The longer Elsevier refuses to make their citations open, the clearer it becomes that their high profit model makes them anti-open
The Open Citations movement is something I’ve been watching with interest. It hasn’t really broken through into the mainstream yet, like Open Access has, because it’s a bit of a more complex topic and the benefits of it aren’t readily apparent. The main idea behind it is that publishers should add all the listed references for an article (i.e. the sources being cited by that article) to that article’s metadata and make this citation data open on platforms like Crossref.
Think about where we typically go now to collect cited-by information for an article. You can go to Citation Index like Scopus or Web of Science, but those only capture a small picture of citations. Both of these tools aim to only index ‘high-quality’ or highly cited journals, which means their citation data will always be incomplete. Estimates I’ve seen suggest there are over 60,000 current scholarly journals being published and Scopus and Web of Science only include around 20,000.
You could use Google Scholar to grab citation information, but their citation data is all text-data mined from PDFs and often involves a lot of duplicate citations and other mistakes. Google’s citation numbers for an article are usually higher than Scopus or Web of Science but the error rate is always much higher.
Having a more complete picture of how much an article has been cited by other articles is an immediate clear benefit of Open Citations. Right now you can get a piece of that via the above tools I’ve listed and, maybe, a piece is all you need. If you’ve got an article that’s been cited 100s of times, likely you aren’t going to look through each of those citing articles. However, if you’ve got an article or a work that only been cited a handful of times, likely you will be much more aware of what those citing articles are saying about your article and how they are using your information.
Being cited only a handful of times is the status quo for most researchers. Especially in specialized and smaller fields which often can be very disconnected. The more we can do to make those interconnections clearer, the more benefits those fields will see.
Another benefit would be citation “trails”, the ability to dive into the link between of who is citing who and build an interconnected web of how ideas get built of each other. This sounds pretty highfalutin, buzzwordy, and impractical. But ‘interconnected web of links and the strength of connections between different links’ is essentially how Google’s PageRank search algorithm functions and artificial intelligence. Until recently citation data has been hard to get at. It’s been locked up on vendor websites with limits on export and analysis. The possibilities of what such data could provide is still hard to imagine because it’s still such a new idea. Imagine, for instance, if you could see a list of ‘negative’ citations for an article. The number and links to articles that disagree with the conclusion of the article you are reading.
I’ve also argued before that ‘citation hopping’ has long been an undersold benefit of citations. It’s always a challenge to find an article on a niche topic. Once one article has been found on that topic, it should be possible to follow citation links between articles to find all the other articles on that topic. Open Citations could make finding all the articles on a topic much easier.
The benefits of Open Citations are clear right? So what’s the hold up? Why don’t we have this now? Well, it’s time we return to an ongoing theme of this blog which is trying to guilt trip Elsevier into doing the right thing. All the large scholarly publishers (Wiley, Springer, Taylor and Francis, Sage) are now providing open references to Crossref. You can search Crossref Participation Reports yourself and see. Elsevier is the only one holding out.
One of the research fields that is particularly interested in having Open Citations is, unsurprisingly, the Scholarly Metrics Community. Open Citation data means more and better data for them to analyze. Elsevier is the largest scholarly article publisher out there, which means their lack of citation data keeps the data incomplete. This has come to a head recently with an Elsevier journal, Journal of Informetrics, one of the leading metric journals, entire editorial board resigning today. While the reason for the mass resignation also included disagreements with Elsevier’s approach to open access policies, the main reason for the resignation appears to be Elsevier’s refusal to open up their citation data.
In response, Elsevier clarified their position of Open Citations (h/t @RickyPo). Their argument is that because they already make their citation data available through their subscription database, Scopus, “[…] Elsevier cannot make such a large corpus of data, which it has added significant value to, available for free”.
They argue that they already invest significantly in citation extraction technology that pulls citation information from their articles and that they can’t give this value away for free. I’m not sure this is a fair argument, because other publishers are also doing this, this is exactly how they provide their open citations to Crossref. They also state that users can access this data via Scopus or contact them if the user’s institution does not subscribe for individual access.
Elsevier is between a rock and hard place here. They’re the only scholarly publisher with a citation database tool like Scopus. Every other publisher gets a clear benefit from opening up their citations, but for Elsevier it threatens the number of subscriptions to Scopus. With Open Citations, a tool like Scopus becomes less valuable, but if they keep their citations closed they’re directly impeding the progress of research.
I think Elsevier should just bite the bullet here. Make their citations open. I don’t actually think subscriptions to Scopus will drop by that much. Even if all citations are open, user friendly interfaces to interact and analyze links between citations are still needed. Scopus can become a subscription based analysis tool on top of free data.
If Scopus is actually, truly, threatened by open citations, then isn’t it clear that this is a tool that only existed because of Elsevier created artificial scarcity to grow their profit margin? We’ve seen a lot of worry in the scholarly communication world lately that Open Access can’t work with the current large publisher profit margins. That publishers will always have an incentive to withhold information because they can sell it. Elsevier and the other publishers have argued against this, saying that it is their large profit margins that ensure high value and quality in publication and it doesn’t mean they are anti-open. Elsevier’s refusal to make their citations open contradicts this argument. Make em open, Elsevier.