At Frankfurt Book Fair 2016 we will be hosting a panel debate on the discoverability of Open Access. In advance of the event we caught up with one of our esteemed panelists, Phill Jones, Director of Publishing Innovation at Digital Science, to discuss his role, how his company is innovating in the academic space and what he thinks are the main challenges posed by Open Access.
Ingenta: What does a typical day in the office of Phill Jones look like?
PJ: My typical day very much depends on where I am. Sometimes I’m in the Digital Science office in London, in which case my day is often stacked with meetings and catch-ups with my fantastic colleagues in the publisher relations, communications and consulting teams. If I’m not there, I might be at a conference or meeting, like Frankfurt Book fair, but for most of the time, I’m at home in Edinburgh. I’m lucky that working from home while travelling a fair amount for work gives me a sense of balance. To put it bluntly, the fact that I don’t spend an hour each way on the London Underground every day, makes 5 hours on a plane every so often a lot more bearable.
My working day starts a little after nine after I get back from walking my two boys to school and settle in for the day with my office manager, Iggy, a black mongrel rescue dog from Romania, but he tends to sleep on the futon most of the time. I’ll check my email and try to read at least a few things from relevant blogs, listserves and news sites, then get to work on whatever tasks I have looming over me. Like most people, those can be varied from helping prepare a proposal for a consulting contract to writing a blog post to organising a seminar. I often have catch-up meetings or calls in the afternoon and by the time those are finished, there’s enough time for a few more emails before my wife and kids get home and my 6 year old will start knocking on my office door asking for me to come down and help him with his homework.
IG: How is Digital Science’s offering innovating the academic publishing space?
PJ: Digital Science was founded on the principle of supporting academics at every stage of the research cycle, from reading literature and getting the initial idea, to maximising the impact of that idea after it has been published. Since then, our mission has evolved a little and today, we see ourselves as supporting the flow of information itself through the research cycle by supplying products and services to all four key stakeholders in scholarly communication; researchers, institutions, funders and if course, publishers.
We’re a portfolio company, meaning that we have both the core of Digital Science, which is essentially a hybrid technology investor and research metrics consultancy and then nine portfolio companies, each of which were hand picked because of their extraordinarily innovative technologies. I won’t list them all, but let me pick out just three. Altmetric, for example, helps all stakeholders understand the broader impact of the work that they produce, publish or support, and to compare themselves against others. figshare helps researcher communities share more of their outputs including data, computer code, visualisations and even grey literature. Intuitions, publishers and even funders can use figshare’s products to support those communities by facilitating that sharing. The last one I’ll mention is Dimensions from Uber research, which is a funding analysis tool built on a unique global grants database. We traditionally sell this product to funders but are increasingly consulting with institutions, governments and publishers to analyse research landscapes and perform predictive analyses. You can read some examples of this work on our website, in the research reports section of our resourced page. https://www.digital-science.com/resources/digital-research-reports/ There’s also a white paper on the effect of Brexit on there that got rather a lot of media attention when it came out.
IG: Would you say Open Access discovery is currently a major issue in scholarly publishing?
PJ: That depends on what you mean by a major issue. I don’t think that a lot of people are talking about it, so from that perspective, I don’t think it is. I’m also not sure that many researchers themselves would identify it as a major workflow issue, but I think that’s because expectations of how easy it is to get from a search result to a legal copy of an article are historically fairly low. For me, this problem pre-dates open access but certainly, green OA in particular has made the issue more pressing.
A good starting point for this is the work of Jason Price, who is Director of Licensing operations at the SCELC library consortium in California. As early as about 8 years ago, he was surveying the accuracy of link resolvers and finding them shockingly poor, due to problems in the meta data workflows and a lack of persistency in URIs. Even without the complication of Green OA copies in institutional repositories, it’s always been the case that a researcher at a university cannot always easily find instances of articles that they have legal access to, for example in aggregator sites or on subject level repositories.
The closest to a working solution so far, has been Google Scholar. Their web-scale approach finds accessible instances of a given article, irrespective of licensing consideration. This apporach has certain draw-backs for publishers, but it’s part of the reason Google has been so successful in the discovery space. I’m not sure how all this will get resolved in the end, but I’m watching what’s happening with both SHARE and CHORUS with interest.
IG: As director of Publishing Innovation at Digital Science, one of your aims is to look at ways of using technology to enhance scientific discovery. What are the most exciting developments going on right now, in your view?
PJ: I mentioned SHARE and CHORUS before. I think that those are very interesting partly because they represent one aspect of a new emerging scholarly infrastructure that we’re only beginning to see the shape of now. Projects like ORCID, Crossref, DataCite, and so on are really just the beginning.
All stakeholders in the scholarly space look to maximise the impact of their investments in terms of effort, resources and money, against a backdrop of ever-increasing scarcity. This means that funders want to give grants to the most effective researchers and projects, institutions want to make sure that they are fulfilling their missions, and publishers seek to remain relevant and publish the most impactful content. As this evolves, infrastructure and metrics will become even more important in the future. It will be imperative that everything has an identifier be it a scholar, an institution, an article or a piece of computer code and that all of those are linkable to their outputs and impact.
So, in a sense, discoverability won’t just be about what an individual researcher can find, but about what the analysis algorithms can find. The outputs of those analyses will start to define what is considered good content, which, in turn will affect what is read. Essentially, it’s a more sophisticated version of the way librarians used to use Impact Factor to select what they considered to be core journals.
IG: In what different ways are Digital Science helping the research community in the area of discovery? And why would you say Digital Science solutions are so effective in this space?
PJ: We have a number of technologies aimed at helping researchers discover better information. ReadCube, which is a reading and discovery environment used by millions of researchers worldwide, has an intelligent content recommendation engine that is highly praised by researchers. We have Altmetric, which I mentioned before, which can be used to help surface content that is having an impact on society, perhaps in the form of news coverage, or policy documents. Figshare, which is our data and research output hosting and sharing environment is used by communities to share data, methods, videos and all manner of content, and links out to the versions of record to help people find and cite the right articles.
Many of our products are built on technologies from actual researchers. That is, the typical Digital Science entrepreneur was a postdoc who suffered frustration with some aspect of scholarly publishing, whether it’s not being able to find and organise literature, or not having a place to put orphan data. While we’ve taken those technologies and expanded their functionality to create products that serve the needs of a variety of stakeholders, they all remain grounded in the workflows and needs of actual researchers and academics.
IG: What would you say are the main benefits, and challenges, for researchers using repositories to publish their work?
PJ: At the very least, a repository makes the article or other research output publicly available, even if it’s not always easily discoverable. For most academics, OA and data sharing remain a way to fulfil an open access or data sharing requirement in a grant application, although that perception is slowly changing. Another key advantage is that at some institutions, inclusion in the institutional repository is needed for work to be included in various research assessment exercises. A good example of this is the Research Excellence Framework (REF) in the UK. I know of more than one university, where outputs have to be put on the specific institutional repository for the institution to include them in the assessment, which in turn is linked to government funding.
The main challenge, frankly, is time. We’re not at the point where a researcher can put content in an institutional repository and expect the same reach and impact as a published article. I’m not sure that we’ll ever get to that point as there are many ways in which publishers really do add value to the content they publish. So researchers effectively have to double-publish. Once in the repository and again in a journal. This extra workload can be onerous, depending on how well designed the repository and CRIS system is.
IG: You were editorial director at JoVE – video journals are a really interesting field, and here at Ingenta we have a number of clients inserting video as supplementary data. What are the discovery challenges posed by video and other supplementary data such as graphs – and how does one overcome them?
PJ: Yes, I was. You’ve done your homework on me. The problem to put it bluntly, is Google can’t hear, it can only read. I remember when JoVE first got into PubMed and the problem that we had was the lack of text to actually index. At that stage, we had to go back to all of our authors and ask them to submit a text protocol to put alongside the video. That was hard work and not all the authors were entirely cooperative, but we got it done well enough to be indexed. We also realised, of course, that SEO only works on text, not on audio or video, so the inclusion of text protocols alongside the articles improved our SEO enormously as well.
So the solution is simple, accompanying text and appropriate html meta tags so that your content is machine readable.
IG: As you pointed out back in 2014, postdoctoral fellows represent the unsung heroes of academia, but increasingly it seems they are taking the law into their own hands in terms of publishing and publicising their research. How is Digital Science helping this process, and what can mainstream – and widely-accessed – aggregators and content-hosting services such as Ingenta Connect do to help?
PJ: Yes. I think I figured out, at one point, that most academic biologists in the US were postdocs, which is worth remembering the next time somebody talks about the ‘faculty and students’ at an institution.
It’s true that some postdocs are becoming their own publicists and taking control of their own academic reputations and I think the reality is that those are the ones that will go on to become successful as academics. At Digital Science, figshare helps researchers make their outputs available and thereby helps them form collaborations and Altmetric provides the vital metrics to enable them to see how effective their reputation management strategies are. We also encourage researchers to learn about social media, open science, and blogging and have performed a range of outreach activities to help young researchers get a handle on how to build and manage their own reputations online. Have a read of our blog, there details of the events we did for Ada Lovelace day, and the recent Altmetrics conference, 3AM. https://www.digital-science.com/blog/
As a whole industry, we need to think of ways to help the next generation of researchers communicate the new types of research and outputs they’re creating. We need to broaden the definition of what is considered a creditable contribution to the scholarly record. By that, I mean it’s often difficult for postdocs to distinguish themselves based on their contributions to articles where the PI in the lab is presumed to be the sole intellectual driving force. Postdocs essentially become beholden to the good will and altruism of a supervisor to give up credit to them, which sometimes doesn’t happen. If early career researchers had more opportunity to highlight their specific contributions, we could remove some of the seniority bias and conflict of interest there.
IG: Vitek Tracz, the founder of F1000, has predicted that academic journals are on the brink of demise because they no longer serve a useful function as a container. To what extent do you agree with this and if so what is the “container” of the future – is it the PLOS-style mega-journal?
PJ: I understand where Vitek is coming from here. I’m not sure I’d agree that they’re on the ‘brink’, though. I see a slightly more drawn out transition that will see journals transforming into something else. Vitek is correct, that the role of journal as ‘container’ is no longer technically necessary. Journal titles are a form of meta-data, they traditionally demarcated research disciplines, which is a role that meta-tags in a mega journal or some kind of repository or platform can replicate and expand on. We no longer have to have just one discipline tag, an article can fit into multiple disciplines. Having said that, journals have another role, that is of presenting a consistent narrative of a field. They’re often headed up by editors-in-chief that are either highly pre-eminent or even the founder of the field that the journal is named for. Communities coalesce around societies, the journal and associated conference for a field. That social aspect could be replaced through other mechanisms, but like all innovations, it’ll happen not when it’s technologically possible, but when the community is ready for it.
IG: How do you see Open Access, particularly where discovery is concerned, evolving within the next five years?
PJ: Slowly. I think that the moral argument around OA is pretty much played out. That’s not to say that everybody agrees, but major funders seem to all be aligning into pro-OA positions and are increasingly willing to use their economic influence to drive scholarship in that direction. Having said that, academia is a big ship that turns slowly, and there are sustainability questions to be addressed. So while I don’t see a reversal of direction any time soon, I think the market will continue it’s slow and inexorable transition.
What I do see happening, is an expansion beyond the discussion of open access and into open science and new research mechanics. As I mentioned before, we’ll see an expansion of the scope of the scholarly record, more content types being perceived as first class research outputs and these objects being increasingly linked and interconnected so that researchers can traverse an ever expanding linked network of research objects, many of which being free at the point of use, but not all.
To hear Phill Jones, and other leading figures in academic publishing, discuss the issues around Open Access discoverability at Frankfurt Book Fair please attend:
Open Access = openly accessible? What can we do to make OA content really discoverable?
9.30am, Wednesday 19th October
Hall 4.2 Hot Spot Stage
The rising tide of Open Access publications – books, monographs and journal articles – brings with it a multitude of difficulties. None more significant than discoverability. Open Access content is buried in numerous silos – in institutional repositories, on sites like OAPEN and DOAJ and on publisher’s own websites – so how can researchers find exactly what they are looking for? In this session, Ingenta will bring together a distinguished panel of experts from across the scholarly publishing landscape to discuss the challenges brought about by the widespread shift to Open Access models, and the ways in which discoverability of Open Access content can be addressed.
* Byron Russell, Head of Ingenta Connect (Chair)
* Phill Jones, Director of Publishing Innovation, Digital Science
* Carrie Calder, Business Operations Director, Springer Nature
* Frances Pinter, Founder of Knowledge Unlatched