|
|
|
|
|
ACM SIGCHI CHI-2000 Workshop #2 Michael G. McKenna mgm@acm.org Please visit http://www.acm.org/sigchi/chi2000/ for the official CHI-2000 site! THE TOPIC: Challenges in Multicultural Human-Computer Interaction (HCI) Design and Development The Problem: Lack of Tools, Knowledge, and Standards A human being, when communicating, uses a multitude of subtle communication mediums to express thought, feeling, identity, and action. Some examples are: language, writing, gesture, art, music, color, artifacts, iconography, protocol, social hierarchy, and spatial relationships, etc. In the field of Human-Computer Interaction (HCI), multilingual and multicultural features are usually not enabled automatically and therefore there is a plethora of applications and interfaces that are monolingual or hard-coded into a single language or cultural interface (usually American English). This is not due to a lack of markets, when over half of all software sold today goes to non-English speaking regions and users, but appears to be due to a lack of effective development frameworks or knowledge of multilingual and multicultural requirements in interaction design by product managers and software developers. The unintentional result can be a view of the average software developer as an American cultural iconoclast. Most interaction development and application building tools provide some mechanism to create localized interfaces with respect to translated messages, and correctly formatted numeric or temporal strings, with some even capable of handling the sticky needs of Semitic right-to-left interface requirements. In most all application building environments, however, internationalization features are optional, and therefore rarely used, unless specifically mandated as part of the requirements of a development project. Even then, multilingual, and especially multi-cultural, functionality may not be implemented due to the high cost of re-engineering for already deployed systems. What are some other issues preventing the implementation of interaction designs to be configurable in their ability to handle the needs of diverse cultures and languages? Most university curriculums do not address cultural interaction issues as part of their Computer Science or Engineering programs. A lack of public terminology-based translation memory (TM) limits affordable machine translation. Financial constraints due to the high cost of human translation limits multilingual deployment. The tools that are available to developers primarily address only the needs of static text-oriented interfaces. There is a lack of multi-lingual testing tools and requirements. Possible Solution: Framework for Modular Multicultural Interaction Generic cultural interaction design is a sub-set of the field of customization. The behavior of each interface should be based on cross-cultural ethnographic study and designed to handle a range of cultural modalities. With component and distributed architectures, each interaction design should be designed with a modular cultural interface that is easily modified without changing core functionality. An interaction should have the goal of Cultural Authenticity, in that it should have a look and feel to the end user as if it were designed in their language and culture from the ground up. With this in mind, the following is a proposed set of areas that need Modular Multicultural HCI Design:
This is by no means an exhaustive list, but is intended to indicate the potential scope of HCI areas affected by cultural interface design and act as a catalyst for Workshop participants to work off of. A Framework should provide a form of modular "Playdough" to aid in easy modification of cultural interfaces. Existing internationalization (I18n) frameworks should be updated beyond the current simplistic locale models to include the wider range of cultural modalities indicated by the above list. Methodologies should include training with respect to cultural and linguistic interaction design, cross-cultural ethnographic studies, multicultural usability heuristics, and multilingual testing. Goal of the Workshop: Towards a Framework for Modular Multicultural Interaction Design and Recommended Methodologies for Effective Multi-cultural Feature Integration There are several works that attempt to help the development community with respect to i18n issues. However they fall into two broad categories of data handling and textual displays [1][3][6][7], or offering a detailed discussion of abstract level sociological and cultural issues, expecting the developer to implement the behavior attributes themselves [2][4][5]. There is very little to cover the area in between. We believe that it is necessary to come up with an acceptable framework that application and HCI development tool vendors can use to ensure that the cultural needs of the global market can be met. To that end, this workshop will attempt to tackle the following issues:
It is expected that the outcome of this workshop will not be a complete corpus on the cultural and linguistic needs of the HCI community and how to solve them, but rather an annotated taxonomy with an informed recommendation of a framework and methodologies that future efforts can leap off from. FORMAT OF THE WORKSHOP Participant Solicitation and Selection We will solicit the CHI community through the usual mechanisms including e-mail, and personal contacts. In addition to the ACM CHI community, the Unicode Consortium, Localisation Industry Standards Association (LISA), and the European Directorate General XIII Human Language Technologies communities will be called. Since CHI 2000 falls in the week immediately following the 16th International Unicode Conference in Amsterdam, many experienced software internationalization practitioners will be able to attend the workshop. Participants will be selected based on position papers or invited based on known experience and demonstrated ability to contribute in this arena. We will be specifically inviting at least two participants to give an informed overview of multicultural and linguistic HCI issues. All participants are expected to be familiar with a least one I18n Framework [1][8] and have experience with HCI design and development issues in general. We would like to get participants from a wide range of backgrounds to increase the richness of participation in the workshop. It is expected that the cross-pollination of the i18n/Unicode and CHI communities will create a valuable output that would be difficult to reproduce in any other setting. Please submit position papers to the Workshop organizers at mgm@acm.org and naftulin@acm.org by February 10, 1999. Method of Interaction and Schedule for the Workshop Day One, Morning: We will start with introductions, logistics and agenda, followed with an overview of cultural and linguistic issues by two invited participants, followed by an opportunity for each participant to share their proposed HCI areas and issues (limit of ten minutes per person initially). Day One, Afternoon: Afterwards, based on HCI interaction and context cluster areas, small groups will be formed to polish the respective cultural and linguistic issues lists. We will then convene as a large group to share our lists of issues and look for additional commonalties and grafting or pruning of the lists. We expect there to be continued social networking among the participants into the evening. Day Two, Morning: We will then break into small groups, again based on HCI interaction and context cluster areas, each small group will devise their own framework. This will be done as a method of increasing the brainstorming and diversity of proposals. To limit the scope of the discussions, we will ask each group to restrict their frameworks to component-based HCI development environments. We will then reconvene and allow each group to present their framework in a facilitated discussion while we attempt to unify similar ideas and integrate unique relevant ideas into a Framework for Modular Multicultural Interaction Design. Day Two, Afternoon: We will then put the final polish on the proposed unified framework. Time permitting, we will then attempt to create a proposed set of Recommended Methodologies for Effective Multi-cultural Feature Integration through facilitated discussion. We will set aside the last ninety minutes to create a list of recommended future work and research, discuss how to best continue the discussions started in the workshop, and conduct synthesis of our findings into the poster as a group. Pre-workshop Activities Participants are expected to come with a set of multicultural HCI issues, and proposed framework, or proposed methodologies to share. Plan for Dissemination The organizers will write up the results of the workshop in a SIGCHI Bulletin paper and solicit more suggested multicultural issues and frameworks from the community. The results will be disseminated in the CHI, Unicode, IETF, ISO, and developer forums. ORGANIZERS' BACKGROUNDS Michael McKenna is a Globalization Strategist at Sybase, Inc., in the Global Products Group, which he founded. The Global Products Group is responsible for the internationalization architecture of all enterprise information management software that Sybase markets. Michael has been responsible for the development of European and Asian releases of enterprise relational database systems, and has recently built up an application tools development team in Singapore. He was the coordinator for the MAITS (Multilingual Application Interface for Telematic Services) Consortium - a project with partial funding from the European Union Fourth Framework Programme. Michael has been a specialist in the field of globalization of applications and distributed environments for over nine years. He has been active in standards activities, spoken at many conferences and seminars on three continents (although not SIGCHI), been an invited speaker at a European Commission Technology Workshop, and co-produces the annual Globalisation Management Strategies Conference. Michael is a licensed professional engineer and has additional background in distributed bibliographic systems, test engineering, and ethnographic research. Henry Naftulin is a Software Engineer at Sybase, Inc., in the Systems Management, Monitoring and Diagnostic Group where he specializes in GUI design. Prior to joining Sybase he worked at SGI and Distrivision Development Corp. specializing in GUI and HCI. Henry received his M.S. degree in Computer Science from the University of California at Davis. He was an assistant teacher for graphical user interfaces, software engineering, AI, and computer architecture classes. Henry has publications in the areas of human computer interaction and security and has previously been a presenter at SIGIR. REFERENCES 1. Hall, P.A.V., and Hudson, R., Software Without Frontiers . Chichester, England: John Wiley & Sons, 1997. 2. Del Galdo, E. M. and Nielsen, J. International User Interfaces . New York: John Wiley & Sons, 1996. 3. O'Donnell, S.M. Programming for the World : a Guide to Internationalization. Englewood Cliffs, New Jersey: P T R Prentice Hall, 1994. 4. Bishop, M. How to Build a Successful International Web Site. Scottsdale, Arizona: The Coriolis Group, Inc., 1998 5. Fernandes, T. Global Interface Design. Boston: Academic Press, Inc., 1995. 6. Jerman-Blazic, B. Transparent Language Processing: User Tool or a Network Service for the Global Information Society. Jozef Stefan Institute, Ljubljana, Slovenia, 1997. 7. Multilingual Application Interface for Telematic Services, CEC funded project from IV Framework, Telematic Programme, LE-1347, November 1995 8. ISO/IEC JTC1/SC22/WG20 ISO/IEC TR 11017:1998: Framework for internationalization, Geneva, 1998. 9. Yeo, A., World-Wide CHI: Cultural User Interfaces, A Silver Lining in Cultural Diversity, SIGCHI Bulletin, Vol.28 No.3, July 1996 |
|
|
| [Home] [Extended Abstract] |
