth to 10 million records takes us one step closer to that goal.

Developed at the University of Michigan’s Library, OAIster is a collection of digital scholarly resources. OAIster is also a service that continually gathers these digital resources to remain complete and fresh. As global digital repositories grow, so do OAIster’s holdings.

Popular search engines don’t have the holdings OAIster does. They crawl web pages and index the words on those pages. It’s an outstanding technique for fast, broad information from public websites. But scholarly information, the kind researchers use to enrich their work, is generally hidden from these search engines.

OAIster retrieves these otherwise elusive resources by tapping directly into the collections of a variety of institutions using harvesting technology based on the Open Archives Initiative (OAI) Protocol for Metadata Harvesting. These can be images, academic papers, movies and audio files, technical reports, books, as well as preprints (unpublished works that have not yet been peer reviewed). By aggregating these resources, OAIster makes it possible to search across all of them and return the results of a thorough investigation of complete, up-to-date resources. . . .

OAIster is good news for the digital archives that contribute material to open-access repositories. "[OAIster has demonstrated that]. . . OAI interoperability can scale. This is good news for the technology, since the proliferation is bound to continue and even accelerate," says Peter Suber, author of the SPARC Open Access Newsletter. As open-access repositories proliferate, they will be supported by a single, well-managed, comprehensive, and useful tool.

Scholars will find that searching in OAIster can provide better results than searching in web search engines. Roy Tennant, User Services Architect at the California Digital Library, offers an example: "In OAIster I searched ‘roma’ and ‘world war,’ then sorted by weighted relevance. The first hit nailed my topic—the persecution of the Roma in World War II. Trying ‘roma world war’ in Google fails miserably because Google apparently searches ‘Rome’ as well as ‘Roma.’ The ranking then makes anything about the Roma people drop significantly, and there is nothing in the first few screens of results that includes the word in the title, unlike the OAIster hit."

OAIster currently harvests 730 repositories from 49 countries on 6 continents. In three years, it has more than quadrupled in size and increased from 6.2 million to 10 million in the past year. OAIster is a project of the University of Michigan Digital Library Production Service.

ScientificCommons.org: Access to Over 13 Million Digital Documents

ScientificCommons.org is an initiative of the Institute for Media and Communications Management at the University of St. Gallen. It indexes both metadata and full-text from global digital repositories. It uses OAI-PMH to identify relevant documents. The full-text documents are in PDF, PowerPoint, RTF, Microsoft Word, and Postscript formats. After being retrieved from their original repository, the documents are cached locally at ScientificCommons.org. It has indexed about 13 million documents from over 800 repositories.

Here are some additional features from the About ScientificCommons.org page:

Identification of authors across institutions and archives: ScientificCommons.org identifies authors and assigns them their scientific publications across various archives. Additionally the social relations between the authors will be extracted and displayed. . . .

Semantic combination of scientific information: ScientificCommons.org structures and combines the scientific data to knowledge areas with Ontology’s. Lexical and statistical methods are used to identify, extract and analyze keywords. Based on this processes ScientificCommons.org classifies the scientific data and uses it e.g. for navigational and weighting purposes.

Personalization services: ScientificCommons.org offers the researchers the possibilities to inform themselves about new publications via our RSS Feed service. They can customize the RSS Feed to a special discipline or even to personalized list of keywords. Furthermore ScientificCommons.org will provide an upload service. Every researcher can upload his publication directly to ScientificCommons.org and assign already existing publications at ScientificCommons.org to his own researcher profile.

Notre Dame Institutional Digital Repository Phase I Final Report

The University of Notre Dame Libraries have issued a report about their year-long institutional repository pilot project. There is an abbreviated HTML version and a complete PDF version.

From the Executive Summary:

Here is the briefest of summaries regarding what we did, what we learned, and where we think future directions should go:

  1. What we did—In a nutshell we established relationships with a number of content groups across campus: the Kellogg Institute, the Institute for Latino Studies, Art History, Electrical Engineering, Computer Science, Life Science, the Nanovic Institute, the Kaneb Center, the School of Architecture, FTT (Film, Television, and Theater), the Gigot Center for Entrepreneurial Studies, the Institute for Scholarship in the Liberal Arts, the Graduate School, the University Intellectual Property Committee, the Provost’s Office, and General Counsel. Next, we collected content from many of these groups, "cataloged" it, and saved it into three different computer systems: DigiTool, ETD-db, and DSpace. Finally, we aggregated this content into a centralized cache to provide enhanced browsing, searching, and syndication services against the content.
  2. What we learned—We essentially learned four things: 1) metadata matters, 2) preservation now, not later, 3) the IDR requires dedicated people with specific skills, 4) copyright raises the largest number of questions regarding the fulfillment of the goals of the IDR.
  3. Where we are leaning in regards to recommendations—The recommendations take the form of a "Chinese menu" of options, and the options are be grouped into "meals." We recommend the IDR continue and include: 1) continuing to do the Electronic Theses & Dissertations, 2) writing and implementing metadata and preservation policies and procedures, 3) taking the Excellent Undergraduate Research to the next level, and 4) continuing to implement DigiTool. There are quite a number of other options, but they may be deemed too expensive to implement.

Will Self-Archiving Cause Libraries to Cancel Journal Subscriptions?

There has been a great deal of discussion of late about the impact of self-archiving on library journal subscriptions. Obviously, this is of great interest to journal publishers who do not want to wake up one morning, rub the sleep from their eyes, and find out over their first cup of coffee at work that libraries have en masse canceled subscriptions because a "tipping point" has been reached. Likewise, open access advocates do not want journal publishers to panic at the prospect of cancellations and try to turn back the clock on liberal self-archiving policies. So, this is not a scenario that any one wants, except those who would like to simply scrap the existing journal publishing system and start over with a digital tabula rosa.

So, deep breath: Is the end near?

This question hinges on another: Will libraries accept any substitute for a journal that does not provide access to the full, edited, and peer-reviewed contents of that journal?

If the answer is "yes," publishers better get out their survival kits and hunker down for the digital nuclear winter or else change business practices to embrace the new reality. Attempts to fight back by rolling back the clock may just make the situation worse: the genie is out of the bottle.

If the answer is "no," preprints pose no threat, but postprints may under some difficult to attain circumstances.

It is unlikely that a critical mass of author created postprints (i.e., author makes the preprint look like the postprint) will ever emerge. Authors would have to be extremely motivated to have this occur. If you don’t believe me, take a Word file that you submitted to a publisher and make it look exactly like the published article (don’t forget the pagination because that might be a sticking point for libraries). That leaves publisher postprints (generally PDF files).

For the worst to happen, every author of every paper published in a journal would have to self-archive the final publisher PDF file (or the publishers themselves would have to do it for the authors under mandates).

But would that be enough? Wouldn’t the permanence and stability of the digital repositories housing these postprints be of significant concern to libraries? If such repositories could not be trusted, then libraries would have to attempt to archive the postprints in question themselves; however, since postprints are not by default under copyright terms that would allow this to happen (e.g., they are not under Creative Commons Licenses), libraries may be barred from doing so. There are other issues as well: journal and issue browsing capabilities, the value-added services of indexing and abstracting services, and so on. For now, let’s wave our hands briskly and say that these are all tractable issues.

If the above problems were overcome, a significant one remains: publishers add value in many ways to scholarly articles. Would libraries let the existing system of journal publishing collapse because of self-archiving without a viable substitute for these value-added functions being in place?

There have been proposals for and experiments with overlay journals for some time, as well other ideas for new quality control strategies, but, to date, none have caught fire. Old-fashioned peer review, copy editing and fact checking, and publisher-based journal design and production still reign, even among the vast majority of e-journals that are not published by conventional publishers. In the Internet age, nothing technological stops tens of thousands of new e-journals using open source journal management software from blooming, but they haven’t so far, have they? Rather, if you use a liberal definition of open access, there are about 2,500 OA journals—a significant achievement; however, there are questions about the longevity of such journals if they are published by small non-conventional publishers such as groups of scholars (e.g., see "Free Electronic Refereed Journals: Getting Past the Arc of Enthusiasm"). Let’s face it—producing a journal is a lot of work, even a small journal that only publishes less than a hundred papers a year.

Bottom line: a perfect storm is not impossible, but it is unlikely.

Results from the DSpace Community Survey

DSpace conducted an informal survey of its open source community in October 2006. Here are some highlights:

  • The vast majority of respondents (77.6%) used or planned to use DSpace for a university IR.
  • The majority of systems were in production (53.4%); pilot testing was second (35.3%).
  • Preservation and interoperability were the highest priority system features (61.2% each), followed by search engine indexing (57.8%) and open access to refereed articles (56.9%). (Percentage of respondents who rated these features "very important.") Only 5.2% thought that OA to refereed articles was unimportant.
  • The most common type of current IR content was refereed scholarly articles and theses/dissertations (55.2% each), followed by other (48.6%) and grey literature (47.4%).
  • The most popular types of content that respondents were planning to add to their IRs were datasets (53.4%), followed by audio and video (46.6% each).
  • The most frequently used type of metadata was customized Dublin Core (80.2%), followed by XML metadata (13.8%).
  • The most common update pattern was to regularly migrate to new versions; however it took a "long time to merge in my customizations/configuration" (44.8%).
  • The most common types of modification were minor cosmetics (34.5%), new features (26.7%), and significant user interface customization (21.6%).
  • Only 30.2% were totally comfortable with editing/customizing DSpace; 56.9% were somewhat comfortable and 12.9% were not comfortable.
  • Plug-in use is light: for example, 11.2% use SRW/U, 8.6% use Manakin, and 5.2% use TAPIR (ETDs).
  • The most desired feature for the next version is a more easily customized user interface (17.5%), closely followed by improved modularity (16.7%).

For information about other recent institutional repository surveys, see "ARL Institutional Repositories SPEC Kit" and "MIRACLE Project’s Institutional Repository Survey."

MIRACLE Project’s Institutional Repository Survey

The MIRACLE (Making Institutional Repositories A Collaborative Learning Environment) project at the University of Michigan’s School of Information presented a paper at JCDL 2006 titled "Nationwide Census of Institutional Repositories: Preliminary Findings."

MIRACLE’s sample population was 2,147 library directors at four-year US colleges and universities. The paper presents preliminary findings from 273 respondents.

Respondents characterized their IR activities as: "(1) implementation of an IR (IMP), (2) planning & pilot testing an IR software package (PPT), (3) planning only (PO), or (4) no planning to date (NP)."

Of the 273 respondents, "28 (10%) have characterized their IR involvement as IMP, 42 (15%) as PPT, 65 (24%) as PO, and 138 (51%) as NP."

The top-ranked benefits of having an IR were: "capturing the intellectual capital of your institution," "better service to contributors," and "longtime preservation of your institution’s digital output." The bottom-ranked benefits were "reducing user dependence on your library’s print collection," "providing maximal access to the results of publicly funded research," and "an increase in citation counts to your institution’s intellectual output."

On the question of IR staffing, the survey found:

Generally, PPT and PO decision-makers envision the library sharing operational responsibility for an IR. Decision-makers from institutions with full-fledged operational IRs choose responses that show library staff bearing the burden of responsibility for the IR.

Of those with operational IRs who identified their IR software, the survey found that they were using: "(1) 9 for Dspace, (2) 5 for bePress, (3) 4 for ProQuest’s Digital Commons, (4) 2 for local solutions, and (5) 1 each for Ex Libris’ DigiTools and Virginia Tech’s ETD." Of those who were pilot testing software: "(1) 17 for DSpace, (2) 9 for OCLC’s ContentDM, (3) 5 for Fedora, (4) 3 each for bePress, DigiTool, ePrints, and Greenstone, (5) 2 each for Innovative Interfaces, Luna, and ETD, and (6) 1 each for Digital Commons, Encompass, a local solution, and Opus."

In terms of number of documents in the IRs, by far the largest percentages were for less than 501 documents (IMP, 41%; and PPT, 67%).

The preliminary results also cover other topics, such as content recruitment, investigative decision-making activities, IR costs, and IR system features.

It is interesting to see how these preliminary results compare to those of the ARL Institutional Repositories SPEC Kit. For example, when asked "What are the top three benefits you feel your IR provides?," the ARL survey respondents said:

  1. Enhance visibility and increase dissemination of institution’s scholarship: 68%
  2. Free, open, timely access to scholarship: 46%
  3. Preservation of and long-term access to institution’s scholarship: 36%
  4. Preservation and stewardship of digital content: 36%
  5. Collecting, organizing assets in a central location: 24%
  6. Educate faculty about copyright, open access, scholarly communication: 8%

ARL Institutional Repositories SPEC Kit

The Institutional Repositories SPEC Kit is now available from the Association of Research Libraries (ARL). This document presents the results of a thirty-eight-question survey of 123 ARL members in early 2006 about their institutional repositories practices and plans. The survey response rate was 71% (87 out of 123 ARL members responded). The front matter and nine-page Executive Summary are freely available. The document also presents detailed question-by-question results, a list of respondent institutions, representative documents from institutions, and a bibliography. It is 176 pages long.

Here is the bibliographic information: University of Houston Libraries Institutional Repository Task Force. Institutional Repositories. SPEC Kit 292. Washington, DC: Association of Research Libraries, 2006. ISBN: 1-59407-708-8.

The members of the University of Houston Libraries Institutional Repository Task Force who authored the document were Charles W. Bailey, Jr. (Chair); Karen Coombs; Jill Emery (now at UT Austin); Anne Mitchell; Chris Morris; Spencer Simons; and Robert Wright.

The creation of a SPEC Kit is a highly collaborative process. SPEC Kit Editor Lee Anne George and other ARL staff worked with the authors to refine the survey questions, mounted the Web survey, analyzed the data in SPSS, created a preliminary summary of survey question responses, and edited and formatted the final document. Given the amount of data that the survey generated, this was no small task. The authors would like to thank the ARL team for their hard work on the SPEC Kit.

Although the Executive Summary is much longer than the typical one (over 5,100 words vs. about 1,500 words), it should not be mistaken for a highly analytic research article. Its goal was to try to describe the survey’s main findings, which was quite challenging given the amount of survey data available. The full data is available in the "Survey Questions and Responses" section of the SPEC Kit.

Here are some quick survey results:

  • Thirty-seven ARL institutions (43% of respondents) had an operational IR (we called these respondents implementers), 31 (35%) were planning one by 2007, and 19 (22%) had no IR plans.
  • Looked at from the perspective of all 123 ARL members, 30% had an operational IR and, by 2007, that figure may reach 55%.
  • The mean cost of IR implementation was $182,550.
  • The mean annual IR operation cost was $113,543.
  • Most implementers did not have a dedicated budget for either start-up costs (56%) or ongoing operations (52%).
  • The vast majority of implementers identified first-level IR support units that had a library reporting line vs. one that had a campus IT or other campus unit reporting line.
  • DSpace was by far the most commonly used system: 20 implementers used it exclusively and 3 used it in combination with other systems.
  • Proquest DigitalCommons (or the Bepress software it is based on) was the second choice of implementers: 7 implementers used this system.
  • While 28% of implementers have made no IR software modifications to enhance its functionality, 22% have made frequent changes to do so and 17% have made major modifications to the software.
  • Only 41% of implementers had no review of deposited documents. While review by designated departmental or unit officials was the most common method (35%), IR staff reviewed documents 21% of the time.
  • In a check all that apply question, 60% of implementers said that IR staff entered simple metadata for authorized users and 57% said that they enhanced such data. Thirty-one percent said that they cataloged IR materials completely using local standards.
  • In another check all that apply question, implementers clearly indicated that IR and library staff use a variety of strategies to recruit content: 83% made presentations to faculty and others, 78% identified and encouraged likely depositors, 78% had library subject specialists act as advocates, 64% offered to deposit materials for authors, and 50% offered to digitize materials and deposit them.
  • The most common digital preservation arrangement for implementers (47%) was to accept any file type, but only preserve specified file types using data migration and other techniques. The next most common arrangement (26%) was to accept and preserve any file type.
  • The mean number of digital objects in implementers’ IRs was 3,844.

dLIST E-Print Archive Adds Use Statistics

Authors who deposit e-prints in dLIST (Digital Library of Information Science and Technology) can now see use statistics for their works (archive users can see use startistics as well). For example, at the record for the "Indian Digital Library in Engineering Science and Technology (INDEST) Consortium: Consortia-Based Subscription to Electronic Resources for Technical Education System in India: A Government of India Initiative," you would click on "View statistics for this eprint" to get the use statistics for this work. You can view use statistics for the past four weeks, this year, last year, or all years.

Archive-wide use statistics are also available from either an e-print record or the dLIST Statistics page. From either one, you can rank all e-prints by use for the same time periods as individual e-prints and show overall archive use by year/month or country.

Disclosure: I am now the Scholarly Communication subject editor for dLIST.

The E-Print Deposit Conundrum

How can scholars be motivated to deposit e-prints in disciplinary archives, institutional repositories, and other digital archives?

In "A Key-Stroke Koan for Our Open-Access Times," Stevan Harnad says:

Researchers themselves have hinted at the resolution to this koan: Yes, they need and want OA. But there are many other demands on their time too, and they will only perform the requisite keystrokes if their employers and/or funders require them to do it, just as it is already their employers and funders who require them to do the keystrokes to publish (or perish) in the first place. It is employers and funders who set researchers’ priorities, because it is employers and funders who reward researchers’ performance. Today, about 15% of research is self-archived spontaneously but 95% of researchers sampled report that they would self-archive if required to do so by their employers and/or funders: 81% of them willingly, 14% reluctantly; only 5% would not comply with the requirement. And in the two objective tests to date of this self-reported prediction, both have fully confirmed it, with over 90% self-archiving in the two cases where it was made a requirement (Southampton-ECS and CERN).

This is a very cogent point, but, if the solution to the problem is to have scholars’ employers compel them to deposit e-prints, the next logical question is: how can university administrators and other key decision makers be convinced to mandate this activity?

In the UK, a debate is raging between OA advocates and publishers about the UK Research Funding Councils’ (RCUK) self-archiving proposal, which would "mandate the web self-archiving of authors’ final drafts of all journal articles resulting from RCUK-funded research." The fact that this national policy debate is occuring at all is an enormous advance for open access. If RCUK mandates e-print deposit, UK university administrators will need no convincing.

In the US, we are a long way from reaching that point, although the NIH’s voluntary e-print deposit policy provides some faint glimmer of hope that key government agencies can be moved to take some kind of action. However, the US does not have an equivalent to RUCK that can make dramatic e-print policy changes that affect research universities in one fell swoop. It does have government agencies, such as NSF, that control federal grant funds, private foundations that control their own grant funds, and thousands of universities and colleges that, in theory, could establish policies. This is a diffuse and varied audience for the OA message to reach and convince, and the message will need to be tailored to the audience to be effective.

While that plays out, we should not forget scholars themselves, however dimly we view the prospects of changing their behavior to be. University librarians and IT staff know their institutions’ scholars and can work with them one-one-one or in groups to gradually influence change. True, it’s "a journey of a thousand miles" approach, but, the number of librarians and IT staff that will be effective on a national stage is small, while the number of them that may be incrementally effective on the local level is large. The efforts are complementary, not mutually exclusive.

I would urge you to read Nancy Fried Foster and Susan Gibbons’ excellent article "Understanding Faculty to Improve Content Recruitment for Institutional Repositories" for a good example of how an IR can be personalized so that faculty have a greater sense of connection to it and how IR staff can change the way they talk about the IR to better match scholars’ world view.

Here are a few brief final thoughts.

First, as is often said, scholars care about the impact of their work, and it is likely that, if scholars could easily see detailed use statistics for their works (e.g., number of requests and domain breakdowns), they might be more inclined to deposit items if those statistics exceed their expectations. So, the challenge here is to incorporate this capability into commonly used archiving software programs if it is absent.

Second, scholars are unlikely to stumble when entering bibliographic data about their works (although it might not be quite as fully descriptive as purists might like), but entering subject keywords is another matter. Sure they know what the work is about, but are they using terms that others would use and that group their work with similar works in retrieval results? Yes, a controlled vocabulary would help, although such vocabularies have their own challenges. But, I wonder if user-generated "tags," such as those used in Technorati, might be another approach. The trick here is to make the tags and the frequency of their use visible to both authors and searchers. For authors, this helps them put their works where they will be found. For searchers, it helps them find the works.

Third, it might be helpful if an author could fill out a bibliographic template for a work once and, with a single keystroke, submit it to multiple designated digital archives and repositories. So, for example, a library author might choose to submit a work to his or her institutional repository, DLIST, and E-LIS all at once. Of course, this would require a minimal level of standardization of template information between systems and the development of appropriate import capabilities. Some will say: "why bother?" True, OAI-PMH harvesting should, in theory, make duplicate deposit unnecessary given OAIster-like systems. But "lots of copies keep stuff safe," and users still take a single-archive searching approach in spite of OAI-PMH systems.

Will You Only Harvest Some?

The Digital Library for Information Science and Technology has announced DL-Harvest, an OAI-PMH service provider that harvests and makes searchable metadata about information science materials from the following archives and repositories:

  • ALIA e-prints
  • arXiv
  • Caltech Library System Papers and Publications
  • DLIST
  • Documentation Research and Training Centre
  • DSpace at UNC SILS
  • E-LIS
  • Metadata of LIS Journals
  • OCLC Research Publications
  • OpenMED@NIC
  • WWW Conferences Archive

DL-Harvest is a much needed, innovative discipline-based search service. Big kudos to all involved.

DLIST also just announced the formation of an advisory board.

The following musings, inspired by the DL-Harvest announcement, are not intended to detract from the fine work that DLIST is doing or from the very welcome addition of DL-Harvest to their service offerings.

Discipline-focused metadata can be relatively easily harvested from OAI-PHM-compliant systems that are organized along disciplinary lines (e.g., the entire archive/repository is discipline-based or an organized subset is discipline-based). No doubt these are very rich, primary veins of discipline-specific information, but how about the smaller veins and nuggets that are hard to identify and harvest because they are in systems or subsets that focus on another discipline?

Here’s an example. An economist, who is not part of a research center or other group that might have its own archive, writes extensively about the economics of the scholarly publishing business. This individual’s papers end up in the economics department section of his or her institutional repository and in EconWPA. They are highly relevant to librarians and information scientists, but will their metadata records be harvested for use in services like DL-Harvest using OAI-PMH since they are in the wrong conceptual bins (e.g., set in the case of the IR)?

Coleman et al. point to one solution in their intriguing "Integration of Non-OAI Resources for Federated Searching in DLIST, an Eprints Repository" paper. But (lots of hand waving here), if using automatic metadata extraction was an easy and simple way to supplement conventional OAI-PMH harvesting, the bottom line question is: how good is good enough? In other words, what’s an acceptable level of accuracy for the automatic metadata extraction? (I won’t even bring up the dreaded "controlled vocabulary" notion.)

No doubt this problem falls under the 80/20 Rule, and the 20 is most likely in the low hanging fruit OAI-PMH-wise, but wouldn’t it be nice to have more fruit?

Not Green Enough

Yesterday, Stevan Harnad took the time to comment extensively on my "How Green Is My Publisher?" posting. Thanks for doing so, Stevan. Here are some further thoughts on the matter.

CB:My publication page, check. We don’t have an institutional repository yet, but I assume that "other external Web site" will cover that when we do, check. Wait a minute, what if I want to deposit the e-print in a disciplinary archive like E-LIS or I want to put it in the Internet Archive’s upcoming "OAI-compliant ‘universal repository‘"? Looks to me like I’m out of luck. No way to immediately deposit the paper in an OAI-PMH compliant archive that will have a longer life than my Website and that can be harvested by OAI-PMH search services, such as OAIster.

SH: "The restrictions on 3rd-party archives are perfectly reasonable and no problem whatsoever at this time. The problem today (just so we keep our eyes on the ball!) is the non-archiving of 85% of articles, hence their inaccessibility to all those would-be users whose universities cannot afford access to the journal’s official version! It is cheap and easy for any university to create an OAI-compliant institutional archive, and OAIster can happily harvest the metadata.
http://archives.eprints.org/eprints.php?action=browse"

eprints.org’s Institutional Archives Registry currently shows a total of 424 archives. When we browse by archive type, we discover that there are 192 "Research Institutional or Departmental" registered archives worldwide. Of course, “Departmental” archives are not institutional repositories. They do not have an institutional scope of coverage, nor are they as likely as institutional archives to be permanent. True, departments are relatively stable, but their commitment to maintaining archives may not be (e.g., the archive may be the pet project of one or a few faculty members). By contrast, once an institution commits to having an archive, it’s likely to be a more permanent arrangement, especially if it is run by a library.

But, let’s wave our hands, and say 100% of them are institutional repositories (IRs). Universities Worldwide, which is "based on the ‘World List of Universities 1997’ published by the International Association of Universities (IAU) and links discovered or posted here," currently lists 7,130 universities in 181 countries. Assuming that this is a good rough approximation, that means that about 6% of all universities have IRs. Meaning, of course, that 94% do not.

And that means that 94% of authors at universities cannot self-archive in an institutional repository (or, given the hand waving, in a departmental archive). True, they can self-archive on personal Web pages. The issues with this strategy are: (a) how may authors have up-to-date publication pages or have publication pages at all?, (b) how long will they last (i.e., authors change jobs, retire, and die)?, and (c) there is no OAI-PMH access to those pages, so they don’t show up in OAIster and similar search engines.

Now, disciplinary archives and the Internet Archive’s universal repository solve these problems. Moreover, they solve another problem: independent scholars, corporate researchers, and other non-academic authors may never have an institutional repository to self-archive in.

I don’t see this as "no problem whatsoever at this time." Quite the contrary. To be "no problem," we would have to believe that it doesn’t matter if articles are archived in OAI-PMH compliant repositories or archives. To be "no problem," we would have to not care whether scholars who will never have an institutional repository at their disposal can self-archive.

As to the question of it being "cheap and easy for any university to create an OAI-compliant institutional archive," I think there is some difference of opinion on that point. Susan Gibbons says [1] the "the costs and efforts involved in maintaining an IR are substantial," and she provides these annual IR cost estimates:

  1. $285,000, MIT
  2. $100,000 (Canadian), Queens University (for staffing only)
  3. $200,000, University of Rochester
  4. between 2,280 and 3,190 staff hours,University of Oregon

But, of course, these differences in perception about costs relate to some degree to Stevan’s next point:

SH: (And worrying about the preservation of non-existent contents is rather putting the cart before the horse. The self-archived OA versions of a goodly portion of the 15% of the articles that have been self-archived in the past 15 years are still online and OA to tell the tale to this day. All their publishers’ official versions are too. So fussing about the permanence of the non-contents of cupboards that are in any case meant to be access-supplements, not the official version of record, is rather misplaced, when what is immediately missing and urgently needed is their presence, not their permanence.)

I think that Stevan will find that few academic libraries are not going to worry about permanence. Not only will they worry about the permeance of digital objects in their repositories, they will also worry about the permanence of publisher’s archives. Librarians know that publishers are corporations, and that corporations change priorities, merge, and fail. As libraries increasingly abandon print subscriptions and go e-only for economic reasons, at some point there will be no permanent distributed print archive of new journal issues in libraries worldwide as there is today, and libraries are going to worry about that a great deal. Moreover, universities are not going to establish institutional repositories just to support OA. That may be one important item on the agenda, but there will be other archiving needs to be met as well, and factors associated with those digital objects will affect the perception of the need for overall IR preservation too.

Libraries are also going to provide new services to provide IR support in addition to technical support, ranging from convicting faculty to self-archive and helping them do so to training users in using IRs (as well as other e-print services worldwide). These services will cost money.

Don’t want libraries to lead the IR effort if this is true?

In the words of Bob Dylan:

I asked the captain what his name was
And how come he didn’t drive a truck
He said his name was Columbus
I just said, "Good luck."

Moving on.

CB: “The agreement also states that the e-print must contain a fair amount of information about the publisher and the paper: the published article’s citation and copyright date, the publisher’s address, information about the publisher’s document delivery service, and a link to the publisher’s home page.”

SH: That’s just fine too. It is only good scholarly practice to provide the full reference information and to link to the official version of record for the sake of all those potential users who can afford it. What is wrong with that, and why would any author not want to do that?

Sure, an author would want to provide a citation to the published paper and a link to it, but I suspect few will be excited about providing a fair amount of advertising information for the publisher in their e-prints, such as the publisher’s address, home page, and document delivery service. It’s not a deal killer, but it’s more work for authors or IR staff. The more individual publisher variations that there are in copyright transfer agreements, the harder it is for scholars and IR staff to meet these varying requirements.

CB: Second, it would be helpful if such directories could identify whether articles can be deposited in key types of archives. I know that we don’t want the color codes to look like SpeedyGrl.com’s Ultimate Color Table, but I think that this is an important factor in addition to the type of e-print permitted.

SH: They already do. The main distinction is the author’s own institutional archive versus central (3rd-party) archives. It is the former that are the critical ones. The rest can be done by metadata harvesting.

The SHERPA colors do not make this distinction. Neither do the otherwise helpful notes. You must look at each specific agreement (if there is a link to it).

CB: Fourth, although copyright transfer agreements have always been a confusing mess, now we want authors to actually read and evaluate them, not just mindlessly sign them like they did when digital archiving wasn’t an issue. And institutional repository managers (or archive managers) need to make sense of them post facto to determine if articles can be legally deposited and what terms apply to those deposits. So, maybe it’s time to tilt at a new windmill: a set of standardized copyright transfer agreements. I know, it’s like trying to herd several thousand hyperactive cats. But, a few years ago, getting standardized use statistics for electronic resources from publishers seemed hopeless, and some progress has been made on that score.

SH: No, it’s not more windmills or red herrings that researchers, their institutions, their funders, and research itself need: What they need is to go ahead and self-archive.

Developing clear, understandable standard copyright transfer agreements is a red herring? Let’s look at just one aspect of the problem: IR managers’ copyright concerns. I offer some quotes:

"One aspect of the survey [baseline survey of research material already held on departmental and personal Web pages in the ed.ac.uk domain] that is not shown in the results is the lack of consistency in dealing with copyright and IPR issues that scholars face when placing material online. Some academic units have responded by not self-archiving any material at all. A rather worrying example of this is the School of Law (—do they know something that we don’t?) A small percentage of individual scholars have responded by using general disclaimers that may or may not be effective. Others, generally well-established professors, have posted material online that is arguably in breach of copyright agreements, e.g. whole book chapters. Most, however, take a middle line of only posting papers from sympathetic publishers who allow some form of self-archiving. It is apparent that if institutional repositories are going to work, then this general confusion over copyright and IPR issues needs to be addressed right at the source." [2]

"Filling a repository for published and peer-reviewed papers is a slow process, and it is clear that it is a task that requires a significant amount of staff input from those charged with developing the repository. Although we have succeeded in adding a reasonable amount of content to the repository we have also been offered significant amounts of content that cannot be added because of restrictive publisher copyright agreements. In some cases academics have offered between ten and twenty articles and we have not been able to add any of them to the repository. This is a clear demonstration that major changes need to take place at a high level in order for repositories to be successful." [3]

Certainly, all OA advocates are eager to get on with the business of doing OA vs. simply reflecting on it, and few have done as much as Stevan to advance the cause, but, in my view, the issues I’ve raised warrant further consideration and action.

Notes

1. Susan Gibbons, "Establishing an Institutional Repository," Library Technology Reports 40, no. 4 (2004): 54, 56.

2. Theo Andrew, "Trends in Self-Posting of Research Material Online by Academic Staff." Ariadne, no. 37 (2003),
http://www.ariadne.ac.uk/issue37/andrew/intro.html.

3. Morag Mackie, "Filling Institutional Repositories: Practical Strategies from the DAEDALUS Project," Ariadne, no. 39 (2004),
http://www.ariadne.ac.uk/issue39/mackie/intro.html.

How Green Is My Publisher?

Back in the early 1990s, I began to fight to retain the copyright to my scholarly writings. First, the publishers thought I was kidding. Then, when it was clear that I wasn’t, they thought I was nuts. Generally, they weren’t willing to negotiate. So, I sought out the few journals that would comply with this strange whim or that had editors who would "forget" to have me sign an author agreement. Unfortunately, some of the more liberal journals got gobbled up by megapublishers, limiting my options and casting some doubt on handshake deals. Once e-only journals by nonconventional publishers took off, they became my venue of choice, since they typically allowed copyright retention by default.

Things have changed, in large part do to the growing influence of the open access movement. Now, many publishers allow self archiving of e-prints (electronic preprints or postprints), and this, in theory, means that authors can cheerfully assign their copyrights to those publishers. How many publishers do this? Well we don’t know for sure, but according to Summary Statistics So Far (whose figures are based on the Romeo Project), 92% of the 8,450 processed journals are "green," (can archive postprint) or "pale green"(can archive preprint). (Gray means you can’t archive either one.)

If you want to self archive a scholarly article, the SHERPA Publisher Copyright Policies & Self-Archiving site is the place to go to determine whether the publisher of the journal you have in mind for your article will permit it. So, when approached recently about writing a paper for a library publisher (let’s call it X), I fired up Mozilla and looked X up. Good news, X is green, meaning "can archive pre-print and post-print." Not the dreaded white ("archiving not formally supported"), not yellow ("can archive pre-print (ie pre-refereeing)"), not even blue ("can archive post-print (ie final draft post-refereeing)"), but green. SHERPA did warn me of two conditions: "Published source must be acknowledged" and "Eprint server is non-profit." No problemo, right? Being ever cautious, I then used the handy link to the actual policy.

Here’s what I found. My "preprint distribution rights" allow "posting as electronic files on the contributor’s own Web site for personal or professional use, or on the contributor’s internal university/corporate intranet or network, or other external Web site at the contributor’s university or institution, but not for either commercial (for-profit) or systematic third party sales or dissemination, by which is meant any interlibrary loan or document delivery systems. The contributor may update the preprint with the final version of the article after review and revision by the journal’s editor(s) and/or editorial/peer-review board."

My publication page, check. We don’t have an institutional repository yet, but I assume that "other external Web site" will cover that when we do, check. Wait a minute, what if I want to deposit the e-print in a disciplinary archive like E-LIS or I want to put it in the Internet Archive’s upcoming "OAI-compliant ‘universal repository‘"? Looks to me like I’m out of luck. No way to immediately deposit the paper in an OAI-PMH compliant archive that will have a longer life than my Website and that can be harvested by OAI-PMH search services, such as OAIster.

The agreement also states that the e-print must contain a fair amount of information about the publisher and the paper: the published article’s citation and copyright date, the publisher’s address, information about the publisher’s document delivery service, and a link to the publisher’s home page. Guess I can do this when I’m modifying the article to incorporate the editorial changes. That should keep me off the streets.

So, what can we conclude from this brief dip into the murky waters of author agreements other than retaining rights may still be a good idea (if you can do it)?

First, There are swirling currents of complexity beneath the placid surface of color-coded copyright transfer agreement directories. This is not to say that such directories are not indispensible (or not doing a good job), but rather that, given the idiosyncratic nature of such agreements, authors still need to read the details if they want to be fully aware of their residual rights. They may not always like what they find, and what they find may affect their willingness to self archive if it’s too limiting or burdensome. "Green" may not always mean "go."

Second, it would be helpful if such directories could identify whether articles can be deposited in key types of archives. I know that we don’t want the color codes to look like SpeedyGrl.com’s Ultimate Color Table, but I think that this is an important factor in addition to the type of e-print permitted.

Third, if claims are going to made about the number of "green" journals, maybe more consideration about what "green" means is in order, and perhaps OA advocates should agree on their color schemes. Is "can archive pre-print and post-print" enough for "green," or should it be "can archive pre-print and post-print on the author’s Website or in any noncommercial archive or repository"? If the latter, the heat should be turned up on publishers that don’t permit it by authors and OA advocates.

Fourth, although copyright transfer agreements have always been a confusing mess, now we want authors to actually read and evaluate them, not just mindlessly sign them like they did when digital archiving wasn’t an issue. And institutional repository managers (or archive managers) need to make sense of them postfacto to determine if articles can be legally deposited and what terms apply to those deposits. So, maybe it’s time to tilt at a new windmill: a set of standardized copyright transfer agreements. I know, it’s like trying to herd several thousand hyperactive cats. But, a few years ago, getting standardized use statistics for electronic resources from publishers seemed hopeless, and some progress has been made on that score.