Algorithmic systems are often made opaque by design, with users being unaware of how much of their data is being gathered [1] and for what purposes. The impact of such algorithmic systems in society has had various reported instances of causing harm and inequality [2].
From both academia and industry different proposals have been emerging, such as frameworks [3] and guidelines [4-5] for algorithmic transparency and accountability.
Although as mentioned by Ananny and Crawford to open the black box may not be enough, algorithms are embedded in complex socio-technical systems and therefore should be observed across the complexity that comprise humans and non-humans assemblages [6].
Different tactics have been emerging for making sense of the complexities in algorithmic systems. The notion of Experiential AI [7] proposes to make the algorithmic mechanisms understandable through felt experience. Graspable AI [8] proposes the use of physical artefacts and material manifestations as a relational way of understanding and interpreting algorithmic systems.
The demonstration being proposed is part of a practice-based PhD research into the development of methods and tools that comprise embodied and choreographic techniques to inform the design of web-based interfaces driven by transparency and legibility. It draws from a body of work [9, 10] that explores how choreographic approaches provide alternative insights on how algorithmic systems shape the creation of web environments.
Central to this research is the understanding of embodiment as physical engagement grounded in and emerging out of everyday experience [11]. Body-centered approaches provide better understanding of the actions we perform and the computational systems we interact with [12] which can further inform and advocate for a system’s transparency.
The methods and tools being developed aim at responding to the opaque algorithmic models embedded in pervasive web-based user-interfaces. For instance, how to disrupt algorithmic exploitation or phenomena such as ‘dark patterns’ [13] and other purposefully designed interactions for the manipulation and alienation of the user.
The tools being made consist of custom interfaces, adaptation of existing plug-ins and extensions to render, for example, machine learning, user tracking and other similarly opaque algorithmic processes visible at the interface level.
By intersecting the tools described above with corporeal awareness and improvisational approaches from live-coding practices, this performative demonstration invites participants to a rehearsal space [14] for reflecting upon, engaging with and challenging user-interface paradigms. And ask how can embodied, critical and situated design practices transform current methods for the development and analysis of web-based user-interfaces that foster wellbeing, inclusion and more equitable algorithmic futures?
[1] S. B. Pold, “New ways of hiding: towards metainterface realism,” Artnodes, 2019, no. Num. 24, pp. 72–82, 2019, doi: https://doi.org/10.7238/a.v0i24.3283.
[2] G. Klumbyte, P. Lücking, and C. Draude, “Reframing AX with Critical Design: The Potentials and Limits of Algorithmic Experience as a Critical Design Concept,” in Proceedings of the 11th Nordic Conference on Human-Computer Interaction: Shaping Experiences, Shaping Society, Tallinn Estonia, Oct. 2020, pp. 1–12. doi: 10.1145/3419249.3420120.
[3] E. Rader, K. Cotter, and J. Cho, “Explanations as Mechanisms for Supporting Algorithmic Transparency,” in Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems, Montreal QC Canada, Apr. 2018, pp. 1–13. doi: 10.1145/3173574.3173677.
[4] S. Amershi et al., “Guidelines for Human-AI Interaction,” in CHI 2019, May 2019. [Online]. Available: https://www.microsoft.com/en-us/research/publication/guidelines-for-human-ai-interaction/
[5] R. Dopplick, “New Statement on Algorithmic Transparency and Accountability by ACM U.S. Public Policy Council,” 2017. https://techpolicy.acm.org/2017/01/new-statement-on-algorithmic-transparency-and-accountability-by-acm-u-s-public-policy-council/
[6] M. Ananny and K. Crawford, “Seeing without knowing: Limitations of the transparency ideal and its application to algorithmic accountability,” New Media & Society, vol. 20, no. 3, pp. 973–989, 2018, doi: 10.1177/1461444816676645.
[7] D. Hemment et al., “Experiential AI,” AI Matters, vol. 5, no. 1, pp. 25–31, Apr. 2019, doi: 10.1145/3320254.3320264.
[8] M. Ghajargar, J. Bardzell, A. M. Smith-Renner, K. Höök, and P. G. Krogh, “Graspable AI: Physical Forms as Explanation Modality for Explainable AI,” in Sixteenth International Conference on Tangible, Embedded, and Embodied Interaction, Daejeon Republic of Korea, Feb. 2022, pp. 1–4. doi: 10.1145/3490149.3503666.
[9] J. Chicau, “Choreo-Graphic-Thinking,” Joanachicau.com. [Online]. Available: https://joanachicau.com/backstage.html. [Accessed: 11-Nov-2022].
[10] X. Xin et al., “Choreographies of the circle & other geometries,” Critical Coding Cookbook, 28-Jun-2022. [Online]. Available: https://criticalcode.recipes/contributions/choreographies-of-the-circle-other-geometries. [Accessed: 11-Nov-2022].
[11] P. Dourish, Where the Action Is: The Foundations of Embodied Interaction. The MIT Press, 2001. doi: 10.7551/mitpress/7221.001.0001.
[12] S. R. Klemmer, B. Hartmann, and L. Takayama, “How bodies matter: five themes for interaction design,” in Proceedings of the 6th ACM conference on Designing Interactive systems - DIS ’06, University Park, PA, USA, 2006, p. 140. doi: 10.1145/1142405.1142429.
[13] Y. Rogers, P. Dourish, P. Olivier, M. Brereton, and J. Forlizzi, “The Dark Side of Interaction Design,” in Extended Abstracts of the 2020 CHI Conference on Human Factors in Computing Systems, Honolulu HI USA, Apr. 2020, pp. 1–4. doi: 10.1145/3334480.3381070.
[14] A. Boal, Teatro do oprimido: e outras poéticas políticas, 1a Edição. São Paulo: Editora 34, 2019.
