Sabrina Sauer is Assistant Professor Media Studies at the University of Groningen, where she also coordinates the Digital Humanities Master and minor programmes. She has an MA in Media Studies and studied as an actor prior to completing her PhD in Science and Technology Studies about user-technology improvisation and co-creation as a source for sustainable ICT innovation within living lab-research. Her current research focuses on data-driven creative media production, the role of digital materiality in co-creation practices, digital humanities, exploratory search technologies, and experiences of serendipity in smart cities. Sauer has published in journals as varied as
Berber Hagedoorn is Assistant Professor Media Studies & Audiovisual Culture at the University of Groningen. Her research interests revolve around screen cultures (representations and crossmedia storytelling practices) and audiovisual cultural memory in Europe. She is the Vice-Chair of ECREA’s Television Studies section and EUscreen Foundation Board Member, organizing cooperation for European research and education into television’s history and its future as a multi-platform storytelling practice. She has extensive experience in Media and Culture Studies and Digital Humanities through large-scale European and Dutch best practice projects on digital audio-visual heritage and cultural memory representation, including
This is the source
Interdisciplinary collaboration within digital humanities research requires brokering and boundary-crossing work. This article maps interdisciplinary exchanges between computer scientists, media scholars and information scientists to explicate how interdisciplinary brokering affects knowledge and tool production in the digital humanities. The analysis of qualitative data collected during a 17 month-long digital humanities research pilot - set to collaboratively test a digital Linked Open Data (LOD) search tool - provides insights into the exploratory search behaviour of a total of 122 (digital) humanities scholars, and how these insights informed brokering work within an interdisciplinary research team. The article argues that interdisciplinary collaboration first and foremost requires disciplinary recognition in order to succeed, and demonstrates how practical, empirical research insights - in this case about user research - mobilize interdisciplinary decision-making processes. Conclusions indicate in what ways computational tools and collaboration affect knowledge production, and suggest that understanding the stakes of digital humanities research with digital audio-visual sources requires an integrated perspective characterized by both theoretical interdisciplinary discussions and their empirical, practical application. The article furthermore concludes that digital humanities brokering and boundary work should not only revolve around tool development, but also requires self-reflexivity in aligning epistemological chasms between disciplines by means of collaborative translation practices.
Mapping interdisciplinary exchanges between computer scientists, media scholars, and information scientists via a collaborative test of a Linked Open Data search tool.
Productive interdisciplinary collaboration is high on research agendas of funding
organisations, especially those organisations looking to stimulate digital humanities
research
Interdisciplinariness
does not warrant immediate celebration. It is important to
acknowledge that doing interdisciplinary work requires translation work:
specifically, an adaptation of research insights to make sense within another
disciplinary perspective, in order to reach conclusions that prove productive to all
researchers and disciplines involved. Overcoming epistemological differences between
disciplines often makes conflict inevitable within interdisciplinary digital
humanities projects practice of
interdisciplinary collaboration
This article presents insights into such interdisciplinary digital humanities
practices. It delves into a particular case of interdisciplinary translation work to
explicate the particularities of what we refer to as interdisciplinary brokering
work. The case centres on the collaboration practices of a team of computer
scientists, information scientists and media scholars who, over the course of 17
months, user tested and fine-tuned a digital tool as part of a digital humanities
pilot study. Throughout this study, the tool became the basis for interdisciplinary
knowledge exchange, much like a boundary object
The tool that became the team’s boundary object, is the Linked Open Data (LOD)
exploratory search browser DIVE+search narratives
as
sequences of search results, connected by scholars in a separate navigation and
bookmarking pane in the search browser (see Figure 2). The CLARIAH Media Suite gives
access to important Dutch national audiovisual archives and is part of the
overarching CLARIAH infrastructure, which is the Dutch national counterpart to the
international CLARIN and DARIAH research infrastructures. CLARIAH develops,
facilitates, and stimulates the use of Digital Humanities resources and
infrastructures
Insights into the team’s interdisciplinary collaboration practices are drawn from qualitative data that was collected during a 17 month-long digital humanities research pilot study (2017-2018) that investigated how effectively DIVE+ supported the digital research practices of humanities scholars. Throughout the research study the authors collected data about humanities scholars’ digital search practices by means of focus groups, search tasks and think-aloud protocols, surveys, and questionnaires of in total 122 participants. The authors then traced how these user insights shaped both the development of DIVE+ as well the interdisciplinarity collaboration within the research team.
As the title of this article suggests, brokering plays a key role in grasping the
interdisciplinary collaboration within this pilot. Brokering is not simply a metaphor
for collaboration. Admittedly borrowed from Innovation Studies
In this case, brokering takes the shape of a brokering community practice [negotiate] their own
activities and identities, [while shaping] the relationships between their
communities of practice
This article’s second goal is to connect our qualitative user studies’ insights to
interdisciplinary concerns about how knowledge is explored, produced, and
disseminated by means of computational tools and digital humanities research
practices. This objective relates to prior studies about scholarly research processes
and how these may inform the development of research environments for digital
scholarship in the humanities
In the next section (Section 2) we present the theoretical framework used to analyse the collected data. The methodological section (Section 3) shows how we related qualitative data collection and grounded theory to answer overarching questions about collaboration and brokering practices. Section 4 presents the analysis, with reflections about interdisciplinary brokering and translation work on two levels: (1) the project level, where team meetings and debates indicate the necessity of both theoretical interdisciplinary discussions and their empirical, practical application when doing digital humanities work and, (2) the tool level, and the role played by insights drawn from the pilot’s user studies in the team collaboration practices. We conclude in section 5 with a discussion of how our analysis at the user study level maps onto larger interdisciplinary concerns.
It is important to note that while the user study provides qualitative insights into the exploratory search behaviour of 122 users, and that the documented research team collaboration was analysed in order to better understand how interdisciplinary research within the digital humanities works, this study does have clear limitations. We used a mixed-method approach to conduct our qualitative research. This choice was informed by both methodological and practical concerns: we wanted to triangulate insights, while simultaneously adapting our methods to specific research settings. This is why think-aloud protocols were supplemented with questionnaires, research diaries and focus group sessions. This mixed-method approach relied on the team’s collective expertise and as a result helped to translate project insights into new knowledge for dissemination within the different researchers’ respective fields of Computer Science, Media Studies and Information Science.
Digital humanities research is characterised as interdisciplinary, and works at
the boundaries of what are considered computational sciences and the humanities.
More often than not, digital humanities are described in terms of how the
computational turn has taken effect in the humanities, thereby mirroring
all-encompassing processes of the datafication of modern society
Masson describes two trends in digital humanities research that relate to the core
objectives of this paper: (1) the critical engagement with digital tools and data,
and (2) the growing tendency to refer to digital humanities research as
transdisciplinary, drawing into focus the work of creating a shared conceptual
framework
fundamental epistemological shifts occur
Our study engages with both of Masson’s digital humanities trends. The analysis shows how, in our project, critical tool engagement informs the collaborative brokering practice of an interdisciplinary research team. Tool user testing not only informs tool development, but also fundamentally shapes the interdisciplinary team’s negotiation practices, especially regarding what is deemed important or within the scope of the research project for each involved researcher and represented discipline. Ultimately, the analysis shows that while this project was interdisciplinary, it never transcended disciplinary concerns to realize a shared conceptual – transdisciplinary – space.
The research project did, however, rely on a shared material space: DIVE+, the
tool that was further developed and tested within the project. Drawing into focus
the brokering practices on both the tool and project levels creates a conceptual
framework that is empirically, practically, observable; shared project objectives
lead to tests and alterations of, and reflections on DIVE+ as a boundary object
working artefact
both adaptable to different viewpoints
and robust enough to maintain identity across them
We work in line with Rieder and Rohle’s (2012) argument that interdisciplinary
digital humanities research should establish areas of collaboration between the
disciplines and to foster the development of practical solutions for interaction
Here, brokering work takes the shape of translation
Drawing out this brokering work shows how collaborators work across disciplines by
material engagement with the tool DIVE+ as a boundary object. DIVE+ is an open
linked exploratory search browser that facilitates the exploration of diverse
archival and museum collections, and allows researchers to investigate how
different media objects, events, people and concepts can be compared, connected,
and contextualized. Within Library and Information Sciences, the process of
information search and retrieval has been described as consisting of different
phases, including exploration. Kuhlthau proposes a six-stage model of information
searching, that relates search stages to phases of uncertainty, and observes that
the model reveals a search process in which a person is seeking meaning in the
course of seeking information
dip
of confidence after
having selected a number of interesting sources. Exploration leads to feelings of
uncertainty, as a user encounters sources that do not fit smoothly with
previously held constructs
aim to solve complex
problems
and that exploratory search itself is open-ended, persistent, and
multi-faceted
(2009). Exploratory search can therefore be regarded as an
iterative process, likened to solving puzzles
In defining how to optimally support exploratory search, Marchionini (2006) describes a system much like DIVE+ that affords user exploration and visualization of relationships between different entities and data facets (topic, time, space, data format). DIVE+ aspires to create ‘search narratives’ by presenting for instance events that are shared across linked data sets. However, as we will discuss in the analysis section, the mismatch between the search narratives afforded by DIVE+’s use of LOD and humanities scholars’ understanding of ‘narrative’ demonstrates the precarity of transdisciplinary presumptions.
Facilitating data exploration by means of linked open data (LOD) allows users to
discover connections across linked data sets
Recent digital humanities research has also focused on discovery by means of exploratory search. Jähnichen et al. (2017) focus on visual topic modelling of large collections of historical digitised materials to enhance exploratory search processes, while Muiser (2017) employed user-centred design case studies to optimize exploration of digital cultural heritage collections. The possibilities explicitly offered by LOD are furthermore documented by Lampert and Southwick (2013) who argue that LOD will not only change search processes, but also shift (librarian) skills and the conception of access to digital collections.
This article presents the results of our user study focused on exploratory search
and LOD, while connecting user study insights to overarching observations made
about interdisciplinary collaboration in terms of brokering practices. We are not
only interested in how users can form new research questions and encounter
research ideas as they engage in exploratory search with LOD, but also draw
attention to a parallel between LOD and interdisciplinary collaboration. Exploring
LOD is similar to exploring interdisciplinary research possibilities; both happen
between what Klein refers to as the Boundary Object and the Interactional
Expertise Trading Zones, that include linguistic exchanges generating
interlanguage and interactional expertise
Digital humanities research is often project-based
The project team, consisting of eight researchers from the fields of Computer Science, Media Studies, and Information Science, met virtually on a bi-weekly basis to discuss project deliverables, tool development, planned user studies, and research insight dissemination efforts. Meetings lasted between 30 and 60 minutes, and were documented in an online shared document. This shared document served as a research diary, chronicling the research project’s issues, concerns, and milestones in a bi-weekly rhythm. This document is analysed to discern recurring themes, ideas, and joint research objectives in order to help understand how the team gave meaning to DIVE+ as the project’s boundary object.
Research diaries were also part of the mixed-methods approach used to undertake user research. A total of 122 (digital) humanities scholars took part in these studies, between May 2017 and January 2018. A number of methods were combined to triangulate results; individually, users were given exploratory search tasks, answered surveys and (post-task) questionnaires about their experiences. This was combined with collaborative tasks, where small groups worked together to create posters about their search experiences and their ideas about exploratory search. User studies were conducted by means of interactive workshops hosted at different universities to include the users’ actual work settings.
The tasks were performed on personal laptops or university computers. During two sessions, the tasks were accompanied by think-aloud protocols which were recorded and subsequently transcribed. Search histories were saved and exported by the user and users filled out a post task questionnaire. In other sessions, tasks were not accompanied by a think-aloud protocol, but by a questionnaire. This was done to not disturb other users in the room, as they were completing tasks in a single, larger, space.
The task assignment was followed by a focus group discussion. During the discussion, users shared their experiences of searching with exploratory search tools, including DIVE+. These sessions were concluded with a poster session. Users worked together in small groups of 3-4 people to draw their research process and trajectory. The main question they all worked with was: How would you visualize your research journey, and what is the (ideal) role of exploratory search in this process? During plenary sessions posters were discussed and research findings displayed (Figure 1).
The first part of the analysis focuses on the research team’s brokering and translation work and then thus turns to how this work was informed by user research insights. These insights formed part of the team’s shared goals, but also became the space in which different research interests and negotiations came to the fore. Lastly, the analysis draws on insights about the collaboration and user research to focus on the necessity of both theoretical interdisciplinary discussions and their empirical, practical application when doing digital humanities work.
The interdisciplinary research team met virtually 27 times on a bi-weekly basis between March 2017 and July 2018. This team consisted of two computer scientists, one information scientist, and two media studies scholars. A data scientist and programmer joined the bi-weekly team meetings on a more ad-hoc basis. Team discussions focused on project goals: (1) how to execute user testing with the DIVE+ browser in order to better understand how (digital) humanities scholars can use exploratory search tools as part of their research practices; (2) how to optimize the DIVE+ browser in terms of its user interface (UI) and its integration into the overarching CLARIAH Media Suite; and (3) tool and project dissemination efforts at conferences and in publications.
Meeting minutes were shared within the team, and archived online. These minutes serve as an overview of the project, project management, and are an artefact onto itself. A grounded analysis of these minutes shows that discussions were for instance not primarily about DIVE+’s UI, but also focused on project management (to indicate, of the 239 coded lines, 50 were labelled ‘project management’, only superseded by 51 times that the UI was topic of discussion).
At a first glance, this interdisciplinary collaboration focused on discussing project timelines, deliverables and key milestones. Issues with DIVE+ functionality, testing and its integration into the Media Suite were also discussed at length. Following Klein, and Star and Griesemer, DIVE+ was the project’s boundary object and its trading zone of interaction from the outset. Discussions revolved around how to investigate the ways in which DIVE+ supports exploration, how to translate insights about use into browser and interface changes, and how to disseminate research insights more broadly (discussed 24 times during the 27 meetings).
Klein discusses the interactional expertise trading zone
where interdisciplinary
work includes linguistic exchange that generates interlanguage and interactional
expertise
The team settled on a mixed-method approach that used exploratory search tasks and think-aloud protocols (used more commonly in human computer interaction studies), as well as research diaries and focus groups (that are employed more readily in qualitative social science and humanities research). The mixed-method approach relied on the team’s collective expertise. This is not trivial: specific experts are involved in the project because they are experts in specific user research methodologies, and are willing to contribute this expertise for the project’s collective benefit. Conversely, these methods were also chosen because these specifically helped the involved researchers translate project insights into new knowledge for dissemination within their respective fields.
Translating the findings from the user studies into requirements for DIVE+ necessitated another step in the collaboration process, in which interdisciplinary collaboration led to clashes in epistemological outlook and understanding. For example, the tensions over the meaning and use of the term ‘narrative’ between researchers based in the humanities and those in computational science exemplify the type of translation problems that can arise during interdisciplinary collaboration.
Whereas the media studies researchers relied on thick descriptions of what
‘narrative’ means to the users in order to reflect on the role of narrativization in
research practices, their user study work also needed to lead to user interface
recommendations, as well as recommendations for the overarching Media Suite. In other
words, user feedback and reflections had to be translated into material suggestions,
as per the project’s objectives. A clear example of this are the team discussions
about and the understanding of what ‘narrative’ means in the context of users’
research practices within DIVE+, and what this difference in meaning-attribution means
in terms of interface functionality and design. For instance, not all users that
partook in the user studies were comfortable with the notion that they create
narratives while doing research. As one (media) historian notes: I believe that the
narrative metaphor does not really apply to my research, because I do not produce
sequential data, but rather a meta-structure, which cannot be told as a story
(Respondent B-P17)
. Yet for media researchers who are interested in audio-visual production
processes, such as image researchers, creating narratives is part of their core
research work. Narrative is thus a multi-interpretable container term that cannot
easily be translated in a one-size fits all UI.
On the level of the search browser, the developers of DIVE+ referred to generated exploration paths as ‘search narratives’, as these ‘retell’ the user’s search history. This, however, lead to user misunderstandings over how they can become storytellers without being able to annotate, add, or shuffle the search results within the exploration path. For the humanities researchers on the interdisciplinary team, ‘narrative’ was also closely connected to the desire to meaningfully edit or comment upon search results. The humanities scholars struggled to view ‘auto-generated narratives’ as research stories; without a way to add an interpretation to the narrative, the humanities users felt a lack of agency over the research narrative.
To the involved information and computer scientists however, the concept of ‘narrative’ was irrevocably tied to a user’s search history. Ultimately, the research team’s media scholars suggested to change this notion of a ‘search narrative’ to a ‘search journey’, thereby preventing future misunderstandings between team members and between the team and the test users. The dispute over a term like ‘narrative’ shows that developing a shared vocabulary means discussing disciplinary differences.
A similar discussion emerged about the idea of how DIVE+ was to be an ‘event-centric’
browser. Initially, the browser included the option to search for events. However, in
large part because of the ambiguity of this concept (when is something an event?),
this idea was eventually abandoned. Discussions over terms such as ‘narrative’ and
‘event’ were followed up with a collective re-interpreting or translating of the term
from the point of view of the users, in order to make alterations to the search
browser and UI that would reflect this user interpretation and to avoid what has been
referred to as the I-methodology
designers consider
themselves as representative of the users
This translation work is part of the brokering that the collaborating researchers engaged in. Beyond translation, brokering also means convincing the other team members of how users seemed to interpret their interactions with DIVE+ and with LOD. In this case, the discussions about the meaning of ‘narrative’ and ‘event’ that users attributed to these terms gained more agency in the brokering practice: the epistemological outlook and interpretations of the users involved in tool testing became key to material or technological changes in the browser. The data that was collected as part of the user studies becomes part of an interactional expertise trading zone; it is not only DIVE+ that connects the members of the research team as a boundary object, but also the data that is collected about DIVE+ and its use (and possible subsequent analysis of that data to publish about the project!) that gives the team discussions, and decisions, its momentum. This data is also part of the negotiation. In order to zoom in further on how this negotiation actually took shape, we now turn to a close analysis of how our collected user research data shaped the interdisciplinary collaborative work.
Using DIVE+, users can explore relations between 1,231.688 entities from 4 digital or
digitized heritage collections
Users saw benefits in using DIVE+ to explore collections, and recognized its
potential to support the discovery of new research ideas and topics:
(Respondent A-G2)DIVE+’s way of linking information allows you to start with a wide
search term and then click through to other or related aspects of it. It enables
some free association, which can trigger research questions
(Respondent X-P12)[It] makes source selection a bit more random, giving the
researcher a chance to find sources that other methods might not reveal
The DIVE+ interface allows users to search for keywords, persons, and locations
within 4 collections. Clicking on a search result provides a more detailed
description of an entity, or, for example, allows the user to play the audio-visual
file that contains the search query. This gives, as one user reports, searching with
DIVE+, a sense of just diving in without knowing what you will end up with
(Respondent B-P12).
The fact that connections between entities are made visible is one of the described
strengths of the browser, as it lets you see connections and thus shows you the
meaning of AV content
(Respondent A-P18).
Connections are deemed valuable when they can be contextualized; researchers see the
tool as useful to gain a quick overview of how entities are related, and what kinds
of materials are accessible and related to particular search queries. At the same
time, they miss context to ground the material as real connections still have to be
made in an old and traditional way… in the mind of the researcher. (...) It is more
about the metadata than about the actual content
(Respondent B-P2). Users also indicate that
while the browser could potentially spark new research questions, this is hampered by
the lack of transparency about
The user-reported need to ground the data, whether within the context of collection,
historical parameters, or a particular label (as a ‘concept’, for example), mirrors
boyd and Crawford’s insights about Big Data analysis (2013). Without context, data
loses its meaning: The ability to represent relationships between people as a graph
does not mean that they convey equivalent information
The overall responses of the users indicate that DIVE+ could potentially support exploration, because it offers an alternative way to interact with different data collections. However, users also felt that the tool lacks direct research benefits because searching with DIVE+ is time consuming, has a steep learning curve, and its search functions are counter-intuitive. Users further questioned what kind of knowledge is afforded by linked data. We will expand on this last observation, because it directly mirrors our findings about brokering as translation and consensus-making within interdisciplinary work.
One of the premises of DIVE+ is that, by offering exploration of entities across 4
different collections, researchers may encounter materials in new ways, which will
allow researchers to ask new questions. This idea was supported by user feedback, as
exploratory search is deemed to [allow] the adaptation of research questions in an
iterative fashion
(Respondent A-P1). However, exploration is limited by the knowledge graph
that these collections, together, create. In other words, search is limited by the
ways in which entities have been made discoverable, and how these have been labelled
(e.g., as persons). The micro-granularity of search that is afforded by linked data
would seem to allow more fine-grained search results. Yet, for the users this meant
that they became more sceptical of what they were finding. Who, after all, is doing
the labelling? In other words, apart from voicing a need for (collection) context to
be able to interpret what they were seeing, users were questioning the transparency
of the search results, and who was defining terms and setting boundaries around these
objects. Within DIVE+ and interdisciplinary research alike, the tensions manifested
within the boundaries of the conversation as much as within its content.
The visualized exploration path, or search narrative (Figure 2), was deemed helpful to keep track of what was found before and to trace how users came to certain materials. Users could see the potential usefulness of this feature, especially to revisit previously examined materials.
However, as stated previously, referring to the collected search history as a
narrative seemed less intuitive to the users than referring to this feature as a
search journey. The actual narrative, it was deemed, is related to the historical
narrative that the user is trying to form about a set of events in the past. Yet,
users saw possibilities to enhance the exploration path functionalities by
incorporating the possibility to add annotations; adding notes about how the user
sees connections between two suggested linked entities enables the researcher to feel
that they are creating narratives. Key to this understanding of narrative creation is
the sense that annotation gives the user the powerful tool and an important strategy
not only to tell events, but above all to interpret them and therefore to attribute
meaning, relevance and to understand reality with more awareness and to transform it
(Respondent E-P7)
. Giving researchers the agency to annotate created exploration paths into
narratives more readily integrates the tool into existing research processes. This
could be useful as during the process of collecting data, the narrative might
change, for media researchers might find information that changes their research
question and primary focus
(Respondent E-P7).
What the above observations indicate about how exploratory search offered by LOD affects knowledge production, is that knowledge production needs to be grounded in a clear context. The user studies show that (digital) humanities scholars greatly value visualizations of historical events in narratives, but still struggle with contextualizing the linked data presented in DIVE+. Adding user-generated annotations to the suggested connections between LOD entities will benefit their contextualization, as it would allow users to reflect on the entity, its source, and role within a possible research narrative.
The user insights about the exploratory search afforded by DIVE+ and the visual representations of the search narrative or journey culminated in three tool recommendations that were devised by the authors, to allow for annotation of search results (to contextualize findings), to allow the export of search results so that a user can keep these as ‘notes’ during research, and to add a feature that allows users to shuffle search results in the search narrative. These features have been designed and implemented in the newest iteration of DIVE+; user research insights thus informed particular changes to the browser to ensure the tool aligns better with research and meaning-making practices of its foreseen users.
The objective of this article was to show how insights into the exploratory search behaviour of (digital) humanities scholars with the DIVE+ browser informed the brokering work of the interdisciplinary research team. The article also argues that successful interdisciplinary collaboration requires disciplinary recognition if it is to be considered a fruitful endeavour. The presented user study insights are a springboard for answers to related questions about knowledge production - and recognition - in the digital humanities. The insights help clarify how the research team project management efforts and collective user study analysis worked to negotiate meaning, methods and outcomes on both the level of the tools’ user interface and on the level of interdisciplinary collaboration.
The user insights culminated in team discussions about how to further refine the concept of narrative in DIVE+. These discussions articulate the important role of critical interdisciplinary dialogue in digital humanities research throughout tool development and application. The recommendations primarily focused on refining DIVE+ to give the user a better sense of data context, as well as suggestions to improve user interaction and experience with the tool’s exploration path (visualizing the search narrative, or, as the scholars agreed to term it, the search journey of the user). Figure 3 and 4 show what these incremental changes look like in the current version of the tool.
While these alterations were the result of user recommendations, they also reflect the outcome of the research team’s brokering - negotiation and translation - work.
On the level of the tool DIVE+, the user studies therefore show that (digital) humanities scholars greatly value visualizations of events in narratives, but struggle with contextualizing the linked data presented in DIVE+. Granting users more agency in the creation of narratives, by allowing them to annotate related entities in the interface’s exploration path, is expected to support this contextualization process. On the level of collaboration between disciplines, the case shows that interdisciplinarity necessitates the ability to work together in a transdisciplinary manner. This means that interdisciplinary collaboration requires the forming of a shared conceptual or even community practice in which brokering takes the shape of not only discussing and negotiating about project management and deliverables, but also requires an almost literal translation of epistemological considerations, in this case about the meaning of a term such as ‘narrative’ or ‘event’.
Overall, this then demands that researchers act in a self-reflexive manner throughout
collaborations: sharing explicit awareness of how their disciplinary background
shapes methodological considerations, their understanding of what is considered to be
knowledge, and how the practice of collaborating shifts or does not shift these
understandings. Masson’s observation that transdisciplinary work demands a shared
conceptual framework
Yet, solely focusing on DIVE+ as a boundary object overlooks the important role
played by the process of developing interactional expertise, of engaging in boundary
work required to realize project deliverables that include an optimized exploratory
search browser as a boundary object. The object seems to, also in project
deliverables, take on a larger role: the tool is at the centre of conference
presentations and publications, whereas the joint discussions and the collaborative
work in producing the object are seemingly left aside. Tool development and project
deliverables trump reflections about process. Admittedly, this seems inherent to work
on any kind of research project. This is also why we feel it is important to connect
a focus on (technical) deliverables to how the interdisciplinary team executed their
project. While this focus on deliverables and tool recommendation realization seems
to align with earlier mentioned ideas about digital humanities as a ‘neoliberal
field’ that is focused on computation and the development of tools for tools’ sake,
the project also illustrates that interpretation and reflections on interpretation
matter within projects such as these. In the end, it’s not what the software does,
it’s what the user does
We would like to conclude with the observation that there is an interesting parallel between the technological affordances of LOD and interdisciplinary digital humanities research. Clearly, semantics play a decisive role in both: linked data allows for semantic linking of entities, and transdisciplinary collaboration is about brokering and translating terms and meanings in such a manner that the entire group understands something in a similar way. Drawing out how this interdisciplinary brokering takes shape in practice requires working towards a sociology of digital humanities; mapping interdisciplinary exchanges in practice. This, in turn, will also bolster an understanding of how semantic connections and linked data can further enrich knowledge creation and collaboration.
This research was supported by the Netherlands Institute for Sound and Vision in the context of Berber Hagedoorn as Sound and Vision Researcher in Residence in 2016-2017 and the Netherlands Organisation for Scientific Research (NWO) under project number CI-14-25 as part of the MediaNow project. This research was also made possible by the CLARIAH-CORE project financed by NWO, with the Research Pilot Narrativizing Disruption.
The authors would like to thank the anonymous reviewers for their helpful comments and suggestions, and Hanne Stegeman for her research assistance during data categorization.