Print Legacies in Online Research Publishing

Digital Disruption in and of Research

Gudrun Frommherz & Justin Matthews, November 2019

‘Digital Transformation’ is one of the killer strategies in today’s digital economy. Companies must embrace new technologies, change their production lines to digital workflows, smarten up their product offers, engage with their customers anytime anywhere, and provide a raft of added value services — including corporate social responsibilities — around their primary business focus. By the end of this decade, over ¾ of businesses in developed economies will have implemented digital transformations in one way or the other. The figure for multinational companies is likely to be even higher.

What does this look like in non-economic activities, such as academia? Taking our own experiences as an indication, academic research as a field of professional activity, and academic research processes as a community of practice, seem to largely evolve outside of digital transformation pressures, seemingly innocent of the challenges and opportunities of digital disruption. Is there a digital disruption of academic research? Where on the digital transformation journey are academic research and publishing?

Across the four broad areas or stages of research, we focus on two aspects: 1) research communication, i.e., the articulation of research in a useful and presentable form and 2) the dissemination of research as an academic publication.

How do most current online journals ‘speak’ digital? Most platforms still rely on a print-era formatting approach to online. In particular, media embeds are integrated as external objects into a page-style layout. This approach continues to disadvantage media formats (image, video, 3D objects, etc.) over written text.

With digital technologies, there is a revival of sensory communication media — i.e. image, sound, spoken language, immediacy and instant response — over abstract encoded communication such as written text. In the historic evolution of communication, visual, aural, pictorial and tactile media dominated for the longest time. With the invention of the printing press about 580 years ago, printed text started to dominate communication — and the encoded, abstracted form of text defined a way of thinking and understanding the world. Knowledge in the form of precise abstracted letters, words and numbers over experiential knowing and its expressions as representational signs such as images and iconographies ended the raw Middle Ages and brought in the rationality of the Enlightenment. Knowledge was no longer a tacit experience that resided in the gut but a crystal clear discerning and labelling in definite terms as words and numbers. The shift from gut to head, from fuzzy to precise, from fluctuation to definite is the actual shift from analog to digital — with digital allowing only for yes|no — it is exactly this and not that.

New knowledge through research is most commonly articulated in the one format that perhaps is least suitable: text. And where non-text objects appear, they are usually ‘straight-jacketed’ into the frame of journal, book, or webpage.

Newer digital technologies, such as virtual reality and mixed-reality environments allow an ‘unbounded’ experience of research.

Why don’t we publish our findings in digitally native formats?

Research articulation extends to research dissemination. How to publish research that, in its very nature, resides outside of the traditional academic publication formats? Academic Publishing is a big industry with a global value of US$26 billion. These are $26 bn of printed WORDS.

A check-in with the etymological meaning of the term ‘publish’ shows that publishing as such does not at all specify the mode of making public. Instead, publishing is about getting knowledge out there and announcing it to the world — in whatever way that might be.

Academic publishing goes back over 350 years to the “Philosophical Transaction” published in March 1665 by the Royal Society UK. The Society was founded five years earlier in 1660 and took “Nullius in Verba” as its defining motto. “Nullius in Verba” is Latin for “on the word of no one” or “take nobody’s word for it”, literally “of not any in words”. This denoted that the spoken word in academia was to be superseded by the published written word to be a valid fact. The sole power of knowledge resided with the printed word. The firm belief that only the printed word is valid reaches back to the first printing of the bible by Johan Gutenberg. Because the bible was now printed and reproduced in a mechanical process — and no longer handwritten by individuals — the words of the text were deemed more reliable and hence ‘more true’ than ever before. There was a new precision and trustworthiness in the printed word — a belief that has shaped academia ever since.

Today, there are over 33,000 English-language journals globally and a further 9,400 journals in other than English languages. Roughly 10,000 publishers output about 3 million articles per annum. It is estimated that there are 7 to 8 million researchers who publish their research in any one year. Until the onset of the WWW, academic publishing was exclusively print-only as hardcopies of books, journals and periodicals.

Despite the electronic dissemination of research, print-only journals still exist. While we have not been able to find reliable figures on this, it seems that up to 50% — i.e. half of all academic publications are still print-only. The fields of medicine, the humanities, and especially creative disciplines such as art, music and theatre, are still notably print-heavy.

Less than 30% of the publishing market have Open Access journals. The actual number of published articles as Open Access is even lower at 15–20%. This is despite that full-text Open Access downloads are 89% higher than subscription-based access, online PDF downloads are 42% higher, and unique visitors to OA publications are 23% higher than for subscription access.

Revenue models in academic publishing have not significantly changed since electronic platforms became available. Advertising, subscriptions and sponsorship are still the leading modes of covering publishing costs. This is accompanied by a volunteer model for peer-reviewing submitted manuscripts. Article Processing Charges average $1,000 per article for a print + online issue with some journals/publishers sitting at the $5,000 higher end of the scale. Charges for colour prints often come extra on top of regular production costs and range between $150 to $1,000 per image. About 53% of science journals charge colour printing fees.

Here is an example of an article I published in a journal for visual research. Note that the term ‘visual’ was in the name of the journal and at the very heart of the focus of the publication. I had supplied 18 images and figures — which was cool because many journals, even those in visual studies do not print figures or restrict the number to two or three. I was surprised when I received the printed edition: no colour! I was even more surprised to see the vertical layout of the image-only pages of the journal. I wrote to the editor about this and they came back explaining that the journal prints a single colour page per issue because of printing costs and the vertical layout allows to flexibly manage the page run: Images would be stripped out of the text and added as additional pages as they were required for spread arrangement and number of pages. The journal wanted to have an even page count across its issues.

The subsequent online edition of the journal also delivered b/w prints although colour prints were available. Although digitally enabled, the publication was colour-blind — to a study all about colour. It appeared that colour was considered a ‘decorative asset’ of an image and b/w reproduction an ‘acceptable limitation’ to accurate research dissemination. All of this occurred in 2010.

Three years later, in 2013, we published another visual study in the same journal. Colour had arrived — both in the Web publication and the online PDF. The journal no longer publishes print issues. Curiously, the ‘full-size’ option of the Web image does not link to the larger resolution that is available in the online PDF but to the same small size. And even more curiously, the vertical layout was still well and alive in the digital PDF. Turning a printed journal or book to its side to view an image is a reasonable expectation. But doing the same to a computer screen? It seems that the ‘colour blind’ of print had shifted to a ‘digitally blind’ in the newer online issues of the journal.

Another significant limitation to non-text research dissemination is prevalent intellectual property laws. Copyright regulations and citing conventions seem text-biased. While citation regulations allow the reliance on 10% of verbatim material from a text source — or one chapter — the same metric does not apply to non-text media.

Take a video clip for instance. Instead of allowing the use of 10% of the source, each single frame is considered a separate whole image with 100% copyright integrity.

Let’s do a comparative calculation here. A non-fiction book is typically between 50,000 and 60,000 words. Let’s take 50k for mathematical ease — 10% of 50k = 5,000 words that legally may be reprinted from that book.

If we were to equal words and frames in a video (which is a somewhat difficult equation but for argument’s sake) we would calculate our full video clip with 50,000 frames (divided by 30 fps makes about 28 min of footage). 105 thereof would permit the use of 5,000 frames or 2.8 minutes. But since copyright regulations are very different for text versus visual sources, we cannot even take a single frame from our clip w/o explicit reprint permission.

Copyright and image publishing are significant limitations to the dissemination of visual research. In one of our research projects, we aggregated a visual database of over 1,000 images of publicly accessible images. Funny — or not — is that we hold full ownership of the database — while not owning a single one of the images therein. Thus, publishing this research in an academic journal requires reprint permission for each of the included images.

Fair use regulations allow the use of visuals under certain conditions including research, education and critique of a work. However, academic journals regularly request explicit copyright clearance for their reprints — despite fair use. Fair use is accepted as a research practice but seems to stop short in research dissemination.

How does the extra pressure on digital visual research weigh? In the above example, 66 images were selected for the publication of four articles. A research assistant helped with the identification and clearing of copyrights. Of the 66 images (which is just 6.6 =% of the total database), we could obtain cost-free reprint permission for just 16 images or 24%/about a quarter. 17% of authors/artists did not respond to enquiries including reminders, 3% declined or the sources were not made available in NZ, and 14% of authors/artists asked for reprint fees for their work. Reprint fees are unheard of in citing, critiquing, or referencing written work.

The costs for making publicly available the research amounted to 3,200 NZ$ for the assistant + 2,730 NZ$ for reprint permissions, adding close to $6k for the planned four articles (or $1,500 per article). With most universities maintaining a no-publishing fees policy when supporting their researchers, these are heavy personal costs.

A summary of Open Access Repository content types lists 10,284 items of which 8,225 were text-based and 2,059 were non-text or mixed. This skewed the balance to 80/20% in favour of text-based research outputs.

It would be easy to assume that most research outputs are naturally text. But to what extent is this indeed correct? What are the opportunities — the pressures for digital transformation — that are not currently taken advantage of?

The latest report of the STM — the International Association of Scientific, Technical and Medical Publishers — that holds under its umbrella 2/3 of all academic publications globally, has adopted the Research Communication Model by the Australian National Data Service (2014) that outlines how digital publishing is thought to impact on the future of research dissemination.

The point, in short, is that through implementing already existing digital technologies and online standards, research will increase in multiplicity, meaning that single data collections can be reused within and across various research projects. There will be an increase in the complexity of research communication through easy integration of multiple data points and interlinking of related research outcomes. Digital publishing will raise the flexibility of time and format, and the speed of publishing will significantly accelerate. Today’s average time-to-publication lingers around an average of 100 days from submission to acceptance and then another several weeks to months from acceptance to publication. Overall, a publication on average takes about 9 months all in all — some people have a baby in that time.

A few digital publishing improvements have already been developed and trialled. Open, executable article content, reviewer self-service, auto-referencing and automated publishing are some areas of experimentation. However, all of these initiatives are spun by publishers in order to maintain their business advantage. Researchers themselves have contributed very little to the digital transformation of academic communication. Academics have been identified as notoriously slow to uptake technological innovation. One study has identified academic researchers almost exclusively communicate via email and not social media — ironically, the study named the science fields as the slowest to adopt 21st-century communication standards. Social Collaboration Networks, such as ResearchGate and Academia, are mostly used for self-profiling and not for research interaction.

Even though US$3.5 bn has been spent on investment in digital publishing systems since 2010, digital transformation in academic research is sluggish at best. Digital technologies are not fully recognised and embraced by both publishers and the academic community. Audience expectations of greatly improved services — that are so powerful in the commercial world — seem to have little changed in academia. And as to the aspect of a digital academic culture, this is a term that we made up because it is yet undefined.