The study heavily relies on qualitative data from a small, non-representative sample (e.g., 12 questionnaire responses and a group interview with 8 employees at BigITCorp). How can the authors justify generalizability of their findings about “mental-emotional experience of work” when the data is drawn from such a limited and potentially biased sample? Given the subjective nature of art interpretation, how did the researchers mitigate confirmation bias in their analysis, especially when coding themes like: “comfort” or “disruption”?
Moreover, the paper introduces the novel concept of a “boundary rogue” to describe artifacts like Multiviews that disrupt contexts. However, the evidence for its distinction from existing concepts (e.g., boundary objects, adversarial design) seem anecdotal and tied to a single case. What empirical or theoretical grounding supports this as a universally applicable concept beyond this specific art-research collaboration? Could the observed effects simply be atributed to novelty of the art piece rather than a broader sociomaterial phenomenon?
Throughout the study, the researchers were not only observers but active collaborators in the RtA process, co-designing questionnaires, facilitating installations, and engaging in ongoing dialogue with the artist and industry partners. This embeddedness raises the question of whether the researchers’ own investment in the success of the collaboration may have shaped how they interpreted participants’ reactions to the artwork, especially in coding and analyzing qualitative data.
For instance, the “potato” is framed as a disruptive, boundary-crossing element that provoked novel reflections on workplace comfort. While the data excerpts are vivid, it is unclear whether alternative, more mundane interpretations of the artwork were equally present but underreported because they did not align with the “boundary rogue” narrative. The analysis seems to emphasize disruption and novelty, but could there also have been moments when the artwork was ignored, misunderstood, or dismissed by participants? If so, how were these accounted for in the theorizing of the boundary rogue?
Additionally, the shift from InterFin to BigITCorp due to bureaucratic delays introduces a selection bias: the company that ultimately hosted Multiviews was perhaps more open to unconventional collaborations, which may have amplified the observed “disruptive” effects. Would the artifact have functioned as a boundary rogue in a less receptive context?
So, while the concept of the boundary rogue is valuable, its formulation might benefit from a more reflexive acknowledgment of how the researchers’ dual roles as collaborators and analysts could have influenced the findings. Future RtA studies might consider strategies for triangulating interpretations, for example, involving external analysts or systematically documenting counter-narratives, to strengthen claims about the artifact’s role across contexts.