I met with Sukumar Srinivas of Cal-IT2 (thanks to Ayhan’s timely introduction). Earlier this year leaders from industry, government and academia from India and United States gathered at a high-level meeting held at Cal-IT2. The focus of this was to explore avenues for collaboration in technology research and education programs between the two counties with emphasis in innovation in science and technology. Sukumar is the manager for India-US initiatives (http://us-india.calit2.net/). Previously, he coordinated the Tribal Digital Village Network program with Hewlett-Packard. He has been involved in several projects connected to developing software for e-inclusion and has worked with User-interface issues such as that I am addressing in my research. I’m excited because I feel I’m in the right place at the right time and I’m more confident of my project turning to a product. I’m beginning to understand better the ground-reality in rural areas, the socio-political influences on communities and the effect of developing more user-friendly technology targeting these groups.

I've put in the finishing touches for my draft and I'm almost ready to turn it in. On thursday I had a draft ready but I felt that a lot was incomplete. For example, Nooface.net has a lot of interesting ideas including 3D techniques which are viable alternatives to traditional iconic representation. I also did not consider including Mobile applications as a part of my research. Now I am convinced that given the dramatic leapfrog that third world communities have made in adopting to mobile technology, it shouldn't be surprising that the industry will target these communities as an emerging market for rich mobile applications. Furthermore, the limitations of visualization real-estate could be beneficial in that we are introducing information to the user screen-by-screen, rather than overloading the user with too much information, something a bigger screen could cause. The linear interactivity could offer better affordance. It would also be a less intimidating appliance to get used to. This would also be a good place to get illiterate user to begin learning to use digital applications. This would eventually ease the task of developers of computer applications who can now make design decisions assuming that their target user is more technology literate.

I also likes Derek's comment on my idea, calling community-oriented digital applications an "appliance". A "Web leaf" metaphor can translate to an appliance whereby each computer can support many web leaves, but will become an appliance to a user whose purpose is specified by the "web leaf". I'm trying to figure out the range of programs that can be expanded by the web leaf macro. But I'm yet to be convinced that this is a sustainable approach. What happens when a "web leaf" becomes redundant to a user who is now well-acquainted with using a computer for the very purpose he began to use a web leaf. What would happen when appliances are no longer needed? How would a user who is now accustomed to an appliance react when he moves to a city with appliances being too advanced for the nomadic user's cognition.

These are some of the issues that confront my proposal. This has caused me to think about the approach is a more unified manner- to make user-interfaces sustainable and well-replicable, the design must keep in mind the cultural and cognitive limitations of not just the target community, but also those of communities with the potential to adopt this technology. The potential user groups need not necessarily have common cultural and literacy characteristics of the group whom the system originally targeted. We will have to look into the pain that the system is solving and the range of communities which experience the same. For example, are only the fishermen in the X district of a state suffer because they do not have a device that will accurately forecast weather for the upcoming week? If fishermen from Y and Z districts, who speak different languages and have different literacy levels can also potentially reap the benefits of such a device, then while designing the interface of the appliance, the spectrum of usability characteristics must be kept in mind while creating the information visualization. Not only will this reduce the learning curve of users from Y and Z districts, it would be easier to deploy and sell these applications to other areas.

In retrospect, I really learned a lot from the discussion we had during my presentation. I wish I could have developed my idea on the Wiki a bit more by then. With that it would have been easier to run through my suggestions, without giving an impression that the ideas are being thrown into the air without supportive explanation. But then again it is only the beginning of a proposal! However, as I have developed it further, I feel that I can give a more substantial example associated with my ideas.

I've been reading up on the research by Andrew Thatcher, a Psychology Professor in University of the Witwatersrand. His more recent work on cognition of technological devices and cognitive ergonomics is specifically of interest to my research. I think I got some useful ideas from his work on user responsiveness to ATMs in South Africa, from a pool of literate and semi-literate users. His research found that semi-literate users had found much less problems with the ATM process than the literate counter-parts. Interestingly, both groups showed a tendency to prefer the icon-based alternative ATM interface to the speech-based alternative interface and the traditional text-based ATM interface.

Donald Z. Osborn is an expert in rural development with extensive experience in West Africa. His paper on African Languages and Information and Communication Technologies:
Literacy, Access, and the Future, calls on to international communities to develop content in African languages. He specifically looks up to the African diaspora to initiate this venture. I found it interesting that efforts to improve ICT use involved knowing either English, Frech or Portugese that are spoken mainly in urban parts of the countries. Many African writing systems are designed to have a meaningful and graphical representation of knowledge. It would be a great idea to incorporate some of these concepts into applications targeting African audience, literate or otherwise.

India could become the world leader in WiMAX (Worldwide Interoperability for Microwave Access), the emerging technology platform for Internet access, according to Dr Craig Barrett, Chairman of Intel.

Dr Barrett was speaking at Baramati in Pune district of Maharashtra state where WiMAX has been implemented on a pilot project basis. Baramati is the constituency of the Union Agriculture Minister, Sharad Pawar.

The commercial implementation of WiMAX for Pune city itself is awaiting regulatory clearances.

To enable the wireless communications infrastructure at the community computer centre and public kiosks, Intel, which has invested $1 billion in developing technologies for WIMAX, and Aircel have installed a WiMAX network at Baramati.

This network allows Internet access speeds of 512 kbps to remote communities.

The PCs for the service, donated by Intel to the community centres and the kiosks, are specifically designed for the Indian environment.

These are the papers and books I've been researching for my upcoming presentation:

• Technology for Social Inclusion, Mark. Warschauer
• Assistive Technology- Shaping the future, Ger M.Craddock
• Designing E-Learning, Saul Carliner
• Pawar, U. S., Pal, J. (accepted, 2006). Multiple mice for computers in education in developing countries, IEEE/ACM Int’l Conf. on Information & Communication Technologies for Development, 2006.
• Rangaswamy, N. and K. Toyama. (2005) Sociology of ICTs: the Myth of the Hybernating Village. HCI International 2005 (Las Vegas), July 2005
• Medhi, I., Sagar, A. and Toyama K. (2006) Text-Free User Interfaces for Illiterate and Semi-Literate Users. International Conference on Information and Communication Technologies and Development (Berkeley, USA), May 2006
• Computers Helping People with Special Needs,10th International Conference, ICCHP 2006, Linz, Austria, July 11-13, 2006, Proceedings
• http://developer.yahoo.com/ypatterns/
• Report of the "Linguistic Support" Working Group, James R. Cordy, Ralph D. Hill, Gurminder Singh, and Brad Vander Zanden

The questions I am trying to answer are as follows-
1. How do we design interfaces for those who can barely read and write?
2. Is education always the answer to overcome the digital divide?
3. Can top-down approaches work, whereby interactive technologies are designed specifically to meet needs of those who are not literate nor economically strong?
4. How different would these interfaces be from those you and I are familiar with?
5. How can the concept of say a job-search application or an ATM be translated to a visual system that allows an illiterate person to interact with, given very little assistance?
6. What cues can we take from the culture of a person's community, while visualizing interfaces for those with special-needs?

My research for COGS 234 with Ed Hutchins strives to answer the first three questions. COGS 220, has given me the opportunity to tackle the last three. The area is very broad and involves plenty of ethnographical study and insight into practises of communities. However I am trying to boil down my understanding to specific groups and specific applications. There are plenty of work going on today, especially targeting the rural population of South Asia. For instance, the Grammteller project in IIT, Chennai used bio-metric fingerprint censors to log-in users to an ATM machine, a slick technology that we in developed nations have still not been able to incorporate into our ATMs! The concept of using "thumb-impression" is very common in rural India, and this analogy is well translated into technology.

This is just one instance of numerous culture-friendly cues to designing user-interfaces. I am really excited to have dived into this area of research. Some of these papers are pretty interesting and offer a bit of an introduction to the area:

1. Bridging or broadening the digital divide: interfacing the experience of learning for the next decade Ben Williamson, Futurelab (Becta website, 2003)
2. Matt Huenerfauth. 2002. Design Approaches for Developing User-Interfaces Accessible to Illiterate Users. Intelligent and Situation-Aware Media and Presentations Workshop. American Association of Artificial Intelligence (AAAI2002) Conference, Edmonton, Alberta, Canada.
3. Marshini Chetty , William Tucker , Edwin Blake, Developing locally relevant software applications for rural areas: a South African example, Proceedings of the 2004 annual research conference of the South African institute of computer scientists and information technologists on IT research in developing countries, p.239-243, October 04-06, 2004, Stellenbosch, Western Cape, South Africa
4. Medhi, I., Pitti, B. and Toyama K. (2005) A Text-Free User Interface for Employment Search Asian Applied Computing Conference (Nepal), December 2005

Readings for 10/19

In the paper titled "Design of 3-D Visualization of Search Results" the authors have carefully analyzed the pros and cons of various visualizations that would aid designers approaching document visualizations. What all the prototypes had in common was that theyseem to lay emphasis on bringing together keywords that shared a meaning. This inevitably led to the problems of clustering and how to represent clusters without imposing a cognitive over-load on the user. The evolution of the 3-D model seemed to solve the problem to a certain extent. Although the direct motivation was to allow the user to see several titles at once, it seems tasking to distinguish so much text that are not very distant, given the limitation of screen space available to represent them. I personally feel many of the prototypes were over-designed for the problem at hand. It seemed to me that the user was not the center of the design but the design itself was in focus, the user's job is moved to making judgements on whether a design worked for him or not.

In the Concept Globe Design, they talk about embedding 2D navigation within the 3D globe representation. This is specially useful when the user wants to inspect a cluster, giving the option to "decide whether to display just an overview of the entire result set, or show details selectively". Also, it would have been a better reading if, in the conclusion, they had elaborated more on the concept of "restructuring", because that seemed to be the key to an better visualization.

I all sold for the idea of adding context to search results, especially if searching for something specific, when I have only a vague idea of where my search fits in. Context gives me cognitive cues that would improve my navigational speed across sites and make me narrow down my choices, accelerating decision-making. What would be cool for me, is that while I'm entering my keyword, categories relevant to my search are displayed and I can narrow down my categories and then continue search. However it could make interesting research to find ways to tackle the situation of results that can appear in multiple categories. This may add to an additional level of complexity for the user to navigate through.