September 2017 open access published on: https://www.pubpub.org/pub/co-design-of-video-abstracts-for-scientific-publications
A number of leading scientific journals now allow researchers to publish video–abstracts alongside their peer–reviewed articles. Many of these video–abstracts are produced as an afterthought to the research process, either internally at the university or by external media companies, a choice that creates a trade–off between the scientific validity of the content and the appropriate use of the affordances of the medium.
This article argues in order for video to reach a level of scientific validity comparable to text, an integration of video into the research process itself is a precondition. Boundary spanners, who float between the rigorous work of scientists and the language of moving images, can transform the use of video from a communication technology to a new immutable mobile at the very heart of the scientific process itself.
Digitization is a recent keyword at universities all around the world. This worldwide transformation from analogue, single-location, to digital, globally accessible data enables videos to play a
number of roles in the scientific process; These roles include ‘show and tell’, tutorials, teaching scientific knowledge, process documentation, visual argumentation, or visualizations of the
invisible and fast dissemination of results. As part of this general turn towards the video from the academic community, there are also a growing number of video abstracts accompanying
peer-reviewed (written) articles from scientific journals (e.g. Taylor & Francis; Nature; Elsevier’s). Video abstracts are typically a summarizing add-on to the written article.
This new form of video abstracts is seen as highly promising and potentially disruptive to current academic routines: ‘Scholarly videos are just one example of the new tools the research community is using to communicate with each other […]. Albeit, historically speaking, scholarly videos are not quite such ‘new tools’ as one might think.
In the production of video abstracts, this apparently widespread lack of knowledge about the historical roots of the use of moving images in science is commonly perceivable when researchers discuss the potential of videos. For example, Harvard University chemist George Whitesides referred to the changes between now and the last twenty years in an interview by Berkowitz in 2013: ‘You had to be able to describe your science in words, or tables, or in plots, in two-dimensions on a piece of paper. [...] With videos, you can now describe dynamic phenomena which are simply too complicated, too complex, too unusual, too full of information to do in words and two-dimensional pictures’. A similar consideration was voiced a few years earlier, namely in 1893, by Ludwig Mach [translated from German by the Author]: ‘It would not be an ungrateful task to capture, e.g. the sprouting of a forest mushroom or the turning of a sunflower over a day in such a series of recordings. The growth etc. could very beautifully be studied in this fashion’. But for sure, video production takes time and has to be learned just as we need to learn writing or reading.
So while film has been frequently used in science since a very long time, two recent technological disruptions altered the complicated status of this old relationship. In addition to the radical change of distribution via broad–band internet and user created video streaming platforms, the camera technology itself also went through a major transformation. Never before was the technical handling so simple and satisfying; cameras today are more affordable, light weight and ‘full-automatic’ than ever before, including the ubiquitous smartphone cameras. As a result, ‘do-it-yourself’ videos are increasingly common in society at large, including also in science.
This article will therefore first examine the procedural conditions for video production present in the current academic context, then discuss how those conditions influence the scientific validity of the product and conclude by describing a pathway towards deeper integration of video into science in a co-design set-up.
As video production in science is in the process of becoming main stream, large numbers of researchers find themselves compelled to use this still – from the perspective of the affordances – rather unknown medium. There are basically two possible approaches for researchers to deal with this unfamiliarity: Researchers can take it on themselves, without any support beyond their own research group. This approach is either limited to the researcher’s knowledge of using the laptop webcam or smartphone cam and some tutorial on the use software’s like iMovie, or includes a student who ‘has done internet videos before’, no matter if it was a gaming-livestream or a funny cat video. Note that there are rare exceptions of a student with an actual background of expertise in film, which is an entirely different case on which we will be elaborated in the last section of this article. If no such student is available and the researchers do not feel comfortable to do it by themselves, they can also turn to the universities video support team. While there is a large diversity of such teams at different universities, they are most frequently part of the Information and Communications Technologies (ICT) unit, and, accordingly, tend to employ experts on the technological aspects of multimedia – and only those. While these two approaches generate visually slightly different results. For example due to inexplicable popularity of greenscreen shots in newly build university studios among ICT teams, they are structurally similar in the sense that the camera is only considered as a technical instrument and used to communicate a finished product. Further, there is neither much consideration for the established knowledge of qualitative media studies, nor for applied knowledge of media designers. The following research video, although investigating educational MOOC videos from universities give further insights into the challenge
The lack of communication between research media/film studies, media design and academic colleagues, including some of those who investigate videos for research dissemination, is exemplified visible in this, self-referential auto-performative video abstract about video abstracts: ‘Exploring Video Abstracts in Science Journals: An Overview and Case Study’. The researcher presents himself as a research expert in the field of video abstracts, while the quality of his own video displays a clear lack of knowledge about the audiovisual medium.
Videostill from the Video Abstract ‘Exploring Video Abstracts in Science Journals’: The immediate questions that come to mind include: Why does he have a black line exactly on top of his head, forming a black square with his hair and the screen? Are we supposed to see the back of his body in the reflection of the screen, and if so, why? What kind of meaning does he want to communicate by showing us that he still uses a phone from the 80's – or should he remove it?
Videostill from the Video Abstract: ‘Exploring Video Abstracts in Science Journals’: The picture shows the immediate transition moment, when the researcher cuts from one picture to the other. This overlaying cross-fade, normally used for soft transition montage, results due the similarity of the imagery into a ‘jumping image’. Usually, a ‘jumping image’ is only used when there is a deliberate effort to confuse the viewer.
Researchers can also take a different route reach out to an external small-scale media production companies. It is a sad reality that most of these companies make the majority of their living of ‘selling their media-artistic soul’ for advertisement, a field very aware, but not necessarily famous for transparency when it comes to subtle altering the spin of their messaging by the usual addition of beauty, youth and fun to anything from toothpaste to car insurance. Universities are a new market segment for such companies and there are various competing offers in the field. While they are rapidly developing, the first and easiest approach for such companies is to follow their usual approaches and generate material that is primarily designed to be *appealing and *attention-driven.
The choice made on the implementing agency obviously has a significant impact on the quality of the product. The following table makes a crude attempt to categorize quality along two dimensions: Content Quality and Video Quality.
Content Quality directly relates to scientific validity. The benchmark here is the content of the written scientific article after peer-review. To ensure content quality, the researchers need to have to be present and engaged with the production in all stages to avoid any misrepresentations. Therefore, both self-produced and ICT videos tend to excel in this category. The primary tool to keep the content quality high is the use of words, graphics, diagrams and presentation slides in the video, in other words sticking to media with which the researchers are deeply familiar with. As an example, a recent study analyzed 448 scientific videos from international universities with the striking result, that the recording of existing presentation slides is with 38% one of the most dominant video styles. Accordingly, such videos frequently end up as a one-to-one copy of an academically codified and normatively–structured paper presentation with slides.
The term Video Quality is used here to denote the deliberate use of the spectrum of available affordances and techniques of the medium, as well as the adherence to established aesthetic standards, unless breaking those standards by deliberate choice. This is where marketing companies are at home, who routinely use the affordances of the medium of moving images to their fullest extent. While it is second nature to a filmmaker to use for example match cuts, rhythms of montage or different camera perspectives to convey subtle cues and connotations of the material presented, they may struggle to identify the validity of those connotations from a scientific point of view without explicit guidance and may have a limited understanding of the delicacies of scientific peer-to-peer communication. \
For most videos produced by those companies, ‘getting high on’ Youtube clicks is a primary success indicator, and therefore competition in the market ensures that their approach optimizes this outcome. The approach taken can have a striking similarity to familiar formats of popular science communication, edutainment television and web clips. It is debatable tough, whether those same criteria, namely aesthetic appeal and simplicity, are also sufficient for internal science videos primarily to be watched by a target audience of peers and advanced degree students.
Category 1 and 2 frequently have in common that the videos are a mere afterthought of the original research process, and not part of it in case they are produced after a research paper is already written and peer-reviewed, respectively a research project reaches its final deliverable for dissemination. Thus the moving images stand in stark contrast to the medium written text, which always is present, from proposal to research plan, to research protocols to article draft, to peer feedback to final article – writing is all over the process. This deep integration of the written word supports consistency and validity throughout the process. According to Bruno Latour, the key to workable and valid scientific processes is captured in the concept of ‘immutable mobiles’, which have the defining ‘properties of being *mobile *but also immutable, presentable, *readable *and combinable with one another’. Scientists have a long tradition in using written notes, reflected e.g. in the practice of keeping a lab journal, in which all numbers, measurements and odd observations occurring in the laboratory are recorded. These text pieces are later condensed and transformed into the final research paper, but survive the process from the actual observation and thus remain valid. Dirk Verdicchio used the terminology of immutable mobiles to describe scientific pictures that are reused for popular science video communication. Further, he argues that such scientific pictures get re-contextualized, but can still be defined as immutable. With modern video technology and the integration of experts from the field of audiovisual (artistic) research, Verdicchio's application of the term can also work for moving images. The remainder of this article will focus on the creation of such video-based immutable mobiles in academic video productions, namely category 3, the co-design of video and science (see table above). This process is typically facilitated by either working with a student who, previously was a filmmaker, or a filmmaker with a scientific background. Such dually educated producers of scientific video shall be defined as boundary spanners.
The terminology of boundary spanners is adapted from Kirby’s (2008) definition as experts who are oscillating between the scientific community and the film industry. Such oscillation has been facilitated by the Bologna Reform, as it created a push towards the production of scientific research at European Universities for Applied Sciences and Arts, including film schools. This resulted in the frequent appointment of traditional academics to these schools, a trend that has been criticized due to the frequent lack of practical experience among these new hires, which also limited the legitimacy to teach at a film school. But those appointments have also facilitated the training of a new generation of boundary spanners with a background in both, applied film and scientific research. The first generation to be educated in a mixed (European) Bologna curriculum is now reaching mid-level academic career positions, and there are many more to come. Such experts are able to work in a co-design with scientists to bridge the gap between science and the affordances of the media. This renewed call for a co-design of scientific video abstracts is in-line with the Virgilio Tosi statement (1977), from the Scientific Film Association who required, that the production of moving images require ‘the need to promote interdisciplinary co-operation […], [and] the creation of a specialized training so as to permit close collaboration between scientists and audiovisual technologists in the field of scientific research’.
When video is used as an immutable mobile, the role of the video-scientific co-design in a research project differs dramatically from the currently prevailing modes of academic video production described above. First, there may be distinct components in the object of research that lend themselves particularly well to the affordances of video. Therefore, after interdisciplinary discussion in the research team, it might be possible to formulate a more detailed research question otherwise not considered. Then, at proposal stage, the work plan of the research project already includes a budget for the optimal type of equipment (epistemological instruments) for the foreseen tasks, as well as sufficient time plan that is required to produce the video data material. This video raw material enters the scientific analysis just as any other data. The boundary spanner might use video content analysis methods, or apply postproduction methods to visualize some of the intermediary and final results.
Once this process is finished, and the write-up period starts, the research project has a wide range of material available to the boundary spanner, who in turn has accompanied the entire research project with all its delicate scientific nuances before recording the last statements from the team needed for a final video abstract. In a process like this, video can become an equally valid form of scientific publication.
 For an example, see Video-Abstract of 'Bifurcation of ensemble oscillations and acoustic emissions from early stage cavitation clouds in focused ultrasound’ from the New Journal of Physics, see: https://www.youtube.com/watch?v=vWaso3tNBHU).
 Notice: This article will not differentiate the terminologies between technological aspects of video, film or cinematography but will use them all regarding their qualities of audiovisual affordances.
 Jon Wagner, "Visible materials, visualised theory and images of social research." Visual Studies 21.01 2006: 55-69.
 Further in-depth discussions about visual representations in science, see: Kathryn, Henderson, On line and on paper: Visual representations, visual culture, and computer graphics in design engineering. MIT press, 1998; Steve, Woolgar, "Representation in scientific practice." 1990.
 “University Affairs”, last modified May 22th, 2017. George Whitesides in Berkowitz, Jacob Blog Article, “Video abstracts, the latest trend in scientific publishing: Will publish or perish soon include video or vanish?” *University Affairs, *2013, http://www.universityaffairs.ca/features/feature–article/video–abstracts–the–latest–trend–in–scientific–publishing/.
 Ludwig Mach „Ueber das Princip der Zeitverkürzung in der Serienphotographie“, Photographische Rundschau, Heft 4, 1893, 121–128, pp. 128.
 For the variety of hand-made scientific videos, please see; Loviscach, J. (2014). Friendly Handmade Explanation Videos. Proceedings of EMOOCs 2014, 240-244; Lankshear, C., & Knobel, M. (2003). Do-It-Yourself Broadcasting: Writing Weblogs in a Knowledge Society; or additionally, see publisher website from Taylor & Francis about creating own video-abstracts, http://authorservices.taylorandfrancis.com/video-abstracts/; for a ‘do-it-yourself’ video practice in society, see Youtube statistics: http://www.statisticbrain.com/youtube-statistics/.
 For the original definition of affordances, please see: Gibson, James J. The ecological approach to visual perception: classic edition. Psychology Press, 2014. For the terminology in relation to design affordances, see: Donald A. Norman, "Affordance, conventions, and design." interactions 6.3 (1999): 38-43.
 Scott Spicer, "Exploring video abstracts in science journals: An overview and case study," Journal of Librarianship and Scholarly Communication 2.2 (2014): 1–13.; ‘Video abstract’ on “Youtube”, last modified May 22th, 2017, https://www.youtube.com/watch?v=7t1pLfedLRY.
 Farocki (2011) describes the principles of jumping images: „Every little cutters' handbook tells you how difficult it is to
match up pictures which are very similar to eachother in terms of objects, composition, and framing (cadrage). The eye
always notices errors, discontinuities; better perhaps to speak of the second and then the first rather than of both at the same
time. There has to be a significant change so that the watchful eye first has to adjust before it can relate to the fresh image and
check the quality of the relationship.“ Harun Farocki, “Shot/Countershot: The Most Important Expression in Filmic Law of
Value,” in Nachdruck/Imprint, ed. S. Gaensheimer & N. Schahausen, Berlin, 2001, pp. 86.
 Jeanine Reutemann, J. "Differences and Commonalities – A comparative report of video styles and course descriptions on edX, Coursera, Futurelearn and Iversity", in: European Stakeholders Summit on experiences and best practices in and around MOOCs, Editors: Khalil Mohammad, Ebner Martin, Michael Kopp, Lorenz Anja, Marco Kalz, 2016, 383-392.
 Although tempting: reaching out to a worldwide audience, comparing this formats to youtube channels like Vsauce, Veratasium or CrashCourse or even the attempt to create a viral video shouldn’t be the most important purpose for video-abstracts. Compare statement e.g on http://iopscience.iop.org/journal/1367-2630/page/How%20to%20make%20a%20good%20video%20abstract, last modified May 23th, 2017.
 For example, the show "Galileo" on the German Pro7/Sat1 television station internally defines their target audience as age 14–49, male, with no previous knowledge on the topic. Accordingly they deliberately avoid scientific terminology. But also online edutainment achieves a much broader spread than scientific peer-to-peer communication or teaching. In an expert interview with the author, the Harvard University Neuroscientist David Cox mentioned that their video ‘producer Nadja did an internship at Buzzfeed. She made a video about synesthesia, which is a phenomena in neuroscience. That one video, a little five minute video [had] three quarters of a million hits in one week. That basically dominated anything we’ve ever done in MCB80x [MOOC about Neuroscience101].’ David Cox in an expert interview mit Jeanine Reutemann, Boston 2015.
 Bruno Latour, “Visualization and Cognition: Drawing things together”, in Knowledge and Society Studies in the Sociology of Culture Past and Present, ed. H. Kuklick, Jai Press, Vol. 6, 2011, 1–40, pp.7.
 Dirk Verdicchio, Das Publikum des Lebens: zur Soziologie des populären Wissenschaftsfilms, Vol. 8. 2010, Bielefeld: Transcript.
 David A. Kirby, “Hollywood knowledge: communication between scientific and entertainment cultures”, in Communicating Science in Social Contexts. Springer Netherlands, 2008. 165–180
 Virgilio Tosi/International Scientific Film Association, Cinematography and Scientific Research, UNESCO, 1977, 1–57, pp. 39.
Berkowitz, Jacob. “Video abstracts, the latest trend in scientific publishing: Will publish or perish soon include video or vanish?” “University Affairs”, (2013), http://www.universityaffairs.ca/features/feature–article/video–abstracts–the–latest–trend–in–scientific–publishing/.
Coopmans, Catelijne, et al. Representation in scientific practice revisited. MIT Press, 2014.
Cox David. “Expert Interview.” Expert Interview with Jeanine Reutemann. Boston. 2015.
Farocki, Harun. “Shot/Countershot: The Most Important Expression in Filmic Law of Value.” Nachdruck/Imprint ed. S. Gaensheimer & N. Schahausen. Berlin: Vorwerk 8. 2001.
Gibson, James J. The ecological approach to visual perception: classic edition. Psychology Press, 2014.
Godard, Jean–Luc. Einführung in eine wahre Geschichte des Kinos. Frankfurt am Main: Fischer Taschenbuch, 1984.
Henderson, Kathryn. On line and on paper: Visual representations, visual culture, and computer graphics in design engineering. MIT press, 1998.
Kirby, David A. “Hollywood knowledge: communication between scientific and entertainment cultures.” Communicating Science in Social Contexts. Netherlands: Springer. (2008): 165–180.
Latour, Bruno. “Visualization and Cognition: Drawing things together.” Knowledge and Society Studies in the Sociology of Culture Past and Present, ed. H. Kuklick, Jai Press, Vol. 6 (2011): 1–40.
Loviscach, J. (2014). Friendly Handmade Explanation Videos. Proceedings of EMOOCs 2014, 240-244; Lankshear, C., & Knobel, M. (2003).
New Journal of Physics: “How to make a good video abstract”, http://iopscience.iop.org/journal/1367-2630/page/How to make a good video abstract, last modified May 23th, 2017.
Norman, Donald A. "Affordance, conventions, and design." interactions 6.3 (1999): 38-43.
Mach, Ludwig. „Ueber das Princip der Zeitverkürzung in der Serienphotographie.“ Photographische Rundschau, Heft 4, (1893): 121–128.
Reutemann, J. (2016), "Differences and Commonalities – A comparative report of video styles and course descriptions on edX, Coursera, Futurelearn and Iversity", in: European Stakeholders Summit on experiences and best practices in and around MOOCs, Editors: Khalil Mohammad, Ebner Martin, Michael Kopp, Lorenz Anja, Marco Kalz, 383-392.
Spicer, Scott. "Exploring video abstracts in science journals: An overview and case study." Journal of Librarianship and Scholarly Communication 2.2 (2014): 1–13; ‘Video Abstract’ on “Youtube”. https://www.youtube.com/watch?v=7t1pLfedLRY, last modified August 6th 2017.
Taylor & Francis, “Do-It-Yourself Broadcasting: Writing Weblogs in a Knowledge Society” from http://authorservices.taylorandfrancis.com/video-abstracts/, last modified August 6th 2017.
Verdicchio, Dirk. Das Publikum des Lebens: zur Soziologie des populären Wissenschaftsfilms Vol. 8. Bielefeld: Transcript. 2010.
Virgilio Tosi/International Scientific Film Association, Cinematography and Scientific Research, UNESCO, 1977, 1–57, pp.39.
Wagner, Jon. "Visible materials, visualised theory and images of social research." Visual Studies 21.01 (2006): 55-69.
Youtube statistics, http://www.statisticbrain.com/youtube-statistics/, last modified August 6th 2017.
Images and Tables
Image 1: Videostill from the research video: Video Styles in MOOCs – A journey into the world of digital education, 2016.
Image 2: Videostill from the Video Abstract ‘Exploring Video Abstracts in Science Journals’: The immediate questions that come to the mind of a filmmaker include: Why does he have a black line exactly on top of his head, forming a black square with his hair and the screen? Are we supposed to see the back of his body in the reflection of the screen, and if so, why? What kind of meaning does he want to communicate by showing us that he still uses a phone from the 80's – or should he remove it?
Image 3: Videostill from the Video Abstract: ‘Exploring Video Abstracts in Science Journals’: The picture shows the immediate transition moment, when the researcher cuts from one picture to the other. This overlaying cross-fade, normally used for soft transition montage, results due the similarity of the imagery into a ‘jumping image’. Usually, a ‘jumping image’ is only used when there is a deliberate effort to confuse the viewer.
Table 1: Overview of Video Production Styles at Universities