Issues in Spatio-Temporal Technologies for the Humanities and Arts

Introduction
Spatio-temporal concepts are so ubiquitous that it is easy for us to forget they are essential to everything we do. All expressions of human culture are related to the dimensions of space and time in the manner of their production and consumption, the nature of their medium and the way in which they express these concepts themselves. The Space/Time Working Group (STWG) of the NeDiMAH [1] network held a full day workshop at the DH2012 conference on the topic of theorising methods that exploit space and time in the Digital Humanities. This session proposal is intended to continue and contribute to that discussion.

Issues
Connectivity. Different types of media provide different affordances (Gibson 1986) for representing space and time. We need more work on connecting space and time as represented in texts, audio, and video with the representations being created in GIS systems and other spatiotemporal databases. Texts, sound, and (moving) images are not simply “media” to be spatiotemporally tagged, but may have narrative structures that represent alternative models of space and time (Jewell 2005).

Uncertainty. There has been a lot of emphasis on ambiguity and vagueness in humanist theorising. Our digital tools depend upon our ability to demarcate boundaries, but do the demarcations allow room for the recognition of vagueness, ambiguity, and uncertainty? Should we abandon borders altogether (Berman 2005)? Places and periods are vague, socially defined constructs and source data always leads to imprecise and/or inaccurate data. Can we find a way of encapsulating ambiguity and uncertainty in metadata itself? How can we model for 'vectors of intensity'—impact—thinking about what we really want to do with what time and space tell us?

Contrasts. Although space and time are closely related, there are significant differences between the two. Bakhtin's concept of the chronotope (Bakhtin 1981) implies a strong connection between time and space, but this is a connection between quite different things. Among the most important differences are the natures of their dimensionality (three dimensions vs. one), their different relationships to the static—dynamic continuum where space is static but changeable, whereas time is a “flow” but the past is unchangeable, and the different methods we use to make the communicative leap across spatial and temporal distance.

Representation. Every medium, whether textual, tactile, illustrative or audible, or some combination of them, exploits space and time differently in order to convey its message. “Every medium has the capacity of mediating only certain aspects of the total reality” (Elleström, 2010, 24). The changes required to express the same concepts in different media are often driven by different spatio-temporal requirements. Authors and artists must decide how to collapse reality into the spatio-temporal limitations of a chosen medium, and the nature of those choices can be as interesting as the expression itself.

Absolute vs. Relative. How do we model for movement, trajectory, fluidity and momentum of events and ideas? How do we allow references to float in time and space? Relevant and foundational works exist, particularly for temporal models (e.g. Doerr and Yiortsou 1998; Plewe 2002, 2003; Grossner 2010), but the DH community would benefit greatly by a targeted effort for continued tool development in this area. Holmen and Ore (2010) suggest a way to handle uncertain time in a conceptual documentation system for archaeology. It is, however, not trivial to extend their model to include space as well.

Technical Literacy. In an area combining old with emerging conventions, how do humanists learn when to read a complex visualisation 'with a grain of salt' and to distinguish the 'truthiness' of something that appears on a screen from the complex process of selecting the sources that underlie it? How can humanists learn to justify and critique tool choice in the same way they justify and critique their selection of sources? The development of tools has to be based on informed cooperation, where the representatives of each discipline are allowed to work together on an equal footing.

Theory vs. Practice. The representational issues discussed above also connect to wider questions. The desire to connect digital tool building with the theoretical discourse of the humanities is often expressed, but it is not clear how to do it or what the utility of this will be. Tool building has its own theories, expressed in the form of encoding schemes, data structures, and ontologies. How can we bring the last few years’ discussions in this area forward? This is more than theory-based practice; it also includes practice as a source of theory on many levels, including the experiment as a theory-testing device as well as experiments creating theoretical ideas, also in serendipitous ways.

Presentations
The four panellists will each make a ten minute introduction, discussing aspects of how spatio-temporal datatypes have been, and can be modelled in relational databases. This includes associated operators and algorithms to enable computation of probabilistic or "fuzzy" extents, in the context of specific cases faced when dealing with their own spatial / temporal data and related materials. The nature of the choices that have to be made in order to represent aspects of the spatiotemporal reality in media expressions is a key issue. The focus of each of the presentations will be:

1. The first presentation will be an introduction to the panel where the problems of media translations, especially between texts and maps, will be highlighted by pointing to specific examples. Fiction as well and non-fiction texts will be shown to include spatial description that makes a translation into a map impossible. However, if one sees each map as a possible interpretation of the text, then the complete text-map expression can be seen as a richer expression than either a text or a map alone.
2. Dealing with dates, time periods, temporal granularity, data formats and asynchronous changes is one of the main issues that needs to be addressed in space-time data models. In the second presentation, we will provide some concrete examples from Chinese history of how using specific dates, qualitative typing of dates, and named time periods can be modelled and queried for spatial objects that change over time. Examples of asynchronous changes in GIS will demonstrate the differences between databases set up for time slices, time series, and temporal networks.
3. The third presentation will discuss data modelling challenges encountered in representing temporal, spatial and thematic dimensions of the lives of ~28,000 Britons, spanning a period of several hundred years—all related by birth or marriage. Lifespans are in many cases bounded by vague or uncertain dates, and geographic associations have varying granularity. The goal is a meaningful "contemporary-of" relation joining the problematic temporal and spatial data with tags for individuals’ professions, aggregated to activity spheres.
4. The concluding presentation will put the examples into the perspective of ontological modelling in culture heritage. Using time as an example, it will be demonstrated how a tool implementing modelling principles from CIDOC-CRM and Allen operators (Allen 1983). The tool can infer conflicting dating, increase precision of starts, ends and durations of events and finally display a chronological overview from a given a dataset of events, their time-spans and relations between events. The system can display all possible chronologies for the events in the set thus adding information to a combined data set.
After these four presentations, we will then invite the audience to a discussion, before we conclude by suggesting a list of important issues for further research. This list will be included in a panel report that will be written by the panellists and published digitally for further discussions, online as well as at physical meetings. Through the practical examples we hope to catalyse future workshops where new methods (such as topologies or qualitative descriptors) can be applied in practice, under the umbrella of the Nedimah Space/Time Working Group. Interested participants will be invited to take part in this continued work together with the panellists.

References
Allen, J. F. (1983). Maintaining Knowledge About Temporal Intervals. Commun. ACM 26(11): 832–43.
Bakhtin, M. M. (1981). Forms of Time and the Chronotope in the Novel. Notes toward a Historical Poetics. In Holquist, M. (ed.), The dialogic imagination: four essays. 84–258. Austin: University of Texas Press.
Berman, M. L. (2005). Boundaries or Networks in Historical GIS: Concepts of Measuring Space and Administrative Geography in Chinese History. Historical Geography 33: 118–133.
Doerr, M., and A. Yiortsou (1998). Implementing a Temporal Datatype, Technical Report ICS-FORTH/TR-236. url: www.ics.forth.gr/isl/publications/paperlink/implementing_a_temporal_datatype.ps.gz (checked 2009-05-27)
Elleström, L. (2010). The Modalities of Media: A Model for Understanding Intermedial Relations. In L. Elleström (Ed.), Media borders, multimodality and intermediality. 11–48. Basingstoke: Palgrave McMillan.
Gibson, J. J. (1986). The ecological approach to visual perception. Hillsdale, N.J.: Lawrence Erlbaum. 1st ed.: 1979.
Grossner, K. (2010). Representing Historical Knowledge in Geographic Information Systems. Ph.D. dissertation, University of California, Santa Barbara, United States -- California. URL: http://www.kgeographer.org/assets/Grossner_dissertation_2010.pdf (checked 2012-11-03)
Holmen, J., and C.-E. Ore (2010). Deducing event chronology in a cultural heritage documentation system. In J. W. Crawford and D. Koller (Eds.), Making History Interactive. Computer Applications and Quantitative Methods in Archaeology (CAA). Proceedings of the 37th International Conference. Oxford: Archaeopress.
Jewell, M. O., K. F. Lawrence, and M. M. Tuffield (2005). OntoMedia: An Ontology for the Representation of Heterogeneous Media. In, Multimedia Information Retrieval Workshop (MMIR 2005) SIGIR, ACM SIGIR.
Plewe, B. S. (2002). The nature of uncertainty in historical geographic information. Transactions in GIS, 6(4): 431-456.
Plewe, B. S. (2003). Representing datum-level uncertainty in historical GIS. Cartography and Geographic Information Science, 30(4): 319-334.
Notes
1. Network for Digital Methods in the Arts and Humanities (http://nedimah.eu). NeDiMAH is funded by the European Science Foundation.