|
|
Action Spécifique : Interculturalité, usages et conception
des NTIC
|
||
|
|
|||
|
INDEX |
|||
|
|
|||
|
PRESENTATION |
|
||
L’évolution rapide des moyens de communication
électronique à l’échelle planétaire constitue un stimulant pour la conception
de systèmes d’interaction destinés à des environnements pluriculturels. Le
contenu et la forme de ces systèmes reflètent l’existence de modèles souvent
implicites qui vont orienter les choix de conception. Les préférences
culturelles (nature du média d’inscription, choix des caractéristiques
graphiques, orientation spatiale, style d’interaction,…) réelles ou supposées
vont donc avoir des conséquences sur l’utilisabilité des systèmes. Cette
problématique prend toute son importance avec la banalisation (relative) de
l’usage du web et l’essor des technologies mobiles qui permettent la constitution
et/ou l’extension de réseaux d’usagers. La question est ici de retraduire un
ensemble de connaissances, d’expériences pratiques, d’études de cas
significatifs en termes d’orientations prescriptives pour la conception
ergonomique de systèmes interactifs (choix de métaphores, formes de navigation,
modèles d’interaction, propriétés graphiques des IHM,…), et d’évaluer l’impact
de ces orientations sur les usages.
Si cette question revêt une importance économique non négligeable pour des activités centrées web de type e-commerce par exemple, les travaux menés dans le contexte particulier de la conception de dispositifs de pilotage et de supervision des systèmes à risque (nucléaire, aéronautique, urgences médicales,…) représentent également une somme d’informations à considérer. La multiplication des études menées à travers le monde dans ce domaine rend en effet possible une confrontation fructueuse des modes de gestion de la diversité culturelle et des approches de conception afférentes. Les enjeux commerciaux se doublent ici d’enjeux humains qui accroissent la visibilité des problèmes rencontrés et conduisent à s’interroger sur les différences d’approche de la prise en compte des spécificités culturelles. Ces différences peuvent êtres appréhendés au niveau international mais également à un niveau plus local (différents sites dans un même pays, différentes équipes sur un même site). Elles peuvent également être considérées sous l’angle des spécificités de « métier ». De ce point de vue, il convient d’identifier les liens possibles avec des concepts tels que par exemple les « communautés de pratique » et d’interroger leur pertinence quant au projet de définition du concept de culture dans les environnements professionnels, et de son incidence sur la conception.
|
OBJECTIFS |
Nous identifions quatre objectifs à atteindre à la fin de l’année 2004. Le livrable sera un rapport rédigé par les membres du pôle.
(1) Terminologie : fixer un ensemble de termes en relation avec la conception de systèmes interactifs prenant en compte les spécificités culturelles des utilisateurs
(2) Etat de l’art : recenser dans un canevas intégrateur l’ensemble des travaux concernant (i) la mise en place de méthodes de conception prenant en compte les spécificités culturelles des utilisateurs (ii) la réalisation, le déploiement et l’usage de systèmes interactifs prenant en compte des spécificités culturelles de leurs utilisateurs
(3) Etude de cas : à partir de données collectées sur les aspects culturels d’utilisateurs, étudier et concevoir un système interactif. L’exercice a pour objectif d’étudier la prise en compte de ces informations liées aux aspects culturels au sein d’une démarche de conception ergonomique
(4) Axes de recherche : dresser une liste d’axes de recherche en relation avec la conception et la réalisation de systèmes interactifs sensibles aux spécificités culturelles de leurs utilisateurs
|
TERMINOLOGIE |
Globalisation: making a product which does not contain any cultural context. Another commonly used term is internationalisation which essentially means the same but implies that the process is restricted to nations.
Localisation: a general term which refers to making a global product usable in a particular market. Can be divided into technical localisation (e.g. covers aspects such as double-byte conversion), national localisation (e.g. language translation, change in units of measurement, changes of symbols and pictures), cultural localisation (e.g. addresses issues of appeal, quality, taste).
Culture: There is no generally accepted definition of culture and within the literature numerous definitions of the term exist (for example from: Hall, 59; Kluchhohn, 62; Krobeber and Parsons, 58; Trompenaars, 93). However, there are several recurrent themes in these definitions: a shared distinctive way of life; commonly held values, ideas, rules and norms; learned behaviour that is common to a group; a common way of viewing and solving a problem (shared learning processes); Vanessa Evers offers one of the most complete definitions: Culture shapes the way people behave, view the world, express themselves and think. It is formed by historical experiences and values, traditions and surroundings.
Cultural Markers: The term ‘cultural markers’ used to denote interface design elements and features that are prevalent and possibly preferred within a particular cultural group. E.g. national symbol, colour, or spatial organisation. Badre also notes that cultural markers can also be genre specific (i.e. certain markers, such as the flag cultural marker, are frequently found on governmental websites). Badre gives a list of cultural markers found during his study of web sites and describes some patterns that reflect cultural practices and preferences. For example, Middle Eastern sites in Arabic and Hebrew have a high frequency of orienting text, links, and graphics from right to left (as opposed to centering or left to right). Also, Badre found that the Brazilian cultural preference for many colours is reflected in their particularly colourful websites. This preference was employed across genres, so the Brazilian governmental website employs many colours, with no one colour being overly dominant. This is unlike government sites from other countries which tend to use national colours on their sites. Indeed there is a delicate interplay between culture and genre as was also shown in Badre’s study where a cultural marker in Lebanon is light graphics and more text. Conversely, the travel genre favours heavy graphics with little text. Lebanese travel sites have few graphics and thus the cultural marker for more text and less graphics dominates over the genre markers. Finally, we should note that it is important to distinguish between internal and external cultural markers – internal markers refer to what the culture itself uses as opposed to external markers which is what people external to that culture believe is a marker.
Models of Culture: From an anthropological point of view, several models have been proposed which evaluate and classify groups of people according to various cultural variables (e.g. Trompenaars, 93; Hofstede, 80; Victor, 92). However, with relation to software and how it is adapted to take into account cultural differences, Hofstede’s model is frequently mentioned as a basis for providing descriptive profiles of culture . Hofstede focuses on cognitive aspects and states that culture is ‘the collective programming of the mind’ where a group of people will think the same way since they share the same learning processes. Hofstede’s (1991) Dimension of Culture: all cultures can be defined through 3 dimensions:
1. power distance (PD) – the degree of emotional dependence between a boss and sub-ordinate
2. collectivism/individualism (IC) – integration into cohesive groups versus being expected to look after oneself
3. femininity/masculinity (MF) – roughly interpreted as toughness or tenderness
and for Western cultures:
4. uncertainty avoidance
and for Eastern cultures:
4. long-term Confucian orientation
For the usability requirements for a new website, Andy Smith suggests two approaches: 1. undertaking local website audits to identify elements that are indigenous to sites in the target culture. 2. Predictively apply theoretical models, such as Hofstede’s, when proposing the design.
Despite their relatively wide acceptance as a way of specifying cultural groups, the use of cultural models for interface design may need to be treated with caution.
|
ETAT DE L’ART |
Culture and cultural values affect our behaviour in the face of interactive technologies. However, most of the work on cultural differences has been in the context of designing culturally sensitive web sites and there has been little treatment of this issue for other interactive applications, or multi-modal interfaces. The initial trend was towards standardisation (develop one version for all cultures) based on the rationale that familiarity with the interface would provide reassurance. However, this strategy could lead to users receiving inappropriate output, which could cause confusion or offence, or could force user to perform counter-intuitive tasks. Thus began a new trend towards globalisation and localisation. Localisation requires that any culture specific features of the software are identified and requires knowledge of the target culture.
Cited motivations for addressing culture: (a) Results in easier to learn the software, thus lower training costs (b) Reduces human error (c) Reduce employee turnover (d) Increase user satisfaction (e) Increase user acceptance of software (f) Allays fears over losing cultural identity / preserves elements of culture.
Culture affects:
User’s acceptance: Includes whether the users find the system easy to use and also the usefulness of the system (i.e. whether the system can be used to achieve some desired goal).
Usability: how well the users can use the functionality of the system. According to Nielsen, usability can be divided into 5 components: Learnability (the system should be easy to use); Efficiency (once the system has been learned, a high level of productivity is possible); Memorability (casual users can return to the system and use it without relearning it); errors (low rate of errors); Satisfaction (users enjoy using the system).
Understandibility
of the system
User’s
perception of the interface
User evaluation methods: Research by Evers suggest that some commonly used evaluation methods, such as observation, questionnaire and interview may not be appropriate for all international users.
Limits to the incorporation of cultural aspects: In the literature, there are some indications that original version applications are preferable to those which incorporate cultural aspects. For example, Evers cites cases of users in Sweden and Romania where the English language version of the software was actually preferred. Similarly, Keniston states that we should not simply assume that everyone in the world "needs" software localized to his or her vernacular language and that other choices may be more productive, more useful, more desired. Here he argues that it might be better for children to be taught programs that use English or some other ‘standard’ language, such as Chinese, Spanish, Arabic, Russian, or French. The rationale behind his argument is that in knowing these languages, with software as one part of the instrumentation of pedagogy, would truly equip students to enter the modern, multinational electronic world as full participants.
Niveau de l’adaptation aux aspects culturels dans un système interactif : Nous proposons d’étudier en détail cet aspect en dressant un état de l’art des approches et systèmes existants selon le niveau d’adaptation. Pour cela, nous nous basons sur le modèle d’architecture logicielle ARCH des systèmes interactifs. Celui-ci permet de définir plusieurs niveaux d’adaptation en considérant le composant logiciel qui doit être changé selon les aspects culturels pris en compte. Le modèle d’architecture logicielle ARCH préconise 5 composants logiciels en forme d’arche.
Nous constatons que la plupart des travaux (en relation avec le web ou non) concernent une adaptation lexicale de l’interaction comme le langage, les couleurs et les icônes. Voici des exemples d’adaptation de niveau lexical donc au sein du composant « bas Niveau d’Interaction » :
i. Language translation.
ii. Change in units of measure, currency, date and number formats, etc.
iii. Redesign of icons, symbols, and pictures. Here, the underlying belief is that these interface items are interpreted differently by users from different cultures. In spite of this widely held belief, the focus tends to be on language translation and changing units of measurement rather than redesigning icons, etc. Inappropriate icons may cause offence (e.g. the ‘thumbs-up’ sign can have radically different meanings depending on the culture) or could cause confusion (e.g. many icons, such as a US house mailbox or a dustbin, which represents a real-world object and is being used as a real-world metaphor through its icon in the virtual world, simply do not exist in many cultures). Thus, a culturally sound icon design should provide an accurate virtual representation of real world objects or actions that occur in the target culture.
iv. Colours. The choice of colour can affect the user’s expectations of the site. Albert Badre gives a colour-culture chart comparing the meanings of red, blue, green, yellow, and white in China, Japan, Egypt, France and the US.
v. Text direction, spatial orientation. This is related to the script direction of the user’s first language, for example there is a high frequency of right-to-left orientation in Egyptian, Israeli and Lebanese sites (Badre)
Nous proposons d’étudier l’existant ou d’envisager de nouveaux systèmes qui prennent en compte d’autres adaptations que celles de niveau lexical. Par exemple une adaptation de niveau syntaxique (Composant Techniques de Présentation) impliquerait une restructuration de l’interface en termes de widgets mais aussi le changement de modalités d’interaction. De même selon certains aspects culturels, le contrôleur de dialogue peut être affecté : les tâches possibles mais aussi leurs enchaînements peuvent être modifiés.
|
ETUDE DE CAS |
Une étude de cas doit être choisie en fonction de la disponibilité de données décrivant les spécificités culturelles des utilisateurs cibles.
|
AXES DE RECHERCHE |
-a- Ingénierie de l’interaction :
- Méthodes et Outils de capture de données relatives aux aspects culturels de l’utilisateur : par exemple un questionnaire n’est peut-être pas un outil possible pour toutes les cultures.
- Méthodes de conception ergonomique prenant en compte les spécificités culturelles des utilisateurs cibles.
- Méthodes et Outils en vue d’une adaptation automatique des interfaces aux spécificités culturelles de l’utilisateur.
-b- How learning processes change from culture to culture when faced with new software
-c- How to take into account cultural aspects in groupware involving people from several different cultures
-d- How it can be used to track changes in culture
|
QUESTIONS A TRAITER |
Si les aspects culturels se traduisent par des connaissances (savoir et savoir faire, pratique) : Quelle différence entre l’adaptation aux aspects culturels et l’adaptation aux modèles de l’utilisateur (domaine étudié en IHM) ?
|
LIENS |
EVENTS:
- 2nd BCS HCI Group Workshop on Culture and HCI - Bridging Cultural and Digital Divides. June 18th 2003 http://www.nimm.demon.co.uk/hci/
- EUPA 2002 Workshop - Culture and Website Usability Strategies For Internationalisation and Localisation. September 2002 South Bank University, London. Presentations available on: http://www.optimum-web.co.uk/eupa/index.shtml
- CULTURAL ISSUES IN HCI; One day workshop, University of Luton. December 2000 The web site of the workshop is no longer available, but a call for participation gives the titles of the presentations and the speakers: http://www.i3net.org/mail/i3news/msg00509.html
PAPERS:
- Designing With Care: Adapting Cultural Probes to Inform Design inSensitive Settings, Paper presented at OZCHI 2003, New Directions In Interaction: Information Environments, Media & Technology, 26-28 November at the University of Queensland, Brisbane, Australia http://www.gdewsbury.ukideas.com/OzCHIProbesfinal.pdf
- Culture and website usability - strategies for internationalisation localisation. Andy Smith, Tim French, Shailey Minocha, Lynne Dunckley http://www.optimum-web.co.uk/eupa/EUPA2002_paper.htm
- Gathering Requirements For Inclusive Design, Paper presented at 2nd BCS HCI Group Workshop on Culture and HCI: Bridging Cultural and Digital Divides, University of Greenwich, UK, June 18th 2003 http://www.gdewsbury.ukideas.com/HCICulturesubmitted2.pdf
- Software Localization: Notes on Technology and Culture. Kenneth Keniston.Massachusetts Institute of Technology. http://web.mit.edu/kken/Public/papers.htm Other publications of Kenneth Keniston http://web.mit.edu/kken/Public/papers.htm
- Publications of Vanessa Evers. The University of Amsterdam. Research interests include: Cultural Aspects of Human Computer Interaction, Culture and ICT, Cross-Cultural User Research http://www.swi.psy.uva.nl/usr/evers/publications.html
- Culturability: The Merging of Culture and Usability Reference Paper by Barber and Badre, 1998 Abstract - http://www.cc.gatech.edu/gvu/people/albert.badre/abstracts.html
- The Cultural Context of Web Genres: Content vs. Style Reference Paper by Badre and Laskowski, 2001 Abstract - http://www.cc.gatech.edu/gvu/people/albert.badre/abstracts.html - genres
- The cultural component of designing and evaluating international user interfaces. Nestor G. Trillo. Proceedings of the 32nd Hawaii International conference on system sciences. 1999 http://www.computer.org/proceedings/hicss/0001/00013/00013058.PDF
- List of Shailey Minocha's publications. Lecturer of Human-Computer Interaction, The Open University, UK. Research interests include: Internationalisation of products and Systems, Developing Design Guidelines /Patterns for E-Business Environments being developed for local / international audience. http://mcs.open.ac.uk/sm577/publications.htm
- List of Elke Duncker's publications . Research interests include: social cultural and gender dynamics of global information and communication systems, and social and cultural scripts of software and systems design. http://www.cs.mdx.ac.uk/staffpages/elke/publications.htm
PROJECTS:
- The Indo European Systems Usability Partnership (IESUP) is a European Commission funded project to facilitate collaboration between European and Indian academics and practitioners interested in the disciplines of usability and human-computer interaction (HCI). One of the main themes is Culture and HCI - debating the role that cultural differences have in systems usability and ways in which effective localisation of systems (specifically in the Indian context) can be achieved. IESUP has a mailing list, see www.jiscmail.ac.uk, to join the list. http://www.iesup.org/index.html
- The Cultural Context of User Interface Design. Georgia Institute of Technology. The premise of this research is that contextual cue interpretation achieved through language and cultural experience has an effect on how easily a person can learn, use, and interact with the interface to software systems. Our objective is threefold: 1) Identify the salient cultural contextual cues which are longitudinally embedded in people's perceptions and behaviors; 2) Specialize the identified cues to specific language-cultures; 3) Study the effects of the identified cues on users interactive performance. Albert Badre and Wendy Newstetter. http://www.cc.gatech.edu/gvu/people/faculty/badre/context.html
ORGANISATIONS:
- Customer Centred Systems group at the University of Luton, UK. A small team is investigating the usability of websites in a global context attempting to quantify the characteristics of websites in different countries and cultures that actually determine success. http://www.luton.ac.uk/depts/comp_info_sys/research_001.shtml
- U K C S I - UK Centre for Software Internationalisation. [contains project information and references] UKCSI has been established to provide a UK focus for research into user requirements in relation to software internationalisation and localisation. "UKCSI is particularly interested in 'soft' issues, those that underpin cultural and cognitive dimensions, and focus on the ways in which people in different cultures interact with computers and websites." http://www.uk-csi.org/
- MIND Labs. A consortium of labs (Michigan State University, USA, Ohio State University, USA, Eindhoven University of Technology, Netherlands, Helsinki School of Economics, Finland. Fernando Pessoa University, Portugal, Russian Academy of Science, Russia, University of Arizona, USA.) One of their research themes covers 'culture, media and mind' - the goal of the project is to explore the relationship between individual cognitive and socio-cultural differences and interaction with online interfaces; and to propose changes in interface design to help close the digital divide. http://www.mindlab.org/web2/research/culture.htm As part of the MIND activity, see HomeNetToo project (adapting interfaces to cultural schema and cognitive styles). Tests three different interfaces (traditional, spatial or interpersonal interface) to test participants cognitive styles and level of knowledge gained. http://www.msu.edu/user/jackso67/homenettoo/main.html http://www.msu.edu/user/jackso67/homenettoo/interface.html
|
REUNIONS |
-
3 novembre 2003 : Réunion de lancement de l’AS - Espace
Découverte France-Télécom Issy les Moulineaux
- 13 mai 2004 : Réunion de travail du pôle - Ordre du jour : (1) Etat de l’art : chaque membre présente un système interactif (2) Etude de cas : choix d’une étude et travail en groupe
Retour à [Haut de page]
[Index]
Laurence Nigay, Laboratoire CLIPS-IMAG, équipe IIHM.
Email : laurence.nigay@imag.fr