The JJPS Extension: Presenting Academic Performance Information
The current state of academic publishing is untenable. From the high cost of journals, to the proliferation of book and journal piracy, the business of scholarly communication is, to use a word that is in vogue these days, in “crisis”. Things must change, but entrenched interests with their resistance to radical reconfigurations of the system are making potential modifications difficult, the efforts of many notwithstanding.
The JJPS Firefox extension is my way of engaging with these issues. Overidentifying with the situation, the extension works to overlay an abundance of information on publisher websites, including bibliometric data, graphs of journal ownership, journal prices, popular words and their value, and publisher stock quotes and headlines. The extension is meant to foreground the absurdity of the present situation, and highlight potential directions the industry could go in if enough resistance is not mobilized. As such, I intend the extension to open up a wider dialogue—and performance—of what scholarly publishing can and should be.
Absurd Pricing
The most well-known problem with scholarly publishing today is the absurd prices charged for many journals. This has recently come again to a head with the call on 4 June by the University of California Libraries for a potential boycott of all Nature Publishing Group (NPG) journals because of NPGs intent to raise prices by over 400%—meaning the average journal price would balloon from $4,465 to $17,479. This is by any reasonable measure preposterous, especially given the present fiscal state of the UC system. Additionally, we have to remember that the majority of the labor of academic publishing is done for free: scholars write articles for free (as in they receive no payment for the article), reviewers review for free, and editors of journals work with little or no institutional support. An exemplar of immaterial labor. Journal publishers, on the other hand, often require their authors to sign over most of their rights, and then turn around and sell the product back to institutions at inflated markups.
Open access publishing is changing the scene and forcing the hands of the major publishers. Projects like arXiv, Public Library of Science, Open Humanities Press, re.press, and Open Journal Systems (upon which this particular journal is built) are providing much needed alternatives to hegemonic for-profit publishers. Yet Open Access cannot and should not be seen as a panacea. This is apparent when we look at the situation of certain open access publishers charging “fees” for publication—and then sometimes accepting fake articles as a result. (See my accompanying article in this issue regarding interventions into scholarly journals for more on the idea of “hoaxes”.) This is not to damn Open Access publishing by any means; rather, it is to say that Open Access publishing, without a concurrent interrogation of the economic underpinnings of the scholarly communication system, will only reform the situation rather than provide a radical alternative. The projects I just mentioned are attempting to do this, but the use of the phrase “Open Access” has unfortunately become as well-worn as the phrases “Open Source” or “Creative Commons”.
Thus the first intention for the JJPS Extension was to present information regarding institutional journal prices on the publisher websites themselves. Presented in a marquee at the bottom of the page, the pricing information, where available, also includes details about the web of relationships of academic publishers. Major businesses such as Elsevier, John Wiley & Sons, Taylor & Francis, and Springer jointly own thousands of journals, a major portion of the more than 16,000 journals I draw from in this project. The pricing information is one way to gauge the enormity of the situation. By summing over all of the pricing information I have (and I only have this information for the major publishers), it would cost institutions over $6 million a year to purchase all of them at their published cost. (Indeed, bigger universities are able to negotiate substantial discounts on bundles of journals, something that is at stake in the UC issue mentioned above. Nevertheless, this again leaves smaller institutions at a disadvantage.)
Networks of Ownership
Mapping which journals are owned by which publishers is an important step in understanding the centralization of the industry. Viewing the abundance of journals owned by any one of the Big Five publishers (Elsevier, John Wiley & Sons, Sage Publications, Springer, and Taylor & Francis) is spectacular. Thus the JJPS extension makes these graphs accessible while browsing. The density of black in the images makes the centralized nature of these publishers clear: there are too many journal names to print legibly.
Yet the graphs in the extension show only one of the subsidiaries of these companies. An organization such as Elsevier might have upwards of ten or more different companies owned under one umbrella. (And Elsevier itself is only part of a much larger media empire, Reed Elsevier NV .) Thus one of the ongoing goals of the JJPS project is to provide legible maps of these publishing behemoths. I present the first iteration of these images below; clicking on the thumbnail will go to a much larger version. (Beware! These images are large, on the order of 10MB each.)
Elsevier
John Wiley & Sons
Sage Publications
Springer
Taylor & Francis
The Impact of the Impact Factor
Bibliometrics is the practice of quantifying bibliographic data, often for the purposes of ranking and measurement. One of the most well-known developments of the field is the impact factor, a value that attempts to quantify the “importance” of a particular journal within a field. While the IF is widely derided and critiqued (as it can be easily manipulated, among other reasons), it is still used extensively, not only by publishers interested in advertising the strength of their journals, but more insidiously by tenure review committees in the evaluation of the “strength” of a particular candidate’s papers. Indeed, Elsevier provides extensive information on their website for working with bibliometric values. The IF and affiliated numbers (such as the Eigenfactor and Article influence score) are, perhaps surprisingly, copyrighted—so the JJPS Extension cannot present them to users, directly that is.
The practice of the quantification of the unquantifiable has lead to a proliferation of other factors that are available to the public, such as the hIndex (as applied to journals) and the SCImago Journal Rank. These last two values are presented to users of the JJPS extension, as well as some specialized ones based on our analysis of the influence of Google; see the section below for my reasoning on this matter.
Bibliometrics might seem like an esoteric science, but people use it everyday by making Google queries. Google’s PageRank algorithm is nothing more than an expansion of what is known as citation analysis. Instead of trying to measure the importance of a particular article by way of the articles that cite it, PageRank measures the importance of a particular site by way of the sites that link to it. Thus the science of bibliometrics is intimately tied to the practices of knowledge production under Google. Understanding how bibliometrics has developed in recent years might provide insights into how Google’s algorithms might develop—and in turn, suggest ways to intervene.
The Influence of Google
Any discussion of the contemporary state of scholarly communication would be incomplete without an analysis of the influence of Google. Already libraries are changing their infrastructure in order to better align themselves with the habits of Google-fed youth. Indeed, you might have noticed, if the switch has happened already at your institution, that gone are the faceted search options on your institution’s library search form—that is, searching by title, author, call number, and so on. In its place is federated search or metasearching single box where free-text keywords can entered and a multitude of databases searched at once. This replacement is meant to model the single search box, single button interface of Google, more familiar to most library patrons than parametrized search. (For an example, see the Cornell Libraries website .)
But this hyper-attention to Google and its algorithms is not limited to the interface. Indeed, journalists are already changing how they write their headlines. Gone are the days of headlines full of witty puns, such as “The marinating of the ancient rhymer”, as they are not “findable”. Instead headlines become merely a precis of the article in question, as Google’s natural language processing (NLP) algorithms no nothing of sarcasm, punnery, or wit.
Lest we think that “findability” has not become a concern for scholarly writers, there are some who are already developing what they call Academic Search Engine Optimization in parallel to the already-existing practices of Search Engine Optimization (SEO), specialized techniques—sometimes quite sketchy—to ensure that one’s page returns at the top of Google’s search results. In their article “Academic Search Engine Optimization (ASEO): Optimizing Scholarly Literature for Google Scholar & Co.”, the authors, Jöran Beel, Bela Gipp, and Erik Wilde, describe a number of modifications to one’s writing to ensure that the article appears as one of the first results in a Google search:
- The use of Google’s tools such as AdWords, Trends, or Insights to choose the most popular words
- The liberal use of these words throughout the article, as well as the distribution of synonyms in the text
While this article has generated a notable critique, its central premise is in fact being taken up by others. The company Mendeley has written a bibliography management tool that, through its online capabilities, works to help you 'Discover the hottest papers, authors and topics in your discipline – right now and in real-time’. The developers of the software are unabashed in their belief that a “crowdsourcing” style model of bibliographic management will help researchers by letting them see what is “popular”—a massive about-face to one of the idealistic goals of academic research, which is to engage in practices irrespective of their “popularity”.
Thus the JJPS Extension projects these trends into the future. While sites like Google Scholar do not presently show advertising, affiliated sites such as Google Books do—and indeed, the monetization of gaze is key to the controversial Google Books settlement agreement now winding its way through the courts. The JJPS Extension takes the journal names and their publishers and pulls the top 1000 most common words. It then uses the Google AdWords Traffic Estimator to determine the popularity and price of each word. Google AdWords is based on an auction system; more popular words cost most to advertise with. The Traffic Estimator provides both estimates of the number of times a particular word is going to be clicked on each day, as well as the price of using a word in that advertisement. With these numbers, then, it is possible to rank the titles of journals by their popularity or their cost. Higher cost journal titles, by this metric then, will be more valuable to advertisers; similarly for journal titles with a larger potential number of clicks. With this information it is possible to replace advertisements on journal websites with ads based on these metrics. Finally, the extension itself provides a list of these words through the “Trending” area overlaid at the top of the page.
While such manipulation of the very basis of scholarly publishing might appear as absurd, there is nothing preventing it from becoming reality. The footings have already been laid; all that is necessary is for certain publishers and/or authors to finish building the structure.
Google’s Renting of Our Labour
One of the most cogent analysts of these trends, on a more general level, is by Matteo Pasquinelli. In his recent text 'Google’s PageRank Algorithm: A Diagram of Cognitive Capitalism and the Rentier of the Common Intellect’, Pasquinelli makes clear the bibliometric lineage of the PageRank algorithm, quoting from Sergey Brin and Lawrence Page’s original paper:
Academic citation literature has been applied to the web, largely by counting citations or backlinks to a given page. This gives some approximation of a page’s importance or quality. PageRank extends this idea by not counting links from all pages equally, and by normalizing by the number of links on a page.
Pasquinelli discerns the ultimate implications of this leap:
This bookish genealogy of PageRank should not be underestimated. A similar way to describe value can be applied to any cognitive object and it is native also to the 'society of the spectacle’ and its wild economy of brands. In a spectacular regime the value of a commodity is produced mainly by a condensation of attention and collective desire driven by mass media and advertisement. From academic publications to commercial brands and the internet ranking itself equivalent processes of condensation of value can be assumed.
For Pasquinelli this is a reappearance of the Marxist concept of rent: rent of our time, rent of our efforts, rent of our intellect. Value accumulation occurs less through the production of physical commodities (although that is still necessary to provide the hardware for the digital services), but rather through the activity of communication. Scholarly work is not immune to this, as the conflicts over journal pricing make clear. In Pasquinelli’s analysis, “PageRank is to the internet, as primitive accumulation and rent are to early capitalism.” The immediate question, then, is not how to live without Google or to develop an alternative in its stead. Rather, and this is all-the-more clear in the case of scholarly work, it is how to hack the system of value production itself, to create new metrics that are not so easily monetized and that are more under the control of the labourers themselves, rather than the landlords.
Journal of Journal Performance Studies is a 2009 commission of New Radio and Performing Arts, Inc., (aka Ether-Ore) for its Turbulence web site. It was made possible with funding from the Jerome Foundation.
Back to turbulence.