2003 Conference Proceedings

Go to previous article 
Go to next article 
Return to 2003 Table of Contents 


Dominique Archambault and Sébastien Sablé
Inserm U483/INOVA, University "Pierre et Marie Curie"
9, quai Saint Bernard - 75252 Paris Cedex 5 - France


The TiM project intends to develop and to adapt computer games for visually impaired children. This paper presents the selected approach for game adaptation, which is to rewrite game scenarii in order to make them accessible through specific devices like tactile boards, Braille displays or speech synthesisers, as well as standard devices like keyboards and sound devices. Then the software architecture for the game engine is described.


Visually impaired children, specific computer games, games design, tactile devices, Braille devices.

1. The TiM Project

The overall aim of the TiM Project [1,2] is to provide young visually impaired children, with or without additional disabilities, with multimedia computer games they can access independently, that is without the assistance of a sighted person.

The TiM project will develop tools that allow:

2. Constraints

2.1. Games should be simple to use

The adapted games have to be considered by children as real games. Adults consent to make concessions in order to be able to use the same tools as their sighted peers via specific Assistive Systems. Usually this requires a long and difficult learning, and a lot of fastidious operations. For instance when using a word-processor with a screen reader, to check if a word is in bold faces in a paragraph, it is necessary to select this word, to open the dialog box corresponding to the characters attributes, and then to seek in this box if the attribute "bold" is checked or not. This cannot be accepted by children, especially when they play.

2.2. Games should be adaptable

The games should be adaptable to the visual possibilities of each child. The main group of children currently concerned is composed of children who cannot use the ordinary graphical interface because they are blind or partially sighted (vision <0,05), with or without additional disabilities (slight to moderate). Their ages vary from 3 to 10 years old.

To reach the needs of these children, the games should be adaptable to all the specific devices they use, each are corresponding to a specific modality:

Games should also support standard devices as speakers, keyboards, mouses, joysticks...

2.3. Games should be device independent

Another important issue is to supply the support for all existing models of each category of device, that are used currently, as well as being able to add drivers easily when a new model comes on the market. Indeed specialised devices are often very expensive (Braille displays starting at $5000). So it is necessary that the games can run with the specific models that each child may have.

3. The TiM approach for adaptation

The mainstream computer games are usually designed especially to be used through a standard multimodal interface (graphical display, mouse and speakers). TiM's approach is to adapt the interface in the most appropriate way for each user according to his specific needs, using specific devices when necessary.

Then the use of specific devices implies that game content should include all the elements which are necessary to correspond to the specific presentation rules of each modality. For instance it is a necessity that all graphical information have alternative content.

Additionally, the specificity of modalities used by visually impaired children often makes it necessary to modify the scenario of interaction. For instance it is not enough to simply replace images by alternative text to be read in Braille.

A lot of games are based upon the global vision of the layout and the visual memory. In the adaptation, the screen cannot be considered as a complementary memory, with a lot of information quickly available.

The approach chosen by the TiM project is to design games using a modality-independent model [3]. In the case of adaptation of mainstream contents, it will be necessary to rewrite the scenarii.

4. The TiM game engine

The game engine has a multi-layered architecture, which main layers are: the Platform API and Low level modules.

4.1 The platform API

The API is organised in many different components. Those components are preexisting functionalities that are provided to the game designer to simplify game design. Given TiM objective of adaptive games, components are very high level objects completely independent of the game representation.

For instance, the API contains a component 'button'. This component includes facilities for being displayed on the screen, at a specific position, with text and image, but it also contains any information needed to be accessed in Braille or with a speech synthesis and with a tactile board. Then if the player uses Braille, the game engine will be able to display the text of this button on his Braille terminal.

The platform API is generated from C++ objects using the Simplified Wrapper And Interface Generator (SWIG [4]). This makes it possible to write the game script in many existing computer languages. At this time, the available languages are the C++, Java and Python languages.

4.2. Low level modules

Those are external libraries used by the platform internals to manage each kind of device. The architecture allows to easily update those modules or add a new one corresponding to a new kind of device.

5. Current status and further works

This game engine is used for designing several games from very different categories that will be evaluated in the next year.


The TiM project is funded by the European Commission, on the program IST 2000 (FP5 - IST - Systems and Services for the Citizen/Persons with special needs) under the reference IST-2000-25298. The contents of this paper is the sole responsibility of the authors and in no way represents the views of the European Commission or its services. The TiM project participants include: Inserm U483/INOVA from Université Pierre and Marie Curie, coordinator, (France), Les Doigts Qui Rêvent (France), Université du Havre (France), Association BrailleNet (France), Halmstad University (Sweden), Sunderland University (United Kingdom), Tomteboda Resource Centre (Sweden) and Virtua Ltd (United Kingdom).

We thank the company 'Mindscape' for allowing the adaptation of the game "Reader rabbit's : Toddler".


  1. Archambault, D. and al. (2000). TIM : Tactile Interactive Multimedia computer games for visually impaired children. European Commission, Information Society Technologies IST-2000-25298. http://www.snv.jussieu.fr/inova/.

  2. Archambault D., Burger D., and Sablé S., The TiM Project: Tactile Interactive Multimedia computer games for blind and visually impaired children, in Assistive Technology -- Added Value to the Quality of Life, In: Marincek C., Bühler C., Knops H., and Andrich R. (eds.), Proceedings of the AAATE'01 Conference, Ljubljana, Slovenia. September 2001, IOS Press Amsterdam, The Netherlands, pp. 359--363

  3. Archambault, D. and Burger, D. From Multimodality to Multimodalities: the need for independent models. In: Constantine Stephanidis (ed.) Proceedings of the UAHCI'01 conference - Universal Access in Human-Computer Interaction (joint with 9th International Conference on Human-Computer Interaction) - Towards an Information Society for All. New-Orleans, Louisiana, USA, August 2001. Lawrence Erlbaum Associates, pp. 227--231.

  4. http://www.swig.org

  5. Sablé S. and Archambault D. libbraille: a portable library to easily access Braille displays. In Miesenberger K., Klaus J. and Zagler W. (eds.) Proceedings of the 8th International Conference ICCHP 2002 - Computer Helping People with Special Needs. Linz Austria July 2002. Springer, LNCS 2398, pp. 345--352. http://libbraille.sourceforge.net

  6. See http://libspeech.sourceforge.net

  7. See http://www.openal.org

  8. See http://libsdl.org

Go to previous article 
Go to next article 
Return to 2003 Table of Contents 
Return to Table of Proceedings

Reprinted with author(s) permission. Author(s) retain copyright.