| 
  • If you are citizen of an European Union member nation, you may not use this service unless you are at least 16 years old.

  • You already know Dokkio is an AI-powered assistant to organize & manage your digital files & messages. Very soon, Dokkio will support Outlook as well as One Drive. Check it out today!

View
 

Theory Breakout Session

Page history last edited by Ryan Shaw 11 years, 8 months ago

THEORY

 

What do we mean by theory and how do think it might be useful?

 

How do we demarcate and define the work we do, in time and space and cyberspace? What are the theories that we can use that can help us better work with these concepts?

 

Is there utility in, or an imperative to, adopt or connect digital tool-building with the discourse of the humanities? How might that work? What translation is necessary? There is a long tradition of theorizing space and time in the humanities. How does that complex body of work get taken up, and criticized, by the digital humanities?

 

Theory is an entrance to a world you want to represent; there are different kinds of entrances.

 

We need to distinguish between theory and concept. Theory is a whole context of cultural influences, while a concept is directed to particular aspects of that context.  Theory is not an entrance. Theory is not necessarily practical.

 

Is theory separable from observation?

 

Are encoding schemes and data structures theories? People use the word theory differently in different disciplines. Are time ontologies and space ontologies theoretical or practical? Ontology is another word used in many different ways...

 

Our ontologies or classifications reflect particular theories. Nomenclatures do not. The OWL-Time ontology is grounded in a particular theory of time. A nomenclature like the Julian calendar is not?

 

Potentially useful theories for thinking about representation of space and time: Levi-Strauss' levels/scales of history; Lessing's Laocoon; the Bakhtinian chronotopephilosophy of history especially Danto and Ankersmit.

 

 

TOOLS THAT NEED THEORIZING

 

What are some of the practical tool-building and tool-using problems we face that we think theory might shed light on?

 

Trying to create real tools that can help historians to work with space, visualize relationships, social background, and processes.

 

What tools do we have, and what questions can we ask that we couldn't ask without them? What predictive power do our models have?

 

Why don't we have a platform for better understanding territories, places, scales and networks? The relations among these different layers are extremely complex, and we need digital tools for managing that complexity.

 

Extracting entities (persons, organizations, places, events) from oral histories: How to visualize them spatially and temporally, showing relationships across multiple timelines, identifying triggering or causal relations.

 

How are tools for the social sciences combining GIS and statistical analysis applicable in the humanities? 

 

Trying to provide a laboratory for medieval historians to try out different theories about, e.g. the movement of kings? Addressing problems with ambiguous and varying placenames.

 

Dealing with different names and boundaries for places and times between Asia and Europe. Encoding of place and time is a problem; identification of places over time is a problem.

 

Modeling and sharing geohistorical information; trying to understand the experiences of other projects.

 

 

MEDIUM SPECIFICITY

 

How do notions of space, time, and "placeness" connect to sound archives? How to use these archives for social history, to enrich the cultural richness of place, while maintaining what is specific and unique to sound media?

 

Video as a medium particularly suited for representing space and time in a unified way.

 

Video can potentially make a better connection between planning and social context. Might it do the same for the relating placial and spatial history?

 

How to integrate media archives into spatial/placial/temporal analysis? How can data and manuscripts from different places can be used in virtual environments.

 

Multimodality: environments for working with video, captions and related texts simultaneously. Toolboxes for working with archives and transcribing audio & video.

 

Video, texts, images etc. effect a split between the time and space of what is actually shown vs. what is supposed to understood (e.g. Vancouver and Cardiff are filmed so much as "other places" and times).

 

Games as yet another medium which present a very different experience of place and time.

 

 

TEXTS

 

Geographic information in texts: texts and maps can express different things about geography. How to combine representations of space and time in different document excerpts in a conceptual model? Some of the information in texts cannot be expressed as maps.

 

Trying to visualize how to bring together scanned texts, e-texts, bibliographies, biographical databases and geographical databases.

 

Seems to be a difference in method and theory between people working with texts vs. people working with cultural heritage.

 

Working with texts as representations of space and time: there is the space and time of the text production vs. the space and time created by the text in the reader's mind.

 

Reading: we read for different purposes. Do we accept the author's authority? If we are reading for different purposes, we can't agree on what should be modeled.

 

Texts leave out information and describe things at different levels of granularity... however to do comparative analyses, we need some common properties.

 

 

NARRATIVE

 

Visualization of narrative: how can we consider not only time and space, but also objects and characters?

 

Ontologies for describing time and events in narrative; time and space in fiction; where is the fiction/fact boundary and how can that be addressed?

 

People sometimes want to look for the historical reality behind mythological stories.

 

Nonlinear patterns within texts: Geoff Ryman's 253: hypertextual novel set in london, where the links among characters are spatial and temporal (e.g. sitting next to people on the tube, living in the same neighborhood) rather than traditionally narrative. 

 

 

REPRESENTATION

 

We are not treating place names and spatial objects the way we should be. We're just attaching names to spatial objects. But place is more complex, it is layers of abstraction on top of that. What is a place, and how does it relate to physical geography?

 

Museum documentation: CIDOC-CRM provides a set of tools for modeling of place and time.

 

There is a need to consider the audience and their perception, how the brain interprets space and place.

 

The paradox of placial history: place is taken as a proxy that conceals all kinds of cultural difference. Every individual perceives place in a different way: how to accomodate varying placial perception? We can't assume that places are proxies for things "everyone knows" (see Gould & White, Mental Maps).

 

Extracting structured knowledge from documents... there are lots of people creating timelines, but the data tends to get embedded in specific tools, making it hard to share with others. We need standards for publishing data so others can use it. Tools die, can we ensure that the data/content doesn't die with it? But what about when tools are tightly integrated with inquiry, and co-developed with it? 

 

What is your data? You can't document the playing of a computer game. There are limits to documentation. 

 

The difficult problem of representing both time and space. You can have precise time and coarsely divided space, or very precise locations and coarsely divided time. There is a tradeoff between representing time and space. This brings problems when you wish to combine data that makes different tradeoffs.

 

Holman & Ore, "Deducing event chronology in a cultural heritage documentation system." Using CIDOC-CRM timespans to model archaeology. Make the start point and end point fuzzy, with each point in time mapped to a 4-tuple. Supports reasoning about ordering of events despite uncertainty. Similar approach works for locations as well: things can be located relative to their containing objects, e.g. people within buildings, buildings within cities, cities within countries.

 

 

UNCERTAINTY & AMBIGUITY & VAGUENESS

 

How do you deal with uncertainty, inconsistency, ambiguity, and vagueness, but still manage to achieve a "distant" view on the whole archive or dataset, despite this fuzziness?

 

Fuzziness and uncertainty; memories start to fade, can we develop tools that support looking for inconsistencies or differences among remembered stories.

 

Lots of emphasis on ambiguity and vagueness in humanist theorizing, looking at how we make demarcations and boundaries. Our digital tools require precision or clarity. How do we reconcile that? Is there space for recognition of ambiguity? CS offers probabilistic theories, fuzzy logic...

 

The Neatline solution is purely visual (a gradient), but can you have a debate about that?

 

There are not just uncertainties and ambiguities but political biases, etc. that fuel debate.

 

Is the logic of the geographical grid and the relationships it defines not in question? If we can be certain of it and put it into geographical databases, if we have that underlying framework, we can potentially combine many different perspectives.

 

On the other hand, chronicle may be more important than geography. Perhaps the chronotope is a more suitable model or framework in that case.

 

There is a spectrum from "pure" fiction to "factual" history, with plenty of stuff in the middle. On one end there are facts but truth is unknown, on the other there are no facts of the matter.

 

Are there tools for assigning place in a fuzzy way? The modifiable areal unit problem: if you name a country, state, locality, neighborhood, places look different at different scales. Analogous problem exists for time (Levi-Strauss's levels of history), and also there is the problem of absolute vs. relative times.

 

Archeological models and how to represent uncertainty: panorama bubbles (Paul Bourke), fuzziness (Arcadia 2003 or VAST 2005), understanding virtual environments based on what they draw... lifeClipper, Swiss augmented reality project that tries to build in fuzziness into its representations.

 

EDTF: proposed standard from the Library of Congress for encoding approximate/uncertain/imprecise times and dates.

 

Once we start defining what place is, we remove information. Vagueness is useful in that it carries (the potential for?) information.

Comments (0)

You don't have permission to comment on this page.