Last month Ingenta’s very own Byron Russell hosted a lively and engaging panel debate on the theme of Open Access discoverability at The London Book Fair’s Faculty stage. The popular session featured a range of authoritative speakers from academic journal and book backgrounds:

  • Rupert Gatti, Director, Open Book Publishers
  • Lara Speicher, Publisher, UCL Press
  • Mark Johnson, Director of Contributor Experience and Product Marketing, PLOS
  • Ros Pyne, Research and Development Manager, Springer Nature

For those who were unable to attend the session, here is the transcript:

BR: Open access publishing is a trend that’s growing. We’re slightly ahead of the curve in the UK.  Worldwide about 17% of academic content that’s published is OA, here in the UK it’s about 18%.

In your experience, what are the main challenges that publishers face when it comes to publishing in OA?

LS: I think one of the main challenges which publishers face when publishing in OA is funding, and the business model. As this session is maybe more about discoverability, our focus is on that as the key challenge. I think discoverability is a challenge which all publishers face, not just OA publishers. We’re all trying to make our titles available to the widest audiences, possibly through the greatest number of platforms, distributors and aggregators. Perhaps one of the challenges for books, in particular, is that many, if not all, distribution models are set up for commercial publishing. So distributing free books via those channels is very difficult, or impossible, so we have to find alternative means. But maybe that’s something we should be looking at as an opportunity rather than a challenge, as there are many ways of distributing OA books including one’s own website, OAPEN, The directory of OA books. Interestingly in a report published recently by Simon Inger on journal discoverability there is a clear pattern of an increase of readers finding articles on publisher platforms via social media. So as well as this wide range of aggregators, libraries, Google, etc. we’re seeing other means that were in the past not as important as publisher platforms becoming more important in discoverability, and that could change more in the coming years and become a very interesting source of readers.

BR: What would you say are the limitations of the OA discovery services that are currently available to OA publishers?

RG: I think on the whole, the discovery services Lara was discussing, the tradition, or dare I say the legacy distribution discovery channels for books, are not really well equipped for OA because the price point of zero is problematic for most of them as they’re looking to make money through that publisher. So I think that’s an issue going forward, as most of the legacy distribution channels use a ‘push’ model. Publishers have this hose and are sort of spraying their content, and people are controlling which hoses they allow to spray their garden. Whereas, I think readers nowadays, especially in a non-OA world, want vacuum cleaners. They want things that are going to suck up lots of content and filter it for them. So there’s a difference in perspective about what readers want and what presently a lot distribution channels are providing.

A couple of other issues as far as books are concerned, there’s nothing to mark them out as OA. The thing about an OA work, as opposed to another work, is that it’s reusable. A reader really wants to know if this is content they can reuse, or a content that’s going to be read. A lot of the distribution channels want to be serving the content as PDFs, for example, while a lot of our content has combined research data with it, links to other online appendices, a lot more content that’s going to go with the book that’s not available front up from the PDF. So once again, existing channels are not making the reader aware that there are lots of different versions of the content out there, and there could be a version that’s more or less appropriate to them. About 40% of our readers, read on Google Books. Now that’s a problem because it doesn’t have all this extra content that goes with the book.

Finally, the other things that books have got are chapters and quite often they’re written by different people and have different DOYs, and it’s almost impossible to get discovery. It’s extremely difficult to make the legacy discovery channels take chapters as seriously as they take journal articles, and to link them in the same way, even though it’s conceptually easy to do. But because it’s found in a book, it becomes administratively problematic, and this would be something that would be great to look to change.

BR: Regarding discoverability over all of OA publishing, what do you think are the main points of difference between books, monographs and journals?

RP: I think there are a number of similarities. We know that Google and Google Scholar are the most important routes for referrals on both our books and our journals, so things like good SEO are really important in both cases. And in both cases you’re looking at depositing and indexing services, they might be slightly different so, something central is really important on the journal side.

On the books side there are some characteristics of OA publishing which do cause some challenges, like chapters for example. And it’s not just chapter level information in an OA book, but what if you want to make a chapter OA within a book that is otherwise pay-walled content? That’s really hard, and the only realistic way of doing it is by releasing OA chapters as mini-books, because none of the discovery systems for books are set up to allow you to make just one chapter freely available. So that’s a really awkward workaround which involves a lot of data manipulation to ensure that your book DOI matches the chapter DOI within the book. On an industry level we’re starting to develop metadata solutions for OA books, at a book level, so we can tag up that content – and some providers are now accepting those metadata feeds. But we don’t yet have metadata formats that will support chapter level information, so I think that’s a big problem.

On the journal side, I think that mega journals pose some particularly interesting challenges, because if you think about the paradigm of journal marketing, and promoting at a journal level, that’s absolutely inappropriate – if you think about scientific reports, it publishes across the full range of life sciences and material sciences, so at that point you’re really starting to look at how you do very targeted marketing to the communities within that range of subjects. In addition to marketing, there are technical implications too, so things like search, but also subject tagging, building a great ontology to drive subject tagging, and making sure that at every stage when the article is submitted you’re gathering the right metadata so that you can then surface that in your discovery mechanisms. These challenges have been created by the systems of mega journals.

MJ: I agree, discoverability is a bit of a challenge from a mega journal perspective, because the content is so broad. I’d take a step back and say that in general the discoverability for OA content is not really as problematic as the content which sits behind a paywall. For example, Google has a policy for indexing only what the end user can read. There’s something of a scholarly exception for journal content, but that doesn’t apply for books usually. So in the case of PLOS where the full text contents of all articles are available, we don’t really have a problem with discoverability through search engines – It’s very easy to discover that content. For the type of content where, as Rupert says, I want the vacuum cleaner approach, there are different tools that people have, like RSS feeds and alerts, for example. And that can be a challenge for a mega journal. At PLOS what we’re thinking about is creating more and more collections of topic specific content to make that content more discoverable to an end user, and we’re taking some baby steps on that front with neuroscience collections and synthetic biology collections, and it seems to be working really well for us.

I also think that the concept of the hybrid journal – a subscription journal which also has some content which is OA – creates a challenge, both from the reader and funder perspective. How do readers know when an article is available OA? Sometimes it may be tagged as OA, but that tag may not follow it through different discovery systems, so that’s a problem. Robert Kylie from the Wellcome Trust has recently published a blog article highlighting their frustrations with hybrid journals, and I think that as funders shift their focus from article level OA, which can be accommodated by hybrid journals, towards full OA journals, we’re going to see a much quicker swing to OA journal publishing.

BR: In addition to making content discoverable, how do you proactively go about disseminating your OA content to increase views and downloads online?

LS: As well as making books available on our own platform, on Amazon and through retailers for the print versions, we try to get on as many OA platforms as we can, such as OAPEN and World Reader, and there are many others in development, traditional hosting platforms which are now going to move into offering OA book hosting. We also try to work in a way that most other university presses work in terms of marketing, and I’ve been asked in the past “why do you need to market OA books, they’re there and available aren’t they?” But of course you do, like any product whether it’s free or paid for, you need to let people know that it’s there. We’re using social media, we may also use Listservs and we have a publishing catalogue in print format PDF that gets mailed out. We use newsletters, we track campaigns to our daily download figures so we can actually see what’s working. We have book launches; we help authors promote their works at conferences, etc. One interesting challenge about promotion and PR which we notice is that when our books get reviewed, whether it’s in print publications or online, it’s really difficult to actually get the reviewers to mention that this is an OA book. They will give the bibliographic details, the print price, the publisher name, but they won’t say that it’s OA. And it’s so easy in an online review to put a link there and get all those readers to link straight through to your content.

BR: How much can publishers rely, if at all, on their authors to promote and market their own content to boost discovery of their publications?

RG: I think OA changes the whole relationship an author has with their own content, in that they own it for the first time. Therefore there is a change in emphasis in the strategic behaviour that an author has with their content, and how they’re going to establish themselves, and an online presence, in the academic community. And OA really eases that. In fact without that, it’s pretty hard to do anything because the author can maintain all sorts of different distribution channels that are relevant to the author and to the author’s discipline. That was something that was not really possible when they didn’t have control over their content. We don’t rely on the author as such, but we do utilise their social networks, connections, knowledge of their discipline, and their awareness of which conferences, blogs and websites the content should be connected to. And without that we can’t do as good a marketing job. But that doesn’t mean that there isn’t a marketing job to be done, there are lots of other ways to do that. So it’s about engaging with the author, which is possible with OA content in a way that I don’t think is possible if it’s not OA.

MJ: With OA the author actually owns the content, and certainly that’s the case at PLOS where all the content is published under Creative Commons attribution license. That means there are a couple of different challenges – first of all you’re promoting the work at an individual article level, not at a journal level or a book level. And in order to promote that, you need a set of tools – and social media is a really great toolset to get the word out about an article or a chapter. But in order for one to see the efficacy of that, you need article level metrics or chapter level metrics. PLOS has article level metrics to help authors see the level of social engagement around their particular work. PLOS is a signatory on the declaration of a research assessment, also known as DORA, which is a commitment by publishers that they will not use journal level metrics like the impact factor as a shorthand for quality in assessing an individual researcher. And we take that very seriously at PLOS. We don’t promote our impact factor or our journal level metrics, even though they’re both very good. We really want to put the emphasis on the article so that we can help the authors promote their content, so we do that through article level metrics, and there are other tools out there which are emerging. For example, there is an organisation called KUDOS, which provides a toolset for researchers for free, which enables them to Tweet and communicate their content through social media and provides a dashboard for publishers showing what level of activity is around at work.

BR: How do you make sure your OA content stands out from the crowd and that important academic work doesn’t get buried?

RP: I don’t think there is a big discoverability problem in OA, so for example if you look at our OA books, we’re finding that our OA books are accessed 12 time more than our non-OA books. So to some extent I think OA content is already standing out from the crowd – and is already showing the benefits of publishing via OA. And after that I think it’s all the things we’ve already talked about – promoting content at an article level, it’s being close to communities, it’s thinking about thematic collections and that’s certainly something we’ve been focusing on in our humanities and social science OA journals. Humanities and social sciences are interesting areas because OA is so much less developed than it is for STM – there’s still a nervousness about it amongst authors. We survey authors every year about their opinions on OA, and the one big concern is still quality, and that’s going down every year, so that’s positive. But we’re still seeing 40% of humanities and social science authors saying they’re concerned about quality, compared with 27% of STM authors. What I’m saying is that in some areas, what we should be doing is persuading OA authors to publish OA in the first place and then that will in turn help promote discoverability

MJ: I agree completely that the more that authors are involved in the publishing process, the better, and I think that OA with a CC BY option gives authors more option to be involved in the process which is absolutely a good thing. CC BY means that the content can live anywhere. For PLOS the content lives on the PLOS journal websites, it lives on PubMed central full text, it lives without a problem in certain academic tools like Academia.edu and ResearchGate and goes on microsoftdocs.com, in fact we don’t have problem with PLOS content going up anywhere. Having the article located in more places certainly helps discoverability and it gives the researcher greater visibility for the work that they’re doing. Specifically, for what PLOS is doing, we’ve invested a lot within the last eight months to build out what we call an editorial media team, which is basically a press releasing service. So we work very closely with the authors and the public information offices at their institutions to create press releases about the research that PLOS is publishing. That lets PLOS work with the authors and the authors institutions to create media coverage about the articles they are publishing. And media coverage can be both mainstream media coverage and also scientific media coverage, and that really helps bring home and distribute the impact and raise the visibility of it. This has generally been a great success and PLOS content was mentioned in the New York Times over 50 times in 2015, which is almost once a week.