“This might be too geeky for a column,” said the subject line of a reader's email, “but just in case …”
It sounded like a challenge, and I took the bait. The topic in question? A new statistical instrument to quantify the degree of open access for scholarly journals. In other words, exactly geeky enough.
The metric can, in principle, be used with journals in any field. At this stage, though, it’s only really being talked about in library and information science (LIS) circles. It represents a challenge to academic librarians to “walk the talk” in regard to their own professional publications. But it's an “inside baseball” discussion that merits attention outside the dugout, given the role of academic librarians in shaping the whole terrain of 21st-century scholarly communication.
That role is crucial but often overlooked. Academic librarians still have the core responsibilities of managing acquisitions and maintaining subscriptions, of course, but must also keep track of the new array (constantly growing, across all disciplines) of digital-format archives, databases, and other repositories. Plus they have the pedagogical task of instructing patrons in how to use new research tools as they become available.
As if that weren’t enough to do, research libraries have been mutating into scholarly publishers in their own right, sometimes in cooperation with their universities’ presses. To borrow a phrase from a recent paper in the journal College & Research Libraries, academic librarians have gone beyond being “gatekeepers of knowledge” -- in charge of its storage and retrieval -- to playing an active role in its promulgation.
Sugimoto et al. sent a survey to their colleagues at 91 academic libraries in the U.S. about how they kept track of developments within library and information science itself. Just over six hundred people filled out all or most of the questionnaire.
The findings reveal a profession that's seriously interested in its own rapidly changing role in scholarly communications: "A vast majority (94.2 percent) consult professional literature" -- defined to include scholarly journals as well as less formal venues such as trade publications and blogs — "on, at the very least, a monthly basis.” More than a quarter of respondents said they did so daily.
Over 80 percent of respondents indicated they followed peer-reviewed LIS journals. More three-quarters kept up with conference papers and proceedings in their field. "Nearly three-quarters," the paper notes, "reported sharing the results of research or reports of best practices" with their colleagues, with more than half (54.2 percent) doing so in peer-reviewed journals."
The other, more granular statistics in the paper are significant, but I want to stress a couple of important big-picture issues suggested by the study. On the one hand, the Indiana researchers describe a kind of virtuous circle. Academic librarians are eager both to produce and to exchange knowledge about their field -- not just to publish but to read one another’s work and to incorporate it into their own activity. (And that is a good thing for the rest of us, prone though we are to taking their efforts for granted.)
The paper also stresses that academic librarians have been advocates for "new (particularly open) systems of scholarly communication." They have shown prescient and growing support for open-access publishing for a number of years now. But here's where things become problematic, because it sounds like the library and information studies people could use some "new (particularly open) systems of scholarly communication” of their own.
Librarians who are also tenure-track faculty need to publish in the field's major peer-reviewed journals. (Forty percent of respondents to the Indiana researchers’ survey were either tenured or on the track.)
But with a prestigious journal, the lag time between between acceptance and publication can run to a year or more. That delay "impede[s] the timeliness and back-and-forth exchanges that are required for effective scholarly communication." And in "technology-related fields ... research may lose its currency if it is not delivered expediently."
Then there is the conundrum assessed in another recent study, Micah Vandergrift and Chealsye Bowley's "Librarian Heal Thyself: A Scholarly Communication Analysis of LIS Journals,” published last month by In the Library With a Lead Pipe, which is probably the best name ever for a peer-reviewed journal. (Vandergrift is a scholarly communications librarian at Florida State University. Bowley is library supervisor at FSU's Florence Study Center in Italy.)
While academic librarians have been strong advocates of open-access publishing, many LIS researchers seem to exempt their own field. One study the authors cite found that half of respondents “cared mostly about publication without considering the policies of the journals in which they published and that only 16 percent had exercised the right to self-archive in the institutional archive.”
Vandegrift and Bowley assembled data on the policies of 111 library and information science journals and found that with a large minority of them (well over a third) the author signs over all copyrights to the publisher — “including but not limited,” as the contracts run, "to the right to publish, republish, transmit, sell, distribute, and otherwise use the [article] in whole or in part … in derivative works throughout the world, in all languages, and in all media of expression now known or later developed.” (You could probably get away with giving the PDF to a close friend, just be very, very quiet about it.)
Just a handful of journals “had direct or implied policies regarding what the author is allowed to do with specific versions of the same work,” including self-archiving in an institutional repository. “A significant percentage of our professional literature,” Vandegrift and Bowley conclude, "is still owned and controlled by commercial publishers whose role in scholarly communication is to maintain ’the scholarly record,’ yes, but also to generate profits at the expense of library budgets by selling our intellectual property back to us.”
A norm doesn’t remain a norm unless nearly everyone involved acquiesces to it. A couple of years ago The Economist referred to the signs of growing unhappiness with the state of scholarly publishing as "The Academic Spring," and Vandegraft and Bowley's paper is part of it.
"A great example of a proactive and outspoken group,” Bowley told me in an email exchange, "was the Journal of Library Administration's Editor-in-Chief Damon Jaggars and entire editorial board who resigned in March 2013 … [over] an author agreement that they thought was "too restrictive and out of step with the expectation of authors.” Vandegrift was among the authors who had requested a Creative Commons license or to retain their copyright — an open-access policy that Taylor & Francis, the journal’s publisher, rejected.
Continuing the effort to bring the publishing practices of LIS researchers into accord with its ethos, Vandegrift and Bowled have created an instrument called the Journal Openness Index. It uses a points system for the various degrees of control over copyright and reuse indicated in a journal’s stated policies. The higher the JOI, the more open-access the publication. Crunching the numbers for several leading LIS titles, the authors find that the journals of professional societies get the highest scores while those from commercial-academic publishers get the lowest, with journals issued by university presses falling somewhere in between.
That is not exactly counterintuitive. But Vandegrift and Bowley offer JOI as a step in the direction of establishing open access as one of the criteria for how colleagues assess the value of scholarship in their own field.
"I imagine all the students that come out of library schools,” Vandegraft said by email, who "go into public librarianship and all of a sudden are cut off from access to the literature that can and should inform the practice of their work, which they were trained to do in library schools where ‘access' is touted as a value. I think we can do better, and I think it will take articles like this one to push librarians to be more proactive and to ask our faculty colleagues to join us."
As for applying JOI to journals in other fields, the idea is feasible but demanding. “Such a project would need proper backing,“ Bowley told me, "whether in the form of a team or institutional and financial support, in order to ensure its long-term upkeep. It could also be partially done through crowdsourcing the information, though. If a professional organization or institution is interested in taking up the project, they would certainly be welcomed to do so.”
I hope their colleagues take them up on it. An informed librarian is a helpful librarian — and it’s a fool who underestimates the value of that.
The carnage and manhunt in Boston last week obliged the Digital Public Library of America to postpone its grand opening festivities at the Boston Public Library until sometime this fall. So sudden a change of plans could only create a logistical nightmare. The roster of museums, archives, and libraries participating in DPLA runs into the hundreds, and the two-day event (Thursday and Friday) was booked to capacity, with scores of people on the standby list. But the finish line for the marathon was just outside the library, and rescheduling unavoidable.
The delay applied only to the gala, not to DPLA itself: the site launched on Thursday at noon, E.S.T., right on schedule. The response online has been, for the most part, enthusiasm just short of euphoria. The collection contains not quite 2.4 million digital “objects,” including books, manuscripts, photographs, recorded sound, and film/video. More impressive than the quantity of material, though, is how much thought has gone into how it’s made available.
That’s true even of the site’s address: DP.LA. I’ve seen at least one grumble about how anomalous this looks. Which it does, but in a good way. Even if you forget the address, it takes no effort to reconstruct. The brevity of the URL makes it convenient to type on a cellphone; when you do, the site’s homepage is readily navigable on the small screen. That demonstrates an awareness of how a good many visitors will actually use the site – more so than is often the case with library catalogs online.
DPLA is the work of people who understand that design is not just icing on the digital cake, but a significant (even decisive) factor in how we engage with content in the first place. They have made available an application program interface (API) for the site, which is a very useful thing indeed, according to my source in the geek community. With the API, users can create new tools for sorting and presenting the library’s materials. Combine it with a geolocation API, for example, and you could put together an application displaying the available photographs of the street you are on, organized decade by decade.
The library’s potential for assembling and integrating an incredible range of documents and knowledge is almost unimaginable. Excitement seems appropriate. But in describing my own impressions of DPLA, I want to be a little more qualified about the enthusiasm it inspires. Things are not nearly as far along as some comments have implied. This isn’t just naysaying. The site is currently in its beta version, and many of my points will probably be nullified in due course. But it’s better to be aware of some of the limitations beforehand than to visit the site expecting a digital Library of Alexandria.
One thing to keep in mind is that DPLA is not so much a library as an enormous card catalog, with the “shelves” of books, photographs, and so forth being the digital collections of libraries and historical societies, large and small, all over the country. The range of material offered through the Digital Public Library of America reflects what people running the local collections have decided to digitize and make available. What DPLA gathers and makes searchable is the metadata: descriptions of what a document contains (its subject, origins, copyright status, and so on) and of its characteristics as a digital object (size and file type).
The DPLA “card” gives the available information about an item, often accompanied by a thumbnail image of the book cover, manuscript, etc. – along with a link taking you to the digital repository in which it appears. DPLA puts the metadata into a standard format. But much of the content-description will inevitably be done by local librarians and archivists, making for a considerable range in detail. Often the DPLA entry will provide a bare minimum of description, though some entries run to a paragraph or two.
But the entry is only as strong as its link. It seemed appropriate to make one of my earliest searches at the Digital Public Library for the quintessential American poet Walt Whitman. There were 52 hits, with 9 of the top 10 being manuscripts of his letters in the Department of Justice collection at the National Archives. Not one of the links for the letters worked. By contrast, I had no trouble getting access to photographs of the poet held by the Smithsonian Institution.
This proved par for the course. Most links worked -- but out of two dozen entries for items in National Archives, only one did. It’s hardly surprising (gremlins have a strong work ethic), but it shows the need for troubleshooting. Users of the library can be expected to point out such glitches, if encouraged to do so. It might be worth adding a widget that would appear in each record allowing users to flag an inoperative link, a typographical error, or some problem with the content description. It's true that the site has a contact page, but people are more likely to report errors if they are encouraged to do so.
Continued thumbing through the catalog demonstrated how early a stage DPLA is in accumulating its collection – and how much fine-tuning its search engine may need.
Entering “Benjamin Franklin,” you get more than 1,400 results. Out of the first 30, all but 3 are documents (usually death certificates) for people named after the inventor and statesman. A toolbar on the left allows the user to refine the search in various ways – but the most useful filter, by subject, is at the very bottom and easy to overlook.
It was encouraging to get 17 results when searching for Phyllis Wheatley, the first published African-American poet, but 15 of them led to records from the 1940 census, by which point she had been dead the better part of 150 years. Only one of the other two was at all germane to her as historical figure. The other concerned an Atlanta branch of the Young Women’s Christian Association named in her honor.
I expected to locate just a few things about the Southern Tenant Farmers Union of the 1930s, but in fact got no hits at all. At the other extreme, DPLA has records for more than 90 items pertaining to the Ku Klux Klan – photographs, handbills, and cartoons, both pro- and anti-. Quite likely these were among the most striking and attention-grabbing items in various collections, and were digitized for use in print publications and online. It's concrete evidence that the Digital Public Library of America's offerings will be only as representative as the decisions made by the contributing institutions.
A number of foundations and government agencies have lent their support to DPLA, and its progress towards incorporation as a 501(c)3 organization should make it an even more appealing destination for the big philanthropic bucks. But important as funding certainly is for the library’s future, what it will ultimately be decisive for its success is a massive infusion of intellectual capital. Some of it will come from code writers hacking out new applications using the library's metadata and API. More than that, though, DPLA will need to encourage the participation and the expertise of people using the site. It's an impressive foundation and scaffold, but it's up to scholars, librarians, and other knowledgeable citizens to build the library, from the ground up.