There’s a new kid on the publication access block at PubMed. It’s a little one, so you might not notice it for a while amongst the giants: a new high-profile spot for icon’s links to free full texts. The icon links to a publication that is not available for free at the journal or in PubMed Central (PMC), but is housed for free public access at an institutional repository (IR).
You can see one by clicking on this logo for the University of Michigan Library’s IR, called Deep Blue:
The icon at PubMed at the top right will take you to Deep Blue – and the full text of a January article on ebola.
You can read more about IR LinkOut in our post today in the NLM Technical Bulletin. IR LinkOut is relatively little at the moment because only a few IRs are participating – you can see which ones here. Just those few expand access to around 25,000 articles.
This is what’s called “green open access”. The article can be released with an embargo – usually a 12-month delay after publication. It’s not likely to be the journal’s typeset version, but the content is the same. Depending on the publication and the IR, it might also come with extras, like datasets.
Green open access is one of the core branches of formal open access (OA) – along with OA at preprint archives, and “gold open access” to the journal versions at OA journals and non-OA journals that sell authors OA for individual publications. There is a spectrum of openness, too, depending on restrictions around how soon the free version is available, whether it can be text-mined, and more. (There’s a quick overview table in this PDF.)
The problem of inaccessible scholarly publication remains, unfortunately, alive and well.
The informal world of article sharing is massive – academics are uploading 2.5 million articles a month into ResearchGate. It’s gotten so big, it recently raised over $50 million more funding and got an academic “Facebook” article in the New York Times.
But people haven’t been as active in uploading their publications to their institution’s repository. As Richard Poynder wrote in a post raising questions about the future of IRs, “author self-archiving has remained a minority sport”:
“as gold OA accelerates so the logic of depositing papers in IRs dissipates… [T]he institutional repository is experiencing an existential crisis”.
My reasons for optimism come from a different perspective. I think there’s important potential here.
People publish a lot in journals that don’t provide gold access, or do so only at unaffordable levels, for reasons that aren’t going away any time soon – like needing to be published in a prestigious journal for their careers, or because a publicly inaccessible journal is “the” one that specializes in their work. For huge numbers of people, even lower cost OA options are a problem. Even embargoed green is far better than no public access at all.
IRs have become an important way for researchers’ work to get noticed and used. Google Scholar picks up and links to full texts from IRs, and that’s a big deal. So does Wikipedia. If your IR joins PubMed’s LinkOut, that’s another big step up in visibility and accessibility: a few million people use PubMed every day.
Reasons to be optimistic IRs will fill up more? Europe’s push for getting more of its research out in the open is a big one. Earlier this year I wrote that it looks likely that there will be a growth spurt in open access by 2020, thanks to Europe, led in particular by the Netherlands and Germany. Part of that will be green growth.
There are signs policies are starting to have an impact. The Higher Education Funding Council for England (HEFCE) provides universities’ core funding. As of 1 April 2017, the only publications that will be taken into account in assessing university productivity and excellence are open access ones: if they are not “gold”, they have to be deposited in a public repository.
Groups of universities are joining together to encourage green open access, for example in Finland. Nationwide IRs for the work of a whole country’s scholars are emerging, for example in the Netherlands, France has had HAL for a long time, and Australia has been consolidating for years too, with a lot of repositories in Australia and New Zealand.
The EU-funded Openaire is aggregating from 2,860 repositories, and the UK’s CORE and Bielefeld’s BASE are big too. The Confederation of OA Repositories (COAR) is working to make IRs “the foundation for a distributed, globally networked infrastructure for scholarly communication”. If you’re a scholar at an institution with an IR, your library people are knocking themselves out to get your work out there: please do your part and encourage your colleagues, too!
How many repositories are there? More than 3,200 according to the DOAR directory. There’s some data here on how much from academics’ online CVs is directed to IRs. The EU’s Open Science Monitor shows about 10% of the publications in Web of Science are in green OA repositories, but that would include PubMed Central (PMC).
The potential pool of literature that could be uploaded into IRs is massive, especially in research fields and countries where the funders don’t mandate public access. Some repositories only make meta-data about publications publicly accessible, though. Much of what’s accessible in repositories is from OA journals or PMC. To truly extend access, an IR needs to include full text that’s not already accessible, in a shareable form – and have a high author participation rate.
That’s a big mountain. High-profile PubMed access is another incentive to climb it.
In addition to the Google Scholar Button, there’s the wonderful Open Access Button which helps you find IR versions – it’s also working to integrate with libraries to save on the cost of interlibrary loans. Unpaywall is a recent addition to this field. I don’t know how these compare for coverage.
Interest declaration: My day job is working on projects at PubMed, and I co-authored the NLM Technical Bulletin article discussed here with my NCBI colleague, Kathy Kwan. This post represents my personal opinion, not the view of NCBI or NLM. As well as blogging at the PLOS Blog Network, I’m a member of PLOS One’s Human Research Advisory Group.
* The thoughts Hilda Bastian expresses here at Absolutely Maybe are personal, and do not necessarily reflect the views of the National Institutes of Health or the U.S. Department of Health and Human Services.