The aim of the GITK project is to provide a solution, which will dramatically improve the adaptivity of software. The vision is to get "user interfaces for all" and even more. To reach this goal an multilayered architecture has been designed, which clearly separates the interface functionality from look and feel concerns. An own markup language - GIML - is used to describe the functionality of dialogs. The final interface will be generated by dialog-rendering modules.
One of the benefits of this approach is that this toolkit equips an application with an user interface which can be exchanged at every time. Thus each user can choose the kind and style of interface he is comfortable with, depending on his abilities, his preferences and his current environment. Even the software author takes advantage by supporting such an approach, as his applications will automatically adapt to new interface technology developments.
Behind the scenes a lot XML technology (see www.w3c.org) namely libxml2 and libxslt (see xmlsoft.org) will be used (like XSLT for doing the transformations).
On the base of the ACM SIGCHI Curricula for Human-Computer Interaction I would classify my work as: C2,C5,D1,D2.

The current situation

Current user interfaces are often hard-coded for one purpose. So one gets either a textual interface or (as usual) a graphical interface. The problem is, that under many situations this is not what the user needs. The interface should adapt to the users abilities and current environmental conditions per request. Current interface engines can not provide that, as they just do not know about such requirements. Those interfaces simply lack of a generalized, well abstracted design, which provides all the fundamental information to dynamically create a interface which matches the current requirements.
Ofcourse the goals that the GITK project is targeting, are not totaly new. There are several project out there which already exists and still new projects are started. I have created a separate page discussing the relation of GITK to other projects.

The vision

My vision is that all people can work effectively with a software in any situation. This includes people with disabilities (such as low vision or tactile limitations) and situations like noisy environments or occupation with parallel tasks (querying the computer for a route while driving a car).
We are technically evolved far enough to try to achieve an improvement over the current situation, where the user still needs to adapt more then necessary.
In this project I will not just deliver a functioning prototype, but will focus especially on providing a wide range of rendering modules, such as a text-mode renderer with braille support for visually impaired people, a gtk-based renderer, a voice i/o renderer and probably even a telephony based voice/keypad i/o based module.

Target audience

This project mainly addresses software-developers. But there is also space for discussions about ergonomics, cognitions and the like. The software packages released by the projects should be seen as prototyes and are not yet useful end-user applications.

Project Information

The project is my doctor thesis, which I write at the TU Dresden and HTWK Leipzig. The work is titled "Dynamically generated multi-modal application interfaces"
GITK is free software and distributed under the GPL. Other licences are available on demand.
The project is funded by the HWP of the BMBF.

Send mail to: Stefan Kost
Last modified: Mon Oct 04 15:32:48 MEST 2004
Valid HTML 4.01! Valid CSS! SourceForge Logo