How HCI can address challenges in ICT4D and Digital Civics

Intro

This blog post will investigate the relationship between HCI and Digital Civics. Specifically, we will be looking at my area of interest in Digital Civics and the input that HCI can have therein. My current research interests lie in scalable platforms for low literacy contexts. We will be looking at design problems, approaches in reflexivity, and then see how some fields within HCI address them or provide ways of rethinking the problems. Primarily, we will be looking at Research through Design and Experience centred design, and see how those fields address concerns within our Digital Civics project.

Concept diagram of Radio Health Dialogues
Concept diagram of Radio Health Dialogues

 

Radio Health Dialogues (RHD) is a scalable IVR (Interactive Voice Response) based platform to enable discussions around health and wellbeing (and other topics in the future) in low-literacy and low-internet connectivity contexts. The radio host may be a health activist or a community representative who facilitates a radio show which includes a health professional and interested community members. Listeners can ask questions and share insights with each other and the host. Its appeal lies in its low requirements: the listeners only need a ‘dumb-phone’ and the show doesn’t cost them anything, while the host can be a low-literate/semi-literate member who facilitates discussions using a mobile application.

RHD is situated within the ICT4D (Information and Communication Technologies for Development) technological field. A significant conference in this field is the ICTD conference (http://ictdconference.org/).

There is usually a focus within HCI to design solutions with a single or few individuals in mind. Especially within the field of Experience-centred design, with its emphasis on the uniqueness of each interaction, scalability is often not addressed and if it is, it is considered a secondary concern. However, when designing platforms for multiple contexts, scalability is one of the main, if not the main concern. Much has been written in the field of Research through Design about the opportunities that interaction design research affords the HCI research community in terms of added perspectives on design thinking. We strongly support this assertion, that design has valuable insights to offer the research process. Over the course of this report, we hope to look at Scalability as a HCI design challenge, and explore how this can be taken forward by digital civics and ICTD researchers (however we will be primarily be addressing Digital Civics challenges in this report for brevity).

Challenges

Radio as a Service

We’ve briefly discussed the concept of RHD above. Here we would like to present the challenges that need to be addressed by Digital Civics researchers in its design. The system is aimed at (rural) contexts with limited digital infrastructure. Areas which have not been penetrated by smartphone prevalence, and where internet connectivity is assumed to be poor. These situations are often a marker of poor socio-economic status, and other issues prevalent in such areas, which need education, discussion and/or levels of empowerment to improve. RHD can be used as such a platform to discuss those three things. The platform needs to be resilient to problems related to the infrastructure.

A common assumption in HCI design is tech literacy. it is assumed that as design principles and technological innovation moves forward, most people will be up to date on them. However, this is often not the case, especially not so in the socio-economically deprived contexts where RHD is going to be used. A challenge for the designers will be on how people with poor tech-literacy will use the power of the RHD platform and the services it offers, without being intimidated by it or unable to get the most out of it. The interface should be uncluttered and unsophisticated. Much work has been published about design principles for users with poor tech-literacy. This needs to be consulted and built into the system to ensure that such users are not marginalised by the system.

Moving away from traditional conceptions of radio shows into a dynamic Radio as a Service Concept.
Moving away from traditional conceptions of radio shows into a dynamic Radio as a Service Concept.

 

Thirdly, we need to investigate the audio interaction design challenges that will be present in the design of this platform. In a typical radio show format, the host is often tasked with controlling a variety of different aspects of the show. We need to find a way to give the host and the listeners the best that the RHD platform can provide without overwhelming the host with sophisticated tasks like listening and screening to listeners who call-in, all the while running a live radio show. This is no mean feat. Another aspect is that by using an IVR system, we can design the system to elicit information and feedback from the listeners. However, the challenge lies in being able to do this while not distracting or overwhelming the listeners, as they will be paying attention to and engaging with the radio show taking place.

Scalability as a challenge

Building scalability into the system is a key feature of RHD. Let us explain what we mean by scalability. Currently, we are thinking of deploying it in three different contexts: a rural community in North India, among Syrian refugees in Lebanon refugee camps, and promoting Maternal mental health awareness among expecting and new mothers in North Yorkshire. One can straightaway see the complexities in building a solution for three different regions (and differing accompanying socio-cultural factors): a rural ‘developed’ country context, a middle-eastern refugee context, and a rural Indian context. Apart from internet connectivity issues, and a health education agenda, the three contexts have few broad commonalities. However, the need for community engagement platforms is clearly there. Furthermore, the platform needs to be accessible, resilient against technological and human factors preventing their existence, and usable. This is the challenge that is facing us: how do we design for scalability, since that is a strong criterion, without minimising the uniqueness of user-computer experience and engagement.

Why Experience Centred Design matters?

Peter Wright and John McCarthy have written extensively about Experience Centred Design. In Daniel Howard’s paraphrase of the seminal work by Wright and McCarthy[1], “Experience-centred design in HCI is a design philosophy that focuses on the experiential relationships that people have with technology”.

The philosophy builds on Graves Peterson’s pragmatist framework for understanding aesthetics as an additional complementary perspective on user-centred design. By utilising the rich philosophical underpinnings of pragmatism, aesthetics-based design enables the design of interfaces that explore the concepts of playfulness, surprise and enchantment. An experience, when seen through this lens, is a unique blend of the interplay between user, context, culture, and history. The experience is a ‘felt’ one. This is in sharp contrast to other approaches which try to make interfaces seductive and appealing regardless of the context of use, culture, history or user.

John Dewey, the influential twentieth century philosopher known for his work on Pragmatism.
John Dewey, the influential twentieth century philosopher known for his work on Pragmatism.

 

Drawing on the work of the stellar 20th century pragmatist philosopher John Dewey, the authors present three themes to understand aesthetic experiences in understanding people’s interactions with technology:

  1. Holistic approach: the intellectual, sensual, and emotional stand as equal partners in experience.
  2. Continuous engagement and sense-making: each self brings personal and cultural meanings to every interaction and completes the experience through acts of sense-making.
  3. Relational approach: the self, object and setting are always in dialogue and thus the experience is never finalised, but always ever-forming.

Using one of the authors’ (Jayne Wallace) own case studies, they demonstrate how experience-centred design principles play out in practice. Wallace designed a piece of digital jewellery for a participant, through an iterative design process while immersing herself thoroughly in the process so as to “engage with the puzzle presented to her” and “connect aspects of interactions that resonate with her”.

Jayne Wallace has an extensive research background in digital jewellery, and incorporating experience-centered design within its development process.
Jayne Wallace has an extensive research background in digital jewellery, and incorporating experience-centered design within its development process.

 

When presenting the case study, the authors show decisively that if they had shunned the experience-centred approach and had treated it like a user-centred one, the participant would not have gained such a rich experience that evoked varied emotions, and guided the participant’s sense-making in unique ways. By utilising ambiguity in design, and evoking existential musings such as the “fleeting qualities of many of our experience”, the designer was able to break the boundaries imposed by traditional design practices. Wright, McCarthy and Wallace present a strong defence for their premise of the value in experience-centred design. Their argument that felt life and human experience need to be placed at the centre of good design theory and practice, is one that needs to be listened to.

Applying this in my own research, when designing cross-cultural technological solutions aimed at low tech-literacy contexts, it is often a temptation to generalise an “average user” scenario, and design a “one size fits all” template. Wright et al. demonstrate the flawed reasoning underlying such a paradigm. Every user is unique, every experience is a unique one, but to cater to each of them, while balancing the challenge of overwhelming the client is a difficult one, but worth pursuing nonetheless. By embracing an iterative prototyping approach as encouraged by the authors, we can design for experience in our solution without eschewing scalability. This will be done by working closely in association with the potential users of the system, and seeing the experiences and ‘desires’ that are currently a part (or they wish was a part) of their life. Then, we develop prototypes with these experiences and desires in mind, take it back to the original users and see what they make of it. We repeat this process until we are satisfied that we haven’t sacrificed enchantment and unique experiences at the expense of scalability.

The Role of Research through Design

Research Through Design as a Method for Interaction Design Research in HCI[2] by Zimmerman et al is another HCI field that has a lot to speak into our project. In the groundbreaking paper by Zimmerman, they look at the HCI Community’s desire to reconcile the sometimes disparate aspects of design in research and design in practice. Zimmerman et al. propose a model of interaction design research for both the HCI research and practice communities. They also propose a set of criteria for evaluating the quality of contributions made by those who practice interaction design research.

As part of their methodology, they conducted semi-structured interviews with leading HCI research academics and interaction designers. They particularly put their focus on addressing “Wicked Problems”, a concept put forward by Melvin Webber that describes a problem that cannot be accurately modeled because of the conflicting perspectives of the stakeholders involved.

While the authors proposed to formalize the many ideas present in HCI community into a single method, I feel that wasn’t accomplished in this paper. While they offer a model that coalesces the HCI research and HCI practice communities, this model is not comprehensive, and only acknowledges the differences present and how an academic in either camp can work with the disparate elements in HCI research as a whole. To present a comprehensive model would take a paper of its own.

However, where they have succeeded is in presenting a good aim for design reserachers to aim for through their work (reframing a problem as they attempt to make the right thing) and the four critical lenses through which to evaluate interaction design research within HCI: Process, Invention, Relevance and Extensibility. Using examples such as the successful Xerox copier, they demonstrate how the design research community doesn’t have to view it as a ‘black art’. These lenses are a good step forward, in dispelling the lack of cohesion due to no agreed upon standard being present in the critical evaluation of interaction design research.

The contributions of this paper make this a must read for anyone undertaking HCI design research. We will be referring to this paper again, at least to gain an understating of how any proposed contributions stack up against the four critical lenses. That is as good a place to start as any. They debunk the notion of design simply providing surface structure or decoration, and this is a useful insight for us. Similar to how the authors used the critical lens, to evaluate popular examples of interaction design, we hope to evaluate our research outputs.

Looking at Annotated Portfolios

Gaver and Bowers’ Annotated Portfolios[3], [4] has much to offer here that can help our conversation. When doing interdisciplinary work like HCI proposes to do, often there can be ‘disciplinary anxieties’: practitioners from different fields within HCI bring their own set of assumptions to the table. These assumptions often contain pre-conceived notions on the ‘proper’ way to do research, or even worse appeal to scientism (“my approach is more scientific than yours”). And this critique of research through one’s preferred lens of research practice leads to an abuse of rationalism.

Below is a video of Professor Gaver talking about Evaluation methods for research outputs. It talks about the key themes behind annotated portfolios.

 

Digital Civics researchers are by definition inter-disciplinary. They cross traditional disciplinary boundaries to utilise the potential of multiple disciplines. This will bring about the ‘disciplinary anxieties’ that we just discussed. A town-planning researcher seeking to use Computer Science related fields will quickly realise the different ways of doing things in the two fields. The researcher will be tempted to critique and analyse the output of this cross-disciplinary endeavour through their epistemological biases borrowed from the Social Sciences.

Similarly, a Computer Scientist used to working in the data analytics field, when working in the field of public health interventions, will be tempted to review the process and any outputs against usability requirements and quantitative methodologies borrowed from Computer Science best practices. In both situations, we are not arguing that the researcher is being intentionally biased, but rather avoiding this situation requires deliberate effort on the part of the researcher.

In our digital civics research area, we are working at utilising digital technologies in resource-poor rural contexts, with an agenda of improving health education and community empowerment. Straightaway, one can see the nuanced mix of studies relating to Computer Science, Cross-cultural studies, Global Health, Health Education etc. are present. The researcher in this position needs to be aware of the ‘disciplinary anxieties’ present and address them adequately in order to do them justice.

We have only talked about the problem so far, and the solution (proposed by Gaver et al.) will (we think) present a way to check the output of our design processes and judge its significance.

Not rushing to an Interpretation

In their seminal HCI paper ‘Staying Open to Interpretation’[5], Sengers and Gaver present a thesis that is very useful for us: multiple, potentially competing interpretations can fruitfully co-exist. We need to move away from rigid testing and analysis methods inherited from the first and second ‘waves’ of HCI. This is especially pertinent because earlier generations of HCI research were concerned with topics like usability and ‘work settings and interaction within well-established communities of practice’[6].

However, when investigating novel cross-cultural interactions where little previous research has been undertaken, we need to hesitate before rushing to traditional ways of analysing the research output (and process). For instance, when analysing the output of RHD, we can measure aspects associated with the lower levels of interpretation e.g. utility, ease of use etc. of the communication platform we develop. However, the higher levels in the interpretation chain ‘involve personal decisions about values and meaning for a specific user’. This higher level interpretation, is what we argue needs to be carefully analysed by placing the interpretation in light of the context and socio-cultural background of the persons utilising the platform. To put it crudely, researchers sitting in a research lab in the western hemisphere may not be the best judges to interpret what is going on in a rural socio-economically challenged context in South Asia.

In light of this danger of misinterpretation (as it were), we will need to design for multiple interpretations. This can be done by designing the RHD platform to cater for the users to ‘make’ their own interpretations of how they see the system being used. This will also cause the designer’s preferred method of interpretation being cast by the wayside as they see the using community reveal novel interpretations which circumvented their existing interpretations.

Conclusion

In this short report we looked at how a digital civics research agenda can be addressed appropriately by engaging with key HCI themes. We argue, the most beneficial HCI themes beneficial to the RHD research space are practical inferences from Experience centred design and Research through Design. We see a lot of merit behind Wright, Wallace and McCarthy’s assertion that one should design for experiences that are contextually situated. However, the real challenge lies in balancing this with scalability across contexts. This is addressed through looking at the work on Annotated Portfolios done by Gaver et al and Bødker’s work on Staying Open to Interpretation.

By avoiding the temptation to rigidly compare research outputs and processes against each other, we can leave space open for interpretation. This space will allow us to avoid an unintentional or intentional bias in how research is evaluated. By allowing the participants and stakeholders in the process to be the ‘judge, jury and executioner’, we can hope to reduce such negative effects. This is especially the case, we argue, in cross cultural research, where researchers will also bring their socio-economic and cultural ‘baggage’, so to speak, with them.

References

[1] P. Wright and J. Mccarthy, Experience-centered design: designers, users, and communities in dialogue., vol. 3, no. 1. 2010.
[2] J. Zimmerman, J. Forlizzi, and S. Evenson, “Research through design as a method for interaction design research in HCI,” Proc. SIGCHI Conf. Hum. factors Comput. Syst. – CHI ’07, pp. 493 – 502, 2007.
[3] B. Gaver and J. Bowers, “Annotated portfolios,” Interactions, vol. 19, no. 4, p. 40, 2012.
[4] J. Bowers, “The Logic of Annotated Portfolios : Communicating the Value of ‘ Research Through Design,” Proc. DIS2012, pp. 68–77, 2012.
[5] P. Sengers and B. Gaver, “Staying open to interpretation: engaging multiple meanings in design and evaluation,” Proc. 6th Conf. Des. …, pp. 99–108, 2006.
[6] S. Bødker, “When second wave HCI meets third wave challenges,” 4th Nord. Conf. Human-Computer Interact. …, no. October, pp. 14–18, 2006.

Leave a Reply