Poster presentation at GrapHNR 2023

The last event we participated in before our summer break end of July was GrapHNR 2023: Graphs and Networks in the fourth dimension – time and temporality as categories of connectedness, which was a joint conference of the Historical Network Research community and Graphs & Networks in the Humanities. And it proved to be a real summer firework display of amazing research as well as wonderful software projects, with a very international and highly engaged line-up of speakers and guests.

The whole four-day program can be found here. Due to the length of the program, we will only concentrate on highlighting some of the main points from the opening and closing keynote presentations here, which were both very inspiring and an absolute treat. Our own poster presentation by Magnus Pfeffer and Tobias Malmsheimer on the JVMG frontend can be found at the end of this blogpost.

The opening keynote Sifting Through the Haystack: From Data to Stories with Visualization by the highly prolific Benjamin Bach of VisHub at the University of Edinburgh was nothing short of a breathtaking roller-coaster ride on the bleeding edge of network visualization. The central statement of “visualization is both an artifact and a methodology” was expounded in detail through the examination of visualization designs, interactive techniques, tools, as well as ways of communicating research through visualizations and teaching visual data literacy.

Regarding visualization design Benjamin Bach started out with the node link diagram that is familiar to most of us working with network analysis, and its incomprehensibility once a critical number of nodes and links are present in the network (the eponymous hairball from the original title of the talk found in the program: Seeing through the Hairball). We were then presented with various ways of ordering links and nodes, that allow for a better grasping of structures inherent in the networks, such as arc diagrams, motif simplification, adjacency matrix representations with ordering, as well as hybrid techniques and designs for multivariate networks. An array of inspiring designs for temporal and geographic networks – including time arcs, fascinating time curves, edge bundling and flowstrates – were also introduced.

As for network visualization tools, beyond the already – at least in part – familiar set of Gephi, Palladio, visone and NodeXL, the exciting new project developed by Benjamin Bach and his colleagues, the Vistorian was also introduced (see also the github page for Vistorian).

Finally, the pros and cons of communicating networks and network research as well as data and research in more general with the help of comics visualization (graph comics and data comics), and how to facilitate the explanation and teaching of visualization techniques and literacy with the help of well-crafted explanatory cheat sheets were also discussed. (The original source of the pdf below can be found here.)

matrix_relative

Benjamin Bach’s opening keynote was not only an exhilarating ride setting both the tone and the bar for the next three days that followed, but, as the amount of links above probably demonstrates, an absolute treasure trove of resources on network and data visualization.

The wow-factor of the closing keynote presentation Modeling Temporality: Visualizing Chronologies & Graphical Modeling as Knowledge Production by Johanna Drucker was just as impactful as that of the opening one, as it took us on a psychedelic-like journey beyond the confines of our Cartesian conceptions of (space and) time towards a more fundamental re-connection with lived human experience and the real and possible visual representations thereof –– the true power of which this blogpost will unfortunately not be able to capture.

Johanna Drucker’s starting points were that a) visualizations are intellectual concepts manifested in graphic form; b) visual epistemology is a primary mode of knowledge production versus just a form of display; and c) in the digital humanities visualization approaches were taken from the sciences and this has been a potential problem. “Time becomes a washing line, which we hang things on” was the very apt metaphor used to draw attention to the unspoken conventions that underpin our understanding of both time and its representations. This is a very limited way of thinking about time and its potential visual representations. Indeed, there have been, and still are, very different ways of keeping time, which in turn imply divergent understandings of reality.

The key to going beyond the confines of our rigid and somewhat lacking Cartesian representations of time is the distinction between time and temporality. In her work spanning more than twenty years and working with various collaborators Johanna Drucker has been mapping out the potential dimensions along which this difference between time and temporality can be understood and then translated into visual representations of temporal events.

From understanding the many complexities that temporality implies – for example the relationship between elements of narratological time (the telling and the told) – the move towards appropriate visual representations is still far from straight forward. The ChronoVis: Paintbox (see also Skye Hoffman’s notes on its development) is an attempt at providing a toolbox for the visual representation of this type of temporality, where contradictory points can live alongside each other, where recursivity is possible, as well as stretchy timelines, where narration can be accounted for, where periodization does not have to be discrete, and which allows for the non-standardization of metrics for the expression of the affective and the experiential.

What’s more, Johanna Drucker and her collaborators have also prepared suggestions for modifying OWL-Time to be able to account for the narrative nature of objects in temporal accounts, as well as a proposed Hermeneutic and Humanistic ChronoJSON Scheme, again incorporating several of the points mentioned above in relation to temporality.

Although the true magic of the closing keynote is nigh impossible to reproduce in a blogpost like this, we would like to end on one of the final points by Johanna Drucker that captures both the depth and the playfulness of her approach: “the dividing instant is the Zen koan of visual modelling.”

If anything, the conference was a very convincing argument for why historical network research is an amazing field to be working in. The hosting and conference organizing institutions, the Academy of Sciences and Literature | Mainz and the Leibniz Institute of European History (IEG) were both wonderful venues; and thanks to the infrastructural setup this was one of those rare conferences where the hybridness of the event was an actual plus and not a burden. Furthermore, the SchUM / Jewish Mainz tour was one of the most memorable and on-theme cultural programs, which was an absolute highlight of the conference experience. We would like to express our heartfelt gratitude to all the conference organizers, presenters and participants for making this such an amazing event, and we are very much looking forward to the next Graphs & Networks in the Humanities and Historical Network Research community events.

Pfeffer_Malmsheimer_GrapHNR_2023