Unpaywall: A New OA Discovery Tool
Some of you may remember oaDOI, a tool that searches for an Open Access (OA) version of an article when provided with a digital object identifier (DOI). Impactstory, the team behind oaDOI, has just announced the upcoming release of a new tool based on the oaDOI API called Unpaywall and it looks to be very useful indeed.
Unpaywall operates on the same principle as oaDOI: it hooks onto an article’s DOI and searches a number of sources for an OA version of that article. The main difference here is that, while oaDOI requires you to enter a DOI into a search box, Unpaywall is a Google Chrome extension and performs that search in the background automatically when you visit the landing page for an article. An unobtrusive tab will pop up on the page, as can be seen in the example to the right, which will be green if an OA link is found and gray if not. Users can then click the tab to be taken to an OA version of the article (if one was found). Finding a DOI and entering into a search isn’t exactly an onerous task, but users are often unwilling to take those few extra steps; the streamlining that Unpaywall provides means that it’s far more likely for researchers to bring into their usual workflow.
Unpaywall officially releases on April 4th, but Impactstory has made the extension available now and I have yet to run into any issues. There are still a few limitations, though, to how useful Unpaywall might prove to be:
As I pointed out when talking about oaDOI, not all articles hosted online have DOIs and most that do will not be available on an OA platform; a more detailed discussion of these issues can be found here. Another important point, one that I failed to notice last time, is that neither oaDOI nor Unpaywall provide information on what version of the article users will be linked to. Many OA versions of published articles are either pre-prints (i.e. the authors’ original manuscript before any revisions suggested by reviewers) or post-prints (i.e. the final accepted version of the manuscript, but often not reflecting final typesetting or copyediting done by the publisher). As you might expect, these versions can differ from the final published version of an article, sometimes substantially.
Most researchers understand this and are familiar with institutional repositories and pre-print servers. Those that aren’t, though, may not know that the version of the article they arrive at through Unpaywall might differ from the published version in some substantive way. If I have one complaint about Unpaywall, it’s that I’d like to see them implement some system for letting users know up front where the article they are being linked to is coming from. This would make it a bit easier for users to follow up on sources and check to see if they are indeed getting a pre- or post-print; their only option currently is to play around with the URL provided to try and arrive at an info page for the article or repository. Justin Priem, one of the founders of Impactstory, has assured me that this is on their radar, though, and that they plan to implement this in future versions of Unpaywall.
Regardless, this is a minor complaint about an overall great product, and I’m very excited to try and get users here at Wayne to start using the extension. This is two great tools in a row from Impactstory, as well, so I’m looking forward to see what they do next.