The Historical Grounding Ratio (HGR) is introduced, defined as the proportion of claim-bearing information units in a generated narrative that are supported by historical archives that are supported by historical archives, and offers an operational measure for comparing information-allocation strategies and their experiential consequences.
Abstract
Location-aware generative systems can now select historical archives and real-time contextual information based on a user's surroundings to automatically generate narratives for urban heritage walks. Yet when multiple sources jointly inform generation, existing systems provide neither a clear representation of how much content from each source actually appears in the output nor an operational means of measuring it. We introduce the Historical Grounding Ratio (HGR), defined as the proportion of claim-bearing information units in a generated narrative that are supported by historical archives. HGR turns the realized share of historical evidence in a narrative into a directly measurable design parameter. In GeoDrama, a mobile narrative system, we created three conditions that used a common retrieval procedure and comparable evidence-bundle sizes while varying the allocation of information from different sources during generation. We evaluated how changes in HGR affected narrative experience through a within-subject walking study with 18 participants. Increasing HGR significantly strengthened the perceived relevance between narrative content and the specific location. However, historical understanding, integration with the visible scene, appropriateness of the amount of information, and intention to explore further did not increase monotonically with HGR; all four measures were highest in the intermediate, balanced condition. These findings show that designing location-aware generative interfaces involves not only retrieving relevant material but also determining how information from different sources composes the final output. HGR offers an operational measure for comparing information-allocation strategies and their experiential consequences.
Abstract. This study investigates the potential of artificial intelligence in translating historical textual registers into graphic reconstruction, exploring interdisciplinary approaches that integrate historical studies and GeoAI. The case study focuses on the “Calcato” of Gandino, a mid-18th-century textual register accompanied by a large-scale territorial map. Produced by land surveyors “treading the ground”, the register describes property boundaries as closed polylines, whose vertices correspond to fixed landscape markers and whose segments are defined by length and orientation. As the map represents a later graphic transposition rather than a direct survey output, it provides the basis for this research: replicating that translation process through computational tools. Following an initial exploratory phase to validate the concept, conversational AI models were used to transcribe the historical text and compute vectors to generate local coordinate tables necessary for reconstructing the geometries of the described areas. The method was tested on cases with varying levels of complexity, comparing the results with the corresponding map representations. A final phase extended the analysis to registers from different historical periods to enable further testing and comparison. The results demonstrate the feasibility of transforming archival textual data into spatial representations, supporting qualitative analysis and the study of historical and territorial transformations.
F. Biolo, Franco Guzzetti, I. Balestreri· The International Archives o...· 0 citations
Digital three-dimensional reconstructions of architectural heritage have become increasingly used as research tools, yet their visually coherent representations often conceal the interpretive processes, uncertainties, and evidential gaps underlying their production. This paper proposes a method for visualising evidential proximity, defined as the distance between the reconstructed geometry and the documented sources that support it, within three-dimensional architectural restitution models. The method is based on the systematic evaluation of heterogeneous historical sources according to their integration modes (transposition, interpretation, and extrapolation), across multiple analytical aspects of the architectural elements. These evaluations are represented directly within the 3D model using a source-based visual encoding approach implemented through native tools in Blender, an open-source 3D modeling software. Its application is demonstrated through the digital reconstruction of the dining room of the Palais Stoclet, a UNESCO World Heritage building that is inaccessible for direct survey and only known through documentary sources. Embedding interpretive metadata within the model itself frames 3D reconstruction as a reflexive space for scientific exchange, contributing to transparency, critical evaluation, and the projection of reconstruction hypotheses across time.
Flavius-Teofil Neaga, Denis Derycke, D. Lo Buglio· SHS Web of Conferences· 0 citations
Visit any living history museum, and you’ll find yourself surrounded by costumed historical interpreters. By demonstrating the appearance of life in the past, these interpreters perform history. While not always considered a form of theatre, first-person historical interpretation contains some of its key hallmarks: liveness, mimesis and storytelling. Specific moments – such as demonstrations and shifts in body language or terminology – can be classified as “twice-behaved behaviours,” an integral process of performance, according to Richard Schechner. By embodying historical figures, interpreters essentially become actors. This research engages with such scholars as Schechner, Alan Filewood and Freddie Rokem. Additionally, I draw on my own experiences as a historical interpreter to look at the ways in which first-person historical interpretation not only falls under the definition of theatre, but also is a unique and successful storytelling tool due to this theatricality.
This study examines how the affected community of the 2025 Sumatra flood disaster is visually and discursively represented in an eight-photograph Instagram carousel published by the international news outlet Radio France Internationale (@rfi). Guided by Kress and van Leeuwen's (2006) Multimodal Discourse Analysis and its tripartite model of representational, interactive and compositional metafunctions, this study undertakes a qualitative close reading of both the visual and verbal layers of the carousel to determine the extent to which the local community is included as a legitimate, voiced participant in the narrative of its own disaster, as opposed to being reduced to a passive recipient of institutional rescue or an anonymous statistic of loss. The analysis finds that visual inclusion in the carousel is graded and frequently contradictory rather than binary: institutional actors such as search-and-rescue personnel and military aircraft are consistently constructed as agents endowed with narrative, interactive and compositional dominance, while the affected community oscillates between brief moments of agency, symbolic abstraction and complete representational absence. The single instance in which a community member directly addresses the viewer through gaze is immediately undercut by a compositional text overlay reporting the death toll across her body, while the two photographs depicting the clearest community agency receive no acknowledgment whatsoever in the accompanying caption. This study identifies these patterns as a false individualization arc and a multimodal disjunction between visual presence and discursive voice, demonstrating that a community can be made visible within a disaster narrative while remaining systematically unheard. As a single-case analysis of one carousel from one outlet, this study does not permit generalization across platforms, disaster events or news organizations and its findings should be read as an in-depth diagnostic instance rather than a representative sample of Global North disaster coverage. Nonetheless, by extending Kress and van Leeuwen's (2006) framework beyond static print and single-image analysis to the sequential, multi-image logic of social media carousels, this study offers a replicable analytical model for detecting the gap between visibility and voice and contributes a concrete methodological entry point for future comparative research into how Global South communities are positioned within Global North disaster journalism.
This article revisits the audiowalk placeless: San’ya, originally produced in 2021 and re-staged in 2025 with geolocated audio technology. Using a practice-led approach and drawing on qualitative interviews and observational field notes, the study examines audio-based walking practices as a form of spatial computing in which digital sound is embedded in physical space and activated through movement. The case is situated in San’ya—a district name that officially no longer exists but remains associated with marginalization—within Tokyo’s continually changing urban fabric, where the past is often only weakly articulated in the built environment. The audiowalk traces layered histories, from the site’s adjacency to the Kozukappara execution grounds in the Edo period, through its history as a settlement for the stigmatized Burakumin caste, and its present-day role as home to an ageing population of day laborers. By keeping the narrative content constant while altering its technical mediation, the re-staging isolates how geolocation and contingency shape participant experience. The study contributes to spatial computing discourse by demonstrating that geolocated audio can function as a spatial interface that reconfigures movement and attention without displacing embodied awareness from the environment. It suggests that such practices can foster engagement with non-monumental urban histories without stabilizing absence into fixed narratives, while also foregrounding ethical questions about mediated presence in socially uneven spaces.
Unlike Heidegger’s claim that architecture reveals its meaning through presence and direct experience, today such encounters are increasingly mediated by images. Architecture circulates first as fragments, photographed, hashtagged, and digitally reshared, before or after being physically experienced. This research introduces Jansang, a Korean term meaning «afterimage», as a lens through which architectural memory is constructed both prior to and following physical encounter, particularly in the context of ruins and fragments. Focusing on the Ruïnekerk, a fifteenth-century church ruin in Bergen, the Netherlands, the study maps how online imagery generates symbolic constellations—windows, crosses, sky openings, and the contours of earth and sky. These recurrent motifs reveal a process of digital recollection through which the ruin is continually rebuilt within shared perception.
The project develops a «kit of parts», a design tool derived from mapping analyses of social-media posts, to translate these visual memories into spatial strategies for adaptive reuse. The approach, inherently project-specific, emerges from how people perceive, document, and share architecture. By shifting focus from material preservation to participatory perception, Jansang proposes a democratic paradigm for reuse: architecture understood not as a fixed object but as a living afterimage, continually shaped by the ways we frame, share, and remember space.
Hyunggyu Kim· Proceedings of the Internati...· 0 citations
A new method, called CW-Net, translates the reasoning process of an autonomous vehicle’s AI system into understandable concepts that explain its behavior.