Event report - Why Are We Not Boycotting Academia.edu?

My PhD is still in that early shapeshifting phase, where the more I read the less I know, and it has moved away from the proposal that was accepted into something else. However, I DO know that it's about how different stakeholders (researchers, library staff) feel about academic social networks (Academia.edu, ResearchGate) and other research sharing infrastructures versus the 'proper' institutional systems for research information and outputs - CRIS (Current Research Information Systems, which may be a collection of systems or homebrew solutions but often commercial products like Symplectic, Converis and Pure) and institutional repositories.


As my PhD is about attitudes to research sharing infrastructures; this event, with its provocative title, therefore became unmissable. I scrambled to get some funding together with short notice, as I wanted to be part of the debate. 

Janneke Adema set the scene for the event by asking the questions presented in the event blurb:

  • Why have researchers been so ready to campaign against for-profit academic publishers such as Elsevier, Springer, Wiley-Blackwell, and Taylor & Francis/Informa, but not against for-profit platforms such as Academia.edu ResearchGate and Google Scholar?
  • Should academics refrain from providing free labour for these publishing companies too?
  • Are there non-profit alternatives to such commercial platforms academics should support instead?
  • Could they take inspiration from the editors of Lingua (now Glossa) and start their own scholar-owned and controlled platform cooperatives for the sharing of research?
  • Or are such ‘technologies of the self’ or ‘political technologies of individuals’, as we might call them following Michel Foucault, merely part of a wider process by which academics are being transformed into connected individuals who endeavour to generate social, public and professional value by acting as microentrepreneurs of their own selves and lives?

Adema suggested that Academia.edu and its rivals are committed not to Open Access (OA) but the business model OA has become. They do not currently charge for 'publication' on their platform, unlike those publishers asking for high APCs (article processing charges) but are experimenting with data (though their business model is currently based on venture capitalist funding and recruitment services, not data exploitation) and services around scholarly communications. Academia.edu's CEO, Richard Price, has indicated that they are interested in being publishers themselves – perhaps for APC income. 

I found this amusing, as Elsevier, who are in battle with the service over the sharing of copyrighted material and charge some of the highest APCs, are currently investing in social (Mendeley) and other services around scholarly communications, including research information and repository software (Pure).

Academia.edu promote themselves as part of the OA movement. Academics think they are doing Open Access, or part of OA, by using social networks to upload and share their work. I think it's important to note here that there has been some 'fightback' on this point recently from librarians, posting links to Kathleen Fitzpatrick's (of which more later) recent blog post, Academia, Not Edu, and pointing out that sharing via these platforms does not count for compliance with institutional and funder OA mandates.

The tone and length of some of these articles and blog posts means they are likely to fall on deaf ears outside the enthusiastic retweets from the library and OA advocacy communities. Additionally, ranting about the ownership model doesn't work when many universities and academics happily use services from venture capital driven companies, such as Google Apps or Facebook, without even beginning to examine who owns the software running the university research systems. The biggest players in CRIS platforms are owned by some of the biggest players in the academic publishing market, whose bad behaviour is also used to sell the benefits of doing Open Access 'properly' -  Elsevier (Pure), Digital Science (Holtzbrinck/Macmillan) and Converis (Thomson Reuters).

Back to Adema's talk. She rightly points out that Open Access is, to its keenest advocates at least, about more than access to free content. It is a radical alternative to current models, addressing related issues of the ethics and politics of publishing. However, this does not change the fact that OA is currently outflanked by Academia.edu, in volume of articles shared and brand recognition, at least.

Pascal Aventurier from INRA spoke next, from a librarian's perspective.  INRA is the French institute for agricultural, food, environmental scientific research. Prodinra is their repository. Visibility in Google Scholar is often higher for content posted on ResearchGate or Academia.edu than for repository copies of the content - this is feedback I have often heard from academics. Aventurier gave us a basic overview of how academic SNS are used, which I will not repeat here.

According to a 2014 Couperin study (the report is in French, I see translation in my future),  86% of French researchers don’t know what the official policies are relating to self-archiving and OA, they just upload their articles to social networks like Academia.edu and ResearchGate without caring about copyright.

Social platforms' marketing on improved citation count is better than anything librarians come up with, and visibility beyond search is better. However, these sites have poor query functions, you can’t export data from those services, and metadata and facets quality and usability are variable. 

Richard Price (Academia.edu's CEO) has an algorithm dream to tell industry which papers are most impactful within a specific time period. So, Aventurier tells us, papers provided are for free but a service like that, built on free labour, would not be free forever.

As a librarian, he suggests researchers deposit in institutional repositories for digital preservation and the link to CRIS/research assessment systems, but put links and abstracts in Academia. However, it was not mentioned that if you do this, you don’t fetch up in the automated emails of interesting new papers in your field and are more likely to frustrate other users in searches (WHY NO FULLTEXT? SHAKES FIST) rather than drive repository traffic.

We are shown a slide full of logos, explaining there are lots of tools that fit under the banner of Open Access and related concepts – ORCID, PubMed, DOAJ, SHERPA, OpenAIRE etc. This took me back to a thought that has recurred for me at many 'Open' events in the past - there are too many tools and too many platforms. Researchers already complain about the labour involved in maintaining multiple logins and having to use multiple services, whereas for academic social networking there are only a couple of main platforms, each offering a range of services, and researchers are drawn to the ones that suits their workflow and academic community best. At this point in Aventurier's talk, as service after service flashes by on the screen, I end up writing "no1 curr" in my notes - if that's me, as someone fully engaged with Open and enjoying new and shiny, saying that, what does your average researcher think?

705 of 1698 participants in the Couperin study used academic social networking services (SNS) and the vast majority reported that they were satisfied or very satisfied, criticising  the seeming lack of permanence and digital preservation, query functions etc. The criticisms feel more like librarian problems - anecdotally the complaints I hear from researchers and read in comment threads are completely different, but then I guess I'll find out in my own research. 

For example Academia.edu's new PageRank part of the service, where academics are asked as unpaid 'experts' to rate papers and make search rankings more authoritative, has met with resistance from more established researchers and enthusiasm, perhaps due to flattery, from PhD students. Whereas suggesting there is no preservation or export function in academic SNS, so they could die tomorrow taking with them all their data, meets with a shrug from anyone I speak to - just like my mum does not believe Facebook will collapse and her photos will disappear.

Aventurier's suggested solutions: use open science tools,  have an ORCID (it's a unique identifier - like a DOI - for researchers, guys, I've had one for years), blog about your research, deposit in the institutional repository.

Kathleen Fitzpatrick from MLA was next. While she is always an interesting speaker, my notes say "reminiscent of a vendor talk", and I think that's fair.

Fitzpatrick admits that everything in her life is based on Google, Slack and Twitter, so she is not in her scholarly or personal life avoiding venture capital funded platforms.

Academia.edu brings with it the primary benefits of start-up culture – democratising image,  free to the user, no regard for employment situation or financial circumstances. Fitzpatrick argues that this makes the precarity of its users worse rather than better – the venture capital business model means the service is provided free to end users, but is limited to a number of potential futures.

The platform has to turn a profit or be sold (to another service or dismantled for parts) or shut down. Option 1 is OK. 2 – who might the buyer be and why are they interested? (Mendeley – bought out by Elsevier). Small and scrappy is rarely subsumed by small and scrappy. Services are subsumed or dismantled by “the monolith they sought to disrupt”. If they are shut down or are bought for a specific part of functionality – the position of scholars precarious as they have no control over where their data, networks etc go.  Academia.edu closure would most hurt the precariously employed, who have nowhere else to go and then nothing to show for the time invested in the platform.

The next part of Fitzpatrick's talk is key to some of the things I've been writing about lately regarding my choice of sample for my PhD research. If you move institution, and feel closer to your field than your institution, how does your stuff move when you move around? The siloed nature of institutional repositories and CRIS is problematic, yes even with harvesting protocols like OAI-PMH to try to help move data around (especially as Elsevier have no interest in making Pure compliant with this or other shared protocols and services). Institutional repositories are great for preservation of work within an institution but not great for the development of a field or sharing within a field. Popular (and non-profit) disciplinary repositories with fare better in the sciences than in arts, humanities, social sciences, professions and interdisciplinary work.

MLA has a repository for its MLA Commons members called CORE (which is confusing, as the aggregator for all UK repositories has the same acronym and common name), which based on FOSS software. Here came the big plug for this paid-for option. It's a member benefit. Searching, reading, and downloading are free – creating and depositing aren’t, because you have to pay for MLA membership.

There has been slow uptake of this option – MLA Commons has been around for just under 3 years and has 5500 registered members. This is a fraction of the total MLA membership, which again is a fraction of the field.

'Free' stops us boycotting Academia.edu – everyone is there. Scholar participation in governance plays a key part in developing the alternative. The community needs to find ways to develop non-VC (venture capital) support for scholarly societies so that this kind of service can be opened up without incurring fees for members – but institutions and funders need to commit to both funding and participation.

I'm not going to write about Gary Hall's 'Ubercapitalism' talk here, as while it was both interesting and entertaining, it wasn't for the most part directly relevant to the topic.

David Parry's work will be familiar to many, and his talk seemed to summarise and validate what the others were saying. The most useful part of his talk was the following relating to academic communities and social networks, which gave me some thinking to do:

Core ethical principles:

  1. Value APIs and shared protocols
  2. Value libraries’ roles
  3. Think about the blockchain

On the blockchain point, Parry suggested it was a model to build trust and interoperability, build in level playing field, a flat system not relying on current power structures and using a distributed database.

I enjoyed all the talks, and the subsequent Q&A discussed some of the points I have been writing about in my PhD notes - why are the 'proper' systems designed for university administrators and librarians but not academics? Why aren't those with relative privilege in permanent jobs leading the charge for change, including influencing buying practices of their own institutions? Why aren't we putting more thought into how knowledge as well as information is communicated - access on its own doesn't make the work make sense to all those who could benefit from it?