SFB Blog “Transformation of Property”


The SFB Blog “Transformation of Property” provides an interdisciplinary public platform through which we share thoughts on property relations in contemporary and past societies. We offer a wide spectrum of views, comments, interpretations, and opinions on all property-related topics from the perspectives of history, law, economics, politics, sociology, and anthropology. We also present and discuss alternative property arrangements and institutions, including indigenous or decolonial perspectives. The blog aims at both an academic community and an interested non-academic public and welcomes wide engagement with our property research from academic and broader public.

 Please send contributions for the blog to: blog-sfb-eigentum@uni-jena.de

Michael McCullough. Source: Flickr, License: CC BY 2.0
Hakan Dahlström. Source: Flickr, License: CC BY 2.0

How the transition to Open Access is changing scientific publishing

by Tilman Reitz

Free and unrestricted access to scientific texts is likely to become standard soon. Scientific publishers are trying different approaches to take advantage of the transition to Open Access, for example by charging publishing fees and tracking user data. Tilman Reitz analyzes what the Open Access transformation means for science and what options there are for shaping it.

Open Access is coming, albeit slowly. Even during the pandemic, where the need for digital text access in research and teaching has become essential, there has not been much significant progress. Recent copyright reforms have seen the criminalization of platforms such as Sci-Hub, designed to skirt paywalls, and yet a legal exception for the purposes of science (i.e., the lifting of copyrights for academic use) has been only marginally established. Currently, only up to 15 percent of a published work may be used unpaid or without publisher approval for research and teaching.
Scientific practice, which would require full access to the overview of existing research, has not been taken into account. Even the approach of requiring the results of publicly funded research projects to be made publicly accessible is still far from institutional realization. 
The EU's "Plan S", which pursues this goal, is not being adequately supported by organizations such as the German Research Foundation. Thus, in large parts of the market, academic publishers are able to cash in twice off research, as described by Leonhard Dobusch:

"Predominantly publicly funded research is [...] peer-reviewed for free by publicly funded scholars and then sold back to publicly funded libraries for expensive money by publishers whose service usually consists of editing, typesetting, and distribution."

In the course of digitization, this problem has expanded and worsened into a "journal crisis" for university libraries, which are running out of money for necessary purchases in the face of rapidly rising subscription prices. Many researchers, universities and research organizations have fought back against this; in the meantime, the leading nations in research are negotiating transitions to Open Access with the major publishers.

The fee burden shifts to authors

But even if scientific publications do become generally accessible in the medium term, the economic scandal of the old paywall oligopoly will continue. Instead of access fees, publishers now earn money from Open Access by charging publication fees, which usually go far beyond the services rendered and can be demanded directly from the authors or their institutions.
Those at fault here are, on the one hand, scientific organizations and science policy, which simply aren’t taking a stand against the market-dominating publishers – and, on the other hand, the reputational needs by which researchers themselves are governed. When publication in a highly cited journal – ScienceNature, or even The British Journal of Sociology – is the ticket to professorships, proposal approvals, and government support for individual institutions, many will pay dearly for the privilege of publication.
The greater the prestige of a magazine, the more inflexible the demand to publish with them, excessive prices notwithstanding. Top publishers with market-dominating positions can easily charge very high prices. As a result, Open Access is allowing an open exploitation of researchers and their institutions, or the explicit extraction of monopoly rents. Critical reactions to approaches such as Plan S argue that funded (early career) researchers should not be prevented from placing their results in well-known journals.   Thus, publication fees are gaining acceptance. However, despite the name "Article Processing Charge" (APC), publication fees have little to do with actual costs. According to calculations by Alexander Grossmann and Björn Brembs, prices of 200 to 1,000 US dollars per article would be sufficient (about 168 to 840 euros; editor's note). However, the standard asking price is currently around 2,000 euros, and leading journals are often well above this. Nature, for example, charges up to 9,500 dollars per article.

What it means for science when Open Access becomes the standard

What can we expect for publication conditions in the medium term if Open Access becomes the norm? First, it’s clear that the new model will coexist with the paywall system for quite a while, as science organizations and policies are negotiating individual contracts with major publishers instead of aggressively setting new standards. The German science organizations have already reached agreements with Springer and Wiley through the Projekt DEAL initiative; Elsevier is still proving particularly resistant.  
Second, these publishers – namely Elsevier, Sage, Springer Nature, Taylor & Francis, and Wiley-Blackwell – will use the long transition and their privileged bargaining positions to consolidate their market power. With proliferation of Open Access in academia, publishers will be increasingly unable to persuade libraries to purchase expensive aggregate journal access portfolios. Rather, they will either charge extremely high APCs for the right to publish or combine these APCs with considerable standard prices for accessing their journals.  
In the initial DEAL agreements, there is an interesting mixed model: Instead of APCs, publish-and-read fees ("PAR fees") of 2,750 euros cover the costs of publishing new work as well as grant access rights to a selection of the publisher’s existing catalogue (see, for example, the DEAL contract details for Wiley and Springer).  
Smaller publishers, however, do not have the bargaining power to negotiate such advantageous transitions. And even after the transition to Open Access, they will probably still be unable to collect high fees. Thus, the problems of monopoly pricing and the marginalization of smaller academic publishers look likely to continue.
There are other threats as well. Universities in the Netherlands have succeeded in negotiating an Open Access package with Elsevier. However, in doing so they made major financial concessions and also granted extensive rights to use data for new business models (described here only in outline, ranging from metadatabases to career support).  
Anyone who thinks of a high-profile, carefully managed catalogue of books and periodicals when they hear the word "publisher" will have to reconsider: In the future, scholarly publishers could turn into data brokers who see publications only as the raw material to profit off user data.  
The emerging new structures are rather advantageous for smaller academic institutions and their researchers in countries such as Germany or the UK, where publication fees are borne by the entire system and thus a poorer institutional position does not limit publication opportunities.  
At the same time, however, researchers from less financially strong countries might find it increasingly difficult to publish their results with any visibilty. And those who work outside the organized academic establishment can hope less and less for the opportunity to bring their results to the scholarly public at all. Such groups will be able to afford reading more and more, but (without individual funds) will be publishing less and less.

Alternatives to the publishing oligopoly – and who could drive them forward

If the prospects are so unfavorable, the question naturally moves to what alternatives there are and who could possibly lead the way. What should be done is not difficult to say: In principle, the transition to Open Access needs to be significantly accelerated, including introducing a legal exception which would make copyrighted material generally free of access under certain conditions (e.g, non-commercial use for research and teaching).  
At the same time, fair criteria for publication fees should be established and these fees capped accordingly. More than 1,000 euros per text cannot be justified. In this way, the public sector would save a lot of money, which could then be better spent (for example, to finance more permanent academic positions). Moreover, publishers beyond the market-dominating five would get a chance to assert themselves in an open price and performance competition.  
It is much more difficult to foresee who could drive such a change. Many researchers in the mathematical and natural sciences have been committed to Open Access publication structures for 20 to 30 years, but their bottom-up initiatives have been slow to bear fruit at best. In the humanities and social sciences, on the other hand, there is still no widespread awareness of the problem or potential for action. This is due partly to their prevailing book culture, and partly to a quiet competitive conformism.  
Here, self-organized Open Access journals and forums are the most likely to take shape; Kathrin Ganz has pointed out that fee-free publications are prevailing in these subjects. In order to not only encourage such approaches, but to generalize them, science organizations and legislators in particular need to get involved. However, they will need to experience much more pressure from below in order to take action and change the situation. Otherwise they will remain on the chosen path of maintaining the oligopoly.  

This post originally appeared in German on irights.info licensed under CC BY-4.0. It has been translated and republished with permission of the author.


Nico Kaiser. Source: Flickr, License: CC BY-2.0

Academic piracy

Researchers and students sometimes take illicit paths to access scientific texts

by Georg Fischer

Over the past two years I have visited my university library only once: the day I had to return a book that I borrowed in January 2020 before the pandemic restrictions began. My university library refers to the library of the university where I graduated some time ago – and where I have been an alumnus ever since. With alumnus status, I can use the services of the library like I did previously as doctoral student – for research, loans, interlibrary loans from other libraries, copy and scan orders, etc. This is a great help, almost a necessity in my scientific and journalistic work. Especially before the pandemic, I made active use of these offers. But what the university library could not fully do even before the pandemic was to provide me with all desired essays, anthologies and monographs at short notice.

Between intellectual desires and material possibilities

As a sociologist and journalist interested in both history and the present, I would like to look at offsite historical sources as well as new publications. I want to identify trends and turnarounds and new fields of research. I want to do targeted research, to stay up to date. I want to know what colleagues publish and where citation cartels are formed. I want to check sources and see what else can be gleaned from them. And, sometimes, I want to drift in the ocean of references without having a goal, to lose myself in rabbit holes, to descend into footnote cellars.
In short: I need open access and a wide selection of scientific literature. But I do not, or cannot, spend 30 euros or more on every book chapter or essay that could be interesting – and which I want to view beyond its abstract – as provided for by the (digital) offerings of the major academic publishers.
This gap between my intellectual desires and material possibilities is a problem. I became aware of the problem during the researching and writing of my dissertation between 2014 and 2018. While working from home due to the pandemic, the problem continued to worsen for me. Colleagues related similar difficulties. Almost all my colleagues who study, research, teach, or are unable to access the physical library seem to be affected.

A fantastically profitable business model

Simon Frith wrote about the music industry at the end of the 1980s:
For the music industry the age of manufacture is now over. Companies (and company profits) are no longer organised around making things but depend on the creation of rights. […] [T]he company task is to exploit as many of these rights as possible […].
For the academic media industry (i.e., the major scientific publishers) this insight is almost ideal. Elsevier, for example, achieves revenue margins of more than 30%. This is now also perceived as a problem in science itself.
The basis for this is a business model that makes maximum use of the copyrights of the published authors: scientists write academic texts, in most cases ensure their quality, and give them to scientific publishers who formally prepare these texts – and then sell back access to the texts to the libraries (or to interested individuals) by means of subscription models.
Unlike the music industry or the newspaper industry, for example, the academic media industry was able to transfer its business model into the digital age relatively unscathed. For some years now, however, cracks have been forming in the walls that scientific publishers have built around the texts they market. Similar to the MP3 crisis of the music industry, users are applying digital tools to overcome the paywalls of publishers.

Designing accesses, laying tunnels, overcoming paywalls

Among these users are not just researchers, students and teaching staff, but also journalists and the interested public who wish for open, free access to scientific findings. To this end, they use resistant, sometimes guerrilla-like strategies to circumvent, undermine and open up the (digital) restrictions of the major publishers.
The best-known example is probably the shadow library SciHub (short for Science Hub), which currently provides access to more than 75 million documents. It was founded in 2011 by Alexandra Elbakyan, who was frustrated that she could not access the required texts from her place of residence in Kazakhstan. Elbakyan decided to program an automatic circumvention of restrictions; in order to obtain a certain text, SciHub pretends to be a library that has already acquired access to this text. SciHub tricks the publisher's website into believing an IP that belongs to the library in question. The text becomes accessible via this digital tunnel and the user can download it as a PDF. Access granted.
To ensure the highest possible availability, SciHub works together with the Russian shadow library LibGen (short for Library Genesis). Beginning in 2008, LibGen aggregated various collections of texts circulating in Russia and put the entire corpus online. In 2014, LibGen offered about 25 million documents, including scientific literature and works of fiction in various languages, resulting from mass downloads of repositories and leaks from university networks and publishers.
Together with LibGen, SciHub is an example of a technically delegated, automated circumvention strategy allowing users to circumvent the copyright issues of accessing literature. This gives them flexibility in the short term but does not solve the underlying problem in the long term.
The whole thing remains a cat-and-mouse game: the sites have to regularly change their top-level domains to escape the access of governmental authorities. Instructions on how to bypass the locks circulate quickly.
Of course, shadow libraries are a thorn in the side of the major publishers – so bad that they have even considered having the university libraries install surveillance software in order to be able to track (illegal) access from university networks. A paper of the German Research Foundation (DFG) recently described the practice of data tracking and addressed it as a problem.
Following a lawsuit by several major publishers, SciHub did not add any new texts to its database for a few months in 2021. On Reddit, a rescue operation has formed for almost 80 terabytes of scientific texts, which should remain available for users via the file sharing system BitTorrent. In addition, the programmer Elbakyan has been targeted by governmental authorities: According to Elbakyan, her Apple account, for example, has been monitored by the FBI for the last two years. In September 2021, SciHub celebrated its ten-year anniversary by uploading more than 2 million new articles.

From peer to peer

SciHub and LibGen are well-versed technical solutions based on a central distribution principle. In contrast, decentralized peer-to-peer-practices have also established themselves in social media.
Under the hashtag #IcanhazPDF users make search queries for scientific texts. Colleagues with access to the desired texts can see who is looking for what and help. The associated Twitter account formulates an etiquette for such search queries containing three rules:

  1. DOI link
  2. Your email address
  3. Delete tweet once PDF received

The third rule in particular serves to protect sources and to prevent possible warnings. In most cases, sending a text violates the copyright of the publisher – although the authors are allegedly less likely to object to it, as they benefit from their work being known, read and quoted by colleagues.
The practice around the hashtag appeals to the community idea of globally networked science and entails a particular principle of the exchange of gifts: the greater the willingness of individuals to provide texts, the higher the general chance of receiving the desired text. At the same time, #IcanhazPDF expands the common practice among researchers to recommend and send each other literature by email.
#IcanhazPDF does not set any limit to scientific disciplines or years of literature sought. In using the hashtag, researchers also provide some insight into their own work, exhibit their own consumption of literature, and show which texts they (want to) receive.
A similar practice takes place in the exchange of literature via Telegram groups. The messenger service was originally developed by Russian programmers and has officially relocated its headquarters to Dubai. Telegram sometimes receives public criticism: not only can short messages be sent free of charge, but the service also provides the technical infrastructure for illegal activities, such as drug trafficking, the spread of conspiracy ideologies, or the organization of political upheavals.
Telegram is obviously weakly regulated. And this is one apparent reason why the sharing of scientific literature also takes place within self-organized groups on the platform. Similar to #IcanhazPDF on Twitter, students and researchers make a text request in one of the various groups, often combined with an indication of the library at which the text is digitally accessible.
The sending of the desired texts usually takes place in private messages. Presumably, this is a safety feature so that the senders cannot be prosecuted for making a copyrighted text publicly available to a large audience.
If you are interested in a text that another person has requested, you can express this via text abbreviations or emoticons. This shows that academic piracy is at best legally questionable but serves the general goal of scientific exchange and can lead to substantive recommendations of literature.
Telegram groups arose in part as a reaction to the pandemic situation, which cut many students off from access to scientific literature and thus made a technical and solidarity-based solution necessary. In some cases, comparable offers existed even before the pandemic.

Shadow libraries as a viable but illegal practice

I admit that I not only research copyright, but also break copyright laws from time to time. Without the techniques and offers listed above, which are largely in the shadow of the law, I would not have been able to write my dissertation, for example. And while I'm thrown back into home office because of the pandemic, for me – as for many others – legal access to scientific literature through libraries is deteriorating. For the year 2020, academic libraries in Germany recorded an average decline of 25 percent in the number of loans compared to the previous year.
Regardless of the pandemic, the shadow libraries and techniques described above are an indication that formal copyright rules (and their application by publishers) collide with the actual wishes of users; students, researchers and the interested public want fast, direct, easy and affordable access to scientific literature. This is needed for studying, research, teaching, and pursuing education.
To achieve their own goals – such as writing a scientific qualification thesis – many students and researchers are willing to take advantage of legally-questionable or even clearly-illegal offers. They value the benefits of the illegal offers more than the resulting damage to themselves or the scientific publishing industry. Organizational researchers refer to this phenomenon as ”useful illegality”, by which they mean the various practices, strategies and mechanisms that are illegitimate or illegal, but very necessary to keep an organization running and give it flexibility in the application of formal rules.
Many people may also see no point in paying out of their own pockets for access to already-publicly-funded research. They bypass the helpful and absolutely necessary, but usually insufficient offers of their university libraries and facilitate new paths that help them better reach their access goals. In the case of peer-to-peer exchanges, users sometimes slip into the role of librarians themselves to do their own colleagues and fellow students a favor, helping to obtain the desired literature and, if necessary, to recommend more. A para-librarian structure is thus created and consolidated.

Desire paths: Not seen as part of the problem, but as part of the solution

Bypassing hurdles relates to a phenomenon that urban planners call “desire paths”: These are defined as “paths and tracks made over time by the wishes and feet of walkers, especially those paths that run contrary to design or planning”. Urban planners have recognized the potential of such desire paths: for example on the university campus in Michigan where users were explicitly allowed to create the paths they wanted between buildings. Later, these organically generated trails and sneaky backways were developed into official paths.
With digital ways of sharing academic texts (such as SciHub or #IcanhazPDF), desire paths also emerge in the digital sphere. They provide shortcuts between the official structures of libraries, albeit through informal, illicit, or even illegal means. The music industry was the first media industry to be turned upside down by the media break of digitization and increased copying possibilities of users; user-generated side paths became the main arms of digital music distribution that could no longer be ignored.
After the music industry initially fought hard against MP3s, illegal file sharing was gradually transformed into a legal business model. iTunes and Spotify appeared as external players with convenient digital offerings, and labels and publishers had to move. Gradually, the understanding spread that MP3 streaming on platforms could fulfill many users' desires for easy and fast access to music – a convenient shortcut to the ways of the CD business. In addition, it activates fans as resources; they can recommend music to their friends, curate in playlists, rate, collect, comment, and pass on music on social media.
In science, a similar process seems to have started with shadow libraries. This is certainly worth considering not as part of the problem, but more as part of a solution that solves an underlying problem in the procurement of scientific literature. For this, the wishes of users for a useful and legal system for obtaining scientific literature would have to be taken much more seriously – and not fought as copyright infringements.

This text was originally published in June 2021 at Verfassungsblog. For the SFB Blog, it was translated into English and slightly adapted. Both original and adapted version are licensed under CC BY-SA-4.0.