Abstract

We present StoryMachine, an international, transdisciplinary project developing a spatial hypertext infrastructure for transcultural digital folkloristics. Contemporary digital folkloristics infrastructure shares a foundational assumption: that knowledge is constituted before it enters a system, and that the system's role is to store, enrich, and surface it. We argue that spatial hypertext is not merely an alternative interface for organizing cultural knowledge, it also enables a different epistemological stance, in which the act of arrangement itself constitutes an epistemic act. While early digital humanities archives often mirrored the organizational logics of print culture, later work has increasingly challenged these inherited epistemologies. StoryMachine contributes to this shift by enabling interpretive relations to emerge through exploratory spatial arrangement rather than requiring relations to be fully specified in advance, aligning with relational and decolonial approaches to cultural data. Building on work in spatial hypertext and theories of distributed and material cognition, we reconceptualize ambiguity not as a deficiency of incomplete structure, but as a productive condition for exploratory interpretation and relational sensemaking. Using StoryMachine as a design context, we develop an account of spatial hypertext as infrastructure for exploratory relation-building prior to semantic stabilization. We conclude by outlining a series of open questions concerning ambiguity, interpretive emergence, computational assistance, and the future design of exploratory knowledge infrastructures for digital folkloristics.

1 Introduction: A Methodological Gap in Digital Folkloristics

Vannevar Bush's 1945 vision of the Memex proposed a device that would store, retrieve, and above all link human knowledge in the associative trails that mirror the mind's own workings [8]. Over the span of eight decades, hypertext research has pursued variants of this vision, from Ted Nelson's Literary Machines [33] through the navigational link-node systems that underpin the Web [10, 26] to the open hypermedia systems and component-based architectures that have extended and formalized the paradigm [4, 34]. Across this history, the dominant question has been: how can we best store and navigate knowledge that already exists? Spatial hypertext also invites a different question: what knowledge can be constituted through spatial practice that cannot be constituted in any other way [28, 44, 45]?

This paper introduces a research project that advances this issue in a domain that has never been a primary testing ground for spatial hypertext: digital folkloristics. Project StoryMachine is developing a spatial hypertext platform for exploring, preserving and providing access to folklore traditions across German- and English-speaking communities, situating itself within the existing landscape of digital folkloristics infrastructure. That landscape shares an epistemological assumption that spatial hypertext is uniquely positioned to challenge. Folklore collections have always grappled with a fundamental asymmetry: the material they preserve is constitutively ambiguous, contextual, and performative, while the systems built to organize and retrieve it are constitutively formal, categorical, and static. This asymmetry is not an oversight but a structural feature of how knowledge about folklore has been produced and institutionalized [37]. Collectors, archivists, and scholars have exercised continuous judgment about what was worth collecting, how to classify it, and what relationships between items were meaningful. In doing so, they have constructed particular visions of tradition as much as they have preserved it. Some of the most far-reaching post paradigm shift critiques of folkloristics were in part responses to this structural asymmetry: the archive's formal organization suppressing the very complexity that performance theory, feminist folklore studies, and postcolonial critique sought to recover [18]. Evolving alongside these disciplinary, methodological, and institutional transformations, the digital turn has invigorated folklore archives as living repositories whose reach and research potential have been profoundly expanded through digitization, structured metadata, and participatory platforms. Over the past two decades, archives across Europe have developed a broad array of digital infrastructures that make collections more accessible, searchable, and interconnected, reshaping archival workflows and enabling new forms of engagement with tradition. However, the underlying epistemological model, the assumption that knowledge about folklore must be formalized before it can be stored, searched, or surfaced, has remained largely intact. It is precisely this model that StoryMachine seeks to rethink. Drawing on the spatial hypertext insight that spatial arrangements can express and computationally support interpretation before formal semantic articulation, the project seeks to build systems in which human associative, tacit, and affective knowledge enters into productive dialogue with computational processes. The aim of hypertext understood as method is the augmentation of human intellect: not the automation of reasoning, but the creation of environments in which human and machine capabilities jointly enable new forms of understanding [3]. Using StoryMachine as a concrete case, this paper argues that spatial hypertext enables a different epistemological orientation toward knowledge organization in digital folkloristics. We develop this argument in dialogue with existing digital folkloristics infrastructures and theories of spatial hypertext before outlining the methodological and epistemological questions that follow from it.

2 The State of Digital Folkloristics Infrastructures

Digital folkloristics infrastructure has developed along four broadly distinguishable methodological strands, each with its own epistemological commitments and its own version of the same foundational limitation.

2.1 Archiving and Classification

The most visible strand consists of digital archives, databases, and classification systems. Over the past two decades, folklore archives across Europe and beyond have undergone a transformation by digitizing millions of manuscript pages, audio recordings, and photographs, building searchable, metadata-rich databases, and connecting collections across national and linguistic borders. In Latvia, the HUMMA platform 1 has consolidated 28 data repositories comprising over 5,786 collections, more than 1.1 million digitized files, 474,000 textual units with metadata, and 36,867 person profiles, supporting citizen science participation, text annotation, and geospatial visualization [25]. In Germany, WossiDiA 2 has transformed Wossidlo's Mecklenburg ethnographic collection into a hypergraph database of over 1.1 million nodes and 524,000 typed relationships, enabling nuanced traversal of dense archival networks that the print collection could not have provided [41]. The ISEBEL project 3 has built a cross-border harvester connecting belief legend archives across Europe through shared metadata schemas and automated translation [30]. The transition from analog collections to digital infrastructures has substantially expanded the accessibility, interoperability, navigability, and collaborative use of folkloristic materials [21]. However, this infrastructure remains object-centric, encoding folklore as document rather than living practice, the limitation UNESCO's 2003 Intangible Cultural Heritage Convention sought to address [19].

The dominant organizing principle of this infrastructure is a deposit-then-query model: knowledge must enter these systems in a formalized state as a metadata field, a keyword, a motif number or a hyperedge type. The Aarne-Thompson-Uther type index (ATU) [49], which underpins most of this classification work, was itself designed as a stable reference system for expert use, assigning each tale type a hierarchical number. Harvilahti has argued that this typological apparatus, however useful as scholarly shorthand, encodes assumptions about narrative structure rooted in European traditions that are not universally applicable, and that indexes need to evolve beyond typological rigidity [14]. Digitizing the ATU has made it faster to query, not more epistemologically flexible. Ryan and Mac Cárthaigh's work on the Irish Folklore Collection points to the extensive ongoing effort required to develop and critically assess flexible approaches to classification for qualitative cultural data beyond the enhancement of metadata [40]. And Skott observes that archival selections tend to present themselves to later users as the totality of the tradition: “selections often become wholes” [46].

2.2 Computational Folkloristics

A second major strand of digital folkloristics is computational folkloristics; the application of algorithmic methods to folklore corpora to reveal patterns, structures, and variation at scales inaccessible to manual analysis. Tangherlini's folklore macroscope program demonstrated that machine learning and network analysis applied to large digitized folklore collections could reveal narrative geography, thematic clustering, and intertextual relationships that no individual scholar could trace [1, 47]. The FILTER 4 project developed a complete computational research environment for Finnish and Estonian runic poetry; an automated preprocessing pipeline combining collections into a unified corpus of 283,571 texts; similarity computation using sequence alignment and vector search; the Runoregi5 close-reading interface; and, a data visualization application, enabling both distant and close reading of intertextual patterns, formulaic variation, and poetic geography [22]. The Dutch Folktale Database has applied automatic enrichment and classification methods to its collections, demonstrating how Natural Language Processing techniques can support folklore typology at scale [31]. These projects have substantially advanced the methodological capabilities of folklore scholarship.

But the FILTER team themselves acknowledge a persistent challenge: how to “systematize or log the versatile, complex and sometimes meandering reading processes” [22] that characterize humanistic research into oral tradition. Computational methods can reveal what has already been formalized: the similarity between classified poem types, the geographic distribution of indexed motifs, the network structure of typed relationships. What they cannot access is the interpretive intuition that precedes classification: the moment when a researcher senses that two tales are related without yet knowing whether the relationship is structural, thematic, historical or performative. The macroscope sees what has already been named; it cannot see the act of naming as it happens.

2.3 Participatory and Citizen Science Approaches

A third strand of digital folkloristics concerns participatory practice: the use of digital platforms to involve communities, volunteers, and members of the public in the co-production of folklore knowledge. A long history of “crowdsourcing” in folklore collecting has been documented [39], and digital platforms have significantly extended the reach of participatory methods. The Latvian manuscript transcription campaigns supported by HUMMA and the garamantas.lv 6 platform, the citizen science functions of Dúchas.ie 7 in Ireland, and the broader iesaisties.lv 8 public engagement platform all invite community involvement in building digital folklore collections [25, 39]. Tolbert and Johnson's digital folkloristics manifesto argues for approaches that integrate participatory design and ethnographic reflexivity with digital methods, treating the communities studied as active partners in knowledge production rather than passive sources [48].

Yet even the most participatory of these platforms invite contribution within predefined frameworks: contributors transcribe, tag, annotate, and classify within schemas designed by archivists and scholars. The framework for participation is given in advance. This is not a criticism of those platforms, as interoperable metadata and sustainable infrastructure require stable schemas. But it does mark a structural limit. Participatory digital folkloristics, as currently practiced, broadens the population of those who can contribute to knowledge production without fundamentally changing how knowledge is produced: still through deposit of formalized claims into a stable system, not through spatial negotiation of relationships that are not yet formal enough to deposit.

2.4 Emergent Digital Folklore

A fourth strand, and the most rapidly growing, concerns emergent digital folklore: the memes, viral narratives, participatory remixes, and algorithmic creativity that constitute what de Seta has called “algorithmic folklore” [11]. These phenomena have become objects of serious scholarly attention, with researchers drawing on digital methods to trace their variation, transmission, and cultural meaning [6]. But they sit uneasily with the archival infrastructure developed for oral tradition. They have no collectors, no field recording dates, no geographic provenance in any traditional sense. They circulate through platforms whose interfaces and algorithms shape their variation in ways that parallel but differ from the social dynamics of oral transmission. ATU type numbers do not fit TikTok 9 trends; and, building a dedicated archive for each emergent form is neither scalable nor epistemologically satisfying. What is needed is an approach capable of exploring the relationships between established tradition and emergent digital folklore without presupposing what those relationships are. This is not currently available in the digital folkloristics toolkit.

Across all four strands of digital folkloristics infrastructure, a common assumption operates: knowledge about folklore must be articulated before it can be systematically processed. The archive receives classified items. The macroscope analyzes typed relationships. The participatory platform accepts tagged contributions. The computational pipeline runs downstream of formalization decisions. This is not a deficiency that more sophisticated technology will overcome; it is a structural feature of the deposit-then-query epistemological model that underlies current digital folkloristics infrastructure at every level. What is missing is a layer that can receive, analyze, and build upon interpretive acts that precede formal articulation; pre-classificatory intuition, and the arrangement that encodes meaning before the analyst has the words for it. This is what StoryMachine seeks to address.

3 Epistemologies of Spatial Hypertext

As classificatory systems accumulate material over time, their underlying assumptions may become increasingly sedimented through continued use, shaping future interpretive possibilities through the recursive stabilization of classificatory relations [7, 38]. While such organization enables consistency, interoperability, and computational analysis at scale, it can also privilege established classificatory perspectives [35] and make more tentative, associative, or emergent interpretive relations difficult to articulate.

Spatial hypertext introduces a different orientation toward knowledge organization. Rather than requiring semantic relations to be fully specified prior to interaction, spatial hypertext allows meaning to emerge through acts of spatial arrangement, associative grouping, and interpretive juxtaposition. Historically, spatial hypertext developed partly in response to the problem of premature formalization, in which systems required users to specify semantic relations and classificatory structures before those users themselves understood the nature of those relations [2, 43]. Early spatial hypertext systems instead emphasized “the ability to leave structure implicit and informal” [29], allowing associations to remain provisional, negotiable, and context-dependent. Meaning therefore emerges not solely through formal symbolic declaration, but through the interpretive organization of materials within a shared spatial field.

This epistemological distinction becomes particularly visible in archival infrastructures such as WossiDiA (cf. Subsection 2.1), which demonstrates how digital folklore archives can move beyond isolated classification toward richer forms of contextual association. Rather than organizing folklore exclusively through rigid hierarchical taxonomies, WossiDiA preserves extensive networks of cross-references, annotations, and associative links created by Richard Wossidlo and later archivists [32]. WossiDiA also introduces an important reflexive dimension by allowing users to retrace editorial selections, textual transformations, and archival interventions through connections between published editions and unpublished source materials [42]. Users may therefore not only access stabilized folklore materials, but also partially reconstruct the processes through which those materials became organized and interpreted within the archive itself. Yet these relational and editorial structures remain largely inherited from a historically sedimented archival system whose semantic organization was established in advance. The interpretive associations available to users are therefore primarily inspectable and traversable rather than dynamically produced through interaction itself.

The same tendency can be observed in computational folkloristics infrastructures such as the FILTER project (cf. Subsection 2.2). Large-scale computational methods enable distant reading, comparative analysis, and algorithmic pattern recognition across extensive corpora. Yet these approaches still depend upon semantic segmentation, normalization, annotation, and encoding. Similarly, participatory infrastructures broaden collective involvement through annotations, transcriptions, metadata enrichment, or polyvocal archival representation [9]. Yet participation typically operates within preconfigured relational structures rather than creating them.

These limitations point toward a different interpretive orientation in which relation-building may occur prior to formal semantic stabilization. Rather than only retrieving, traversing, or computationally analyzing predefined symbolic relations, users may construct provisional interpretive constellations through exploratory spatial association, negotiating what may count as a meaningful relation prior to formal semantic articulation. Associations may remain overlapping, revisable, ambiguous, and partially unresolved rather than converging toward fixed semantic organization. Such exploratory openness becomes particularly significant in rapidly evolving forms of digital vernacular culture, where motifs circulate through social media, recommendation systems, memes, and other algorithmically mediated environments, continuously reshaping interpretive relations and resisting stable taxonomic closure.

This movement away from fully stabilized semantic relations aligns with broader hermeneutic critiques of classificatory knowledge organization [15]. Traditional systems of knowledge organization privilege semantic closure and fixed organizational structures over interpretive plurality, reducing ambiguity through explicit classificatory determination. Spatial hypertext does not reject organization altogether, but supports interpretive relations that remain fluid and revisable within evolving contexts. Meaning emerges relationally through acts of exploratory association rather than solely through predefined classificatory frameworks.

From this perspective, spatial hypertext is not simply an alternative interface paradigm, but a different hermeneutic model of knowledge production. Conventional archival and semantic infrastructures largely assume that structure produces interpretation: classificatory frameworks determine the semantic field within which meaning can be retrieved and understood. Spatial hypertext partially reverses this relationship. Here, interpretation produces structure. By postponing commitment to stabilized classificatory systems, spatial hypertext allows interpretive relations to emerge through spatial association. Spatial arrangement therefore functions not merely as presentation, but as semantic proposition through which interpretive structures are actively produced, negotiated, and revised.

StoryMachine explores this epistemological orientation through a cognitive-map-based interface in which users spatially arrange motifs, narrative fragments, and generated suggestions. Here, spatial organization is not secondary to knowledge production but constitutive of it. The evolving arrangement both represents interpretation and enables new interpretive relations to emerge.

4 Ambiguity and Interpretive Emergence in Spatial Hypertext

If spatial hypertext postpones commitment to fully stabilized semantic relations, interpretive structures are no longer fully determined prior to interaction. Meaning instead emerges progressively through exploratory engagement with environments whose relations remain provisional, negotiable, and incomplete. Ambiguity here does not imply the absence of organization, but partial semantic indeterminacy in which interpretive relations remain suggestive without becoming fully specified in advance, making ambiguity a constitutive rather than deficient condition of interpretive activity. Early spatial hypertext research accordingly treated interpretive openness not as a deficiency of incomplete structure, but as a productive condition supporting exploratory sensemaking [5, 29].

Spatial hypertext should therefore not be understood merely as an external representation of already-formed knowledge. Rather, spatial arrangements participate directly in interpretive activity, supporting comparison, inference, and the gradual emergence of conceptual structure. This understanding aligns with theories from distributed cognition, according to which interpretation emerges through interaction between users and external representations rather than solely within the individual mind [20]. It also resonates with epistemological accounts of folklore that emphasize practical, distributed, and non-testimonial forms of cultural knowledge transmission over formalized propositional structures [12]. Spatial arrangements do not simply display relations; they organize interpretive attention by making some associations salient while leaving others latent or unresolved, thereby affording interpretive exploration [16, 17, 24]. Rearranging, clustering, separating, and juxtaposing materials therefore function as epistemic activities through which understanding emerges through interaction. Thinking in spatial hypertext thus becomes inseparable from thinking with and through things, in which material structures participate in interpretation rather than merely representing it [27].

Exploratory engagement with provisional spatial arrangements also enables interpretive structures to emerge that are not fully contained in any individual fragment or relation alone. As partially related motifs, annotations, or narrative fragments are brought into provisional association, new meanings emerge through their integration into new interpretive structures [13]. Meaning therefore derives not simply from retrieving predefined symbolic relations, but from the dynamic integration of partially specified associations.

If interpretive structures emerge through the ongoing integration of provisional associations, stabilization can no longer be understood simply as the resolution of ambiguity into explicit structure. Concerning spatial hypertext, ambiguity was often understood as a transitional condition that could gradually give way to more stable and explicit organizational structures through processes of incremental formalization [43]. Instead, stabilization becomes partial, local, and contingent: a temporary closure within an ongoing interpretive process rather than its final endpoint. Interpretive structures can remain open because meaning continues to emerge through ongoing material engagement rather than terminating in finalized symbolic closure. Stabilization thus becomes a situated, relational process rather than the establishment of a fixed semantic object [36].

This perspective also carries broader implications for spatial hypertext theory and system design. If interpretive structures remain open, revisable, and emergent rather than gradually converging toward finalized semantic organization, spatial hypertext systems cannot be understood merely as transitional environments for informal knowledge management prior to formalization. Instead, they become infrastructures for ongoing interpretive negotiation in which ambiguity, provisionality, and overlapping relations remain constitutive features of knowledge production itself. The role of the system therefore shifts from clarifying structure to sustaining ongoing interpretive negotiation.

StoryMachine therefore explores spatial hypertext not simply as a visualization technique for folkloristic materials, but as an interpretive infrastructure designed to support the formation and provisional stabilization of relations that may remain unresolved within conventional archival and computational systems.

5 Open Questions

Contemporary digital folkloristic infrastructures have substantially expanded the accessibility, interoperability, and computational analyzability of folkloric materials. Yet these infrastructures frequently depend upon the prior stabilization of interpretive relations into classificatory, semantic, or algorithmically processable structures. This paper has argued that spatial hypertext offers a different epistemological orientation in which interpretive relations may emerge provisionally through exploratory spatial interaction rather than solely through predefined semantic formalization. Ambiguity, from this perspective, does not simply indicate incomplete organization, but may function as a productive condition for exploratory interpretation, relational sensemaking, and the gradual emergence of meaning.

Such an orientation raises a number of unresolved questions for both digital folkloristics and spatial hypertext research. One central question concerns stabilization itself. In the context of spatial hypertext, stabilization refers not necessarily to the elimination of ambiguity, but to the progressive formalization of interpretive relations into structures that become increasingly explicit, communicable, computationally processable, or institutionally established. If interpretive relations remain provisional, revisable, and context-dependent, how much stabilization is necessary for meaningful archival organization, interoperability, or computational analysis? Under what conditions should interpretive structures become formalized, and can multiple partially stabilized interpretations coexist without collapsing into incoherence? Spatial hypertext challenges the assumption that semantic closure necessarily represents the final or ideal state of knowledge organization, yet infrastructures cannot operate without some degree of stabilization. The problem is therefore not whether stabilization should occur, but how systems might support temporary, local, and revisable forms of stabilization while preserving interpretive flexibility.

Closely related to stabilization is the question of granularity. Across the landscape of digital folkloristics infrastructures, different systems embody different answers to the problem of how much contextual specificity should be sacrificed for tractability. WossiDiA demonstrates what highly fine-grained formalization can preserve, while projects such as FILTER reveal the analytical possibilities opened by corpus-scale standardization and automation. At the same time, genre-specific infrastructures often encounter difficulties when attempting to integrate heterogeneous collections at larger scales [22, 23, 41]. Spatial hypertext does not eliminate the need for formalization, but allows commitments to particular interpretations to be postponed while exploratory relations remain negotiable. At the same time, many infrastructural functions, including storage, retrieval, interoperability, and computational assistance, require some form of representational commitment. The question is therefore not whether interpretation must eventually become fixed, but how provisional and revisable structures can be represented without prematurely foreclosing alternative possibilities. At some point, however, interpretive structures must be represented in forms that computational systems can store, query, and reuse. Where that point should lie, and what kinds of knowledge become compressed or transformed in the process, is not solely a technical question. It requires ongoing negotiation between system designers, researchers, and the communities whose knowledge these infrastructures seek to represent.

Within StoryMachine, this broader question acquires a particularly concrete form through the interaction between spatial arrangement, computational parsing, and recommendation. The system must eventually translate user-generated spatial configurations into machine-readable structures that can be stored, queried, and used for recommendation. Yet when multiple interpretations remain plausible, should the system commit to a single representation, maintain several competing interpretations, or preserve ambiguity as such? More fundamentally, can computational systems distinguish between insufficient information and genuine interpretive plurality? This is not merely a technical problem but an epistemological one. As Harvilahti's critique of motif classification suggests, the consequences of resolving interpretive ambiguity prematurely are not trivial: classificatory decisions may narrow the range of possible relations that remain visible within a tradition [14]. Similar risks arise in computational systems when exploratory interpretations are formalized too early, potentially reproducing at the level of machine inference the same forms of epistemological closure that have long been debated at the level of collection, classification, and archival organization.

This also raises broader questions concerning the nature of the archival object itself. Traditional folklore archives primarily preserve stabilized artifacts: texts, recordings, motifs, classifications, and metadata. Yet if interpretive relations emerge dynamically through exploratory interaction, then interpretive processes themselves may become culturally and epistemically significant. Evolving configurations of annotations, spatial arrangements, associative trajectories, and histories of rearrangement may themselves constitute meaningful archival materials. Projects such as WossiDiA already suggest this possibility insofar as they preserve not only folkloric materials themselves, but also the relational and editorial structures through which those materials were historically organized, annotated, and interpreted. StoryMachine raises this question in a particularly concrete form. As user-generated spatial arrangements become available to future users through computational assistance, interpretive histories themselves may begin to function as archival resources. Should such arrangements be preserved, revised, superseded, or accumulated? When different users construct divergent readings of the same narrative materials, emphasizing, for example, different historical, structural, political, or performative dimensions, the design of interpretive infrastructures becomes consequential. Should competing interpretations be selectively privileged, combined into generalized representations, or maintained as distinct alternatives with their respective interpretive contexts intact? The first two approaches risk reproducing a dynamic long noted in archival practice, in which selections gradually come to appear as wholes and institutional choices become naturalized as facts rather than recognized as contingent interpretive decisions [37, 46]. This question extends longstanding concerns about interpretive plurality from archival and classificatory practice into the computational layer itself.

These questions become particularly significant as contemporary digital environments increasingly incorporate algorithmic recommendation, generative systems, and forms of AI-assisted interpretation. Systems capable of suggesting associations, generating narrative continuations, or surfacing latent relations may substantially expand exploratory interpretive possibilities. Yet computational assistance also risks prematurely constraining interpretation by stabilizing relations too early, privileging statistically dominant patterns, or obscuring the situated and negotiable character of interpretive meaning-making.

StoryMachine makes this tension directly visible. Recommendations are generated partly from previously formalized spatial structures and therefore participate in shaping what becomes perceptible, salient, or associatively available during subsequent acts of interpretation. Computational assistance may thus expand exploratory possibilities while simultaneously influencing their trajectory. The challenge is not simply how to improve recommendation quality, but how to design systems that support exploratory interpretation without foreclosing ambiguity, alternative relational configurations, or the possibility that multiple interpretations remain simultaneously valid.

A further question concerns the communities whose traditions such infrastructures seek to represent. Participatory approaches in digital folkloristics have substantially expanded who contributes to folklore knowledge production [39]. Yet even highly participatory platforms often invite contribution within organizational frameworks that have already been institutionally defined [25]. Systems that preserve ambiguity and interpretive openness may offer new possibilities for negotiating meaning, representing plurality, and accommodating alternative perspectives. At the same time, productive ambiguity may not always be experienced as empowering. For community members whose relationship to tradition is lived rather than primarily scholarly, ambiguity may be experienced as uncertainty, instability, or a lack of guidance. StoryMachine raises this tension directly by exploring forms of participation in which interpretive relations themselves remain open to negotiation. As O'Carroll argues, meaningful engagement with folk cultural materials requires forms of participation that remain attentive to their depth, intimacy, and complexity [37]. How such commitments can be translated into usable interpretive infrastructures remains largely unexplored.

6 Conclusion

StoryMachine explores one possible response to these tensions through a cognitive-map-based spatial hypertext environment designed to support exploratory relation-building prior to semantic stabilization. Rather than treating spatial hypertext as a transitional stage preceding formal organization, the system approaches spatial arrangement itself as an interpretive infrastructure through which meanings, associations, and provisional structures may emerge dynamically through interaction. In this sense, StoryMachine contributes to digital folkloristics not by replacing existing archival, computational, or participatory infrastructures, but by addressing a methodological gap they share: the absence of a system capable of receiving, analyzing, and building upon interpretive acts that precede formal articulation. At the same time, it suggests understanding of spatial hypertext as an interpretive infrastructure in which ambiguity, exploratory arrangement, and provisional stabilization are not transitional conditions, but constitutive features of knowledge production. Yet the broader implications of spatial hypertext infrastructures remain unresolved. What is gained by preserving interpretive openness, and what may be lost when abandoning stable classificatory structures? Whether infrastructures can hold knowledge in a state of productive ambiguity without losing it remains one of the central questions raised by StoryMachine. The archive that can hold knowledge in a state of productive ambiguity without losing it has not yet been built; StoryMachine is an attempt to begin.

Acknowledgments

StoryMachine is a project funded by the Arts and Humanities Research Council (grant ID “AH/Z507222/1”) and the Deutsche Forschungsgemeinschaft (grant ID “547532269”) under their sixth UK–German Funding Initiative in the Humanities.

Notes

https://humma.lv/, accessed June 1, 2026

https://apps.wossidia.de/webapp/run, accessed May 21, 2026

https://isebel.eu/, accessed June 1, 2026

https://blogs.helsinki.fi/filter-project/, accessed June 1, 2026

https://runoregi.fi/, accessed June 1, 2026

http://garamantas.lv, accessed June 1, 2026

https://www.duchas.ie/en, accessed June 1, 2026

http://iesaisties.lv, accessed June 1, 2026

https://www.tiktok.com, accessed June 1, 2026

Source

ACM full-text HTML

ACM Digital Library

Imported from ACM’s structured HTML source. ACM Reference Format: Lisa Roßner, Sabine Slowik, Claus Atzenbeck, and Peter Nürnberg. 2026. Spatially Arranging is Knowing: Spatial Hypertext as Method for Digital Folkloristics. In 37th ACM Conference on Hypertext (HT '26), September 14--18, 2026, London, United Kingdom. ACM, New York, NY, USA 7 Pages. https://doi.org/10.1145/3800935.3830886

References

[1] James Abello, Peter Broadwell, and Timothy Tangherlini. 2012. Computational Folkloristics. Communications of The ACM - CACM 55 (07 2012), 60–70. https://doi.org/10.1145/2209249.2209267

[2] Mark W. R. Anderson. 2025. W(h)ither Spatial Hypertext?. In Proceedings of the 36th ACM Conference on Hypertext and Social Media(HT ’25). Association for Computing Machinery, New York, NY, USA, 100–112. https://doi.org/10.1145/3720553.3746683

[3] Claus Atzenbeck and Peter J. Nürnberg. 2019. Hypertext as Method. In Proceedings of the 30th ACM Conference on Hypertext and Social Media (HT ’19). ACM, New York, NY, USA, 29–38. https://doi.org/10.1145/3342220.3343669

[4] Claus Atzenbeck, Daniel Roßner, and Manolis Tzagarakis. 2018. Mother – An Integrated Approach to Hypertext Domains. In Proceedings of the 29th ACM Conference on Hypertext and Social Media. ACM, New York, NY, USA, 145–149. https://doi.org/10.1145/3209542.3209570

[5] Mark Bernstein. 2011. Can we talk about spatial hypertext. In Proceedings of the 22nd ACM Conference on Hypertext and Hypermedia (Eindhoven, The Netherlands) (HT ’11). Association for Computing Machinery, New York, NY, USA, 103–112. https://doi.org/10.1145/1995966.1995983

[6] T.J. Blank. 2012. Folk Culture in the Digital Age: The Emergent Dynamics of Human Interaction. Utah State University Press, Logan. 1–262 pages.

[7] Geoffrey C. Bowker and Susan Leigh Star. 1999. Sorting Things Out: Classification and Its Consequences. The MIT Press, Cambridge, Massachusetts. https://doi.org/10.7551/mitpress/6352.001.0001

[8] Vannevar Bush. 1945. As we may think. The Atlantic Monthly 176, 1 (7 1945), 101–108. http://www.theatlantic.com/doc/194507/bush

[9] Kaeko Chiba. 2026. Participatory digital archives with senses: looking at local culture from multiple voices and voices unheard. Asian Anthropology 25, 2 (2026), 115–132. https://doi.org/10.1080/1683478X.2026.2613615

[10] Jeff Conklin. 1987. Hypertext: an introduction and survey. Computer 20, 9 (1987), 17–41. https://doi.org/10.1109/MC.1987.1663693

[11] Gabriele de Seta. 2024. An Algorithmic Folklore: Vernacular Creativity in Times of Everyday Automation. Institute of Network Cultures, Amsterdam, 233–253.

[12] Til Eyinck. 2025. Epistemology of Folk-Lore. Social Epistemology 0, 0 (2025), 1–16. https://doi.org/10.1080/02691728.2025.2496199

[13] Gilles Fauconnier and Mark Turner. 2003. Conceptual Blending, Form and Meaning. Recherches en Communication; No 19: Sémiotique cognitive — Cognitive Semiotics; 57-86 19 (03 2003). https://doi.org/10.14428/rec.v19i19.48413

[14] Finnish Literature Society and Lauri Harvilahti. 2022. Motif Indexes in the Age of the Internet. Folklore: Structure, Typology, Semiotics 5, 4 (2022), 80–90. https://doi.org/10.28995/2658-5294-2022-5-4-80-90

[15] Antonio García Gutiérrez. 2014. Declassification in Knowledge Organization: A Post-epistemological Essay. Knowledge Organization 41 (09 2014), 393–409. https://doi.org/10.5771/0943-7444-2014-5-393

[16] James J. Gibson. 2014. The Ecological Approach to Visual Perception. Psychology Press, New York. Original work published 1979. https://doi.org/10.4324/9781315740218

[17] Charles Goodwin. 1994. Professional Vision. American Anthropologist 96, 3 (1994), 606–633. https://doi.org/10.1525/aa.1994.96.3.02a00100

[18] Eldar Heide. 2018. New perspectives on archival shortcomings: A note on the history of a discipline, on self-perception and on methodology within folkloristics, ethnology and cultural studies. Folklore Fellows’ Communications (FFC), Vol. 315. Suomalainen Tiedeakatemia, Helsinki, 240–256.

[19] Yumeng Hou, Sarah Kenderdine, Davide Picca, Mattia Egloff, and Alessandro Adamou. 2022. Digitizing Intangible Cultural Heritage Embodied: State of the Art. J. Comput. Cult. Herit. 15, 3, Article 55 (Sept. 2022), 20 pages. https://doi.org/10.1145/3494837

[20] Edwin Hutchins. 2006. The Distributed Cognition Perspective on Human Interaction. Routledge, Abingdon, UK, 375–398.

[21] Emese Ilyefalvi. 2018. The Theoretical, Methodological and Technical Issues of Digital Folklore Databases and Computational Folkloristics. Acta Ethnographica Hungarica 63 (2018), 209–258. https://doi.org/10.1556/022.2018.63.1.11

[22] Maciej Janicki, Kati Kallio, Mari Väina, and Eetu Mäkelä. 2024. Developing a Digital Research Environment for Finnic Oral Poetry. Baltic Journal of Modern Computing 12, 4 (2024), 535–547. https://doi.org/10.22364/bjmc.2024.12.4.15

[23] Risto Järv. 2017. The Singing Wolf Meets His Kin Abroad. Web-based databases of the Estonian Folklore Archives. Estudis de Literatura Oral Popular / Studies in Oral Folk Literature 5, 5 (June 2017), 29–44. https://doi.org/10.17345/elop201629-44

[24] David Kirsh and Paul Maglio. 1994. On Distinguishing Epistemic from Pragmatic Action. Cognitive Science 18, 4 (1994), 513–549. https://doi.org/10.1207/s15516709cog18041

[25] Sandis Laime and Sanita Reinsone. 2024. HUMMA.LV: Towards a Collaborative Digital Platform for Humanities and Arts in Latvia. Baltic Journal of Modern Computing 12, 4 (2024), 487–492. https://doi.org/10.22364/bjmc.2024.12.4.10

[26] George P. Landow. 2006. Hypertext 3.0: Critical Theory and New Media in an Era of Globalization. Johns Hopkins University Press, Baltimore, MD, USA.

[27] Lambros Malafouris. 2019. Mind and Material Engagement. Phenomenology and the Cognitive Sciences 18 (02 2019), 1–17. https://doi.org/10.1007/s11097-018-9606-7

[28] Catherine C. Marshall, Frank G. Halasz, Russell A. Rogers, and William C. Janssen. 1991. Aquanet: a hypertext tool to hold your knowledge in place. In Proceedings of the 3rd ACM Conference on Hypertext. ACM, New York, NY, USA, 261–275. https://doi.org/10.1145/122974.123000

[29] Catherine C. Marshall and Frank M. Shipman. 1995. Spatial hypertext: Designing for Change. Commun. ACM 38, 8 (Aug. 1995), 88–97. https://doi.org/10.1145/208344.208350

[30] Theo Meder, Petra Himstedt-Vaid, and Holger Meyer. 2023. The ISEBEL Project: Collecting International Narrative Heritage in a Multilingual Search Engine. Fabula 64, 1-2 (July 2023), 107–127. https://doi.org/10.1515/fabula-2023-0006

[31] Theo Meder, Folgert Karsdorp, Dong Nguyen, Mariet Theune, Dolf Trieschnigg, and Iwe Muiser. 2016. Automatic Enrichment and Classification of Folktales in the Dutch Folktale Database. The Journal of American Folklore 129 (02 2016), 78. https://doi.org/10.5406/jamerfolk.129.511.0078

[32] Holger Meyer, Christoph Schmitt, Stefanie Janssen, and Alf-Christian Schering (Eds.). 2014. Corpora ethnographica online: Strategien der Digitalisierung kultureller Archive und ihrer Präsentation im Internet. Waxmann Verlag GmbH, Münster. https://doi.org/10.18453/rosdokid00002265

[33] Theodor Holm Nelson. 1982. Literary Machines. Mindful Press, Sausalito, CA, USA.

[34] Peter J. Nürnberg and Helen Ashman. 1999. What was the question? Reconciling open hypermedia and World Wide Web research. In Proceedings of the 10th ACM Conference on Hypertext and Hypermedia. ACM, New York, NY, USA, 83–90. https://doi.org/10.1145/294469.294492

[35] Hope A. Olson. 2001. The Power to Name: Representation in Library Catalogs. Signs 26, 3 (2001), 639–668.

[36] Walter J. Ong. 2005. Orality and Literacy. The Technologizing of the World. Routledge, London, England. Original work published 1982.

[37] Clíona O'Carroll. 2018. Tradition Archives as Memory Institutions in the Past and in the Future: Foreword. Folklore Fellows’ Communications (FFC), Vol. 315. Suomalainen Tiedeakatemia, Helsinki, 12–23.

[38] Mark S. Peacock. 2009. Path Dependence in the Production of Scientific Knowledge. Social Epistemology 23, 2 (2009), 105–124. https://doi.org/10.1080/02691720902962813

[39] Sanita Reinsone. 2018. Participatory Practices and Tradition Archives. FF Communications 2018-January (01 2018), 279–296.

[40] Catherine Ryan and Críostóir Mac Cárthaigh. 2018. Legacy Folklore Collections and the Use of Thesauri. Folklore Fellows’ Communications (FFC), Vol. 315. Suomalainen Tiedeakatemia, Helsinki, 297–322.

[41] Christoph Schmitt and Alf-Christian Schering. 2023. WossiDiA: A Pioneer of Digitized Tradition Archives and Its Hypergraph Technology. Fabula 64, 1-2 (July 2023), 128–149. https://doi.org/10.1515/fabula-2023-0007

[42] Christoph Schmitt and Alf-Christian Schering. 2023. WossiDiA: A Pioneer of Digitized Tradition Archives and Its Hypergraph Technology. Fabula 64 (07 2023), 128–149. https://doi.org/10.1515/fabula-2023-0007

[43] Frank M. Shipman and Catherine C. Marshall. 1999. Formality Considered Harmful: Experiences, Emerging Themes, and Directions on the Use of Formal Representations in Interactive Systems. Computer Supported Cooperative Work (CSCW) 8 (1999), 333–352. https://doi.org/10.1023/A:1008716330212

[44] Frank M. Shipman and Catherine C. Marshall. 1999. Spatial hypertext: an alternative to navigational and semantic links. Comput. Surveys 31, 4es (dec 1999), 14–es. https://doi.org/10.1145/345966.346001

[45] Frank M. Shipman, J. Michael Moore, Preetam Maloor, Haowei Hsieh, and Raghu Akkapeddi. 2002. Semantics happen: Knowledge building in spatial hypertext. In Proceedings of the 13th Conference on Hypertext and Hypermedia. ACM, New York, NY, USA, 25–34. http://doi.acm.org/10.1145/513338.513350

[46] Fredrik Skott. 2018. History in the Making: The Ethics and Politics of Net Publishing. Folklore Fellows’ Communications (FFC), Vol. 315. Suomalainen Tiedeakatemia, Helsinki, 323–336.

[47] Timothy Tangherlini. 2013. The Folklore Macroscope Challenges for a Computational Folkloristics. Western Folklore 72 (12 2013), 7–27.

[48] Jeffrey A. Tolbert and Eric D. M. Johnson. 2019. Digital Folkloristics: Text, Ethnography, and Interdisciplinarity. Western Folklore 78, 4 (2019), 327–356.

[49] Hans-Jörg Uther. 2024. Animal Tales, Tales of Magic, Religious Tales, and Realistic Tales, with an Introduction (nouv. édition revue et augmentée ed.). Number 284 in Folklore Fellows Communications. The Kalevala Society Foundation, Helsinki.