The information theories developed by Claude Shannon and Warren Weaver in their experiments on communication have had a significant influence on the discipline. Originally published in 1948, Shannon and Weaver’s engineering model is primarily concerned with effective message transmission across an information channel and factors influencing message reception (such as channel capacity, noisiness, or degrees of uncertainty). The model is explicit about not incorporating meaning into a definition of information, claiming that the ‘semantic aspects of communication are irrelevant to the engineering problem’ (Shannon, 1948).

Scholars from a variety of disciplines have long argued that information theory and its reductions a la the ‘conduit metaphor’ have hindered the development of robust understandings of information as a dynamic, emergent process (Reddy, 1977; Soto and Sonnenschein, 2020). LIS (library and information science) does not define concepts, attributes or material properties of information and tends to defer to a larger framework concerned with behavior around information. Within the discipline, information-theoretic influences on the operationalization of the field’s target value resolve primarily as either painting information use as a causal process of linear transmission and/or portraying information as a variable ‘thing’ that can be represented for retrieval and/or preservation. Both conceptions have significant consequences for how we study and understand information in its different contexts of use. (Day 2000, p.808) argues that the conduit approach has structured ‘how we conceive and define the nature and scope of the field, what methods are allowable, what researchers, researchers, and disciplines are valuable for information studies.’ According to (Capurro 1991), proponents of information theory conceive of information as a singular, fixed representational value that flows linearly between two points. Information thus becomes a one-dimensional message, a ‘substance that is transported between the sender and the receiver’ (Mai, 2013, p.676). Consequently, there is very little qualitative understanding about the evidential practices that make something information (Furner, 2004).

Cultural heritage institutions have long acknowledged difficulties tracking and evaluating the use of their collections (Dearstyne, 1987; Chapman et al., 2015) and the transformation to digital collections has only amplified an existing problem space for assessment. Namely, what constitutes the use of a [digital] collection? Does clicking on an artifact indicate use? Should it? In LIS studies on digital collections use, the so-called ‘conduit approach’ has influenced the kinds of methodologies employed. In the first approach, collection use is quantifiably assessed through measuring access to digital resources. Researchers use transaction logs or web site analytics as tools to evaluate resource use (Prom, 2011; Lapworth, 2020). A second approach assesses use through the occurrence of citations and/or references in publications (Sinn, 2012; Harris and Hepburn, 2013). In the third approach, scholarly preferences for accessing collections are analysed, with an emphasis on whether scholars access materials through print or online methods (Tibbo, 2003; Anderson, 2004; Duff, Craig, and Cherry, 2004).

These studies provide very little insight into the multi-faceted, interactional space of scholarly information use in digital collection spaces. There are inherent limitations with employing quantitatively driven frameworks, theories, and methodologies to frame questions of information use - itself a highly interpretive, and often undefined sphere of information activities. In the first and third approaches, researchers are studying access to (rather than use of) information sources (Spink and Cole, 2006). Citation analysis is only partially helpful as an indicator of use; users might consult multiple sources in the course of their research but may not necessarily cite them. A further problem with citation analysis is the relative newness of standardized practices for citing digital resources, likely painting an inaccurate picture of use. As Sheble and Wildemuth (2013) note, transaction logs ‘cannot tell us anything about users’ cognitive or affective responses during the system interaction’ (p.174). Treating resource access and retrieval mechanisms simply as a proxy for information use obscures important aspects of the interaction space. How do factors such as digitization quality, the availability of technical metadata, or navigation options impact scholarly interpretation and use? What contextual qualities support meaningful discovery, access, and use of digital collections, and how does the interaction process influence the experience or outcomes of use? Answers to these questions can help inform the successful design of dynamic collection spaces and influence how LIS proceeds with theoretical developments.

The dynamics of information flows within larger computer-mediated communication ecosystems introduce heuristic and practical complexities. Scholars working in digital collections now encounter artifacts through an interface where form, content, and interpretation converge in a particular space-time context (Hedstrom, 2002). This convergence meshes together ‘previously separated realms’ (Lowood, 2002, p.9) into singular knowledge representations conditioned by ongoing mediations. As the factors which mediate content change (e.g., device, modality), so will the interpretation. Writing in the 1960s, the sociologist John McHale warned:

we are faced…with the emergence of a new and powerful fusion of technological capabilities which not only potentially amplifies our capabilities to deal with our social and physical environment transactions but which by its function as screen, channel and multiplexer of information actually reshapes the information content and perception of society – in ways that our conventional wisdom and traditional institutional means may not be able to foresee, comprehend or effectively control (p.16-17).

Studies on digital collections use capture a specific instantiation of a space-time bound relational link – an enactment of a viewing modality coupled with a representational level of content. But the specificity of this capture can also obfuscate the process of information. What gets lost in translation? In studying use in dynamic environments, being able to capture the organic, dynamic, and evolutionary nature of interpretation is essential. Let us consider ‘information use’ on social media. What users see on social media such as Twitter (X), Instagram and Tik Tok are algorithmically mediated; for example, Tik Tok’s for you page offers content tailored to the user. The videos – Tik Toks – that each person sees, their order, and how they experience (interpret) them differs. In addition, a person’s use experience is shaped dialogically in an ongoing interpretive process, as future content is based on users’ past and current interactions with content. Tik Tok users are aware of the impacts of the algorithmic nature of the platform, such as the creation of filter bubbles and the promotion of potentially harmful content, and users change their behaviours to reflect how they understand themselves (Karizat et al., 2021). User behavior, then, can only ever be understood in the context of how they perceive system behavior, themselves, and indeed, even their views of the human experience.

As a user, different social technologies may be associated with different meanings, e.g., Facebook as a space for presentation of the socially acceptable; Instagram, for stylized self-presentation; and Twitter as a venue for informal information exchange (Boczkowski et al., 2018). Companies and users co-create experiences with these platforms over time through features with socially embedded meanings, such as retweets on Twitter (X) and stories on Instagram. Considering users’ interactions in context, and through the language and culture of the media can facilitate a more nuanced understanding of meaning-making and use within different social media platforms.

The lens of experienced information provides a conceptual window into environments where people build understanding by engaging with documentation (Gorichanaz, 2017). Focusing on experienced information emphasizes the interactions between individuals and their life-worlds rather than only capturing the outcomes of those interactions (Bruce, 2014). Holistic accounts that capture how information is experienced prompt attention to processes underlying ‘what information a given artifact imparts’ (Hoyte, 2019, p.415).

Information pathways is one methodological approach for studying how information is experienced and has been proposed as a guiding metaphor from which to characterize the design of cultural heritage collections. Goodale et al. (2012) observe that information pathways can have various conceptualizations, including as information-seeking journeys, a starting point or way in, a route through, a learning process, and others. Wray et al. (2013) emphasizes two dimensions of pathways, one involving creative, non-linear journeys (e.g., a search history) and one supporting learning and meaning making (e.g., linked metadata). Assessment of the similarity between works in a collection affords a sense of the potential for non-linear serendipitous explorations. In a similar vein, Adams, and Blandford (2005) performed a grounded theory analysis of the information journeys taken by users of digital libraries in two domains, health, and academia, identifying three main stages – information initiation, facilitation (or gathering) and interpretation – and opportunities to better support users in these experiences.

Though log data collected concerning digital collection use are often quantitative in nature, they can be considered more interpretively, as pathways for understanding how information is experienced. Behavioral metrics could serve as a temporal view of the path that one takes – their experience – of an artifact or collection. Though the manifest meaning of page views, click-throughs, and other user interactions may be unclear, an alternate way of conceptualizing log data are as the artifacts of an experiential process. When one views the traces of activity through a website, one is viewing a process of sense-making. Considering users’ behaviours sequentially could lead to informative re-constructions of how users learn and engage in meaning-making with collections, as well as paths that perhaps designers had not anticipated, facilitating the imagining of new potentialities.

Given the right conditions and environments, all kinds of entities (artifacts, people, events) can inform or ‘make something known’ (Briet, 1951), but it is the processing of that entity for ‘informational purposes’ (Buckland, 2001) that signifies a transition to documentation status. Taylor (1986;1991) is one LIS scholar who recognized the importance of approaching information from a situated perspective, arguing that specific meaning(s) and value are established in the context of use. He developed a conceptual framework for studies on information use environments, aiming to capture influential elements that impact flow and determination of informational value. According to Taylor, information always exists as a phenomenon of study in a situational context. He explains, ‘Data becomes information when we establish relationships among data...it is the structure and relations among data, and most important, the rules and conditions used to establish those relations’ (1991, p. 7-8). Viewed through this lens, artifacts can serve as prompts for informational study and in practice approaches, as ‘epistemic objects’ always in the process of unfolding (Knorr-Cetina, 2000). The situated nature of information interactions was also taken up in LIS by Blandford and Attfield (2010), elaborating on previous work theorizing information journeys. They describe how sites of situated information interactions are mediated contexts that are ‘physical, social, and activity-centred’ and can overlap situationally (2010, p.2).

In affiliated knowledge communities, the intersecting spheres of information studies, human-computer interaction, and organizational work studies have produced relevant theoretical work on ‘the situated entanglement of the technical and the social’ (Orlikowski, 2005, p.185). Interrogating processes of information from this perspective can be fruitful for unpacking the ‘multiple logics that configure together social and material realities (Barad, 2007). Technology itself can be interpreted as a social practice and efforts to study knowledge-generative environments must account for the ‘constitutive, co-dependent relations of complexity’ that underpin all digital information environments (Drucker, para. 30, 2013). Studying conditions of situated information permits attention to the ‘the individual work and its activation of the particular conventions as its point of departure’ (Doane, p. 131). Rather than focusing on the medium and its corresponding material presence, though, information is studied in the context of its performative becoming; though not the sole, liminal focus of interest, the medium is an inherent part of the (informational) context.

While in principle, studying situated information or information in its ‘performative becoming’ may seem intuitive, questions may arise concerning how this might be achieved in practice. There are methods from different disciplines and approaches that can be helpful. For example, the concept of information horizon introduced by Sonnenwald (1999), referring to the perceived horizon of potential information resources, affords an intuitive representation of the socially constructed and situated use of information by different users. Often used in contextual inquiry (Beyer & Holtzblatt, 1999), visual tools such as information flow diagrams, cultural models, sequence models and other diagrams might also be used to understand and envision potential information uses. Jo Bates et al. (2016) introduces a methodological approach they call data journeys for exploring the socio-material, constitutive nature of digital data in the context of its use. The value of this situated perspective is the ability to attend to how information flows through complex environments, making visible ‘potential movement, blocked movement and lack of movement.’

In this paper, we argue that participatory approaches to information practice research can be an effective means of contending with ‘information-as-potentiality’ and meaning-making. Let us consider the scenario of information creation and use in heritage sites such as archives and museums. Rather than viewing these as static, participatory information practices invite people to interact with sites as a ‘means for social interaction, participation, co-creation and contribution’ (Stuedahl, 2018, p. 219). Incorporating users’ information practices and potentially leveraging available technologies to assess feelings and reactions in the moment (e.g., through ecological momentary assessment) could facilitate richer engagement with and design of heritage environments. For example, the Acropolis Museum invited museum visitors to evaluate mobile-based storytelling guides designed to facilitate personalized experiences (Roussou & Katifori, 2018). They incorporated data collected through ethnography (i.e., shadow observation), interviews, questionnaires, and data logging into analysis, and ultimately considering themes such as interactive story plot and narration, staging, and wayfinding, personalisation, and social interaction.

Participatory information approaches such as decentralized curation focus on usability and findability (‘radical user orientation’) and contextualisation of both records and the entire archival process (Huvila, 2008); and engaging community members in the appraisal, arrangement, and description phases of creating an archival record, are needed to ensure meaningful representation (Shilton & Srinivasan, 2007). Examples of participatory archives projects are #NorthVanStories, a social media collection project in which residents of North Vancouver were encouraged to tag social media with this hashtag, and #NorthVanStories–Living Histories, a digital video oral history project (Bushey, 2023). #NorthVanStories–Living Histories involved multiple steps, including identification of community assets, selection of projects relevant to the communities, contact with storytellers, sharing a questionnaire with questions to support the aims of the project, and creation of five videos with storytellers from diverse and vulnerable communities. This multi-step process illustrates consideration of the information to be included and scaffolding (the questionnaire) to support, but not define the knowledge representation, thus preserving voice and flexibility for storytellers to tell stories their own way. In both projects, meaning was defined and evolved through dialectical engagement, thus highlighting the need to: a) design with the participation of communities involved; and b) develop designs that allow for information to be defined and evolve as it is being produced. Thus, participatory information practices facilitate the realization of collective negotiation of the meaning of information as an emergent process.

Information can become embodied in and through a variety of forms that extend beyond the human body. Different contextual factors catalyse the informational process, including affective motivations and goals (Nahl, 2007) and align with domain-specific meanings (Hjorland & Alberton, 1995). Working in the vein of digital information studies, Hayles (2002) argues that the materiality of information arises from the combination of user interactions and interpretive strategies. In other words, information emerges in its material form through the coupling of interpretation with interaction. (Dourish 2001) also proposed a theory of embodied interaction that emphasizes context-aware approaches to studies of social interactions with physical and/or technical artifacts. (Bates 2018) helpfully reminds us that as humans ‘operating in a world of possible affordances’ (p.255) we have the embodied capacity to perceive, experience, and communicate information in meaningful ways.

The relational links between perception and embodiment were made in early critiques of information theory, initially by the British scientist Donald MacKay who noted the significant role of perception in the transfer of information. For MacKay, the act of perceiving messages is made possible by an existing internal state of conditional readiness. He writes,

the mechanical energy of a message must be sufficient to do the mechanical job that eventually resets the brain-levels; but the selective job, of determining which levels shall move, depends on the form of the message, and on the state of your brain before you hear it (1969, p.24).

Drawing attention to the job of selection is critical for MacKay’s argument; while all sorts of stimuli might be received, only stimuli that evoke a matching response in the organism can be perceived. This distinction is critical to understanding how perception itself is embodied in the activity of information, or, as MacKay puts it, the ‘activity of knowing is an activity of internal adaptive response’ (p. 61).

More recently, the concept of embodied information has received attention by LIS, often by way of studies on information practices. According to Olsson and Lloyd (2016),

embodied information practices (social, physical, and material) therefore act as complex and rich sources of information that are referenced through the social, physical, and material conditions of the site and its ways of knowing.

We also observe embodied ways of knowing in serious leisure, including in ultra-running (Gorichanaz, 2015) and enthusiast car restoration (Lloyd-Zantiotis & Olsson, 2017). Similarities in these examples include the complexities of information practices, their social and oral nature, and the important role of acquired knowledge. However, there are also differences in individual ways of knowing. Whereas, in the context of ultra-running, the body was a key source of information (Gorichanaz, 2015), a participant in Lloyd-Zantiotis and Olsson’s (2017) study developed a feel for different types of materials and how to work with them in the context of the car restoration process.

Studies of embodied information have employed a variety of methods, such as interviews, diaries, and observation, that might be useful for capturing the richness of information behavior that occur – ‘information-as–potentiality’ – with digital collections. For example, the experience of historical artifacts in digital form, can be embodied. Green (2018) writes that social media affords ‘curatorial hands’ through haptic experiences of digitized manuscripts. Though even the physical access to artifacts is often limited to a select few who are able to travel to view artifacts behind glass display cases, today social media such as Instagram can afford novel forms of engagement manipulating a digital representation via tapping, clicking or swiping, or scrolling back and forth through an set of images curated as a story, thus enabling viewers to interact with artifacts in ways not hitherto possible using traditional viewers. How is such engagement to be measured? One might conceive of the utilization of different ethnographic methods, but also the development of novel instruments that incorporate consideration of the embodied nature of information.

Media archaeology is one possible theoretical approach for augmenting embodied information research in interacting with digital collections. For example, a media archaeological approach could prove useful in presenting scholars with technical (e.g., EXIF) metadata alongside the display of digital artifacts from online collections. How do scholarly interpretations of digital artifacts change in these different contexts? How might such factors inform or contribute to the design and use of digital collections?