Thursday, April 25, 2013

It's My Fault (At Least Partially)

I want to talk game theory. Specifically, coordination games. Carl T. and Theodore C. Bergstrom (2006) identify the scholarly publication business as a coordination game. In such a game, there is an existing convention, a way of doing things, that is practiced by a large group. While that convention may no longer be the ideal way of doing things, no one individual can move away from the convention and benefit, because unless everyone else does. So, to steal their example, suppose we have three scholars in a field. There is the legacy journal, with publishing practices no one likes, and a new journal appears. The new journal is cheaper and it has the potential for more circulation, so one author considers publishing in it. However, if the other two stick with the legacy, the legacy will still be preferred by most readers, academic libraries will only buy the legacy if they have to make a choice, and that first scholar's work will actually receive less attention. All three would have to jump ship. Now change 3 to however many scholars there are in a given field, and you see how difficult this can be.

So, I bring all that up to give some context to my interpretation of this blog post, detailing how Keven Smith accidentally published an article at a Taylor & Francis journal, whose author's rights policies he disapproves of.  Smith, as the Scholarly Communications Officer at Duke, is exactly the sort of person you expect to be making careful, informed decisions on what publishers he chooses to publish with, and no doubt he usually does. The story he spins on how this publication came to be is not, I think, the standard method one usually goes about getting something published. Still, he draws many good lessons from this experience, one that really stuck out to me: "authors choose journals, not publishers."

Scholars, when determining where to publish, don't think about who the publisher is but whether the topic they are writing about fits the journal they are writing for, and this is what Smith considered when he signed. He thought his work would be a good fit for the journal. He's almost certainly right, as well: Serials Librarian, T&F journal in question, probably is "a proper venue" for a write-up of the talk he gave. Why? It's where there are readers interested in these topics. Which makes me think, why do authors choose journals, not publishers? Probably because readers choose journals, not publishers. Bergstrom and Bergstrom stated that readers choose journals because quality scholars publish there, but I think it's also true that quality scholars choose journals because there are many readers. It's mutually reinforcing.

If I am honest about my own inchoate practices as a scholar, as I determine what I should be reading to keep track of my interests I don't consider who is actually publishing these journals. I care about what is in them. I am an OA advocate and talk a big game about it, but it in no way changes my research practice. I don't prioritize finding information in an OA journal. I prioritize finding information in the best journal, and OA doesn't factor into my definition of "best." As a budding librarian I care about publishing practice, but as a budding scholar I really only care about best information,* and sometimes (often) that information is in journals with publishing practices I don't approve of. In thinking this way, I'm helping to keep the convention of the coordination game stable. As long as I, and my fellow readers, keep reading journals without interest in how they publish, what incentive is there for authors to try to disrupt the convention?

* Actually, as a librarian I only care about publishing practice when advocating for new publishing practices. When doing reference or information literacy instruction, I don't care about publishing practice when I guide patrons and students towards resources. I don't think I'm unusual in this. The only one of the ACRL Information Literacy Competency Standards that could be used for considering publication practices is Standard Five, maybe, if you squint at it and turn your head sideways a little bit.

Bergstrom, C. T., & Bergstron, T. C. (2006). The economics of ecology journals. Frontiers in Ecology and the Environment, 4(9), 488-495. Retrieved from

Sunday, April 21, 2013

Google Scholar

I was reading a 2010 article from Brian D. Edgar and John Willinsky surveying journals published on the Open Journals System. These in many ways are the sorts of journals that the new, digital, OA order makes possible: small, non-commercial, run by dedicated scholars, international, interdisciplinary, investigating new areas of inquiry. It was interesting stuff, but one quote, basically made offhandedly, stuck out to me:
These journals owe a debt, if perhaps more so than other journals, to the indirect support of Google Scholar for indexing the contents of these journals on publication, making them open to discovery on the same grounds as other journals, while providing readers with a degree of quality control, through Google’s page ranking and the citation counts it provides in its search results.
This is not the first time I've seen this sentiment come up. Actually, over the course of my scholarly communications class, it has been brought up in article after article how OA journals and IRs get most of their traffic from GS. This...may not be a good thing. First, is GS really going to be around in a few years? GS is currently not providing ads, and I'm presuming that the information Google learns about its users from GS are not really helping it all that much with its main money-makers, ads for the finance & insurance industries, retail, and tourism. Perhaps it's not surprising that after they announced Google Reader was dead, there was a fair amount of speculation that Google Scholar was next.

Still, let's assume that Google feels like keeping GS around for a while. Should the scholarly community, from scholars to librarians, really be content with reliance on GS? I do not think so, and I think why is well summed up by Charles Potter (2008): "We should not be fooled that any technology used by Google (or any search technology, really) is a neutral force in the information seeking process." (10)

Potter then goes on to lay out multiple ways in which Google Scholar is influencing the search process in ways that we, as librarians, generally do not approve of. Google is, of course, an advertising company, not an information company (that Venture Beat article linked above notes that Google makes 96% of its revenue from ads). To sell these ads, Google data mines, and then it keeps its information about you basically forever. Librarians would probably consider many of Google's practices to be in violation of our Core Value of Privacy/Confidentiality if we were doing them, and yet so many of our libraries link to GS. Google, whether through the main page or GS, is also dependent on a mass-model view of information in order to mass advertise. The single search bar is built on the idea that our information needs are homogeneous, that you can just passively put in a couple keywords, and that Google, through its algorithm, can find your information for you (Go ahead, try to find the Advanced Search). This is anathema to what librarians do: we try to meet the specific needs of our users, and we empower them to control and adjust their search results. 

In addition to Potter's reasons, there is also, of course, the infamous filter bubble. For those unaware, the filter bubble is the idea that Google (and other online services) are personalizing the information they deliver to you by your biases, so that potentially a searcher would never find information that they disagree with. The actual extent of Google's personalization is debated, but that actually helps the argument: Google is a black box. Nobody knows what they are doing, what they are planning to do, and what they could do in the future. Do we really trust them to do the right thing? 

I don't think so. We need a DPLA for Open Access materials. The White House OA memo is going to put a lot more OA material out there. Let's not make it a trade from the monopoly of the current scholarly commercial giants to a monopoly of Google.

Edgar, B. D., & Willinsky, J. (2010). A survey of scholarly journals using open journal systems. Scholarly and Research Communication, 1(2). Retrieved from

Potter, C. (2008). Standing on the shoulders of libraries: A holistic and rhetorical approach to teaching Google Scholar. Journal of Library Administration, 47(1/2), 5-28. doi:10.1080/01930820802110563

Tuesday, April 9, 2013

Institutional Repositories: What Are They Good For?

Edit: all of the following presumes that we define success of IRs in terms of holding peer-reviewed scholarly works, as Steve Harnad does. Whether that is a good idea or not is its own argument.

Discussing institutional repositories yesterday in my scholarly communications class and how to convince faculty to archive in them, a question was posed: how do you convince a faculty member that submits their work to gold OA publications that participate in CLOCKSS to self-archive in the IR? Is there a good reason for them to?

Well, to answer that question, we need to look at the benefits of a IR versus that of said publication. Cullen and Chawner (2011) identify four key aspects of scholarly communication: registration (who  does this belong to?)*, certification (is this any good?), awareness (how do we let others know about this?), and archiving (how to we ensure this is always available in the future?). A traditional publication does all four of these things to varying degrees of success: they identify the author, they certify the work through peer review, they raise awareness through marketing the publication, and they keep back-copies, though they really rely a lot on research libraries for archiving. When librarians and others who promote and maintain IRs stress the value of them, we generally focus on awareness and archiving. We increase awareness of the publication through making it freely accessible in the IR with a permanent link and searchable online, through standards such as OAI-PMH, and build trust among scholars that we will maintain those documents online (Kim 2011).

But, Cullen and Chawner (2011) find in their research that scholars are considered more with registration and certification than awareness and archiving, which are two things that IRs do not do at this moment. Now, let's compare it to that gold OA publication: it has the registration and certification (peer-review) part down, and it compares favorably with the IR on awareness and archiving. Like the IR, it's online, for free, with a stable link, and is indexed by search engines, making it findable and accessible by more than the traditional article. Additionally, with CLOCKSS it ensures that even if the publisher goes under the article is preserved. It does it all in one. How can the IR compete with that, especially long-term?   

Well, we could draw on the idea of the decoupled journal championed by Priem and Hemminger (2012). Putting aside the good objections with their ideas on implementation and looking at the idea in the abstract, they identify functions of scholarly communication nearly identical to Cullen and Chawner: archiving, registration, dissemination, and certification. IRs have the archiving down, as well as part of the dissemination through their metadata. Now, why would a scholar want to disengage their certification from archiving and dissemination?** Priem and Hemminger hope for an increased diversity of certification/assessment under a decoupled system, and I agree that there is a lot of potential benefit here. A multidisciplinary work could be certified by groups in multiple disciplines. There could be certification as to how well the work follows accepted research practice, as well as those for "significance." Those representing traditionally disadvantaged perspectives could have certifications for disciplines, so a mainstream economics paper could be evaluated by feminist economics as to how well is avoids andocentric ideas. There can be anonymous and open review. Some of these the author would submit to, while others may take the Journal of the Digital Humanities model and go looking for things to certify. Thus, the object could investigated and certified from a variety of perspectives, allowing for a more nuanced understanding of quality than now (where something is either accepted or rejected). The object in the IR could display the varied certifications that the scholarly work received with it, perhaps using an open badge framework, so while the IR does not certify, it archives certifications.

Cullen, R. and Chawner, B. (2011). ' Institutional Repositories, Open Access, and Scholarly Communication: A Study of Conflicting Paradigms.' The Journal of ACademic Librarianship 37:6, 460-470. DOI:10.1016/j.acalib.2011.07.002

Kim, J. (2011). 'Motivations of Faculty Self-Archiving in Institutional Repositories.'The Journal of Academic Librarianship 37:3, 246-254. DOI:10.1016/j.acalib.2011.02.017

Priem, J. and Hemminger, B.M. (2012). 'Decoupling the Scholarly Journal.' Frontiers in Computational Neuroscience, 6, 1-13. DOI:10.3389/fncom.2012.00019

*Cullen and Chawner, as well as Roosendaal and Geurts, whom they borrow this framework from, never do a particularly good job of explaining what they mean by registration, other than in involves intellectual property and copyright, so I'm going to focus less on it.
**Remember, I'm still not touching registration.
Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 3.0 Unported License.